Version number and release date for 2.4.4

- Lexer needed additional lookbehind criteria for ::isTag (thanks to Brad
  Collette).

.gitignore

@@ -2,6 +2,7 @@ cmake.h
 commit.h
 Makefile
 src/task
+src/taskd
 src/libtask.a
 src/commands/libcommands.a
 src/columns/libcolumns.a

AUTHORS

@@ -1,4 +1,4 @@
-The development of taskwarrior was made possible by the significant
+The development of Taskwarrior was made possible by the significant
 contributions of the following people:
 
   Paul Beckingham    (Principal Author)
@@ -13,6 +13,7 @@ contributions of the following people:
   Louis-Claude Canon (Contributing Author)
   Scott Kostyshak    (Contributing Author)
   Renato Alves       (Contributing Author)
+  Tomas Babej        (Contributing Author)
 
 The following submitted code, packages or analysis, and deserve special thanks:
 
@@ -112,6 +113,10 @@ The following submitted code, packages or analysis, and deserve special thanks:
   Jeremy John Reeder
   Roman Inflianskas
   Łukasz Panek
+  V.Krishn
+  Jens Erat
+  Peter Rochen
+  Oota Toshiya
 
 Thanks to the following, who submitted detailed bug reports and excellent
 suggestions:
@@ -233,10 +238,16 @@ suggestions:
   dev-zero
   Petteri
   Black Ops Testing
-  Jens Erat
   Leon Feng
-  Tomas Babej
   Nick Person
   Arnoud K
   Ozgur Akgun
   David Costa
+  Sujeevan Vijayakumaran
+  Scott Carter
+  Taisuke Hachimura
+  Martin
+  Alexandre de Verteuil
+  Scott M
+  Stefan Frühwirth
+  Pierre Campet

CMakeLists.txt

@@ -1,5 +1,8 @@
 cmake_minimum_required (VERSION 2.8)
 set(CMAKE_LEGACY_CYGWIN_WIN32 0) # Remove when CMake >= 2.8.4 is required
+if(POLICY CMP0037)
+  cmake_policy(SET CMP0037 OLD)
+endif()
 
 include (CheckFunctionExists)
 include (CheckStructHasMember)
@@ -7,12 +10,40 @@ include (CheckStructHasMember)
 set (HAVE_CMAKE true)
 
 project (task)
-set (PROJECT_VERSION "2.4.0")
+set (PROJECT_VERSION "2.4.4")
+
+OPTION(USE_GNUTLS "Build gnutls support." ON)
+
+message ("CMAKE_SYSTEM_NAME ${CMAKE_SYSTEM_NAME}")
+
+include (CheckCXXCompilerFlag)
+
+# NOTE: If we are to actually use C++11 features, we should either require
+#       a compiler that supports the -std=c++11 flag or check for the
+#       features used.
+#       Relying on -std=c++0x or even -std=gnu++0x is highly volatile.
+
+CHECK_CXX_COMPILER_FLAG("-std=c++11" _HAS_CXX11)
+CHECK_CXX_COMPILER_FLAG("-std=c++0x" _HAS_CXX0X)
+CHECK_CXX_COMPILER_FLAG("-std=gnu++0x" _HAS_GNU0X)
+
+if (_HAS_CXX11)
+  set (_CXX11_FLAGS "-std=c++11")
+elseif (_HAS_CXX0X)
+  message (WARNING "Enabling -std=c++0x draft compile flag. Your compiler does not support the standard '-std=c++11' option.  Consider upgrading.")
+  set (_CXX11_FLAGS "-std=c++0x")
+elseif (_HAS_GNU0X)
+  message (WARNING "Enabling -std=gnu++0x draft compile flag. Your compiler does not support the standard '-std=c++11' option. Consider upgrading.")
+  set (_CXX11_FLAGS "-std=gnu++0x")
+else (_HAS_CXX11)
+ message (FATAL_ERROR "C++11 support missing. Try upgrading your C++ compiler. If you have a good reason for using an outdated compiler, please let us know at support@taskwarrior.org.")
+endif (_HAS_CXX11)
 
 if (${CMAKE_SYSTEM_NAME} MATCHES "Linux")
   set (LINUX true)
 elseif (${CMAKE_SYSTEM_NAME} MATCHES "Darwin")
   set (DARWIN true)
+  set (_CXX11_FLAGS "${_CXX11_FLAGS} -stdlib=libc++")
 elseif (${CMAKE_SYSTEM_NAME} MATCHES "kFreeBSD")
   set (KFREEBSD true)
 elseif (${CMAKE_SYSTEM_NAME} MATCHES "FreeBSD")
@@ -25,10 +56,20 @@ elseif (${CMAKE_SYSTEM_NAME} MATCHES "SunOS")
   set (SOLARIS true)
 elseif (${CMAKE_SYSTEM_NAME} STREQUAL "GNU")
   set (GNUHURD true)
+elseif (${CMAKE_SYSTEM_NAME} STREQUAL "CYGWIN")
+  set (CYGWIN true)
+  # NOTE: Not setting -std=gnu++0x leads to compile errors even with
+  #       GCC 4.8.3, and debugging those leads to insanity. Adding this
+  #       workaround instead of fixing Cygwin.
+  set (_CXX11_FLAGS "-std=gnu++0x")
 else (${CMAKE_SYSTEM_NAME} MATCHES "Linux")
   set (UNKNOWN true)
 endif (${CMAKE_SYSTEM_NAME} MATCHES "Linux")
 
+set (CMAKE_CXX_FLAGS "${_CXX11_FLAGS} ${CMAKE_CXX_FLAGS}")
+
+set (CMAKE_CXX_FLAGS "-Wall -Wsign-compare -Wreturn-type ${CMAKE_CXX_FLAGS}")
+
 if (NETBSD)
    # Since readline, etc likely to be in /usr/pkg/lib, not standard library
    # Otherwise will remove links during install
@@ -65,23 +106,15 @@ set (PACKAGE_TARNAME "${PACKAGE}")
 set (PACKAGE_VERSION "${VERSION}")
 set (PACKAGE_STRING "${PACKAGE} ${VERSION}")
 
-message ("-- Looking for GnuTLS")
-find_package (GnuTLS)
-if (GNUTLS_FOUND)
-  set (HAVE_LIBGNUTLS true)
-  set (TASK_INCLUDE_DIRS ${TASK_INCLUDE_DIRS} ${GNUTLS_INCLUDE_DIR})
-  set (TASK_LIBRARIES    ${TASK_LIBRARIES}    ${GNUTLS_LIBRARIES})
-endif (GNUTLS_FOUND)
-
-#message ("-- Looking for pthread")
-#find_path (PTHREAD_INCLUDE_DIR pthread.h)
-#find_library (PTHREAD_LIBRARY NAMES pthread)
-#if (PTHREAD_INCLUDE_DIR AND PTHREAD_LIBRARY)
-#  message ("-- Found pthread: ${PTHREAD_LIBRARY}")
-#  set (HAVE_LIBPTHREAD true)
-#  set (TASK_INCLUDE_DIRS ${TASK_INCLUDE_DIRS} ${PTHREAD_INCLUDE_DIR})
-#  set (TASK_LIBRARIES    ${TASK_LIBRARIES}    ${PTHREAD_LIBRARIES})
-#endif (PTHREAD_INCLUDE_DIR AND PTHREAD_LIBRARY)
+if (USE_GNUTLS)
+   message ("-- Looking for GnuTLS")
+   find_package (GnuTLS)
+   if (GNUTLS_FOUND)
+     set (HAVE_LIBGNUTLS true)
+     set (TASK_INCLUDE_DIRS ${TASK_INCLUDE_DIRS} ${GNUTLS_INCLUDE_DIR})
+     set (TASK_LIBRARIES    ${TASK_LIBRARIES}    ${GNUTLS_LIBRARIES})
+   endif (GNUTLS_FOUND)
+endif (USE_GNUTLS)
 
 check_function_exists (timegm  HAVE_TIMEGM)
 check_function_exists (get_current_dir_name HAVE_GET_CURRENT_DIR_NAME)
@@ -130,6 +163,7 @@ set (LANGUAGE_ITA_ITA 5)
 set (LANGUAGE_POR_PRT 6)
 set (LANGUAGE_EPO_RUS 7)
 set (LANGUAGE_POL_POL 8)
+set (LANGUAGE_JPN_JPN 9)
 
 message ("-- Configuring cmake.h")
 configure_file (

COPYING

@@ -1,4 +1,4 @@
-taskwarrior - a command line task list manager.
+Taskwarrior - a command line task list manager.
 
 Copyright 2006 - 2015, Paul Beckingham, Federico Hernandez.
 

ChangeLog

@@ -1,10 +1,148 @@
-2.4.0 (2015-01-01) -
+2.4.4 (2015-05-10) -
+
+- TW-69   wait dates relative to due date (thanks to John Florian).
+- TW-1285 I'd like to use relative dates combined with times (thanks to Adam
+          Gibbins).
+- TW-1474 Documentation is confusing with respect to user/uuid on the server
+          (thanks to Tomas Babej).
+- TW-1596 taskwarrior can't compile FreeBSD 9.3 32bit environment (thanks to
+          ribbon)
+- TW-1603 Priority color precedence changed since it is a UDA, should be lowered
+          again (thanks to Jens Erat).
+- TW-1605 Japanese translation for Taskwarrior (thanks to Oota Toshiya).
+- TW-1606 scheduled.any filter (thanks to Peter Rochen).
+- TW-1608 The recur/recurring report shows tasks without a recur interval
+          (thanks to Brad Collette).
+- TW-1610 Disabling GC can lead to editing the wrong task (thanks to Scott M).
+- The 'obfuscate' setting, if set to '1' will replace all text with 'xxx'.
+- POSIX file locking mechanism, eliminating platform-specific code.
+
+------ current release ---------------------------
+
+2.4.3 (2015-04-19) 499044b9b6bdbc95338ea585204e949d80b24a09
+
+- TW-57   user defined attribute sort order (thanks to Max Muller).
+- TW-70   urgency.user.keyword.<keyword>.coefficient=...
+- TW-111  User-defined priorities.
+- TW-1279 Make default.* not apply to recurring tasks.
+- TW-1287 Make default.* not apply to synced tasks.
+- TW-1539 user-defined urgency coefficients for priority values.
+- TW-1541 Priority should be converted to UDA (in default taskrc) (thanks to
+          Tomas Babej).
+- TW-1556 task hangs when modifying uda field with percent-encoded (url-encoded)
+          value (thanks to Stefan Frühwirth).
+- TW-1578 Bash tab completion problems on first run
+          (thanks to Renato Alves and Ptolemarch).
+- TW-1580 "modified" attribute no longer updated (thanks to David Patrick).
+- TW-1581 Tasks with dependencies show wrong urgency values for the first
+  report run after a task in the dependency chain is completed/deleted (thanks
+  to Ulf Eliasson).
+- TW-1583 Invalid ID displayed for first report after done/delete (thanks to
+  Ulf Eliasson).
+- TW-1584 attr.{isnt,not} use partial matching.
+- TW-1587 Fix and improve example on-exit hook, adjust to new hooks API
+  (thanks to Jochen Sprickerhof).
+- TW-1588 Most Export scripts cannot deal with new export format (thanks to
+  Scott Carter).
+- TW-1590 syntax of rcfile not documented (whitespace, line continuation)
+          (thanks to Scott M).
+- TW-1591 add an option to see non-pending project with command task summary
+          (thanks to Pierre Campet).
+- TW-1595 info command doesn't print urgency details, if urgency is negative
+          (thanks to Peter Rochen).
+- Setting 'bulk' to zero is interpreted as infinity, which means there is no
+  amount of changes that is considered dangerous (thanks to Tomas Babej).
+- Disable hooks in bash completion script.  Hooks were previously able to
+  abort processing or output interfering data, breaking completion.
+- Fix "task add due:tomorrow+3days" failing to work without spaces.
+- Performance improvements:
+  + Stops after measuring a fixed-width column format.
+  + Reduced number of std::string copies.
+
+2.4.2 (2015-03-15) b9dc0813d9a8922b4cef9595033f133f9fbabf44
+
+- TW-41   Tasks in subprojects are not counted in project completion (thanks
+          to Renato Alves).
+- TW-1450 Projects command should trigger running garbage collector (thanks to
+          Tomas Babej).
+- TW-1535 move default listing-break from list to ls (thanks to David Patrick).
+- TW-1545 cc1plus: error: unrecognized command line option '-std=c++11' (thanks
+          to Petteri).
+- TW-1546 column type due.remaining breaks colors on due tasks (thanks to
+          Renato Alves).
+- TW-1547 Recur column is always shown even if no recurring task is displayed
+          (thanks to Renato Alves).
+- TW-1549 task annotate hangs with specific text pattern (thanks to Alexandre
+          de Verteuil).
+- TW-1550 _contexts helper-command (thanks to David Patrick).
+- TW-1551 Unable to get a UDA value from DOM (thanks to Tomas Babej).
+- Eliminated some code that is not UTF8-safe.
+- Removed pthreads linkage.
+- Implemented the context feature.
+- Closed dangling pipes in execute (), resolving problems when a hook script
+  forks (thanks to Jens Erat).
+- Re-enabled hook script feedback when exiting with 0 exit status.
+- The 'info' command now shows virtual tags.
+- Fixed major on-modify hooks regression where hooks could no longer modify
+  the tasks handed to them.
+- 'task _version' now outputs "2.4.2 (git-ref)" when built from git. "2.4.2"
+  when built from release tarballs (thanks to Renato Alves).
+
+2.4.1 (2015-02-16) 82e019a4a8b20de63d53b51d59b8d1c89d3c05b2
+
+- TW-1457 Non-existent attributes are not properly handled (thanks to Tomas
+          Babej).
+- TW-1484 The 'history' and 'ghistory' reports do not obey rc.color.label.
+- TW-1486 task wait shows completed tasks which has a wait attribute (thanks to
+          Sujeevan Vijayakumaran).
+- TW-1487 Task export exports some numeric attributes as strings (thanks to
+          Tomas Babej).
+- TW-1491 Regression in deleting due dates (thanks to Jens Erat).
+- TW-1492 compiling v2.4.0 using musl(libc) (thanks to V.Krishn).
+- TW-1495 German translation for taskwarrior (thanks to Jens Erat).
+- TW-1498 Filtering for presence of UDA matches all tasks (thanks to Ralph
+          Bean).
+- TW-1501 Calc can't handle multi-digit numbers in some expressions (thanks to
+          Jeremy John Reeder).
+- TW-1502 Successful on-add hook with no output confuses TW (thanks to Tomas
+          Babej).
+- TW-1504 On-modify hook does not accept correct JSON format (thanks to Tomas
+          Babej).
+- TW-1505 completely ignore mis-named hook scripts (thanks to Tomas Babej,
+          David Patrick).
+- TW-1509 Hooks modifications performed on sync not sycning back (thanks to
+          Tomas Babej).
+- TW-1510 Task can save empty value in the data backlog (thanks to Tomas Babej).
+- TW-1517 Hook performance should be measured individually for each hook (thanks
+          to Tomas Babej).
+- TW-1518 Misbehaving hooks silently break task processing (thanks to Wim
+          Schuermann).
+- TW-1519 Testing suite forces taskd.trust="ignore hostname" (thanks to Renato
+          Alves).
+- TW-1522 Date format doesn't like hyphens (thanks to Scott Carter).
+- TW-1524 Build Broken (thanks to Jack).
+- TW-1530 Multiple on-add hooks generating new tasks are ignored (thanks to
+          Tomas Babej).
+- TW-1531 'task export' should handle recurrence (thanks to Tomas Babej).
+- TW-1532 Hooks does not execute any script on Cygwin (thanks to Taisuke
+          Hachimura).
+- TW-1534 Urgency coefficient for user project disables 'info' output (thanks to
+          Martin).
+- TW-1542 Large numeric UDA values get rendered in scientific notation on export
+          (thanks to Ralph Bean).
+- Fixed assorted color theme problems.
+- Changed assorted reports so they do not use '.age' format for dates that are
+  in the future, because those are never shown with this format (thanks to
+  Sujeevan Vijayakumaran).
+- New 'recurrence' configuration setting can disable recurring task generation.
+
+2.4.0 (2015-01-01) 670102842c39bdc62ef84ae4b679a8f5a2d89523
 
 - TD-42 Cannot compile taskd - GNUTLS_VERSION undefined in diag.cpp (thanks
         to Michele Vetturi).
 - TD-45 Fix preprocessor define (thanks to Jochen Sprickerhof).
-- TD-55 TLSServer/Client need to include <errno.h> on Solaris (thanks to Tatjana
-        Heuser). Also applied to NetBSD.
+- TD-55 TLSServer/Client need to include <errno.h> on Solaris (thanks to
+        Tatjana Heuser). Also applied to NetBSD.
 - TD-56 File.cpp needs to include <string.h> on Solaris (thanks to Tatjana
         Heuѕer).
 - TD-57 taskdctl script assumes /bin/sh is /bin/bash (thanks to Tatjana Heuser).
@@ -45,6 +183,7 @@
 - TW-52 "task add ... recur:2 months" interpreted as "2s" (thanks to jwhisnant).
 - TW-55 Bulk edit recurring tasks without answering yes/no for each? (thanks to
         Max Muller).
+- TW-63 indicators for UDAs (thanks to David Patrick).
 - TW-71 task ls/list/long/etc. should match contents of projects too (thanks to
         Cory Donnelly).
 - TW-72 extend info report with urgency column.
@@ -286,8 +425,6 @@
 - Supports 'debug.parser' configuration setting.
 - Supports 'color.label.sort' for column labels of sort columns.
 
------- current release ---------------------------
-
 2.3.0 (2014-01-15) c4eb46507031b7dee839dcb932bb2a22b2f0d3a2
 
 Features

DEVELOPER

@@ -1,9 +1,15 @@
 How to Build Taskwarrior
 
+  Satisfy the Requirements:
+    - gcc 4.7 or later, clang 3.3 or later or equivalent. This is because 2.4.4
+      will be utilizing C++11.
+    - libuuid
+    - gnutls (optional)
+
   Obtain and build code:
     $ git clone https://git.tasktools.org/scm/tm/task.git task.git
     $ cd task.git
-    $ git checkout 2.4.0                 # Latest dev branch
+    $ git checkout 2.4.4                 # Latest dev branch
     $ cmake -DCMAKE_BUILD_TYPE=debug .   # debug or release. Default: neither.
     $ make VERBOSE=1                     # Shows details
 
@@ -18,8 +24,9 @@ How to Build Taskwarrior
   current development branch. The source tarballs do not reflect HEAD, and do
   not contain the test suite.
 
-  If you send us a patch, make sure that patch is made against git HEAD. We
-  cannot apply patches made against the tarball source.
+  If you send a patch (support@taskwarrior.org), make sure that patch is made
+  against git HEAD on the development branch. We cannot apply patches made
+  against the tarball source, or master.
 
 General Statement
   This file is intended to convey the current efforts, priorities and needs of
@@ -181,12 +188,11 @@ Work in Progress
 Current Codebase Condition
 
   'master' branch:
-    - 2.3.0 Current release, locked.
+    - 2.4.3 Current release, locked.
 
-  '2.4.0' branch:
-    - Current development branch with new command line parser, expressions,
-      and heading for beta soon.
+  '2.4.4' branch:
+    - Current development branch no plans yet.
 
 ---
 
-2014-10-12 Updated for 2.4.0
+2015-04-21 Updated for 2.4.4

EXPOSITION

@@ -24,7 +24,7 @@ Command Line Parsing
   determines whether subsequent arguments are interpreted as part of a filter or
   set of modifications.
 
-  The Parser object is fed command line arguments, then through a succession of
+  The CLI object is fed command line arguments, then through a succession of
   calls builds and annotates a parse tree.  To help with this, the Lexer is
   used to break up strings into tokens.
 

INSTALL

@@ -1,16 +1,26 @@
 Installation Instructions
 -------------------------
 
-Please follow the instructions below to build task with cmake.
+Please follow the instructions below to build and install Taskwarrior from
+source.
 
 
 Pre-requisites
 --------------
 
-You will need the CMake build system installed in order to build taskwarrior
-from source.
+You will need the CMake build system installed in order to build Taskwarrior
+from source. More information on cmake can be obtained at http://cmake.org
 
-More information on cmake can be obtained at http://cmake.org
+You will also need:
+  - make
+
+You will need a C++ compiler that supports full C++11, which includes:
+  - gcc 4.7   (released 2012-03-23)
+  - clang 3.3 (released 2013-01-07)
+
+You will need the following libraries:
+  - libuuid
+  - gnutls    (optional - for syncing)
 
 It is HIGHLY RECOMMENDED that you build with a library that provides uuid_*
 functions, such as libuuid.
@@ -19,14 +29,14 @@ functions, such as libuuid.
 Basic Installation
 ------------------
 
-Briefly, these shell commands will unpack, build and install taskwarrior:
+Briefly, these shell commands will unpack, build and install Taskwarrior:
 
-  $ tar xzf task-X.Y.Z.tar.gz                     [1]
-  $ cd task-X.Y.Z                                 [2]
-  $ cmake .                                       [3]
-  $ make                                          [4]
-  $ sudo make install                             [5]
-  $ cd .. ; rm -r task-X.Y.Z                      [6]
+  $ tar xzf task-X.Y.Z.tar.gz                   [1]
+  $ cd task-X.Y.Z                               [2]
+  $ cmake .                                     [3]
+  $ make                                        [4]
+  $ sudo make install                           [5]
+  $ cd .. ; rm -r task-X.Y.Z                    [6]
 
 These commands are explained below:
 
@@ -38,7 +48,7 @@ These commands are explained below:
   3. Invokes CMake to scan for dependencies and machine-specific details, then
      generate the makefiles.  This may take a minute.
 
-  4. Builds taskwarrior.  This may take a minute.
+  4. Builds Taskwarrior.  This may take a minute.
 
   5. Installs the program, documentation and other data files.
 
@@ -92,29 +102,31 @@ Currently the defined languages are:
 
   eng_USA 1
   esp_ESP 2
+  deu_DEU 3
   fra_FRA 4
   ita_ITA 5
   por_PRT 6
+  epo_RUS 7
+  pol_POL 8
 
 
 Uninstallation
 --------------
 
-To uninstall taskwarrior, you need the Makefiles, so if you deleted them in
-step 7 above, they must first be regenerated by following steps [1], [2] and
-[3].  Then simply run:
+There is no uninstall option in CMake makefiles. This is a manual process.
 
-  $ sudo make uninstall TODO TODO TODO -- cmake doesn't include uninstall
+To uninstall Taskwarrior, remove the files listed in the install_manifest.txt
+file that was generated when you built Taskwarrior.
 
 
 Taskwarrior Build Notes
 -----------------------
 
-Taskwarrior has dependencies that are detected by cmake in almost all cases, but
+Taskwarrior has dependencies that are detected by CMake in almost all cases, but
 there are situations and operating systems that mean you will need to offer a
 little help.
 
-If taskwarrior will not build on your system, first take a look at the Operating
+If Taskwarrior will not build on your system, first take a look at the Operating
 System notes below.  If this doesn't help, then go to the Troubleshooting
 section, which includes instructions on how to contact us for help.
 
@@ -131,9 +143,7 @@ Cygwin
 
 
 Darwin, 32bit
-  The taskwarrior packages will not work on a 32-bit OSX installation on Core
-  Duo hardware.  You will need to build Taskwarrior from source, and use this
-  configure command:
+  Taskwarrior needs to be built from source, and use this configure command:
 
     cmake -DCMAKE_C_FLAGS=-m32 -DCMAKE_CXX_FLAGS=-m32 -DCMAKE_EXE_LINKER_FLAGS=-m32 .
 
@@ -144,6 +154,18 @@ Darwin, 32bit
     http://stackoverflow.com/questions/6077414/cmake-how-to-set-the-ldflags-in-cmakelists-txt
 
 
+OpenBSD
+  In order to build Taskwarrior 2.4.2+, you might need to install a newer GCC
+  version from ports or packages.
+
+  Afterwards, run
+    cmake -DCMAKE_CXX_COMPILER=eg++ .
+
+  and build normally.
+
+  See: https://bug.tasktools.org/browse/TW-1579
+
+
 Troubleshooting
 ---------------
 
@@ -160,7 +182,7 @@ If a build does not succeed, please send the contents of the 'CMakeCache.txt'
 and 'CMakeFiles/CMakeOutput.log' files to support@taskwarrior.org, or post a
 message in the support forums at taskwarrior.org along with the information.
 
-If CMake runs but taskwarrior does not build, please send the contents of the
+If CMake runs but Taskwarrior does not build, please send the contents of the
 above files as well as a transcript from the build, which is not written to a
 file and must be captured from the terminal.
 

LICENSE

@@ -1,4 +1,4 @@
-taskwarrior - a command line task list manager.
+Taskwarrior - a command line task list manager.
 
 Copyright 2006 - 2015, Paul Beckingham, Federico Hernandez.
 

NEWS

@@ -1,98 +1,40 @@
 
-New Features in taskwarrior 2.4.0
+New Features in Taskwarrior 2.4.4
 
-  - The 'show' command displays default configuration values, when appropriate.
-  - Removed deprecated commands 'push', 'pull' and 'merge'.
-  - Portuguese (por-PRT), French (fra-FRA), Esperanto (epo-RUS) and Polish
-    (pol-POL) localizations.
-  - Better handling for deletion of recurring tasks.
-  - New virtual tags: YESTERDAY, TOMORROW, READY, PENDING, COMPLETED, DELETED,
-    TAGGED.
-  - The '_get' command properly uses exit codes.
-  - Regular expressions are now enabled by default.
-  - The 'filter' verbosity token shows the complete filter used for the last
-    command.
-  - The 'new-uuid' verbosity token shows the UUID of newly created tasks.
-  - The 'info' report now breaks down urgency values.
-  - New 'color.until' color rule.
-  - Using a non-zero 'urgency.inherit.coefficient' value means a task inherits
-    the urgency values of the dependency chain.
-  - Listing breaks now supported.  See 'man taskrc'.
-  - New fish shell completion script.
-  - The filter form 'name:value' now maps to the partial match operator '=',
-    rather than the exact match operator, '=='.  This means that dates now
-    match on the day by default, not the time also.
-  - UDA indicator format.
-  - Hooks.
-  - New and updated holidays.rc files.
+  - Taskwarrior now requires GCC 4.7+ / Clang 3.3+, for full C++11 support.
+  - Taskwarrior now builds properly on 32-bit platforms.
+  - The beginnings of a Japaneѕe translation.
 
-New commands in taskwarrior 2.4.0
+New commands in Taskwarrior 2.4.4
 
-  - New 'calc' command (and standalone utility) for quick command line
-    calculations.
+  - None
 
-New configuration options in taskwarrior 2.4.0
+New configuration options in Taskwarrior 2.4.4
 
-  - The 'taskd.trust' setting is now a tri-state, supporting values 'strict',
-    'ignore hostname' and 'allow all', for server certificate validation.
-  - New theme: dark-gray-blue-256.theme
-  - The 'allow.empty.filter' setting can be set to 'no' to disallow the
-    potentially dangerous combination of write commands and empty filters.
-  - New truncated_count column style for the description field which as the
-    name says is a combination of the existing truncated and count styles.
-  - New 'hooks' setting is a master control switch for hook processing.
-  - New 'debug.hooks' for debugging hook scripts.
-  - New 'debug.parser' for debugging parser issues scripts.
-  - New 'color.label.sort' is used to color the column label of sort columns.
-  - New 'urgency.uda.<name>.<value>' allows specific UDA values to affect
-    urgency.
-  - New 'recurrence.confirmation' which allows bypassing confirmation for
-    changes to recurring tasks, by accepting, rejecting or prompting.
-  - New 'uda.<name>.indicator' to override the UDA indicator.
+  - The 'obfuscate' setting, if set to '1' will replace all text with 'xxx'.
+    This is useful when sharing data for bug reporting purposes.
 
-Newly deprecated features in taskwarrior 2.4.0
+Newly deprecated features in Taskwarrior 2.4.4
 
-  - The alias '_query' is deprecated.
-  - Bare word search terms are deprecated, so 'task pattern list' will not
-    be supported in future releases, instead requiring 'task /pattern/ list'.
-  - The 'limit' pseudo attribute will change to 'rc.limit' in a future release.
+  - None
 
-Removed features in 2.4.0
+Removed features in 2.4.4
 
-  - Version 1.x column names no longer supported.
-  - Version 1.x sort columns no longer supported.
-  - Old-style color names including underscores are no longer supported.
-  - Removed priority counts from the 'projects' commands.
-  - Removed the unused 'locale' configuration variable.
-  - Removed the unused 'patterns' confguration variable.
-  - Removed the obsolete 'task-faq.5' man page.
-  - Removed the obsolete 'task-tutorial.5' man page.
-  - Removed the obsolete 'tasksh.1' man page.
-  - Removed the 'tasksh' program from the Taskwarrior distribution. It is now
-    a separate project.
-  - Removed the 'complete.all.projects' setting that was not used.
-  - Removed support for 'report.X.limit', which can now be set in the report
-    filter 'limit:N'.
-  - The 'total active time' information is removed from the 'info' report.
-    This was being misinterpreted as support for time tracking.
-  - The 'shadow file' feature is removed, but replaced with an example hook
-    script that performs the same function..
+  - None
 
 Known Issues
 
-  - On Cygwin, case-insensitive regex searches are broken. The default is
-    now case-sensitive for Cygwin only.
-  - https://bug.tasktools.org
+  - https://bug.tasktools.org/
 
 Taskwarrior has been built and tested on the following configurations:
 
-  * OS X 10.10 Yosemite, 10.9 Mavericks
-  * Fedora 20 Heisenbug
-  * Ubuntu 14.04 Trusty Tahr
-  * Debian 7.0 Wheezy (stable)
-  * Arch 2014.05
-  * FreeBSD 10
-  * Cygwin 1.7.29, 1.7.32
+  * OS X
+  * Fedora
+  * Ubuntu
+  * Debian
+  * Arch
+  * FreeBSD
+  * Cygwin
 
 ---
 

README.md

@@ -39,7 +39,7 @@ Please send your code patches to:
     support@taskwarrior.org
 
 Consider joining bug.tasktools.org, answers.tasktools.org and participating in
-the future of taskwarrior.
+the future of Taskwarrior.
 
 ---
 

cmake.h.in

@@ -27,6 +27,7 @@
 #define LANGUAGE_POR_PRT ${LANGUAGE_POR_PRT}
 #define LANGUAGE_EPO_RUS ${LANGUAGE_EPO_RUS}
 #define LANGUAGE_POL_POL ${LANGUAGE_POL_POL}
+#define LANGUAGE_JPN_JPN ${LANGUAGE_JPN_JPN}
 
 /* git information */
 #cmakedefine HAVE_COMMIT
@@ -51,12 +52,6 @@
 /* Found the GnuTLS library */
 #cmakedefine HAVE_LIBGNUTLS
 
-/* Found the Readline library */
-#cmakedefine HAVE_READLINE
-
-/* Found the pthread library */
-#cmakedefine HAVE_LIBPTHREAD
-
 /* Found tm_gmtoff */
 #cmakedefine HAVE_TM_GMTOFF
 

cmake/Modules/FindReadline.cmake

@@ -1,81 +0,0 @@
-# - Find the readline library
-# This module defines
-#  READLINE_INCLUDE_DIR, path to readline/readline.h, etc.
-#  READLINE_LIBRARIES, the libraries required to use READLINE.
-#  READLINE_FOUND, If false, do not try to use READLINE.
-# also defined, but not for general use are
-# READLINE_readline_LIBRARY, where to find the READLINE library.
-# READLINE_ncurses_LIBRARY, where to find the ncurses library [might not be defined]
-
-# Apple readline does not support readline hooks
-# So we look for another one by default
-IF (APPLE OR FREEBSD)
-  FIND_PATH (READLINE_INCLUDE_DIR NAMES readline/readline.h PATHS
-    /usr/include/
-    /sw/include
-    /opt/local/include
-    /opt/include
-    /usr/local/include
-    NO_DEFAULT_PATH
-    )
-ENDIF (APPLE OR FREEBSD)
-FIND_PATH (READLINE_INCLUDE_DIR NAMES readline/readline.h)
-
-
-# Apple readline does not support readline hooks
-# So we look for another one by default
-IF (APPLE OR FREEBSD)
-  FIND_LIBRARY (READLINE_readline_LIBRARY NAMES readline PATHS
-    /usr/lib
-    /sw/lib
-    /opt/local/lib
-    /opt/lib
-    /usr/local/lib
-    NO_DEFAULT_PATH
-    )
-ENDIF (APPLE OR FREEBSD)
-FIND_LIBRARY (READLINE_readline_LIBRARY NAMES readline)
-
-# Sometimes readline really needs ncurses
-IF (APPLE OR FREEBSD)
-  FIND_LIBRARY (READLINE_ncurses_LIBRARY NAMES ncurses PATHS
-    /usr/lib
-    /sw/lib
-    /opt/local/lib
-    /opt/lib
-    /usr/local/lib
-    /usr/lib
-    NO_DEFAULT_PATH
-    )
-ENDIF (APPLE OR FREEBSD)
-FIND_LIBRARY (READLINE_ncurses_LIBRARY NAMES ncurses)
-
-MARK_AS_ADVANCED (
-  READLINE_INCLUDE_DIR
-  READLINE_readline_LIBRARY
-  READLINE_ncurses_LIBRARY
-  )
-
-SET (READLINE_FOUND "NO" )
-IF (READLINE_INCLUDE_DIR)
-  IF (READLINE_readline_LIBRARY)
-    SET (READLINE_FOUND "YES" )
-    SET (READLINE_LIBRARIES
-      ${READLINE_readline_LIBRARY} 
-      )
-
-    # some readline libraries depend on ncurses
-    IF (READLINE_ncurses_LIBRARY)
-      SET (READLINE_LIBRARIES ${READLINE_LIBRARIES} ${READLINE_ncurses_LIBRARY})
-    ENDIF (READLINE_ncurses_LIBRARY)
-
-  ENDIF (READLINE_readline_LIBRARY)
-ENDIF (READLINE_INCLUDE_DIR)
-
-IF (READLINE_FOUND)
-  MESSAGE (STATUS "Found readline library")
-ELSE (READLINE_FOUND)
-  IF (READLINE_FIND_REQUIRED)
-    MESSAGE (FATAL_ERROR "Could not find readline -- please give some paths to CMake")
-  ENDIF (READLINE_FIND_REQUIRED)
-ENDIF (READLINE_FOUND)

doc/man/task-color.5.in

@@ -1,7 +1,7 @@
-.TH task-color 5 2015-01-01 "${PACKAGE_STRING}" "User Manuals"
+.TH task-color 5 2015-05-10 "${PACKAGE_STRING}" "User Manuals"
 
 .SH NAME
-task-color \- A color tutorial for the taskwarrior command line todo manager.
+task-color \- A color tutorial for the Taskwarrior command line todo manager.
 
 .SH SETUP
 The first thing you need is a terminal program that supports color.  All
@@ -18,8 +18,8 @@ which shell you use).  If this is a new setting, you will need to either run
 that profile script, or close and reopen the terminal window (which does the
 same thing).
 
-Now tell taskwarrior that you want to use color.  This is the default for
-taskwarrior, so the following step may be unnecessary.
+Now tell Taskwarrior that you want to use color.  This is the default for
+Taskwarrior, so the following step may be unnecessary.
 
     $ task config color on
 
@@ -28,11 +28,11 @@ like:
 
     color=on
 
-Now taskwarrior is ready.
+Now Taskwarrior is ready.
 
 .SH AUTOMATIC MONOCHROME
-It should be mentioned that taskwarrior is aware of whether its output is going
-to a terminal, or to a file or through a pipe.  When taskwarrior output goes to
+It should be mentioned that Taskwarrior is aware of whether its output is going
+to a terminal, or to a file or through a pipe.  When Taskwarrior output goes to
 a terminal, color is desirable, but consider the following command:
 
     $ task list > file.txt
@@ -268,7 +268,7 @@ from the example):
 include ${CMAKE_INSTALL_PREFIX}/${TASK_RCDIR}/dark-256.theme
 .RE
 
-You can use any of the standard taskwarrior themes:
+You can use any of the standard Taskwarrior themes:
 
 .RS
 dark-16.theme
@@ -323,7 +323,7 @@ http://www.opensource.org/licenses/mit-license.php for more information.
 .BR taskrc(5),
 .BR task-sync(5)
 
-For more information regarding taskwarrior, the following may be referenced:
+For more information regarding Taskwarrior, the following may be referenced:
 
 .TP
 The official site at
@@ -339,5 +339,5 @@ You can contact the project by writing an email to
 
 .SH REPORTING BUGS
 .TP
-Bugs in taskwarrior may be reported to the issue-tracker at
+Bugs in Taskwarrior may be reported to the issue-tracker at
 <http://taskwarrior.org>

doc/man/task-sync.5.in

@@ -1,4 +1,4 @@
-.TH task-sync 5 2015-01-01 "${PACKAGE_STRING}" "User Manuals"
+.TH task-sync 5 2015-05-10 "${PACKAGE_STRING}" "User Manuals"
 
 .SH NAME
 task-sync \- A discussion and tutorial for the various task(1) data
@@ -80,24 +80,25 @@ The Taskserver was designed for this purpose to be secure, fast and conflict
 free, allowing data interchange between assorted Taskwarrior clients, and
 tolerant of network connectivity problems.
 
-There is a 'sync' command built in to Taskwarrior, and with a server account
-and client configuration, syncing is done on demand.
+There is a 'sync' command built in to Taskwarrior (provided the GnuTLS library
+is installed), and with a server account and client configuration, syncing is
+done on demand.
 
 Setup is a matter of creating an account on a Taskserver (see your Taskserver
 provider or operate your own - see
 http://taskwarrior.org/docs/server_setup.html)
 
-Once you have an account, you'll receive a certificate, key and a password.
+Once you have an account, you'll receive a certificate, key, and credentials.
 You'll need to put the certificate and key somewhere like this:
 
     $ cp <name>.cert.pem ~/.task
     $ cp <name>.key.pem ~/.task
 
-Then you configure Taskwarrior, using the account details:
+Then you configure Taskwarrior, using the provided details:
 
     $ task config taskd.certificate ~/.task/<name>.cert.pem
     $ task config taskd.key         ~/.task/<name>.key.pem
-    $ task config taskd.credentials <organization>/<name>/<password>
+    $ task config taskd.credentials <organization>/<name>/<UUID>
     $ task config taskd.server      <server domain>:<port>
 
 If you are using a private server, you are likely also using a self-signed
@@ -108,7 +109,7 @@ certificate, which means you will need one of the following additional entries:
 The CA (Certificate Authority) will be used to verify the server certificate.
 Alternatively, you can override the cert verification process using:
 
-    $ task config taskd.trust       yes
+    $ task config taskd.trust       'allow all'
 
 This is an insecure option that should be used with caution, because it directs
 Taskwarrior to trust any certificate.

doc/man/task.1.in

@@ -1,4 +1,4 @@
-.TH task 1 2015-01-01 "${PACKAGE_STRING}" "User Manuals"
+.TH task 1 2015-05-10 "${PACKAGE_STRING}" "User Manuals"
 
 .SH NAME
 task \- A command line todo manager.
@@ -14,7 +14,7 @@ that you want to do, allowing you to add/remove, and otherwise manipulate them.
 Taskwarrior has a rich set of subcommands that allow you to do various things
 with it.
 
-At the core, taskwarrior is a list processing program. You add text and
+At the core, Taskwarrior is a list processing program. You add text and
 additional related parameters and redisplay the information in a nice way.  It
 turns into a todo list program when you add due dates and recurrence. It turns
 into an organized todo list program when you add priorities, tags (one word
@@ -98,7 +98,7 @@ generated, as with verbose=nothing).
 .SH READ SUBCOMMANDS
 
 Reports are read subcommands. There are several reports currently predefined in
-taskwarrior. The output and sort behavior of these reports can be configured in
+Taskwarrior. The output and sort behavior of these reports can be configured in
 the configuration file. See also the man page taskrc(5).  There are also other
 read subcommands that are not reports.
 
@@ -379,9 +379,25 @@ time from the specified task.
 Miscellaneous subcommands either accept no command line arguments, or accept
 non-standard arguments.
 
+.TP
+.B task calc <expression>
+Evaluates an algebraic expression. Can be used to test how Taskwarrior
+parses and evaluates the expression given on the command line.
+
+Examples:
+
+    task calc 1 + 1
+    2
+
+    task calc now + 8d
+    2015-03-26T18:06:57
+
+    task calc eom
+    2015-03-31T23:59:59
+
 .TP
 .B task config [name [value | '']]
-Add, modify and remove settings directly in the taskwarrior configuration.
+Add, modify and remove settings directly in the Taskwarrior configuration.
 This command either modifies the 'name' setting with a new value of 'value',
 or adds a new entry that is equivalent to 'name=value':
 
@@ -396,6 +412,46 @@ Finally, this command removes any 'name=...' entry from the .taskrc file:
 
     task config name
 
+.TP
+.B task context <name>
+Sets the currently active context. See the CONTEXT section.
+
+Example:
+
+    task context work
+
+.TP
+.B task context delete <name>
+Deletes the context with the name <name>. If the context being deleted is currently
+set as active, it will be unset.
+
+Example:
+
+    task context delete work
+
+.TP
+.B task context define <name> <filter>
+Defines a new context with name <name> and definition <filter>. This command
+does not affect the currently set context, just adds a new context definition.
+
+Examples:
+
+    task context define work project:Work
+    task context define home project:Home or +home
+    task context define superurgent due:today and +urgent
+
+.TP
+.B task context list
+Outputs a list of available contexts along with their definitions.
+
+.TP
+.B task context none
+Unsets the currently active context, if any was set.
+
+.TP
+.B task context show
+Shows the currently active context, along with its definition.
+
 .TP
 .B task diagnostics
 Shows diagnostic information, of the kind needed when reporting a problem.
@@ -454,7 +510,7 @@ Reverts the most recent action.  Obeys the confirmation setting.
 
 .TP
 .B task version
-Shows the taskwarrior version number.
+Shows the Taskwarrior version number.
 
 .SH HELPER SUBCOMMANDS
 
@@ -474,6 +530,10 @@ Generates a list of all commands, for autocompletion purposes.
 .B task _config
 Lists all supported configuration variables, for completion purposes.
 
+.TP
+.B task _context
+Lists all available context variables, for completion purposes.
+
 .TP
 .B task <filter> _ids
 Shows only the IDs of matching tasks, in the form of a list.
@@ -506,7 +566,7 @@ Displays the urgency measure of a task.
 
 .TP
 .B task _version
-Shows only the taskwarrior version number.
+Shows only the Taskwarrior version number.
 
 .TP
 .B task _zshcommands
@@ -810,7 +870,7 @@ The original example therefore must be entered as:
   task '( project:Home or project:Garden )' list
 
 This includes quotes to escape the parentheses, so that the shell doesn't
-interpret them and hide them from taskwarrior.
+interpret them and hide them from Taskwarrior.
 
 There is redundancy between operators, attribute modifiers and other syntactic
 sugar.  For example, the following are all equivalent:
@@ -962,8 +1022,68 @@ biannual, biyearly, 2yr
 Every two years.
 .RE
 
+.SH CONTEXT
+Context is a user-defined filter, which is automatically applied to all commands
+that filter the task list. In particular, any report command will have its
+result affected by the current active context.
+
+    $ task list
+    ID Age Project  Description        Urg
+    1  2d  Sport    Run 5 miles        1.42
+    2  1d  Home     Clean the dishes   1.14
+
+    $ task context home
+    Context 'home' set. Use 'task context none' to remove.
+
+    $ task list
+    ID Age Project  Description        Urg
+    2  1d  Home     Clean the dishes   1.14
+    Context 'home' set. Use 'task context none' to remove.
+
+As seen in the example above, context is applied by specifying its name to the
+"context" command. To change the currently applied context, just pass the
+new context's name to the 'context' command.
+
+To unset any context, use the 'none' subcommand.
+
+    $ task context none
+    Context unset.
+
+    $ task list
+    ID Age Project  Description        Urg
+    1  2d  Sport    Run 5 miles        1.42
+    2  1d  Home     Clean the dishes   1.14
+
+Context can be defined using the 'define' subcommand, specifying both the name
+of the new context, and it's assigned filter.
+
+    $ task context define home
+    Are you sure you want to add 'context.home' with a value of 'project:Home'? (yes/no) yes
+    Context 'home' successfully defined.
+
+To remove the definition, use the 'delete' subcommand.
+
+    $ task context delete home
+    Are you sure you want to remove 'context.home'? (yes/no) yes
+    Context 'home' successfully undefined.
+
+To check what is the currently active context, use the 'show' subcommand.
+
+    $ task context show
+    Context 'home' with filter 'project:Home' is currently applied.
+
+Contexts can store arbitrarily complex filters.
+
+    $ task context define family project:Family or +paul or +nancy
+    Are you sure you want to add 'context.home' with a value of 'project:Family or +paul or +nancy'? (yes/no) yes
+    Context 'family' successfully defined.
+
+Contexts are permanent, and the currently set context name is stored in the
+"context" configuration variable. The context definition is stored in the
+"context.<name>" configuration variable.
+
 .SH COMMAND ABBREVIATION
-All taskwarrior commands may be abbreviated as long as a unique prefix is used,
+All Taskwarrior commands may be abbreviated as long as a unique prefix is used,
 for example:
 
 .RS
@@ -1002,7 +1122,7 @@ $ task add "quoted ' quote"
 $ task add escaped \\' quote
 .RE
 
-The argument \-\- (a double dash) tells taskwarrior to treat all other args
+The argument \-\- (a double dash) tells Taskwarrior to treat all other args
 as description:
 
 .RS
@@ -1057,11 +1177,11 @@ the 'data.location' configuration setting of the task data directory.
 For examples please see the online documentation starting at
 
 .RS
-<http://taskwarrior.org/projects/taskwarrior/wiki>
+<http://taskwarrior.org/docs>
 .RE
 
-Note that the online documentation is more detailed and more current than this
-man page.
+Note that the online documentation can be more detailed and more current than
+this man page.
 
 .SH FILES
 
@@ -1099,7 +1219,7 @@ http://www.opensource.org/licenses/mit-license.php for more information.
 .BR task-color(5),
 .BR task-sync(5)
 
-For more information regarding taskwarrior, see the following:
+For more information regarding Taskwarrior, see the following:
 
 .TP
 The official site at
@@ -1115,6 +1235,6 @@ You can contact the project by emailing
 
 .SH REPORTING BUGS
 .TP
-Bugs in taskwarrior may be reported to the issue-tracker at
+Bugs in Taskwarrior may be reported to the issue-tracker at
 <http://taskwarrior.org>
 

doc/man/taskrc.5.in

@@ -1,4 +1,4 @@
-.TH taskrc 5 2015-01-01 "${PACKAGE_STRING}" "User Manuals"
+.TH taskrc 5 2015-05-10 "${PACKAGE_STRING}" "User Manuals"
 
 .SH NAME
 taskrc \- Configuration details for the task(1) command
@@ -11,7 +11,7 @@ taskrc \- Configuration details for the task(1) command
 .B TASKRC=<directory-path>/.taskrc task ...
 
 .SH DESCRIPTION
-.B taskwarrior
+.B Taskwarrior
 obtains its configuration data from a file called
 .I .taskrc
 \&. This file is normally located in the user's home directory:
@@ -49,43 +49,56 @@ $ task rc.<name>=<value> ...
 .RE
 
 If
-.B taskwarrior
+.B Taskwarrior
 is run without an existing configuration file it will ask if it should create a
 default, sample
 .I .taskrc
 file in the user's home directory.
 
-The taskwarrior configuration file consists of a series of assignments in each
-line. The assignments have the syntax:
+The .taskrc file follows a very simply syntax defining name/value pairs:
 
 .RS
-<name-of-configuration-variable>=<value-to-be-set>
+<name> = <value>
 .RE
 
-where:
-.RS
-.TP
-<name-of-configuration-variable>
-is one of the variables described below
+There may be whitespace around <name>, '=' and <value>, and it is ignored.
+Whitespace within the <value> is left intact.
+Whitespace is not permitted in comma-separated lists.
+The entry must be on a single line, no continuations.
+Values support UTF8 as well as JSON encoding, such as \\uNNNN.
 
-.TP
-<value-to-be-set>
-is the value the variable is to be set to.
-.RE
-
-and set a configuration variable to a certain value. The equal sign ("=") is
-used to separate the variable name from the value to be set.
-
-The hash mark, or pound sign ("#") is used as a comment character. It can be
-used to annotate the configuration file. All text after the character to the end
-of the line is ignored.
-
-The configuration file supports UTF8 as well as JSON encoding, such as \\uNNNN.
-
-Note that taskwarrior is flexible about the values used to represent Boolean
+Note that Taskwarrior is flexible about the values used to represent Boolean
 items. You can use "on", "yes", "y", "1" and "true".
 Anything else means "off".
 
+.RS
+include <file>
+.RE
+
+There may be whitespace around 'include' and <file>. The file may be an
+absolute or relative path, and the special character '~' is expanded to mean
+$HOME.
+The entry must be on a single line, no continuations.
+
+.RS
+# <comment>
+.RE
+
+A comment consists of the character '#', and extends from the '#' to the end
+of the line. There is no way to comment a multi-line block. There may be
+blank lines.
+
+Almost every value has a default setting, and an empty .taskrc file is one
+that makes use of every default. The contents of the .taskrc file therefore
+represent overrides of the default values. To remove a default value completely
+there must be an entry like this:
+
+.RS
+<name> =
+.RE
+
+This entry overrides the default value with a blank value.
+
 .SH EDITING
 You can edit your .taskrc file by hand if you wish, or you can use the 'config'
 command. To permanently set a value in your .taskrc file, use this command:
@@ -136,7 +149,7 @@ include ${CMAKE_INSTALL_PREFIX}/${TASK_RCDIR}/holidays.en-US.rc
 include ${CMAKE_INSTALL_PREFIX}/${TASK_RCDIR}/dark-16.theme
 .RE
 
-This includes two standard files that are distributed with taskwarrior, which
+This includes two standard files that are distributed with Taskwarrior, which
 define a set of US holidays, and set up a 16-color theme to use, to color the
 reports and calendar.
 
@@ -145,7 +158,7 @@ These environment variables override defaults and command line arguments.
 
 .TP
 .B TASKDATA=~/.task
-This overrides the default path for the taskwarrior data files.
+This overrides the default path for the Taskwarrior data files.
 
 .TP
 .B TASKRC=~/.taskrc
@@ -158,7 +171,7 @@ Valid variable names and their default values are:
 
 .TP
 .B data.location=$HOME/.task
-This is a path to the directory containing all the taskwarrior files. By
+This is a path to the directory containing all the Taskwarrior files. By
 default, it is set up to be ~/.task, for example: /home/paul/.task
 
 Note that you can use the
@@ -281,15 +294,14 @@ and project. The others are sent to standard output.
 
 .TP
 .B confirmation=yes
-May be "yes" or "no", and determines whether taskwarrior will ask for
-confirmation before deleting a task, performing bulk changes, or the undo
-command. The default value is "yes". Consider leaving this setting as "yes",
-for safety.
+May be "yes" or "no", and determines whether Taskwarrior will ask for
+confirmation before deleting a task or performing the undo command. The default
+value is "yes". Consider leaving this setting as "yes", for safety.
 
 .TP
 .B allow.empty.filter=yes
 An empty filter combined with a write command is potentially a way to modify
-all tasks by mistkae, and when this is detected, confirmation is required.
+all tasks by mistake, and when this is detected, confirmation is required.
 Setting this to 'no' means that it is an error to use a write command with no
 filter.
 
@@ -313,25 +325,30 @@ Controls padding between columns of the report output. Default is "1".
 .TP
 .B bulk=3
 Is a number, defaulting to 3. When this number or greater of tasks are modified
-in a single command, confirmation will be required, unless the
+in a single command, confirmation will be required, regardless of the value of
 .B confirmation
-variable is "no".
+variable. The special value bulk=0 is treated as an infinity.
 
 This is useful for preventing large-scale unintended changes.
 
 .TP
 .B nag=You have more urgent tasks.
-This may be a string of text, or blank. It is used as a prompt when a task is
-started or completed that is not considered high priority. Default value is:
-You have more urgent tasks. It is a gentle reminder that you are contradicting
-your own priority settings.
+This may be a string of text, or blank.  It is used as a prompt when a task is
+started of completed, when there are other tasks with a higher urgency.
+Default value is: 'You have more urgent tasks'.
+It is a gentle reminder that you are contradicting your own urgency settings.
 
 .TP
-.B list.all.projects=yes
+.B list.all.projects=no
 May be yes or no, and determines whether the 'projects' command lists all the project
 names you have used, or just the ones used in active tasks. The default value is
 "no".
 
+.TP
+.B summary.all.projects=no
+If set to yes, shows all projects in the summary report, even if there are no
+pending tasks.  The default value is "no".
+
 .TP
 .B complete.all.tags=yes
 May be yes or no, and determines whether the tab completion scripts consider all
@@ -371,7 +388,7 @@ Defaults to infix.
 
 .TP
 .B dom=on
-Enables or disables access to taskwarrior internals and task metadata on the
+Enables or disables access to Taskwarrior internals and task metadata on the
 command line. Defaults to on.
 
 .TP
@@ -412,6 +429,11 @@ The character or string to show in the depends.indicator column. Defaults to +.
 .B uda.<name>.indicator=U
 The character or string to show in the <uda>.indicator column. Defaults to U.
 
+.TP
+.B recurrence=yes
+Controls whether recurrence is enable, and whether recurring tasks continue to
+generate new task instances. Defaults to 'yes'.
+
 .TP
 .B recurrence.confirmation=prompt
 Controls whether changes to recurring tasks propagates to other child tasks with
@@ -432,7 +454,7 @@ and one for a week from tomorrow.
 
 .TP
 .B undo.style=side
-When the 'undo' command is run, taskwarrior presents a before and after
+When the 'undo' command is run, Taskwarrior presents a before and after
 comparison of the data. This can be in either the 'side' style, which compares
 values side-by-side in a table, or 'diff' style, which uses a format similar to
 the 'diff' command.
@@ -480,6 +502,12 @@ shows parse trees from every phase of the parse.
 Controls the GnuTLS diagnostic level. For 'sync' debugging. Level 0 means no
 diagnostics. Level 9 is the highest. Level 2 is a good setting for debugging.
 
+.TP
+.B obfuscate=1
+When set to '1', will replace all report text with 'xxx'.
+This is useful for sharing report output in bug reports.
+Default value is '0'.
+
 .TP
 .B alias.rm=delete
 Taskwarrior supports command aliases. This alias provides an alternate name
@@ -503,7 +531,7 @@ the 'history' report, for example, or 'export'.
 .B dateformat.annotation=
 .TP
 .B report.X.dateformat=Y-M-D
-This is a string of characters that defines how taskwarrior formats date values.
+This is a string of characters that defines how Taskwarrior formats date values.
 The precedence order for the configuration variable is report.X.dateformat then
 dateformat.report then dateformat for formating the due dates in reports.
 If both report.X.dateformat and dateformat.report are not set then dateformat
@@ -616,6 +644,10 @@ field that is set. Otherwise, they are set to the corresponding values of
 .RE
 .RE
 
+.TP
+.B date.iso=yes
+Enables ISO-8601 date support.  The default value is "yes".
+
 .TP
 .B weekstart=Sunday
 Determines the day a week starts. Valid values are Sunday or Monday only. The
@@ -729,7 +761,7 @@ holiday.eastersunday.date=easter
 .RE
 .RE
 
-Note that the taskwarrior distribution contains example holiday files that can
+Note that the Taskwarrior distribution contains example holiday files that can
 be included like this:
 
 .RS
@@ -743,7 +775,7 @@ include ${CMAKE_INSTALL_PREFIX}/${TASK_RCDIR}/holidays.en-US.rc
 .B monthsperline=3
 Determines how many months the "task calendar" command renders across the
 screen. Defaults to however many will fit. If more months than will fit are
-specified, taskwarrior will only show as many that will fit.
+specified, Taskwarrior will only show as many that will fit.
 
 .SS DEPENDENCIES
 
@@ -759,7 +791,7 @@ Determines whether dependency chain repair requires confirmation.
 
 .TP
 .B color=on
-May be "on" or "off". Determines whether taskwarrior uses color. When "off",
+May be "on" or "off". Determines whether Taskwarrior uses color. When "off",
 will use dashes (-----) to underline column headings.
 
 .TP
@@ -820,18 +852,6 @@ Task has at least one tag.
 .B color.recurring
 Task is recurring.
 .br
-.B color.pri.H
-Task has priority H.
-.br
-.B color.pri.M
-Task has priority M.
-.br
-.B color.pri.L
-Task has priority L.
-.br
-.B color.pri.none
-Task has no priority.
-.br
 .B color.completed
 Task is completed.
 .br
@@ -1050,10 +1070,6 @@ Urgency inherited from dependency chain
 .RS
 Urgency coefficient for due dates
 .RE
-.B urgency.priority.coefficient=6.0
-.RS
-Urgency coefficient for priorities
-.RE
 .B urgency.waiting.coefficient=-3.0
 .RS
 Urgency coefficient for waiting status
@@ -1094,6 +1110,10 @@ Specific tag coefficient.
 .RS
 Specific project coefficient.
 .RE
+.B urgency.user.keyword.<keyword>.coefficient=...
+.RS
+Specific description keyword coefficient.
+.RE
 .B urgency.uda.<name>.coefficient=...
 .RS
 Presence/absence of UDA data.
@@ -1116,13 +1136,6 @@ Provides a default project name for the
 .I task add
 command, if you don't specify one. The default is blank.
 
-.TP
-.B
-default.priority=M
-Provides a default priority for the
-.I task add
-command, if you don't specify one. The default is blank.
-
 .TP
 .B
 default.due=...
@@ -1141,7 +1154,7 @@ command, if you don't specify values. The default is blank.
 .TP
 .B
 default.command=next
-Provides a default command that is run every time taskwarrior is invoked with no
+Provides a default command that is run every time Taskwarrior is invoked with no
 arguments. For example, if set to:
 
 .RS
@@ -1151,7 +1164,7 @@ default.command=project:foo list
 .RE
 
 .RS
-then taskwarrior will run the "project:foo list" command if no command is
+then Taskwarrior will run the "project:foo list" command if no command is
 specified. This means that by merely typing
 .RE
 
@@ -1328,19 +1341,56 @@ Provides a default report label for the UDA called '<name>'.
 For type 'string' UDAs only, this provides a comma-separated list of acceptable
 values. In this example, the '<name>' UDA may only contain values 'A', 'B',
 or 'C', but may also contain no value.
+
+Note that the order of the value is important, and denotes the sort order from
+highest ('A') to lowest ('C').
+
+Note that a blank value is permitted.
+.RE
+
+.TP
+.B uda.<name>.default=...
+.RS
+Provides a default value for the UDA called '<name>'.
 .RE
 
 .TP
 .B Example 'estimate' UDA
 This example shows an 'estimate' UDA that stores specific values for the size
-of a task.
+of a task.  Note the blank value after 'trivial'.
 
 .RS
 .B uda.estimate.type=string
 .br
 .B uda.estimate.label=Size Estimate
 .br
-.B uda.estimate.values=trivial,small,medium,large,huge
+.B uda.estimate.values=huge,large,medium,small,trivial,
+.RE
+
+.RS
+Note that the value are sorted
+
+    huge > large > medium > small > trivial > ''
+.RE
+
+.SS CONTEXT
+Context setting is a mechanism which allows the user to set a permanent filter,
+thus avoiding the need to specify one filter repeatedly. More details on usage
+can be found in the task(1) manpage.
+
+The current context is stored in the taskrc file, along with definitions for
+all user provided contexts.
+
+.TP
+.B context=<name>
+.RS
+Stores the value of the currently active context.
+.RE
+
+.TP
+.B context.<name>=<filter>
+.RS
+Stores the definition of the context with the name <name>.
 .RE
 
 .SS SYNC
@@ -1413,7 +1463,7 @@ http://www.opensource.org/licenses/mit-license.php for more information.
 .BR task-color(5),
 .BR task-sync(5)
 
-For more information regarding taskwarrior, see the following:
+For more information regarding Taskwarrior, see the following:
 
 .TP
 The official site at
@@ -1429,6 +1479,6 @@ You can contact the project by emailing
 
 .SH REPORTING BUGS
 .TP
-Bugs in taskwarrior may be reported to the issue-tracker at
+Bugs in Taskwarrior may be reported to the issue-tracker at
 <http://taskwarrior.org>
 

doc/misc/themes/README

@@ -1,15 +1,24 @@
 Themes
 
-To generate samples of themes, first execute the 'run' script to generate the
+To generate samples of themes, first execute the 'setup' script to generate the
 sample data.  Note that this data may need to be tweaked to include qualities
 that need to be illustrated in theme sample.
 
-Then edit the 'rc' file to include the desired theme file.
+Using a dark-background terminal (black recommended), run the following:
 
-Then run 'per' to run a few commands for each theme.
+  run.dark
 
-Note that this will require that the terminal window be switched between a black
-and white background to properly show the light and dark themes.
+Using a light-background terminal, run the following:
+
+  run.light
+
+Using a solarized dark terminal, run the following:
+
+  run.solar.dark
+
+Using a solarized light terminal, run the following:
+
+  run.solar.light
 
 Note that for the solarized themes, the terminal color palette needs to be set
 to specific colors.

doc/misc/themes/run

@@ -1,23 +0,0 @@
-
-
-for theme in $PWD/../../rc/*.theme
-do
-  cat <<EOF >>x
-data.location=.
-confirmation=off
-_forcecolor=on
-include $theme
-EOF
-
-  echo "--- $theme -----------------------------------------------------"
-  echo '$ task list'
-  task rc:x list
-  echo '$ task summary'
-  task rc:x summary
-  echo '$ task ghistory'
-  task rc:x ghistory
-  echo '$ task calendar'
-  task rc:x calendar
-  echo '$ task burndown.daily'
-  task rc:x burndown.daily
-done

doc/misc/themes/run.dark

@@ -21,6 +21,8 @@ include $theme
 EOF
 
   echo "--- $theme -----------------------------------------------------"
+  echo '$ task color legend'
+  task rc:x color legend
   echo '$ task list'
   task rc:x list
   echo '$ task summary'

doc/misc/themes/run.default

@@ -0,0 +1,24 @@
+#!/bin/bash
+
+cat <<EOF >x
+data.location=.
+confirmation=off
+detection=off
+_forcecolor=on
+default.height=24
+verbose=off
+EOF
+
+echo "--- DEFAULT -----------------------------------------------------"
+echo '$ task color legend'
+task rc:x color legend
+echo '$ task list'
+task rc:x list
+echo '$ task summary'
+task rc:x summary
+echo '$ task ghistory'
+task rc:x ghistory
+echo '$ task calendar'
+task rc:x calendar
+echo '$ task burndown.daily'
+task rc:x burndown.daily

doc/misc/themes/run.light

@@ -14,6 +14,8 @@ include $theme
 EOF
 
   echo "--- $theme -----------------------------------------------------"
+  echo '$ task color legend'
+  task rc:x color legend
   echo '$ task list'
   task rc:x list
   echo '$ task summary'

doc/misc/themes/run.solar.dark

@@ -13,6 +13,8 @@ include $theme
 EOF
 
   echo "--- $theme -----------------------------------------------------"
+  echo '$ task color legend'
+  task rc:x color legend
   echo '$ task list'
   task rc:x list
   echo '$ task summary'

doc/misc/themes/run.solar.light

@@ -13,6 +13,8 @@ include $theme
 EOF
 
   echo "--- $theme -----------------------------------------------------"
+  echo '$ task color legend'
+  task rc:x color legend
   echo '$ task list'
   task rc:x list
   echo '$ task summary'

doc/rc/dark-16.theme

@@ -24,7 +24,7 @@
 #
 ###############################################################################
 
-rule.precedence.color=deleted,completed,active,keyword.,tag.,uda.,project.,overdue,scheduled,due.today,due,blocked,blocking,recurring,tagged,pri.
+rule.precedence.color=deleted,completed,active,keyword.,tag.,project.,overdue,scheduled,due.today,due,blocked,blocking,recurring,tagged,uda.
 
 # General decoration
 color.label=
@@ -33,15 +33,15 @@ color.alternate=
 color.header=yellow
 color.footnote=yellow
 color.warning=bold red
-color.error=yellow
-color.debug=yellow
+color.error=white on red
+color.debug=blue
 
 # Task state
-color.completed=green on white
-color.deleted=red on white
+color.completed=
+color.deleted=
 color.active=black on bright green
 color.recurring=magenta
-color.scheduled=on green
+color.scheduled=white on green
 color.until=
 color.blocked=black on white
 color.blocking=black on bright white
@@ -50,10 +50,9 @@ color.blocking=black on bright white
 color.project.none=
 
 # Priority
-color.pri.H=bold white
-color.pri.M=white
-color.pri.L=
-color.pri.none=
+color.uda.priority.H=bold white
+color.uda.priority.M=white
+color.uda.priority.L=
 
 # Tags
 color.tag.next=bold yellow

doc/rc/dark-256.theme

@@ -24,36 +24,35 @@
 #
 ###############################################################################
 
-rule.precedence.color=deleted,completed,active,keyword.,tag.,uda.,project.,overdue,scheduled,due.today,due,blocked,blocking,recurring,tagged,pri.
+rule.precedence.color=deleted,completed,active,keyword.,tag.,project.,overdue,scheduled,due.today,due,blocked,blocking,recurring,tagged,uda.
 
 # General decoration
 color.label=
 color.label.sort=
-color.alternate=on gray1
+color.alternate=on gray2
 color.header=color3
 color.footnote=color3
 color.warning=bold red
-color.error=color3
-color.debug=color3
+color.error=white on red
+color.debug=color4
 
 # Task state
-color.completed=rgb010 on white
-color.deleted=rgb100 on white
+color.completed=
+color.deleted=
 color.active=rgb555 on rgb410
 color.recurring=rgb013
 color.scheduled=on rgb001
 color.until=
 color.blocked=white on color8
-color.blocking=white on color15
+color.blocking=black on color15
 
 # Project
 color.project.none=
 
 # Priority
-color.pri.H=color255
-color.pri.L=color245
-color.pri.M=color250
-color.pri.none=
+color.uda.priority.H=color255
+color.uda.priority.L=color245
+color.uda.priority.M=color250
 
 # Tags
 color.tag.next=rgb440

doc/rc/dark-blue-256.theme

@@ -24,24 +24,24 @@
 #
 ###############################################################################
 
-rule.precedence.color=deleted,completed,active,keyword.,tag.,uda.,project.,overdue,scheduled,due.today,due,blocked,blocking,recurring,tagged,pri.
+rule.precedence.color=deleted,completed,active,keyword.,tag.,project.,overdue,scheduled,due.today,due,blocked,blocking,recurring,tagged,uda.
 
 # General decoration
 color.label=
 color.label.sort=
-color.alternate=on gray1
+color.alternate=on gray2
 color.header=rgb013
 color.footnote=rgb013
 color.warning=bold red
-color.error=rgb013
+color.error=white on red
 color.debug=rgb013
 
 # Task state
-color.completed=rgb001 on white
-color.deleted=rgb100 on white
+color.completed=
+color.deleted=
 color.active=rgb045 on rgb015
 color.recurring=rgb115
-color.scheduled=on rgb011
+color.scheduled=on rgb012
 color.until=
 color.blocked=white on rgb001
 color.blocking=white on rgb002
@@ -50,10 +50,9 @@ color.blocking=white on rgb002
 color.project.none=
 
 # Priority
-color.pri.H=rgb035
-color.pri.L=rgb015
-color.pri.M=rgb025
-color.pri.none=
+color.uda.priority.H=rgb035
+color.uda.priority.L=rgb015
+color.uda.priority.M=rgb025
 
 # Tags
 color.tag.next=rgb550
@@ -76,8 +75,8 @@ color.history.delete=color0 on rgb035
 color.history.done=color0 on rgb025
 
 # Report: summary
-color.summary.background=white on color0
-color.summary.bar=white on rgb003
+color.summary.background=on rgb001
+color.summary.bar=on rgb114
 
 # Command: calendar
 color.calendar.due.today=color0 on color252

doc/rc/dark-gray-256.theme

@@ -24,21 +24,21 @@
 #
 ###############################################################################
 
-rule.precedence.color=deleted,completed,active,keyword.,tag.,uda.,project.,overdue,scheduled,due.today,due,blocked,blocking,recurring,tagged,pri.
+rule.precedence.color=deleted,completed,active,keyword.,tag.,project.,overdue,scheduled,due.today,due,blocked,blocking,recurring,tagged,uda.
 
 # General decoration
 color.label=
 color.label.sort=
-color.alternate=on gray1
+color.alternate=on gray2
 color.header=color0 on gray11
 color.footnote=on gray5
 color.warning=bold red
-color.error=red on white
-color.debug=black on white
+color.error=white on red
+color.debug=blue
 
 # Task state
-color.completed=black on white
-color.deleted=black on white
+color.completed=
+color.deleted=
 color.active=black on gray18
 color.recurring=
 color.scheduled=on gray8
@@ -50,10 +50,9 @@ color.blocking=bold white
 color.project.none=
 
 # Priority
-color.pri.H=gray21
-color.pri.M=gray18
-color.pri.L=gray12
-color.pri.none=gray18
+color.uda.priority.H=gray21
+color.uda.priority.M=gray18
+color.uda.priority.L=gray12
 
 # Tags
 color.tag.next=bold white
@@ -61,7 +60,7 @@ color.tag.none=
 color.tagged=
 
 # Due
-color.due=on gray2
+color.due=on gray3
 color.due.today=on gray4
 color.overdue=on gray6
 
@@ -76,12 +75,12 @@ color.history.delete=black on gray10
 color.history.done=gray5 on gray23
 
 # Report: summary
-color.summary.bar=on gray15
-color.summary.background=on black
+color.summary.bar=on gray12
+color.summary.background=on gray5
 
 # Command: calendar
 color.calendar.due=on gray8
-color.calendar.due.today=on gray15
+color.calendar.due.today=black on gray15
 color.calendar.holiday=black on gray20
 color.calendar.overdue=gray2 on gray10
 color.calendar.today=bold white
@@ -89,7 +88,7 @@ color.calendar.weekend=on gray2
 color.calendar.weeknumber=gray6
 
 # Command: sync
-color.sync.add=gray15 on gray5
+color.sync.added=gray15 on gray5
 color.sync.changed=black on gray10
 color.sync.rejected=gray5 on gray23
 

doc/rc/dark-gray-blue-256.theme

@@ -24,17 +24,17 @@
 #
 ###############################################################################
 
-rule.precedence.color=deleted,completed,active,keyword.,tag.,uda.,project.,overdue,scheduled,due.today,due,blocked,blocking,recurring,tagged,pri.
+rule.precedence.color=deleted,completed,active,keyword.,tag.,project.,overdue,scheduled,due.today,due,blocked,blocking,recurring,tagged,uda.
 
 # General decoration
 color.label=
 color.label.sort=
-color.alternate=on gray0
+color.alternate=on gray2
 color.header=gray10
 color.footnote=gray10
 color.warning=
-color.error=rgb500
-color.debug=rgb500
+color.error=white on red
+color.debug=blue
 
 # Task state
 color.completed=
@@ -44,16 +44,15 @@ color.recurring=
 color.scheduled=
 color.until=
 color.blocked=bold gray10 on gray4
-color.blocking=bold gray18 on gray6
+color.blocking=gray18 on gray6
 
 # Project
 color.project.none=
 
 # Priority
-color.pri.H=
-color.pri.L=
-color.pri.M=
-color.pri.none=
+color.uda.priority.H=
+color.uda.priority.L=
+color.uda.priority.M=
 
 # Tags
 color.tag.next=
@@ -66,18 +65,18 @@ color.due.today=color0 on rgb024
 color.overdue=color0 on rgb035
 
 # Report: burndown
-color.burndown.pending=on gray9
-color.burndown.started=on gray16
-color.burndown.done=on rgb013
+color.burndown.pending=white on gray9
+color.burndown.started=black on gray16
+color.burndown.done=white on rgb013
 
 # Report: history
-color.history.add=on gray9
-color.history.delete=black on gray23
-color.history.done=black on rgb013
+color.history.add=white on gray6
+color.history.delete=black on gray18
+color.history.done=black on rgb024
 
 # Report: summary
 color.summary.bar=on rgb012
-color.summary.background=on color0
+color.summary.background=on gray2
 
 # Command: calendar
 color.calendar.due=color0 on gray10
@@ -94,6 +93,6 @@ color.sync.changed=gray15
 color.sync.rejected=gray23
 
 # Command: undo
-color.undo.before=green
-color.undo.after=red
+color.undo.before=rgb013
+color.undo.after=rgb035
 

doc/rc/dark-green-256.theme

@@ -24,24 +24,24 @@
 #
 ###############################################################################
 
-rule.precedence.color=deleted,completed,active,keyword.,tag.,uda.,project.,overdue,scheduled,due.today,due,blocked,blocking,recurring,tagged,pri.
+rule.precedence.color=deleted,completed,active,keyword.,tag.,project.,overdue,scheduled,due.today,due,blocked,blocking,recurring,tagged,uda.
 
 # General decoration
 color.label=
 color.label.sort=
-color.alternate=on gray1
+color.alternate=on gray2
 color.header=rgb031
 color.footnote=rgb031
-color.warning=bold red
-color.error=rgb031
-color.debug=rgb031
+color.warning=rgb020
+color.error=white on red
+color.debug=blue
 
 # Task state
-color.completed=rgb020 on white
-color.deleted=rgb200 on white
+color.completed=
+color.deleted=
 color.active=rgb050 on rgb010
 color.recurring=rgb151
-color.scheduled=on rgb011
+color.scheduled=black on rgb031
 color.until=
 color.blocked=white on rgb010
 color.blocking=white on rgb020
@@ -50,10 +50,9 @@ color.blocking=white on rgb020
 color.project.none=
 
 # Priority
-color.pri.H=rgb050
-color.pri.L=rgb010
-color.pri.M=rgb030
-color.pri.none=
+color.uda.priority.H=rgb050
+color.uda.priority.L=rgb010
+color.uda.priority.M=rgb030
 
 # Tags
 color.tag.next=rgb440
@@ -76,7 +75,7 @@ color.history.delete=color0 on rgb050
 color.history.done=color0 on rgb030
 
 # Report: summary
-color.summary.background=white on color0
+color.summary.background=white on gray3
 color.summary.bar=white on rgb030
 
 # Command: calendar
@@ -95,4 +94,4 @@ color.sync.rejected=rgb010
 
 # Command: undo
 color.undo.after=rgb053
-color.undo.before=rgb031
+color.undo.before=rgb021

doc/rc/dark-red-256.theme

@@ -24,24 +24,24 @@
 #
 ###############################################################################
 
-rule.precedence.color=deleted,completed,active,keyword.,tag.,uda.,project.,overdue,scheduled,due.today,due,blocked,blocking,recurring,tagged,pri.
+rule.precedence.color=deleted,completed,active,keyword.,tag.,project.,overdue,scheduled,due.today,due,blocked,blocking,recurring,tagged,uda.
 
 # General decoration
 color.label=
 color.label.sort=
-color.alternate=on gray1
+color.alternate=on gray2
 color.header=rgb100
 color.footnote=rgb100
-color.warning=bold red
-color.error=rgb100
-color.debug=rgb100
+color.warning=red
+color.error=white on red
+color.debug=blue
 
 # Task state
-color.completed=rgb020 on white
-color.deleted=rgb200 on white
+color.completed=
+color.deleted=
 color.active=rgb500 on rgb100
 color.recurring=rgb511
-color.scheduled=on rgb201
+color.scheduled=white on rgb311
 color.until=
 color.blocked=white on rgb100
 color.blocking=white on rgb200
@@ -50,13 +50,12 @@ color.blocking=white on rgb200
 color.project.none=
 
 # Priority
-color.pri.H=rgb500
-color.pri.L=rgb300
-color.pri.M=rgb400
-color.pri.none=
+color.uda.priority.H=rgb500
+color.uda.priority.L=rgb300
+color.uda.priority.M=rgb400
 
 # Tags
-color.tag.next=rgb440
+color.tag.next=rgb511
 color.tag.none=
 color.tagged=color246
 
@@ -94,6 +93,6 @@ color.sync.changed=rgb411
 color.sync.rejected=rgb200
 
 # Command: undo
-color.undo.after=rgb503
-color.undo.before=rgb301
+color.undo.after=rgb511
+color.undo.before=rgb200
 

doc/rc/dark-violets-256.theme

@@ -24,7 +24,7 @@
 #
 ###############################################################################
 
-rule.precedence.color=deleted,completed,active,keyword.,tag.,uda.,project.,overdue,scheduled,due.today,due,blocked,blocking,recurring,tagged,pri.
+rule.precedence.color=deleted,completed,active,keyword.,tag.,project.,overdue,scheduled,due.today,due,blocked,blocking,recurring,tagged,uda.
 
 # General decoration
 color.label=
@@ -33,15 +33,15 @@ color.alternate=on gray2
 color.header=rgb013
 color.footnote=rgb013
 color.warning=
-color.error=rgb013
-color.debug=rgb013
+color.error=white on red
+color.debug=blue
 
 # Task state
 color.completed=
 color.deleted=
 color.active=rgb445 on rgb213
 color.recurring=rgb115
-color.scheduled=
+color.scheduled=white on rgb113
 color.until=
 color.blocked=white on rgb101
 color.blocking=white on rgb202
@@ -50,10 +50,9 @@ color.blocking=white on rgb202
 color.project.none=
 
 # Priority
-color.pri.H=rgb404
-color.pri.M=rgb304
-color.pri.L=rgb325
-color.pri.none=
+color.uda.priority.H=rgb404
+color.uda.priority.M=rgb304
+color.uda.priority.L=rgb325
 
 # Tags
 color.tag.next=
@@ -61,9 +60,9 @@ color.tag.none=
 color.tagged=rgb334
 
 # Due
-color.due=rgb055
-color.due.today=rgb533
-color.overdue=color9
+color.due=rgb015
+color.due.today=rgb125
+color.overdue=color5
 
 # Report: burndown
 color.burndown.pending=on rgb103
@@ -76,14 +75,14 @@ color.history.done=color0 on rgb205
 color.history.delete=color0 on rgb305
 
 # Report: summary
-color.summary.bar=white on rgb103
-color.summary.background=white on color0
+color.summary.bar=white on rgb104
+color.summary.background=white on rgb001
 
 # Command: calendar
 color.calendar.due=color0 on rgb325
 color.calendar.due.today=color0 on rgb404
-color.calendar.holiday=color15 on rgb022
-color.calendar.overdue=color0 on color9
+color.calendar.holiday=color15 on rgb102
+color.calendar.overdue=color0 on color5
 color.calendar.today=color15 on rgb103
 color.calendar.weekend=gray12 on gray3
 color.calendar.weeknumber=rgb104

doc/rc/dark-yellow-green.theme

@@ -24,7 +24,7 @@
 #
 ###############################################################################
 
-rule.precedence.color=deleted,completed,active,keyword.,tag.,uda.,project.,overdue,scheduled,due.today,due,blocked,blocking,recurring,tagged,pri.
+rule.precedence.color=deleted,completed,active,keyword.,tag.,project.,overdue,scheduled,due.today,due,blocked,blocking,recurring,tagged,uda.
 
 # General decoration
 color.label=
@@ -33,15 +33,15 @@ color.alternate=on gray2
 color.header=rgb031
 color.footnote=rgb031
 color.warning=
-color.error=rgb031
-color.debug=rgb031
+color.error=white on red
+color.debug=blue
 
 # Task state
 color.completed=
 color.deleted=
-color.active=rgb451 on rgb310
+color.active=rgb451 on rgb320
 color.recurring=rgb343
-color.scheduled=
+color.scheduled=black on rgb441
 color.until=
 color.blocked=white on rgb110
 color.blocking=white on rgb220
@@ -50,10 +50,9 @@ color.blocking=white on rgb220
 color.project.none=
 
 # Priority
-color.pri.H=rgb450
-color.pri.M=rgb030
-color.pri.L=rgb010
-color.pri.none=
+color.uda.priority.H=rgb450
+color.uda.priority.M=rgb030
+color.uda.priority.L=rgb010
 
 # Tags
 color.tag.next=
@@ -61,9 +60,9 @@ color.tag.none=
 color.tagged=rgb342
 
 # Due
-color.due=rgb420
-color.due.today=rgb410
-color.overdue=rgb400
+color.due=rgb440
+color.due.today=rgb430
+color.overdue=rgb420
 
 # Report: burndown
 color.burndown.pending=on rgb110
@@ -71,19 +70,19 @@ color.burndown.started=on rgb430
 color.burndown.done=on gray4
 
 # Report: history
-color.history.add=color0 on rgb010
-color.history.done=color0 on rgb030
-color.history.delete=color0 on rgb050
+color.history.add=color0 on rgb110
+color.history.done=color0 on rgb430
+color.history.delete=white on gray4
 
 # Report: summary
-color.summary.bar=white on rgb030
-color.summary.background=white on color0
+color.summary.bar=white on rgb330
+color.summary.background=white on rgb110
 
 # Command: calendar
-color.calendar.due=color0 on rgb430
-color.calendar.due.today=color0 on rgb410
+color.calendar.due=color0 on rgb440
+color.calendar.due.today=color0 on rgb430
 color.calendar.holiday=rgb151 on rgb020
-color.calendar.overdue=color0 on rgb400
+color.calendar.overdue=color0 on rgb420
 color.calendar.today=color15 on rgb110
 color.calendar.weekend=on color235
 color.calendar.weeknumber=rgb110
@@ -94,6 +93,6 @@ color.sync.changed=rgb430
 color.sync.rejected=rgb110
 
 # Command: undo
-color.undo.before=rgb031
-color.undo.after=rgb053
+color.undo.before=rgb021
+color.undo.after=rgb042
 

doc/rc/light-16.theme

@@ -24,7 +24,7 @@
 #
 ###############################################################################
 
-rule.precedence.color=deleted,completed,active,keyword.,tag.,uda.,project.,overdue,scheduled,due.today,due,blocked,blocking,recurring,tagged,pri.
+rule.precedence.color=deleted,completed,active,keyword.,tag.,project.,overdue,scheduled,due.today,due,blocked,blocking,recurring,tagged,uda.
 
 # General decoration
 color.label=
@@ -33,12 +33,12 @@ color.alternate=
 color.header=bold white on bright black
 color.footnote=bold cyan on bright black
 color.warning=bold red
-color.error=red on white
-color.debug=white on black
+color.error=white on red
+color.debug=blue
 
 # Task state
-color.completed=green on black
-color.deleted=red on black
+color.completed=
+color.deleted=
 color.active=bold yellow on bright black
 color.recurring=
 color.scheduled=on bright cyan
@@ -50,10 +50,9 @@ color.blocked=blue
 color.project.none=
 
 # Priority
-color.pri.H=
-color.pri.M=
-color.pri.L=bright black
-color.pri.none=
+color.uda.priority.H=
+color.uda.priority.M=
+color.uda.priority.L=bright black
 
 # Tags
 color.tag.next=bold yellow on bright black
@@ -63,17 +62,17 @@ color.tagged=
 # Due
 color.due=on bright green
 color.due.today=on bright yellow
-color.overdue=on bright magenta
+color.overdue=on bright red
 
 # Report: burndown
-color.burndown.pending=on bright green
+color.burndown.pending=on bright red
 color.burndown.started=on bright yellow
 color.burndown.done=on green
 
 # Report: history
-color.history.add=blue on bright yellow
-color.history.done=green on bright green
-color.history.delete=black on red
+color.history.add=black on bright red
+color.history.done=black on bright green
+color.history.delete=black on yellow
 
 # Report: summary
 color.summary.bar=on bright green
@@ -82,11 +81,11 @@ color.summary.background=on white
 # Command: calendar
 color.calendar.due=on bright green
 color.calendar.due.today=blue on bright yellow
-color.calendar.holiday=yellow
-color.calendar.overdue=on bright magenta
+color.calendar.holiday=on yellow
+color.calendar.overdue=on bright red
 color.calendar.today=blue
 color.calendar.weekend=on white
-color.calendar.weeknumber=white on bright black
+color.calendar.weeknumber=blue
 
 # Command: sync
 color.sync.added=green

doc/rc/light-256.theme

@@ -24,7 +24,7 @@
 #
 ###############################################################################
 
-rule.precedence.color=deleted,completed,active,keyword.,tag.,uda.,project.,overdue,scheduled,due.today,due,blocked,blocking,recurring,tagged,pri.
+rule.precedence.color=deleted,completed,active,keyword.,tag.,project.,overdue,scheduled,due.today,due,blocked,blocking,recurring,tagged,uda.
 
 # General decoration
 color.label=
@@ -33,13 +33,13 @@ color.alternate=on gray22
 color.header=color15 on gray8
 color.footnote=on gray18
 color.warning=color9
-color.error=red on white
-color.debug=color7 on color0
+color.error=white on red
+color.debug=rgb025
 
 # Task state
-color.completed=rgb353 on rgb000
-color.deleted=rgb533 on rgb000
-color.active=rgb420
+color.completed=
+color.deleted=
+color.active=rgb510
 color.recurring=
 color.scheduled=on rgb345
 color.until=
@@ -50,10 +50,9 @@ color.blocked=rgb003
 color.project.none=
 
 # Priority
-color.pri.H=gray0
-color.pri.M=gray5
-color.pri.L=gray10
-color.pri.none=gray5
+color.uda.priority.H=gray0
+color.uda.priority.M=gray5
+color.uda.priority.L=gray10
 
 # Tags
 color.tag.next=rgb420
@@ -66,14 +65,14 @@ color.due.today=on rgb353
 color.overdue=on rgb544
 
 # Report: burndown
-color.burndown.pending=on rgb141 
-color.burndown.started=on rgb440
+color.burndown.pending=on rgb411 
+color.burndown.started=on rgb550
 color.burndown.done=on rgb151
 
 # Report: history
-color.history.add=rgb005 on rgb440
-color.history.done=rgb020 on rgb343
-color.history.delete=rgb300 on rgb533
+color.history.add=color0 on rgb411
+color.history.done=color0 on rgb151
+color.history.delete=color0 on rgb550
 
 # Report: summary
 color.summary.bar=on rgb141
@@ -82,7 +81,7 @@ color.summary.background=on gray20
 # Command: calendar
 color.calendar.due=on rgb343
 color.calendar.due.today=on rgb353
-color.calendar.holiday=rgb420
+color.calendar.holiday=color0 on rgb530
 color.calendar.overdue=on rgb533
 color.calendar.today=rgb005
 color.calendar.weekend=on gray21

doc/rc/no-color.theme

@@ -24,7 +24,7 @@
 #
 ###############################################################################
 
-rule.precedence.color=deleted,completed,active,keyword.,tag.,uda.,project.,overdue,scheduled,due.today,due,blocked,blocking,recurring,tagged,pri.
+rule.precedence.color=deleted,completed,active,keyword.,tag.,project.,overdue,scheduled,due.today,due,blocked,blocking,recurring,tagged,uda.
 
 # This is a theme that disables all default colors
 # It can be used as template for other themes
@@ -53,10 +53,9 @@ color.blocking=
 color.project.none=
 
 # Priority
-color.pri.H=
-color.pri.L=
-color.pri.M=
-color.pri.none=
+color.uda.priority.H=
+color.uda.priority.L=
+color.uda.priority.M=
 
 # Tags
 color.tag.next=

doc/rc/solarized-dark-256.theme

@@ -41,7 +41,7 @@
 #
 # Taskwarrior solarized-dark-256 theme created by Stefan A. Keel.
 
-rule.precedence.color=deleted,completed,active,keyword.,tag.,uda.,project.,overdue,scheduled,due.today,due,blocked,blocking,recurring,tagged,pri.
+rule.precedence.color=deleted,completed,active,keyword.,tag.,project.,overdue,scheduled,due.today,due,blocked,blocking,recurring,tagged,uda.
 
 # General decoration
 color.label=
@@ -67,10 +67,9 @@ color.blocking=color15 on color10
 color.project.none=
 
 # Priority
-color.pri.H=bold blue #color14
-color.pri.M=bold yellow #color12
-color.pri.L=bold green #color11
-color.pri.none=
+color.uda.priority.H=bold blue #color14
+color.uda.priority.M=bold yellow #color12
+color.uda.priority.L=bold green #color11
 
 # Tags
 color.tag.next=
@@ -83,14 +82,14 @@ color.due.today=color1
 color.overdue=color5
 
 # Report: burndown
-color.burndown.done=color0 on color4
+color.burndown.done=color0 on color6
 color.burndown.pending=color0 on color1
-color.burndown.started=color0 on color9
+color.burndown.started=color0 on color3
 
 # Report: history
 color.history.add=color0 on color1
 color.history.delete=color0 on color3
-color.history.done=color0 on color10
+color.history.done=color0 on color6
 
 # Report: summary
 color.summary.background=on color0

doc/rc/solarized-light-256.theme

@@ -41,12 +41,12 @@
 #
 # Taskwarrior solarized-dark-256 theme created by Stefan A. Keel.
 
-rule.precedence.color=deleted,completed,active,keyword.,tag.,uda.,project.,overdue,scheduled,due.today,due,blocked,blocking,recurring,tagged,pri.
+rule.precedence.color=deleted,completed,active,keyword.,tag.,project.,overdue,scheduled,due.today,due,blocked,blocking,recurring,tagged,uda.
 
 # General decoration
 color.label=
 color.label.sort=
-color.alternate=on white #color7 (allows bold for alternate rows)
+color.alternate=on color7
 color.header=color2
 color.footnote=color2
 color.warning=
@@ -56,21 +56,20 @@ color.debug=color3
 # Task state
 color.completed=
 color.deleted=
-color.active=bold red #color9
+color.active=color9
 color.recurring=color4
 color.scheduled=
 color.until=
-color.blocked=on color14
-color.blocking=on color14
+color.blocked=color0 on color14
+color.blocking=color15 on color0
 
 # Project
 color.project.none=
 
 # Priority
-color.pri.H=bold black #color0
-color.pri.M=bold yellow #color11
-color.pri.L=bold cyan #color14
-color.pri.none=
+color.uda.priority.H=bold color0
+color.uda.priority.M=bold color11
+color.uda.priority.L=bold color14
 
 # Tags
 color.tag.next=
@@ -83,14 +82,14 @@ color.due.today=color1
 color.overdue=color5
 
 # Report: burndown
-color.burndown.done=color0 on color4
+color.burndown.done=color0 on color6
 color.burndown.pending=color0 on color1
-color.burndown.started=color0 on color9
+color.burndown.started=color0 on color3
 
 # Report: history
 color.history.add=color0 on color1
 color.history.delete=color0 on color3
-color.history.done=color14 on color0
+color.history.done=color0 on color6
 
 # Report: summary
 color.summary.background=on color7

scripts/add-ons/export-csv.pl

@@ -48,7 +48,7 @@ if ($command =~ /No matches/)
 print "'uuid','status','tags','entry','start','due','recur','end','project',",
       "'priority','fg','bg','description'\n";
 
-for my $task (split /,$/ms, qx{$command})
+for my $task (split "\n", qx{$command})
 {
   my $data = from_json ($task);
 

scripts/add-ons/export-html.pl

@@ -60,7 +60,7 @@ print "<html>\n",
       "      <tbody>\n";
 
 my $count = 0;
-for my $task (split /,$/ms, qx{$command})
+for my $task (split "\n", qx{$command})
 {
   ++$count;
   my $data = from_json ($task);

scripts/add-ons/export-ical.pl

@@ -49,7 +49,7 @@ print "BEGIN:VCALENDAR\n",
       "VERSION:2.0\n",
       "PRODID:=//GBF/taskwarrior 1.9.4//EN\n";
 
-for my $task (split /,$/ms, qx{$command})
+for my $task (split "\n", qx{$command})
 {
   my $data = from_json ($task);
 

scripts/add-ons/export-sql.py

@@ -29,7 +29,7 @@ export-sql.py -- Export the taskwarrior database as a series of SQL commands.
 
 Example usage::
 
-    $ ./export-sql.py | sqlite3 mytasks.db
+    $ PYTHONIOENCODING=UTF-8 ./export-sql.py | sqlite3 mytasks.db
     $ /usr/bin/sqlite3 mytasks.db "select * from annotations;"
 
 This script has only been tested with sqlite3, but in theory, it could be
@@ -120,8 +120,8 @@ def to_sql(task):
         values = template.format(**annot)
         return "INSERT INTO \"annotations\" VALUES(%s)" % values
 
-    template = "{uuid}, {description}, {entry}, {end}, " + \
-           "{priority}, {project}, {status}"
+    template = u"{uuid}, {description}, {entry}, {end}, " + \
+           u"{priority}, {project}, {status}"
 
     nullables = ['end', 'priority', 'project', 'status']
     defaults = dict([(key, None) for key in nullables])
@@ -140,10 +140,10 @@ def main():
     """ Return a list of SQL statements. """
 
     # Use the taskwarrior 2.0+ export command to filter and return JSON
-    command = "task rc.verbose=nothing rc.json.array=no export " + " ".join(sys.argv[1:])
+    command = "task rc.verbose=nothing rc.json.array=yes " + " ".join(sys.argv[1:]) + " export"
 
     # Load each task from json to a python dict
-    tasks = map(json.loads, commands.getoutput(command).split(",\n"))
+    tasks = json.loads(commands.getoutput(command))
 
     # Mangle datetime strings into python datetime objects
     tasks = map(parse_datetime, tasks)
@@ -162,7 +162,7 @@ if __name__ == '__main__':
     sql = table_definitions + ";\n".join(lines) + ';'
 
     # Print them out, decorated with sqlite3 trappings
-    print """
+    print u"""
 BEGIN TRANSACTION;
 {sql}
 COMMIT;""".format(sql=sql)

scripts/add-ons/export-tsv.pl

@@ -48,7 +48,7 @@ if ($command =~ /No matches/)
 print "uuid\tstatus\ttags\tentry\tstart\tdue\trecur\tend\tproject\t",
       "priority\tfg\tbg\tdescription\n";
 
-for my $task (split /,$/ms, qx{$command})
+for my $task (split "\n", qx{$command})
 {
   my $data = from_json ($task);
 

scripts/add-ons/export-xml.pl

@@ -46,7 +46,7 @@ if ($command =~ /No matches/)
 
 # Generate output.
 print "<tasks>\n";
-for my $task (split /,$/ms, qx{$command})
+for my $task (split "\n", qx{$command})
 {
   my $data = from_json ($task);
 

scripts/add-ons/export-yad.pl

@@ -48,7 +48,7 @@ if ($command =~ /No matches/)
 print "'uuid','status','tags','entry','start','due','recur','end','project',",
       "'priority','fg','bg','description'\n";
 
-for my $task (split /,$/ms, qx{$command})
+for my $task (split "\n", qx{$command})
 {
   my $data = from_json ($task);
 

scripts/bash/task.sh

@@ -50,7 +50,7 @@
 #
 ################################################################################
 #the following variable is substituted for by ../../test/bash_completion.t
-taskcommand='task rc.verbose:nothing'
+taskcommand='task rc.verbose:nothing rc.confirmation:no rc.hooks:off'
 
 _task_get_tags() {
     $taskcommand _tags
@@ -68,7 +68,13 @@ _task_offer_projects() {
     COMPREPLY=( $(compgen -W "$($taskcommand _projects)" -- ${cur/*:/}) )
 }
 
-_task() 
+_task_offer_contexts() {
+    COMPREPLY=( $(compgen -W "$($taskcommand _context) define delete list none show" -- $cur) )
+}
+
+_task_context_alias=$($taskcommand show | grep alias.*context | cut -d' ' -f1 | cut -d. -f2)
+
+_task()
 {
     local cur prev opts base
 
@@ -91,6 +97,10 @@ _task()
     opts="$commands_aliases $($taskcommand _columns)"
 
     case "${prev}" in
+        $_task_context_alias|cont|conte|contex|context)
+                _task_offer_contexts
+                return 0
+                ;;
         :)
             case "${prev2}" in
                 pri|prior|priori|priorit|priority)

scripts/fish/task.fish

@@ -15,6 +15,24 @@
 #  * Tags
 #  * Attribute names and modifiers
 #
+#
+# You can override some default options in your config.fish:
+#
+# # Tab-completion of task descriptions.
+# # Warning: This often creates a list of suggestions which spans several pages,
+# # and it usually pushes some of the commands and attributes to the end of the
+# # list.
+# set -g task_complete_task yes
+#
+# # Tab-completion of task IDs outside of the "depends" attribute.
+# # Warning: This often creates a list of suggestions which spans several pages,
+# # and it pushes all commands and attributes to the end of the list.
+# set -g task_complete_id yes
+#
+# # Attribute modifiers (DEPRECATED since 2.4.0)
+# set -g task_complete_attribute_modifiers yes
+#
+#
 # Copyright 2014 - 2015, Roman Inflianskas <infroma@gmail.com>
 #
 # Permission is hereby granted, free of charge, to any person obtaining a copy
@@ -150,8 +168,7 @@ function __fish.task.list.attr_value_by_name
       __fish.task.combos_simple $attr (__fish.task.list $attr)
     # case 'description' 'due' 'entry' 'end' 'start' 'project' 'recur' 'until' 'wait'
     case '*'
-      # BUG: remove in 2.4.0
-      if echo (commandline -ct) | grep -q '\.'
+      if [ "$task_complete_attribute_modifiers" = 'yes' ]; and echo (commandline -ct) | grep -q '\.'
         __fish.task.combos_with_mods $attr (__fish.task.list $attr)
       else
         __fish.task.combos_simple $attr (__fish.task.list $attr)
@@ -195,7 +212,7 @@ function __fish.task.list.id
   end
 end
 
-# BUG: remove in 2.4.0
+# Attribure modifiers (DEPRECATED since 2.4.0)
 function __fish.task.list.mod
   for mod in 'before' 'after' 'over' 'under' 'none' 'is' 'isnt' 'has' 'hasnt' 'startswith' 'endswith' 'word' 'noword'
     echo $mod
@@ -266,7 +283,7 @@ function __fish.task.combos_simple
   end
 end
 
-# BUG: remove in 2.4.0
+# Attribure modifiers (DEPRECATED since 2.4.0)
 function __fish.task.combos_with_mods
   __fish.task.combos_simple $argv
   for mod in (__fish.task.list.mod)
@@ -290,16 +307,10 @@ __fish.task.complete attr_value
 __fish.task.complete attr_name
 __fish.task.complete config
 
-# Uncomment the following line if you want tab-completion of task descriptions.
-# Warning: This often creates a list of suggestions which spans several pages,
-# and it usually pushes some of the commands and attributes to the end of the
-# list.
+if [ "$task_complete_task" = 'yes' ]
+  __fish.task.complete task
+end
 
-#__fish.task.complete task
-
-# Uncomment the following line if you want tab-completion of task IDs outside
-# of the "depends" attribute.  Warning: This often creates a list of
-# suggestions which spans several pages, and it pushes all commands and
-# attributes to the end of the list.
-
-#__fish.task.complete id with_description
+if [ "$task_complete_id" = 'yes' ]
+  __fish.task.complete id with_description
+end

scripts/hooks/on-add

@@ -1,20 +1,19 @@
-#!/bin/bash
+#!/bin/sh
 
-# The on-add event is triggered separately for each task added
+# The on-add event is triggered separately for each task added. This hook
+# script can accept/reject the addition. Processing will continue.
 
 # Input:
-# - line of JSON for the task added
-
-# Output:
-# - all emitted JSON lines are added/modified as tasks, if the exit code is
-#   zero, otherwise ignored.
-# - minimal new task:  {"description":"Buy milk"}
-# - to modify a task include complete JSON
-# - all emitted non-JSON lines are considered feedback messages if the exit
-#   code is zero, otherwise they are considered errors.
-
+# - Line of JSON for proposed new task.
 read new_task
 
+# Output:
+# - JSON, modified or unmodified.
+# - Optional feedback/error.
 echo $new_task
 echo 'on-add'
+
+# Status:
+# - 0:     JSON accepted, non-JSON is feedback.
+# - non-0: JSON ignored, non-JSON is error.
 exit 0

scripts/hooks/on-add.the

@@ -1,4 +1,4 @@
-#!/bin/bash
+#!/usr/bin/env bash
 
 read new_task
 

scripts/hooks/on-exit

@@ -1,20 +1,20 @@
-#!/bin/bash
+#!/bin/sh
 
-# The on-exit event is triggered once, after all processing is complete, i.e.
-# last
-
-# Input:
-# - read-only line of JSON for each task added/modified
+# The on-exit event is triggered once, after all processing is complete.
+# This hooks script has no effect on processing.
 
 # Output:
-# - any emitted JSON is ignored
-# - all emitted non-JSON lines are considered feedback messages if the exit
-#   code is zero, otherwise they are considered errors.
+# - Optional feedback/error.
 
-while read -t 1 modified_task
+n=0
+while read modified_task
 do
-  # Scan task
+  n=$(($n + 1))
 done
 
-echo 'on-exit'
+echo "on-exit: Counted $n added/modified tasks."
+
+# Status:
+# - 0:     Non-JSON is feedback.
+# - non-0: Non-JSON is error.
 exit 0

scripts/hooks/on-exit.shadow-file

@@ -1,4 +1,4 @@
-#!/bin/bash
+#!/usr/bin/env bash
 
 # This hook script replaces the shadow file feature, found in Taskwarrior
 # prior to version 2.4.0.

scripts/hooks/on-launch

@@ -1,18 +1,17 @@
-#!/bin/bash
+#!/bin/sh
 
 # The on-launch event is triggered once, after initialization, before any
-# processing occurs, i.e first
+# processing occurs. This hooks script has no effect on processing.
 
 # Input:
-# - none
+# - None
 
 # Output:
-# - all emitted JSON lines are added/modified as tasks, if the exit code is
-#   zero, otherwise ignored.
-# - minimal new task:  {"description":"Buy milk"}
-# - to modify a task include complete JSON
-# - all emitted non-JSON lines are considered feedback messages if the exit
-#   code is zero, otherwise they are considered errors.
-
+# - Optional feedback/error.
 echo 'on-launch'
+
+# Status:
+# - 0:     JSON ignored, non-JSON is feedback.
+# - non-0: JSON ignored, non-JSON is error.
 exit 0
+

scripts/hooks/on-modify

@@ -1,22 +1,21 @@
-#!/bin/bash
+#!/bin/sh
 
-# The on-modify event is triggered separately for each task added or modified
+# The on-modify event is triggered separately for each task modified. This hook
+# script can accept/reject the modification. Processing will continue.
 
 # Input:
 # - line of JSON for the original task
 # - line of JSON for the modified task, the diff being the modification
-
-# Output:
-# - all emitted JSON lines are added/modified as tasks, if the exit code is
-#   zero, otherwise ignored.
-# - minimal new task:  {"description":"Buy milk"}
-# - to modify a task include complete JSON
-# - all emitted non-JSON lines are considered feedback messages if the exit
-#   code is zero, otherwise they are considered errors.
-
 read original_task
 read modified_task
 
+# Output:
+# - JSON, modified or unmodified.
+# - Optional feedback/error.
 echo $modified_task
 echo 'on-modify'
+
+# Status:
+# - 0:     JSON accepted, non-JSON is feedback.
+# - non-0: JSON ignored, non-JSON is error.
 exit 0

scripts/vim/syntax/taskrc.vim

@@ -72,10 +72,9 @@ syn match taskrcGoodKey '^\s*\Vcolor.history.add='he=e-1
 syn match taskrcGoodKey '^\s*\Vcolor.history.delete='he=e-1
 syn match taskrcGoodKey '^\s*\Vcolor.history.done='he=e-1
 syn match taskrcGoodKey '^\s*\Vcolor.overdue='he=e-1
-syn match taskrcGoodKey '^\s*\Vcolor.pri.H='he=e-1
-syn match taskrcGoodKey '^\s*\Vcolor.pri.L='he=e-1
-syn match taskrcGoodKey '^\s*\Vcolor.pri.M='he=e-1
-syn match taskrcGoodKey '^\s*\Vcolor.pri.none='he=e-1
+syn match taskrcGoodKey '^\s*\Vcolor.uda.priority.H='he=e-1
+syn match taskrcGoodKey '^\s*\Vcolor.uda.priority.L='he=e-1
+syn match taskrcGoodKey '^\s*\Vcolor.uda.priority.M='he=e-1
 syn match taskrcGoodKey '^\s*\Vcolor.recurring='he=e-1
 syn match taskrcGoodKey '^\s*\Vcolor.scheduled='he=e-1
 syn match taskrcGoodKey '^\s*\Vcolor.summary.background='he=e-1
@@ -154,7 +153,9 @@ syn match taskrcGoodKey '^\s*\Vurgency.blocking.coefficient='he=e-1
 syn match taskrcGoodKey '^\s*\Vurgency.due.coefficient='he=e-1
 syn match taskrcGoodKey '^\s*\Vurgency.inherit.coefficient='he=e-1
 syn match taskrcGoodKey '^\s*\Vurgency.next.coefficient='he=e-1
-syn match taskrcGoodKey '^\s*\Vurgency.priority.coefficient='he=e-1
+syn match taskrcGoodKey '^\s*\Vurgency.uda.priority.H.coefficient='he=e-1
+syn match taskrcGoodKey '^\s*\Vurgency.uda.priority.M.coefficient='he=e-1
+syn match taskrcGoodKey '^\s*\Vurgency.uda.priority.L.coefficient='he=e-1
 syn match taskrcGoodKey '^\s*\Vurgency.project.coefficient='he=e-1
 syn match taskrcGoodKey '^\s*\Vurgency.scheduled.coefficient='he=e-1
 syn match taskrcGoodKey '^\s*\Vurgency.tags.coefficient='he=e-1

src/.gitignore

@@ -3,3 +3,4 @@ Makefile.in
 tw-*
 args
 calc
+lex

src/CLI.cpp

@@ -220,16 +220,14 @@ const std::string A::dump () const
 // Static method.
 void CLI::getOverride (int argc, const char** argv, std::string& home, File& rc)
 {
-  bool terminated = false;
   for (int i = 0; i < argc; ++i)
   {
     std::string raw = argv[i];
 
     if (raw == "--")
-      terminated = true;
+      return;
 
-    if (! terminated      &&
-        raw.length () > 3 &&
+    if (raw.length () > 3 &&
         raw.substr (0, 3) == "rc:")
     {
       rc = raw.substr (3);
@@ -351,7 +349,6 @@ void CLI::initialize (int argc, const char** argv)
   _terminated = false;
 
   _original_args.push_back (argv[0]);
-  bool terminated = false;
   for (int i = 1; i < argc; ++i)
   {
     if (isTerminator (argv[i]))
@@ -374,6 +371,63 @@ void CLI::add (const std::string& arg)
   analyze ();
 }
 
+////////////////////////////////////////////////////////////////////////////////
+void CLI::addContextFilter ()
+{
+  // Detect if any context is set, and bail out if not
+  std::string contextName = context.config.get ("context");
+  if (contextName == "")
+  {
+    context.debug ("No context applied.");
+    return;
+  }
+
+  // Detect if UUID or ID is set, and bail out
+  if (_args.size ())
+  {
+    std::vector <A>::const_iterator a;
+    for (a = _args.begin (); a != _args.end (); ++a)
+    {
+      // TODO This looks wrong.
+      if (a->hasTag ("FILTER") &&
+          a->hasTag ("ATTRIBUTE") &&
+          ! a->hasTag ("TERMINATED") &&
+          ! a->hasTag ("WORD") &&
+          (a->attribute ("raw") == "id" || a->attribute ("raw") == "uuid"))
+      {
+        context.debug (format ("UUID/ID lexeme found '{1}', not applying context.", a->attribute ("raw")));
+        return;
+      }
+    }
+  }
+
+  // Apply context
+  context.debug ("Applying context: " + contextName);
+  std::string contextFilter = context.config.get ("context." + contextName);
+
+  if (contextFilter == "")
+    context.debug ("Context '" + contextName + "' not defined.");
+  else
+  {
+    addRawFilter ("( " + contextFilter + " )");
+    if (context.verbose ("context"))
+      context.footnote (format ("Context '{1}' set. Use 'task context none' to remove.", contextName));
+  }
+}
+
+////////////////////////////////////////////////////////////////////////////////
+// Process raw string into parsed filter.
+void CLI::addRawFilter (const std::string& arg)
+{
+  std::string lexeme;
+  Lexer::Type type;
+  Lexer lex (arg);
+  lex.ambiguity (false);
+
+  while (lex.token (lexeme, type))
+    add (lexeme);
+}
+
 ////////////////////////////////////////////////////////////////////////////////
 // Intended to be called after ::initialize() and ::add(), to perform the final
 // analysis. Analysis is also performed directly after the above, because there
@@ -388,7 +442,7 @@ void CLI::analyze (bool parse /* = true */, bool strict /* = false */)
   // For propagation.
   _strict = strict;
 
-  for (int i = 0; i < _original_args.size (); ++i)
+  for (unsigned int i = 0; i < _original_args.size (); ++i)
   {
     std::string raw = _original_args[i];
     A a ("arg", raw);
@@ -474,8 +528,14 @@ void CLI::applyOverrides ()
 
 ////////////////////////////////////////////////////////////////////////////////
 // Extract all the FILTER-tagged items.
-const std::string CLI::getFilter ()
+const std::string CLI::getFilter (bool applyContext /* = true */)
 {
+  // Handle context setting
+  // Commands that don't want to respect current context should leverage
+  // the applyContext argument
+  if (applyContext)
+    addContextFilter ();
+
   std::string filter = "";
   if (_args.size ())
   {
@@ -500,6 +560,7 @@ const std::string CLI::getFilter ()
       filter = "( " + filter + " )";
   }
 
+  context.debug("Derived filter: '" + filter + "'");
   return filter;
 }
 
@@ -627,7 +688,7 @@ const std::string CLI::dump (const std::string& title /* = "CLI Parser" */) cons
 //       be lexed from those that need to be left alone.
 //
 // Either the arg is appended to _original_args intact, or the lexemes are.
-void CLI::addArg (const std::string& arg)
+void CLI::addArg (const std::string& arg, Lexer::Type type /* = Lexer::Type::word */)
 {
   std::string raw = trim (arg);
 
@@ -653,8 +714,8 @@ void CLI::addArg (const std::string& arg)
     _original_args.push_back (raw);
   }
 
-  // The argument may require lexing.  Lex anyway, and analyze before comitting
-  // that.
+  // The argument may require lexing. Lex anyway, and analyze before comitting
+  // to that.
   else
   {
     // Lex each remaining argument.  The apply a series of disqualifying tests
@@ -665,7 +726,7 @@ void CLI::addArg (const std::string& arg)
     Lexer lex (raw);
     lex.ambiguity (false);
 
-    std::vector <std::pair <std::string, Lexer::Type> > lexemes;
+    std::vector <std::pair <std::string, Lexer::Type>> lexemes;
     while (lex.token (lexeme, type))
       lexemes.push_back (std::pair <std::string, Lexer::Type> (lexeme, type));
 
@@ -681,7 +742,7 @@ void CLI::addArg (const std::string& arg)
     {
       // How often have I said to you that when you have eliminated the
       // impossible, whatever remains, however improbable, must be the truth?
-      std::vector <std::pair <std::string, Lexer::Type> >::iterator l;
+      std::vector <std::pair <std::string, Lexer::Type>>::iterator l;
       for (l = lexemes.begin (); l != lexemes.end (); ++l)
         _original_args.push_back (l->first);
     }
@@ -713,9 +774,7 @@ void CLI::aliasExpansion ()
       {
         if (_aliases.find (raw) != _aliases.end ())
         {
-          std::vector <std::string> lexed;
-          Lexer::token_split (lexed, _aliases[raw]);
-
+          std::vector <std::string> lexed = Lexer::split (_aliases[raw]);
           std::vector <std::string>::iterator l;
           for (l = lexed.begin (); l != lexed.end (); ++l)
           {
@@ -851,7 +910,7 @@ void CLI::categorize ()
       a->tag ("MODIFICATION");
 
       // If the argument contains a space, it was quoted.  Record that.
-      if (! noSpaces (raw))
+      if (! Lexer::isOneWord (raw))
         a->tag ("QUOTED");
 
       changes = true;
@@ -861,7 +920,7 @@ void CLI::categorize ()
       a->tag ("FILTER");
 
       // If the argument contains a space, it was quoted.  Record that.
-      if (! noSpaces (raw))
+      if (! Lexer::isOneWord (raw))
         a->tag ("QUOTED");
 
       changes = true;
@@ -1175,7 +1234,7 @@ void CLI::desugarFilterAttributeModifiers ()
                 }
                 else if (modifier == "isnt" || modifier == "not")
                 {
-                  op.attribute ("raw", "!=");
+                  op.attribute ("raw", "!==");
                   rhs.attribute ("raw", "'" + value + "'");
                   rhs.tag ("LITERAL");
                 }
@@ -1324,14 +1383,12 @@ void CLI::findIDs ()
   {
     if (a->hasTag ("FILTER"))
     {
-      bool found = false;
-
       // IDs have a limited character set.
       std::string raw = a->attribute ("raw");
       if (raw.find_first_not_of ("0123456789,-") == std::string::npos)
       {
         // Container for min/max ID ranges.
-        std::vector <std::pair <int, int> > ranges;
+        std::vector <std::pair <int, int>> ranges;
 
         // Split the ID list into elements.
         std::vector <std::string> elements;
@@ -1347,7 +1404,7 @@ void CLI::findIDs ()
 
           if (terms.size () == 1)
           {
-            if (! digitsOnly (terms[0]))
+            if (! Lexer::isAllDigits (terms[0]))
             {
               is_an_id = false;
               break;
@@ -1368,8 +1425,8 @@ void CLI::findIDs ()
           }
           else if (terms.size () == 2)
           {
-            if (! digitsOnly (terms[0]) ||
-                ! digitsOnly (terms[1]))
+            if (! Lexer::isAllDigits (terms[0]) ||
+                ! Lexer::isAllDigits (terms[1]))
             {
               is_an_id = false;
               break;
@@ -1410,7 +1467,7 @@ void CLI::findIDs ()
           a->tag ("ID");
 
           // Save the ranges.
-          std::vector <std::pair <int, int> >::iterator r;
+          std::vector <std::pair <int, int>>::iterator r;
           for (r = ranges.begin (); r != ranges.end (); ++r)
             _id_ranges.push_back (*r);
         }
@@ -1519,7 +1576,7 @@ void CLI::insertIDExpr ()
         reconstructed.push_back (openParen);
 
         // Add all ID ranges.
-        std::vector <std::pair <int, int> >::iterator r;
+        std::vector <std::pair <int, int>>::iterator r;
         for (r = _id_ranges.begin (); r != _id_ranges.end (); ++r)
         {
           if (r != _id_ranges.begin ())
@@ -1814,8 +1871,7 @@ void CLI::injectDefaults ()
       if (defaultCommand != "")
       {
         // Split the defaultCommand into separate args.
-        std::vector <std::string> tokens;
-        Lexer::token_split (tokens, defaultCommand);
+        std::vector <std::string> tokens = Lexer::split (defaultCommand);
 
         // Modify _args to be:   <args0> [<def0> ...] <args1> [...]
         std::vector <A> reconstructed;
@@ -2209,7 +2265,7 @@ bool CLI::isIDSequence (const std::string& raw) const
 ////////////////////////////////////////////////////////////////////////////////
 bool CLI::isID (const std::string& raw) const
 {
-  return digitsOnly (raw);
+  return Lexer::isAllDigits (raw);
 }
 
 ////////////////////////////////////////////////////////////////////////////////
@@ -2303,11 +2359,11 @@ bool CLI::isName (const std::string& raw) const
 {
   if (raw != "")
   {
-    for (int i = 0; i < raw.length (); ++i)
+    for (unsigned int i = 0; i < raw.length (); ++i)
     {
-      if (i == 0 && ! Lexer::is_ident_start (raw[i]))
+      if (i == 0 && ! Lexer::isIdentifierStart (raw[i]))
         return false;
-      else if (! Lexer::is_ident (raw[i]))
+      else if (! Lexer::isIdentifierNext (raw[i]))
         return false;
     }
 
@@ -2319,19 +2375,19 @@ bool CLI::isName (const std::string& raw) const
 
 ////////////////////////////////////////////////////////////////////////////////
 bool CLI::disqualifyInsufficientTerms (
-  const std::vector <std::pair <std::string, Lexer::Type> >& lexemes) const
+  const std::vector <std::pair <std::string, Lexer::Type>>& lexemes) const
 {
   return lexemes.size () < 3 ? true : false;
 }
 
 ////////////////////////////////////////////////////////////////////////////////
 bool CLI::disqualifyNoOps (
-  const std::vector <std::pair <std::string, Lexer::Type> >& lexemes) const
+  const std::vector <std::pair <std::string, Lexer::Type>>& lexemes) const
 {
   bool foundOP = false;
-  std::vector <std::pair <std::string, Lexer::Type> >::const_iterator l;
+  std::vector <std::pair <std::string, Lexer::Type>>::const_iterator l;
   for (l = lexemes.begin (); l != lexemes.end (); ++l)
-    if (l->second == Lexer::typeOperator)
+    if (l->second == Lexer::Type::op)
       foundOP = true;
 
   return ! foundOP;
@@ -2339,16 +2395,16 @@ bool CLI::disqualifyNoOps (
 
 ////////////////////////////////////////////////////////////////////////////////
 bool CLI::disqualifyOnlyParenOps (
-  const std::vector <std::pair <std::string, Lexer::Type> >& lexemes) const
+  const std::vector <std::pair <std::string, Lexer::Type>>& lexemes) const
 {
   int opCount      = 0;
   int opSugarCount = 0;
   int opParenCount = 0;
 
-  std::vector <std::pair <std::string, Lexer::Type> >::const_iterator l;
+  std::vector <std::pair <std::string, Lexer::Type>>::const_iterator l;
   for (l = lexemes.begin (); l != lexemes.end (); ++l)
   {
-    if (l->second == Lexer::typeOperator)
+    if (l->second == Lexer::Type::op)
     {
       ++opCount;
 
@@ -2375,7 +2431,7 @@ bool CLI::disqualifyOnlyParenOps (
 // as there are no operators in between, which includes syntactic sugar that
 // hides operators.
 bool CLI::disqualifyFirstLastBinary (
-  const std::vector <std::pair <std::string, Lexer::Type> >& lexemes) const
+  const std::vector <std::pair <std::string, Lexer::Type>>& lexemes) const
 {
   bool firstBinary = false;
   bool lastBinary  = false;
@@ -2394,7 +2450,7 @@ bool CLI::disqualifyFirstLastBinary (
 ////////////////////////////////////////////////////////////////////////////////
 // Disqualify terms when there operators hidden by syntactic sugar.
 bool CLI::disqualifySugarFree (
-  const std::vector <std::pair <std::string, Lexer::Type> >& lexemes) const
+  const std::vector <std::pair <std::string, Lexer::Type>>& lexemes) const
 {
   bool sugared = true;
   for (unsigned int i = 1; i < lexemes.size () - 1; ++i)

src/CLI.h

@@ -77,9 +77,11 @@ public:
   void entity (const std::string&, const std::string&);
   void initialize (int, const char**);
   void add (const std::string&);
+  void addContextFilter ();
+  void addRawFilter (const std::string& arg);
   void analyze (bool parse = true, bool strict = false);
   void applyOverrides ();
-  const std::string getFilter ();
+  const std::string getFilter (bool applyContext = true);
   const std::vector <std::string> getWords ();
   bool canonicalize (std::string&, const std::string&, const std::string&) const;
   std::string getBinary () const;
@@ -88,7 +90,7 @@ public:
   const std::string dump (const std::string& title = "CLI Parser") const;
 
 private:
-  void addArg (const std::string&);
+  void addArg (const std::string&, Lexer::Type type = Lexer::Type::word);
   void aliasExpansion ();
   void findOverrides ();
   void categorize ();
@@ -126,11 +128,11 @@ private:
   bool isOperator       (const std::string&) const;
   bool isName           (const std::string&) const;
 
-  bool disqualifyInsufficientTerms (const std::vector <std::pair <std::string, Lexer::Type> >&) const;
-  bool disqualifyNoOps             (const std::vector <std::pair <std::string, Lexer::Type> >&) const;
-  bool disqualifyOnlyParenOps      (const std::vector <std::pair <std::string, Lexer::Type> >&) const;
-  bool disqualifyFirstLastBinary   (const std::vector <std::pair <std::string, Lexer::Type> >&) const;
-  bool disqualifySugarFree         (const std::vector <std::pair <std::string, Lexer::Type> >&) const;
+  bool disqualifyInsufficientTerms (const std::vector <std::pair <std::string, Lexer::Type>>&) const;
+  bool disqualifyNoOps             (const std::vector <std::pair <std::string, Lexer::Type>>&) const;
+  bool disqualifyOnlyParenOps      (const std::vector <std::pair <std::string, Lexer::Type>>&) const;
+  bool disqualifyFirstLastBinary   (const std::vector <std::pair <std::string, Lexer::Type>>&) const;
+  bool disqualifySugarFree         (const std::vector <std::pair <std::string, Lexer::Type>>&) const;
 
 public:
   std::multimap <std::string, std::string> _entities;
@@ -138,7 +140,7 @@ public:
   std::vector <std::string>                _original_args;
   std::vector <A>                          _args;
 
-  std::vector <std::pair <int, int> >      _id_ranges;
+  std::vector <std::pair <int, int>>       _id_ranges;
   std::vector <std::string>                _uuid_list;
   bool                                     _strict;
   bool                                     _terminated;

src/CMakeLists.txt

@@ -48,16 +48,19 @@ set (task_SRCS CLI.cpp CLI.h
 add_library (task STATIC ${task_SRCS})
 add_executable (task_executable main.cpp)
 add_executable (calc_executable calc.cpp)
+add_executable (lex_executable lex.cpp)
 
 # Yes, 'task' is included twice, otherwise linking fails on assorted OSes.
 target_link_libraries (task_executable task commands columns task ${TASK_LIBRARIES})
 target_link_libraries (calc_executable task commands columns task ${TASK_LIBRARIES})
+target_link_libraries (lex_executable task commands columns task ${TASK_LIBRARIES})
 
 set_property (TARGET task_executable PROPERTY OUTPUT_NAME "task")
 
 install (TARGETS task_executable DESTINATION ${TASK_BINDIR})
 
 set_property (TARGET calc_executable PROPERTY OUTPUT_NAME "calc")
+set_property (TARGET lex_executable PROPERTY OUTPUT_NAME "lex")
 
 #SET(CMAKE_BUILD_TYPE gcov)
 #SET(CMAKE_CXX_FLAGS_GCOV "--coverage")

src/Config.cpp

@@ -82,6 +82,7 @@ std::string Config::_defaults =
   "#                                              # Comma-separated list.  May contain any subset of:\n"
   "#verbose=blank,header,footnote,label,new-id,new-uuid,affected,edit,special,project,sync,filter\n"
   "confirmation=yes                               # Confirmation on delete, big changes\n"
+  "recurrence=yes                                 # Enable recurrence\n"
   "recurrence.confirmation=prompt                 # Confirmation for propagating changes among recurring tasks (yes/no/prompt)\n"
   "allow.empty.filter=yes                         # An empty filter gets a warning and requires confirmation\n"
   "indent.annotation=2                            # Indent spaces for annotations\n"
@@ -116,6 +117,7 @@ std::string Config::_defaults =
   "dateformat.info=Y-M-D H:N:S                    # Preferred display date format for information\n"
   "dateformat.report=                             # Preferred display date format for reports\n"
   "dateformat.annotation=                         # Preferred display date format for annotations\n"
+  "date.iso=yes                                   # Enable ISO date support\n"
   "weekstart="
              STRING_DATE_SUNDAY_LONG
                   "                               # Sunday or Monday only\n"
@@ -145,7 +147,6 @@ std::string Config::_defaults =
   "urgency.next.coefficient=15.0                  # Urgency coefficient for 'next' special tag\n"
   "urgency.due.coefficient=12.0                   # Urgency coefficient for due dates\n"
   "urgency.blocking.coefficient=8.0               # Urgency coefficient for blocking tasks\n"
-  "urgency.priority.coefficient=6.0               # Urgency coefficient for priorities\n"
   "urgency.active.coefficient=4.0                 # Urgency coefficient for active tasks\n"
   "urgency.scheduled.coefficient=5.0              # Urgency coefficient for scheduled tasks\n"
   "urgency.age.coefficient=2.0                    # Urgency coefficient for age\n"
@@ -163,38 +164,37 @@ std::string Config::_defaults =
   "\n"
   "# Color controls.\n"
   "color=on                                       # Enable color\n"
-#ifdef LINUX
+#if defined(LINUX) || defined(DARWIN)
   "\n"
-  "rule.precedence.color=deleted,completed,active,keyword.,tag.,uda.,project.,overdue,scheduled,due.today,due,blocked,blocking,recurring,tagged,pri.\n"
+  "rule.precedence.color=deleted,completed,active,keyword.,tag.,project.,overdue,scheduled,due.today,due,blocked,blocking,recurring,tagged,uda.\n"
   "\n"
   "# General decoration\n"
   "color.label=\n"
   "color.label.sort=\n"
-  "color.alternate=on gray1\n"
+  "color.alternate=on gray2\n"
   "color.header=color3\n"
   "color.footnote=color3\n"
   "color.warning=bold red\n"
-  "color.error=color3\n"
-  "color.debug=color3\n"
+  "color.error=white on red\n"
+  "color.debug=color4\n"
   "\n"
   "# Task state\n"
-  "color.completed=rgb010 on white\n"
-  "color.deleted=rgb100 on white\n"
+  "color.completed=\n"
+  "color.deleted=\n"
   "color.active=rgb555 on rgb410\n"
   "color.recurring=rgb013\n"
   "color.scheduled=on rgb001\n"
   "color.until=\n"
   "color.blocked=white on color8\n"
-  "color.blocking=white on color15\n"
+  "color.blocking=black on color15\n"
   "\n"
   "# Project\n"
   "color.project.none=\n"
   "\n"
-  "# Priority\n"
-  "color.pri.H=color255\n"
-  "color.pri.L=color245\n"
-  "color.pri.M=color250\n"
-  "color.pri.none=\n"
+  "# Priority UDA\n"
+  "color.uda.priority.H=color255\n"
+  "color.uda.priority.L=color245\n"
+  "color.uda.priority.M=color250\n"
   "\n"
   "# Tags\n"
   "color.tag.next=rgb440\n"
@@ -240,7 +240,7 @@ std::string Config::_defaults =
   "\n"
 #else
   "\n"
-  "rule.precedence.color=deleted,completed,active,keyword.,tag.,uda.,project.,overdue,scheduled,due.today,due,blocked,blocking,recurring,tagged,pri.\n"
+  "rule.precedence.color=deleted,completed,active,keyword.,tag.,project.,overdue,scheduled,due.today,due,blocked,blocking,recurring,tagged,uda.\n"
   "\n"
   "# General decoration\n"
   "color.label=\n"
@@ -249,15 +249,15 @@ std::string Config::_defaults =
   "color.header=yellow\n"
   "color.footnote=yellow\n"
   "color.warning=bold red\n"
-  "color.error=yellow\n"
-  "color.debug=yellow\n"
+  "color.error=white on red\n"
+  "color.debug=blue\n"
   "\n"
   "# Task state\n"
-  "color.completed=green on white\n"
-  "color.deleted=red on white\n"
+  "color.completed=\n"
+  "color.deleted=\n"
   "color.active=black on bright green\n"
   "color.recurring=magenta\n"
-  "color.scheduled=on green\n"
+  "color.scheduled=white on green\n"
   "color.until=\n"
   "color.blocked=black on white\n"
   "color.blocking=black on bright white\n"
@@ -265,11 +265,10 @@ std::string Config::_defaults =
   "# Project\n"
   "color.project.none=\n"
   "\n"
-  "# Priority\n"
-  "color.pri.H=bold white\n"
-  "color.pri.M=white\n"
-  "color.pri.L=\n"
-  "color.pri.none=\n"
+  "# Priority UDA\n"
+  "color.uda.priority.H=bold white\n"
+  "color.uda.priority.M=white\n"
+  "color.uda.priority.L=\n"
   "\n"
   "# Tags\n"
   "color.tag.next=bold yellow\n"
@@ -317,22 +316,33 @@ std::string Config::_defaults =
   "color.undo.before=red\n"
   "\n"
 #endif
+  "# UDA priority\n"
+  "uda.priority.type=string                       # UDA priority is a string type\n"
+  "uda.priority.label=Priority                    # UDA priority has a display label'\n"
+  "uda.priority.values=H,M,L,                     # UDA priority values are 'H', 'M', 'L' or ''\n"
+  "                                               # UDA priority sorting is 'H' > 'M' > 'L' > '' (highest to lowest)\n"
+  "#uda.priority.default=M                        # UDA priority default value of 'M'\n"
+  "urgency.uda.priority.H.coefficient=6.0         # UDA priority coefficient for value 'H'\n"
+  "urgency.uda.priority.M.coefficient=3.9         # UDA priority coefficient for value 'M'\n"
+  "urgency.uda.priority.L.coefficient=1.8         # UDA priority coefficient for value 'L'\n"
+  "\n"
   "# Here is the rule precedence order, highest to lowest.\n"
   "# Note that these are just the color rule names, without the leading 'color.'\n"
   "#      and any trailing '.value'.\n"
-  "rule.precedence.color=deleted,completed,active,keyword.,tag.,uda.,project.,overdue,scheduled,due.today,due,blocked,blocking,recurring,tagged,pri.\n"
+  "rule.precedence.color=deleted,completed,active,keyword.,tag.,project.,overdue,scheduled,due.today,due,blocked,blocking,recurring,tagged,uda.\n"
   "\n"
   "#default.project=foo                           # Default project for 'add' command\n"
-  "#default.priority=M                            # Default priority for 'add' command\n"
   "#default.due=eom                               # Default due date for 'add' command\n"
   "default.command=next                           # When no arguments are specified\n"
   "\n"
   "_forcecolor=no                                 # Forces color to be on, even for non TTY output\n"
   "complete.all.tags=no                           # Include old tag names in '_ags' command\n"
   "list.all.projects=no                           # Include old project names in 'projects' command\n"
+  "summary.all.projects=no                        # Include old project names in 'summary' command\n"
   "list.all.tags=no                               # Include old tag names in 'tags' command\n"
   "print.empty.columns=no                         # Print columns which have no data for any task\n"
   "debug=no                                       # Display diagnostics\n"
+  "obfuscate=no                                   # Obfuscate data for error reporting\n"
   "fontunderline=yes                              # Uses underlines rather than -------\n"
   "shell.prompt=task>                             # Prompt used by the shell command\n"
   "\n"
@@ -358,19 +368,19 @@ std::string Config::_defaults =
   "\n"
   "report.long.description=All details of tasks\n"
   "report.long.labels=ID,A,Created,Mod,Deps,P,Project,Tags,Recur,Wait,Sched,Due,Until,Description\n"
-  "report.long.columns=id,start.active,entry,modified.age,depends,priority,project,tags,recur,wait.age,scheduled,due,until,description\n"
+  "report.long.columns=id,start.active,entry,modified.age,depends,priority,project,tags,recur,wait.remaining,scheduled,due,until,description\n"
   "report.long.filter=status:pending\n"
   "report.long.sort=modified-\n"
   "\n"
   "report.list.description=Most details of tasks\n"
   "report.list.labels=ID,Active,Age,D,P,Project,Tags,R,Sch,Due,Until,Description,Urg\n"
-  "report.list.columns=id,start.age,entry.age,depends.indicator,priority,project,tags,recur.indicator,scheduled.countdown,due,until.age,description.count,urgency\n"
+  "report.list.columns=id,start.age,entry.age,depends.indicator,priority,project,tags,recur.indicator,scheduled.countdown,due,until.remaining,description.count,urgency\n"
   "report.list.filter=status:pending\n"
-  "report.list.sort=start-,due+,project+/,urgency-\n"
+  "report.list.sort=start-,due+,project+,urgency-\n"
   "\n"
   "report.ls.description=Few details of tasks\n"
   "report.ls.labels=ID,A,D,Project,Tags,R,Wait,S,Due,Until,Description\n"
-  "report.ls.columns=id,start.active,depends.indicator,project,tags,recur.indicator,wait.age,scheduled.countdown,due.countdown,until.countdown,description.count\n"
+  "report.ls.columns=id,start.active,depends.indicator,project,tags,recur.indicator,wait.remaining,scheduled.countdown,due.countdown,until.countdown,description.count\n"
   "report.ls.filter=status:pending\n"
   "report.ls.sort=start-,description+\n"
   "\n"
@@ -382,13 +392,13 @@ std::string Config::_defaults =
   "\n"
   "report.newest.description=Newest tasks\n"
   "report.newest.labels=ID,Active,Created,Age,Mod,D,P,Project,Tags,R,Wait,Sch,Due,Until,Description\n"
-  "report.newest.columns=id,start.age,entry,entry.age,modified.age,depends.indicator,priority,project,tags,recur.indicator,wait.age,scheduled.countdown,due,until.age,description\n"
-  "report.newest.filter=(status:pending or status:waiting) and recur.none:\n"
+  "report.newest.columns=id,start.age,entry,entry.age,modified.age,depends.indicator,priority,project,tags,recur.indicator,wait.remaining,scheduled.countdown,due,until.age,description\n"
+  "report.newest.filter=(status:pending or status:waiting)\n"
   "report.newest.sort=entry-\n"
   "\n"
   "report.oldest.description=Oldest tasks\n"
   "report.oldest.labels=ID,Active,Created,Age,Mod,D,P,Project,Tags,R,Wait,Sch,Due,Until,Description\n"
-  "report.oldest.columns=id,start.age,entry,entry.age,modified.age,depends.indicator,priority,project,tags,recur.indicator,wait.age,scheduled.countdown,due,until.age,description\n"
+  "report.oldest.columns=id,start.age,entry,entry.age,modified.age,depends.indicator,priority,project,tags,recur.indicator,wait.remaining,scheduled.countdown,due,until.age,description\n"
   "report.oldest.filter=(status:pending or status:waiting)\n"
   "report.oldest.sort=entry+\n"
   "\n"
@@ -400,7 +410,7 @@ std::string Config::_defaults =
   "\n"
   "report.active.description=Active tasks\n"
   "report.active.labels=ID,Started,Active,Age,D,P,Project,Tags,Recur,W,Sch,Due,Until,Description\n"
-  "report.active.columns=id,start,start.age,entry.age,depends.indicator,priority,project,tags,recur,wait.indicator,scheduled.age,due,until,description\n"
+  "report.active.columns=id,start,start.age,entry.age,depends.indicator,priority,project,tags,recur,wait.indicator,scheduled.remaining,due,until,description\n"
   "report.active.filter=status:pending and +ACTIVE\n"
   "report.active.sort=project+,start+\n"
   "\n"
@@ -412,30 +422,30 @@ std::string Config::_defaults =
   "\n"
   "report.recurring.description=Recurring Tasks\n"
   "report.recurring.labels=ID,Active,Age,D,P,Project,Tags,Recur,Sch,Due,Until,Description,Urg\n"
-  "report.recurring.columns=id,start.age,entry.age,depends.indicator,priority,project,tags,recur,scheduled.countdown,due,until.age,description,urgency\n"
+  "report.recurring.columns=id,start.age,entry.age,depends.indicator,priority,project,tags,recur,scheduled.countdown,due,until.remaining,description,urgency\n"
   "report.recurring.filter=(status:pending or status:waiting) and (+PARENT or +CHILD)\n"
   "report.recurring.sort=due+,urgency-,entry+\n"
   "\n"
   "report.waiting.description=Waiting (hidden) tasks\n"
   "report.waiting.labels=ID,A,Age,D,P,Project,Tags,R,Wait,for,Sched,Due,Until,Description\n"
-  "report.waiting.columns=id,start.active,entry.age,depends.indicator,priority,project,tags,recur.indicator,wait,wait.age,scheduled,due,until,description\n"
+  "report.waiting.columns=id,start.active,entry.age,depends.indicator,priority,project,tags,recur.indicator,wait,wait.remaining,scheduled,due,until,description\n"
   "report.waiting.filter=+WAITING\n"
   "report.waiting.sort=due+,wait+,entry+\n"
   "\n"
   "report.all.description=All tasks\n"
   "report.all.labels=ID,St,UUID,A,Age,Done,D,P,Project,Tags,R,Wait,Sch,Due,Until,Description\n"
-  "report.all.columns=id,status.short,uuid.short,start.active,entry.age,end.age,depends.indicator,priority,project.parent,tags.count,recur.indicator,wait.age,scheduled.age,due,until.age,description\n"
+  "report.all.columns=id,status.short,uuid.short,start.active,entry.age,end.age,depends.indicator,priority,project.parent,tags.count,recur.indicator,wait.remaining,scheduled.remaining,due,until.remaining,description\n"
   "report.all.sort=entry-\n"
   "\n"
   "report.next.description=Most urgent tasks\n"
   "report.next.labels=ID,Active,Age,Deps,P,Project,Tag,Recur,S,Due,Until,Description,Urg\n"
-  "report.next.columns=id,start.age,entry.age,depends,priority,project,tags,recur,scheduled.countdown,due.age,until.age,description,urgency\n"
+  "report.next.columns=id,start.age,entry.age,depends,priority,project,tags,recur,scheduled.countdown,due.remaining,until.remaining,description,urgency\n"
   "report.next.filter=status:pending limit:page\n"
-  "report.next.sort=start-,urgency-\n"
+  "report.next.sort=urgency-\n"
   "\n"
   "report.ready.description=Most urgent actionable tasks\n"
   "report.ready.labels=ID,Active,Age,D,P,Project,Tags,R,S,Due,Until,Description,Urg\n"
-  "report.ready.columns=id,start.age,entry.age,depends.indicator,priority,project,tags,recur.indicator,scheduled.countdown,due.countdown,until.age,description,urgency\n"
+  "report.ready.columns=id,start.age,entry.age,depends.indicator,priority,project,tags,recur.indicator,scheduled.countdown,due.countdown,until.remaining,description,urgency\n"
   "report.ready.filter=+READY\n"
   "report.ready.sort=start-,urgency-\n"
   "\n"
@@ -453,7 +463,7 @@ std::string Config::_defaults =
   "\n"
   "report.blocking.description=Blocking tasks\n"
   "report.blocking.labels=ID,UUID,A,Deps,Project,Tags,R,W,Sch,Due,Until,Description,Urg\n"
-  "report.blocking.columns=id,uuid.short,start.active,depends,project,tags,recur,wait.indicator,scheduled.age,due.age,until.age,description.count,urgency\n"
+  "report.blocking.columns=id,uuid.short,start.active,depends,project,tags,recur,wait.indicator,scheduled.remaining,due.remaining,until.remaining,description.count,urgency\n"
   "report.blocking.sort=urgency-,due+,entry+\n"
   "report.blocking.filter= status:pending +BLOCKING\n"
   "\n";

src/Context.cpp

@@ -65,33 +65,6 @@ static const char* modifierNames[] =
 
 #define NUM_MODIFIER_NAMES       (sizeof (modifierNames) / sizeof (modifierNames[0]))
 
-static const char* attributeNames[] =
-{
-  "depends",
-  "description",
-  "due",
-  "end",
-  "entry",
-  "id",
-  "imask",
-  "mask",
-  "modified",
-  "parent",
-  "priority",
-  "project",
-  "recur",
-  "scheduled",
-  "start",
-  "status",
-  "tags",
-  "until",
-  "urgency",
-  "uuid",
-  "wait"
-};
-
-#define NUM_ATTRIBUTE_NAMES       (sizeof (attributeNames) / sizeof (attributeNames[0]))
-
 ////////////////////////////////////////////////////////////////////////////////
 Context::Context ()
 : rc_file ("~/.taskrc")
@@ -101,6 +74,7 @@ Context::Context ()
 , dom ()
 , determine_color_use (true)
 , use_color (true)
+, run_gc (true)
 , verbosity_legacy (false)
 , terminal_width (0)
 , terminal_height (0)
@@ -275,7 +249,7 @@ int Context::initialize (int argc, const char** argv)
 
     ////////////////////////////////////////////////////////////////////////////
     //
-    // [8] Run on.launch hooks.
+    // [8] Initialize hooks.
     //
     ////////////////////////////////////////////////////////////////////////////
 
@@ -478,7 +452,14 @@ int Context::dispatch (std::string &out)
     // GC is invoked prior to running any command that displays task IDs, if
     // possible.
     if (c->displays_id () && !tdb2.read_only ())
+    {
+      run_gc = config.getBoolean ("gc");
       tdb2.gc ();
+    }
+    else
+    {
+      run_gc = false;
+    }
 
 /*
     // Only read-only commands can be run when TDB2 is read-only.
@@ -504,7 +485,7 @@ bool Context::color ()
     use_color = config.getBoolean ("color");
 
     // Only tty's support color.
-    if (! isatty (fileno (stdout)))
+    if (! isatty (STDOUT_FILENO))
     {
       // No ioctl.
       config.set ("detection", "off");
@@ -652,18 +633,32 @@ void Context::staticInitialization ()
   CLI::minimumMatchLength   = config.getInteger ("abbreviation.minimum");
 
   Task::defaultProject      = config.get ("default.project");
-  Task::defaultPriority     = config.get ("default.priority");
   Task::defaultDue          = config.get ("default.due");
 
   Task::searchCaseSensitive = Variant::searchCaseSensitive = config.getBoolean ("search.case.sensitive");
   Task::regex               = Variant::searchUsingRegex    = config.getBoolean ("regex");
   Lexer::dateFormat         = Variant::dateFormat          = config.get ("dateformat");
+  Lexer::isoEnabled         = Variant::isoEnabled          = config.getBoolean ("date.iso");
+
+  Config::const_iterator rc;
+  for (rc = config.begin (); rc != config.end (); ++rc)
+  {
+    if (rc->first.substr (0, 4) == "uda." &&
+        rc->first.substr (rc->first.length () - 7, 7) == ".values")
+    {
+      std::string name = rc->first.substr (4, rc->first.length () - 7 - 4);
+      std::vector <std::string> values;
+      split (values, rc->second, ',');
+
+      for (auto r = values.rbegin(); r != values.rend (); ++r)
+        Task::customOrder[name].push_back (*r);
+    }
+  }
 
   std::map <std::string, Column*>::iterator i;
   for (i = columns.begin (); i != columns.end (); ++i)
     Task::attributes[i->first] = i->second->type ();
 
-  Task::urgencyPriorityCoefficient    = config.getReal ("urgency.priority.coefficient");
   Task::urgencyProjectCoefficient     = config.getReal ("urgency.project.coefficient");
   Task::urgencyActiveCoefficient      = config.getReal ("urgency.active.coefficient");
   Task::urgencyScheduledCoefficient   = config.getReal ("urgency.scheduled.coefficient");
@@ -770,7 +765,7 @@ void Context::clear ()
 // this output?'.
 void Context::updateXtermTitle ()
 {
-  if (config.getBoolean ("xterm.title") && isatty (fileno (stdout)))
+  if (config.getBoolean ("xterm.title") && isatty (STDOUT_FILENO))
   {
     std::string command = cli.getCommand ();
     std::string title;

src/Context.h

@@ -94,6 +94,8 @@ public:
   bool                                determine_color_use;
   bool                                use_color;
 
+  bool                                run_gc;
+
   bool                                verbosity_legacy;
   std::vector <std::string>           verbosity;
   std::vector <std::string>           headers;

src/DOM.cpp

@@ -116,16 +116,16 @@ bool DOM::get (const std::string& name, Variant& value)
     }
     else if (name == "context.width")
     {
-      value = Variant (context.terminal_width
-                         ? context.terminal_width
-                         : context.getWidth ());
+      value = Variant (static_cast<int> (context.terminal_width
+                                           ? context.terminal_width
+                                           : context.getWidth ()));
       return true;
     }
     else if (name == "context.height")
     {
-      value = Variant (context.terminal_height
-                         ? context.terminal_height
-                         : context.getHeight ());
+      value = Variant (static_cast<int> (context.terminal_height
+                                           ? context.terminal_height
+                                           : context.getHeight ()));
       return true;
     }
     else
@@ -218,7 +218,7 @@ bool DOM::get (const std::string& name, const Task& task, Variant& value)
   // <attr>
   if (task.size () && name == "id")
   {
-    value = Variant (task.id);
+    value = Variant (static_cast<int> (task.id));
     return true;
   }
 
@@ -240,6 +240,12 @@ bool DOM::get (const std::string& name, const Task& task, Variant& value)
       Column* column = context.columns[canonical];
       if (column)
       {
+        if (column->is_uda () && ! task.has (canonical))
+        {
+          value = Variant ("''");
+          return true;
+        }
+
         if (column->type () == "date")
           value = Variant (task.get_date (canonical), Variant::type_date);
         else if (column->type () == "duration" || canonical == "recur")
@@ -290,7 +296,7 @@ bool DOM::get (const std::string& name, const Task& task, Variant& value)
     {
       if (elements[1] == "id")
       {
-        value = Variant (ref.id);
+        value = Variant (static_cast<int> (ref.id));
         return true;
       }
       else if (elements[1] == "urgency")
@@ -307,6 +313,12 @@ bool DOM::get (const std::string& name, const Task& task, Variant& value)
           Column* column = context.columns[canonical];
           if (column)
           {
+            if (column->is_uda () && ! ref.has (canonical))
+            {
+              value = Variant ("''");
+              return true;
+            }
+
             if (column->type () == "date")
               value = Variant (ref.get_date (canonical), Variant::type_date);
             else if (column->type () == "duration")
@@ -341,15 +353,15 @@ bool DOM::get (const std::string& name, const Task& task, Variant& value)
             // <date>.minute
             // <date>.second
             Date date (ref.get_date (canonical));
-                 if (elements[2] == "year")    { value = Variant (date.year ());      return true; }
-            else if (elements[2] == "month")   { value = Variant (date.month ());     return true; }
-            else if (elements[2] == "day")     { value = Variant (date.day ());       return true; }
-            else if (elements[2] == "week")    { value = Variant (date.week ());      return true; }
-            else if (elements[2] == "weekday") { value = Variant (date.dayOfWeek ()); return true; }
-            else if (elements[2] == "julian")  { value = Variant (date.dayOfYear ()); return true; }
-            else if (elements[2] == "hour")    { value = Variant (date.hour ());      return true; }
-            else if (elements[2] == "minute")  { value = Variant (date.minute ());    return true; }
-            else if (elements[2] == "second")  { value = Variant (date.second ());    return true; }
+                 if (elements[2] == "year")    { value = Variant (static_cast<int> (date.year ()));      return true; }
+            else if (elements[2] == "month")   { value = Variant (static_cast<int> (date.month ()));     return true; }
+            else if (elements[2] == "day")     { value = Variant (static_cast<int> (date.day ()));       return true; }
+            else if (elements[2] == "week")    { value = Variant (static_cast<int> (date.week ()));      return true; }
+            else if (elements[2] == "weekday") { value = Variant (static_cast<int> (date.dayOfWeek ())); return true; }
+            else if (elements[2] == "julian")  { value = Variant (static_cast<int> (date.dayOfYear ())); return true; }
+            else if (elements[2] == "hour")    { value = Variant (static_cast<int> (date.hour ()));      return true; }
+            else if (elements[2] == "minute")  { value = Variant (static_cast<int> (date.minute ()));    return true; }
+            else if (elements[2] == "second")  { value = Variant (static_cast<int> (date.second ()));    return true; }
           }
         }
       }
@@ -408,15 +420,15 @@ bool DOM::get (const std::string& name, const Task& task, Variant& value)
               // <annotations>.<N>.entry.minute
               // <annotations>.<N>.entry.second
               Date date (i->first.substr (11));
-                   if (elements[4] == "year")    { value = Variant (date.year ());      return true; }
-              else if (elements[4] == "month")   { value = Variant (date.month ());     return true; }
-              else if (elements[4] == "day")     { value = Variant (date.day ());       return true; }
-              else if (elements[4] == "week")    { value = Variant (date.week ());      return true; }
-              else if (elements[4] == "weekday") { value = Variant (date.dayOfWeek ()); return true; }
-              else if (elements[4] == "julian")  { value = Variant (date.dayOfYear ()); return true; }
-              else if (elements[4] == "hour")    { value = Variant (date.hour ());      return true; }
-              else if (elements[4] == "minute")  { value = Variant (date.minute ());    return true; }
-              else if (elements[4] == "second")  { value = Variant (date.second ());    return true; }
+                   if (elements[4] == "year")    { value = Variant (static_cast<int> (date.year ()));      return true; }
+              else if (elements[4] == "month")   { value = Variant (static_cast<int> (date.month ()));     return true; }
+              else if (elements[4] == "day")     { value = Variant (static_cast<int> (date.day ()));       return true; }
+              else if (elements[4] == "week")    { value = Variant (static_cast<int> (date.week ()));      return true; }
+              else if (elements[4] == "weekday") { value = Variant (static_cast<int> (date.dayOfWeek ())); return true; }
+              else if (elements[4] == "julian")  { value = Variant (static_cast<int> (date.dayOfYear ())); return true; }
+              else if (elements[4] == "hour")    { value = Variant (static_cast<int> (date.hour ()));      return true; }
+              else if (elements[4] == "minute")  { value = Variant (static_cast<int> (date.minute ()));    return true; }
+              else if (elements[4] == "second")  { value = Variant (static_cast<int> (date.second ()));    return true; }
             }
           }
         }

src/Date.cpp

@@ -31,13 +31,13 @@
 #include <assert.h>
 #include <string.h>
 #include <stdlib.h>
-#include <ctype.h>
 #include <Nibbler.h>
 #include <Date.h>
 #include <Variant.h>
 #include <Dates.h>
 #include <text.h>
 #include <util.h>
+#include <utf8.h>
 #include <i18n.h>
 #include <Context.h>
 
@@ -823,7 +823,7 @@ void Date::operator++ (int)
 ////////////////////////////////////////////////////////////////////////////////
 bool Date::isEpoch (const std::string& input)
 {
-  if (digitsOnly (input) &&
+  if (Lexer::isAllDigits (input) &&
       input.length () <= 10 )
   {
     _t = (time_t) atoi (input.c_str ());

src/Dates.cpp

@@ -30,6 +30,7 @@
 #include <text.h>
 #include <Dates.h>
 #include <Date.h>
+#include <Lexer.h>
 #include <i18n.h>
 
 ////////////////////////////////////////////////////////////////////////////////
@@ -354,7 +355,7 @@ bool namedDates (const std::string& name, Variant& value)
   // 4th
   else if ((
             name.length () == 3 &&
-            isdigit (name[0])   &&
+            Lexer::isDigit (name[0])   &&
             ((name[1] == 's' && name[2] == 't') ||
              (name[1] == 'n' && name[2] == 'd') ||
              (name[1] == 'r' && name[2] == 'd') ||
@@ -363,8 +364,8 @@ bool namedDates (const std::string& name, Variant& value)
            ||
            (
             name.length () == 4 &&
-            isdigit (name[0])   &&
-            isdigit (name[1])   &&
+            Lexer::isDigit (name[0])   &&
+            Lexer::isDigit (name[1])   &&
             ((name[2] == 's' && name[3] == 't') ||
              (name[2] == 'n' && name[3] == 'd') ||
              (name[2] == 'r' && name[3] == 'd') ||
@@ -375,7 +376,7 @@ bool namedDates (const std::string& name, Variant& value)
     int number;
     std::string ordinal;
 
-    if (isdigit (name[1]))
+    if (Lexer::isDigit (name[1]))
     {
       number = strtol (name.substr (0, 2).c_str (), NULL, 10);
       ordinal = lowerCase (name.substr (2));

src/Directory.cpp

@@ -34,7 +34,7 @@
 #include <stdlib.h>
 #include <Directory.h>
 
-#if defined SOLARIS || defined NETBSD
+#if defined SOLARIS || defined NETBSD || defined FREEBSD
 #include <limits.h>
 #endif
 

src/Duration.cpp

@@ -61,31 +61,49 @@ static struct
   {"fortnight",   14 * DAY,    true},
   {"hours",        1 * HOUR,   false},
   {"hour",         1 * HOUR,   true},
+  {"hrs",          1 * HOUR,   false},
+  {"hr",           1 * HOUR,   true},
   {"h",            1 * HOUR,   false},
   {"minutes",      1 * MINUTE, false},
-  {"minute",       1 * MINUTE, false},
-  {"min",          1 * MINUTE, false},
+  {"minute",       1 * MINUTE, true},
+  {"mins",         1 * MINUTE, false},
+  {"min",          1 * MINUTE, true},
   {"monthly",     30 * DAY,    true},
   {"months",      30 * DAY,    false},
   {"month",       30 * DAY,    true},
-  {"mo",          30 * DAY,    false},
+  {"mnths",       30 * DAY,    false},
+  {"mths",        30 * DAY,    false},
+  {"mth",         30 * DAY,    true},
+  {"mos",         30 * DAY,    false},
+  {"mo",          30 * DAY,    true},
+  {"m",           30 * DAY,    false},
   {"quarterly",   91 * DAY,    true},
   {"quarters",    91 * DAY,    false},
   {"quarter",     91 * DAY,    true},
+  {"qrtrs",       91 * DAY,    false},
+  {"qrtr",        91 * DAY,    true},
+  {"qtrs",        91 * DAY,    false},
+  {"qtr",         91 * DAY,    true},
   {"q",           91 * DAY,    false},
   {"semiannual", 183 * DAY,    true},
   {"sennight",    14 * DAY,    false},
   {"seconds",      1 * SECOND, false},
   {"second",       1 * SECOND, true},
+  {"secs",         1 * SECOND, false},
+  {"sec",          1 * SECOND, true},
   {"s",            1 * SECOND, false},
   {"weekdays",     1 * DAY,    true},
   {"weekly",       7 * DAY,    true},
   {"weeks",        7 * DAY,    false},
   {"week",         7 * DAY,    true},
+  {"wks",          7 * DAY,    false},
+  {"wk",           7 * DAY,    true},
   {"w",            7 * DAY,    false},
   {"yearly",     365 * DAY,    true},
   {"years",      365 * DAY,    false},
   {"year",       365 * DAY,    true},
+  {"yrs",        365 * DAY,    false},
+  {"yr",         365 * DAY,    true},
   {"y",          365 * DAY,    false},
 };
 
@@ -107,7 +125,7 @@ Duration::Duration (time_t input)
 Duration::Duration (const std::string& input)
 : _secs (0)
 {
-  if (digitsOnly (input))
+  if (Lexer::isAllDigits (input))
   {
     time_t value = (time_t) strtol (input.c_str (), NULL, 10);
     if (value == 0 || value > 60)
@@ -284,10 +302,11 @@ bool Duration::parse (const std::string& input, std::string::size_type& start)
   std::string::size_type original_start = start;
   Nibbler n (input.substr (start));
 
-  // TODO This can be made static, and so preserved between calls.
-  std::vector <std::string> units;
-  for (int i = 0; i < NUM_DURATIONS; i++)
-    units.push_back (durations[i].unit);
+  // Static and so preserved between calls.
+  static std::vector <std::string> units;
+  if (units.size () == 0)
+    for (unsigned int i = 0; i < NUM_DURATIONS; i++)
+      units.push_back (durations[i].unit);
 
   std::string number;
   std::string unit;
@@ -295,13 +314,12 @@ bool Duration::parse (const std::string& input, std::string::size_type& start)
   if (n.getOneOf (units, unit))
   {
     if (n.depleted () ||
-        Lexer::is_ws (n.next ()))
+        Lexer::isWhitespace (n.next ()))
     {
       start = original_start + n.cursor ();
 
       // Linear lookup - should be logarithmic.
-      double seconds = 1;
-      for (int i = 0; i < NUM_DURATIONS; i++)
+      for (unsigned int i = 0; i < NUM_DURATIONS; i++)
       {
         if (durations[i].unit == unit &&
             durations[i].standalone == true)
@@ -319,14 +337,14 @@ bool Duration::parse (const std::string& input, std::string::size_type& start)
     if (n.getOneOf (units, unit))
     {
       if (n.depleted () ||
-          Lexer::is_ws (n.next ()))
+          Lexer::isWhitespace (n.next ()))
       {
         start = original_start + n.cursor ();
         double quantity = strtod (number.c_str (), NULL);
 
         // Linear lookup - should be logarithmic.
         double seconds = 1;
-        for (int i = 0; i < NUM_DURATIONS; i++)
+        for (unsigned int i = 0; i < NUM_DURATIONS; i++)
         {
           if (durations[i].unit == unit)
           {

src/Eval.cpp

@@ -127,7 +127,7 @@ void Eval::evaluateInfixExpression (const std::string& e, Variant& v) const
   // Reduce e to a vector of tokens.
   Lexer l (e);
   l.ambiguity (_ambiguity);
-  std::vector <std::pair <std::string, Lexer::Type> > tokens;
+  std::vector <std::pair <std::string, Lexer::Type>> tokens;
   std::string token;
   Lexer::Type type;
   while (l.token (token, type))
@@ -155,7 +155,7 @@ void Eval::evaluatePostfixExpression (const std::string& e, Variant& v) const
   // Reduce e to a vector of tokens.
   Lexer l (e);
   l.ambiguity (_ambiguity);
-  std::vector <std::pair <std::string, Lexer::Type> > tokens;
+  std::vector <std::pair <std::string, Lexer::Type>> tokens;
   std::string token;
   Lexer::Type type;
   while (l.token (token, type))
@@ -179,7 +179,7 @@ void Eval::compileExpression (const std::string& e)
   while (l.token (token, type))
   {
     if (_debug)
-      context.debug ("Lexer '" + token + "' " + Lexer::type_name (type));
+      context.debug ("Lexer '" + token + "' " + Lexer::typeToString (type));
     _compiled.push_back (std::pair <std::string, Lexer::Type> (token, type));
   }
 
@@ -236,7 +236,7 @@ void Eval::getBinaryOperators (std::vector <std::string>& all)
 
 ////////////////////////////////////////////////////////////////////////////////
 void Eval::evaluatePostfixStack (
-  const std::vector <std::pair <std::string, Lexer::Type> >& tokens,
+  const std::vector <std::pair <std::string, Lexer::Type>>& tokens,
   Variant& result) const
 {
   if (tokens.size () == 0)
@@ -245,11 +245,11 @@ void Eval::evaluatePostfixStack (
   // This is stack used by the postfix evaluator.
   std::vector <Variant> values;
 
-  std::vector <std::pair <std::string, Lexer::Type> >::const_iterator token;
+  std::vector <std::pair <std::string, Lexer::Type>>::const_iterator token;
   for (token = tokens.begin (); token != tokens.end (); ++token)
   {
     // Unary operators.
-    if (token->second == Lexer::typeOperator &&
+    if (token->second == Lexer::Type::op &&
         token->first == "!")
     {
       if (values.size () < 1)
@@ -262,7 +262,7 @@ void Eval::evaluatePostfixStack (
       if (_debug)
         context.debug (format ("Eval {1} ↓'{2}' → ↑'{3}'", token->first, (std::string) right, (std::string) result));
     }
-    else if (token->second == Lexer::typeOperator &&
+    else if (token->second == Lexer::Type::op &&
              token->first == "_neg_")
     {
       if (values.size () < 1)
@@ -278,7 +278,7 @@ void Eval::evaluatePostfixStack (
       if (_debug)
         context.debug (format ("Eval {1} ↓'{2}' → ↑'{3}'", token->first, (std::string) right, (std::string) result));
     }
-    else if (token->second == Lexer::typeOperator &&
+    else if (token->second == Lexer::Type::op &&
              token->first == "_pos_")
     {
       // The _pos_ operator is a NOP.
@@ -287,7 +287,7 @@ void Eval::evaluatePostfixStack (
     }
 
     // Binary operators.
-    else if (token->second == Lexer::typeOperator)
+    else if (token->second == Lexer::Type::op)
     {
       if (values.size () < 2)
         throw std::string (STRING_EVAL_NO_EVAL);
@@ -338,24 +338,28 @@ void Eval::evaluatePostfixStack (
       Variant v (token->first);
       switch (token->second)
       {
-      case Lexer::typeNumber:
-      case Lexer::typeHex:
-        v.cast (Variant::type_integer);
-        if (_debug)
-          context.debug (format ("Eval literal number ↑'{1}'", (std::string) v));
+      case Lexer::Type::number:
+        if (Lexer::isAllDigits (token->first))
+        {
+          v.cast (Variant::type_integer);
+          if (_debug)
+            context.debug (format ("Eval literal number ↑'{1}'", (std::string) v));
+        }
+        else
+        {
+          v.cast (Variant::type_real);
+          if (_debug)
+            context.debug (format ("Eval literal decimal ↑'{1}'", (std::string) v));
+        }
         break;
 
-      case Lexer::typeDecimal:
-        v.cast (Variant::type_real);
-        if (_debug)
-          context.debug (format ("Eval literal decimal ↑'{1}'", (std::string) v));
-        break;
 
-      case Lexer::typeOperator:
+      case Lexer::Type::op:
         throw std::string (STRING_EVAL_OP_EXPECTED);
         break;
 
-      case Lexer::typeIdentifier:
+      case Lexer::Type::dom:
+      case Lexer::Type::identifier:
         {
           bool found = false;
           std::vector <bool (*)(const std::string&, Variant&)>::const_iterator source;
@@ -380,20 +384,33 @@ void Eval::evaluatePostfixStack (
         }
         break;
 
-      case Lexer::typeDate:
+      case Lexer::Type::date:
         v.cast (Variant::type_date);
         if (_debug)
           context.debug (format ("Eval literal date ↑'{1}'", (std::string) v));
         break;
 
-      case Lexer::typeDuration:
+      case Lexer::Type::duration:
         v.cast (Variant::type_duration);
         if (_debug)
           context.debug (format ("Eval literal duration ↑'{1}'", (std::string) v));
         break;
 
       // Nothing to do.
-      case Lexer::typeString:
+/*
+      case Lexer::Type::uuid:
+      case Lexer::Type::hex:
+      case Lexer::Type::list:
+      case Lexer::Type::url:
+      case Lexer::Type::pair:
+      case Lexer::Type::separator:
+      case Lexer::Type::tag:
+      case Lexer::Type::path:
+      case Lexer::Type::substitution:
+      case Lexer::Type::pattern:
+      case Lexer::Type::word:
+*/
+      case Lexer::Type::string:
       default:
         if (_debug)
           context.debug (format ("Eval literal string ↑'{1}'", (std::string) v));
@@ -427,26 +444,26 @@ void Eval::evaluatePostfixStack (
 //   Primitive   --> "(" Logical ")" | Variant
 //
 void Eval::infixParse (
-  std::vector <std::pair <std::string, Lexer::Type> >& infix) const
+  std::vector <std::pair <std::string, Lexer::Type>>& infix) const
 {
-  int i = 0;
+  unsigned int i = 0;
   parseLogical (infix, i);
 }
 
 ////////////////////////////////////////////////////////////////////////////////
 // Logical     --> Regex {( "and" | "or" | "xor" ) Regex}
 bool Eval::parseLogical (
-  std::vector <std::pair <std::string, Lexer::Type> >& infix,
-  int &i) const
+  std::vector <std::pair <std::string, Lexer::Type>>& infix,
+  unsigned int &i) const
 {
   if (i < infix.size () &&
       parseRegex (infix, i))
   {
     while (i < infix.size () &&
+           infix[i].second == Lexer::Type::op &&
            (infix[i].first == "and"  ||
             infix[i].first == "or"   ||
-            infix[i].first == "xor") &&
-           infix[i].second == Lexer::typeOperator)
+            infix[i].first == "xor"))
     {
       ++i;
       if (! parseRegex (infix, i))
@@ -462,16 +479,16 @@ bool Eval::parseLogical (
 ////////////////////////////////////////////////////////////////////////////////
 // Regex       --> Equality {( "~" | "!~" ) Equality}
 bool Eval::parseRegex (
-  std::vector <std::pair <std::string, Lexer::Type> >& infix,
-  int &i) const
+  std::vector <std::pair <std::string, Lexer::Type>>& infix,
+  unsigned int &i) const
 {
   if (i < infix.size () &&
       parseEquality (infix, i))
   {
     while (i < infix.size () &&
+           infix[i].second == Lexer::Type::op &&
            (infix[i].first == "~" ||
-            infix[i].first == "!~") &&
-           infix[i].second == Lexer::typeOperator)
+            infix[i].first == "!~"))
     {
       ++i;
       if (! parseEquality (infix, i))
@@ -487,18 +504,18 @@ bool Eval::parseRegex (
 ////////////////////////////////////////////////////////////////////////////////
 // Equality    --> Comparative {( "==" | "=" | "!==" | "!=" ) Comparative}
 bool Eval::parseEquality (
-  std::vector <std::pair <std::string, Lexer::Type> >& infix,
-  int &i) const
+  std::vector <std::pair <std::string, Lexer::Type>>& infix,
+  unsigned int &i) const
 {
   if (i < infix.size () &&
       parseComparative (infix, i))
   {
     while (i < infix.size () &&
+           infix[i].second == Lexer::Type::op &&
            (infix[i].first == "=="  ||
             infix[i].first == "="   ||
             infix[i].first == "!==" ||
-            infix[i].first == "!=") &&
-           infix[i].second == Lexer::typeOperator)
+            infix[i].first == "!="))
     {
       ++i;
       if (! parseComparative (infix, i))
@@ -514,18 +531,18 @@ bool Eval::parseEquality (
 ////////////////////////////////////////////////////////////////////////////////
 // Comparative --> Arithmetic {( "<=" | "<" | ">=" | ">" ) Arithmetic}
 bool Eval::parseComparative (
-  std::vector <std::pair <std::string, Lexer::Type> >& infix,
-  int &i) const
+  std::vector <std::pair <std::string, Lexer::Type>>& infix,
+  unsigned int &i) const
 {
   if (i < infix.size () &&
       parseArithmetic (infix, i))
   {
     while (i < infix.size () &&
+           infix[i].second == Lexer::Type::op &&
            (infix[i].first == "<=" ||
             infix[i].first == "<"  ||
             infix[i].first == ">=" ||
-            infix[i].first == ">") &&
-           infix[i].second == Lexer::typeOperator)
+            infix[i].first == ">"))
     {
       ++i;
       if (! parseArithmetic (infix, i))
@@ -541,16 +558,16 @@ bool Eval::parseComparative (
 ////////////////////////////////////////////////////////////////////////////////
 // Arithmetic  --> Geometric {( "+" | "-" ) Geometric}
 bool Eval::parseArithmetic (
-  std::vector <std::pair <std::string, Lexer::Type> >& infix,
-  int &i) const
+  std::vector <std::pair <std::string, Lexer::Type>>& infix,
+  unsigned int &i) const
 {
   if (i < infix.size () &&
       parseGeometric (infix, i))
   {
     while (i < infix.size () &&
+           infix[i].second == Lexer::Type::op &&
            (infix[i].first == "+" ||
-            infix[i].first == "-") &&
-           infix[i].second == Lexer::typeOperator)
+            infix[i].first == "-"))
     {
       ++i;
       if (! parseGeometric (infix, i))
@@ -566,17 +583,17 @@ bool Eval::parseArithmetic (
 ////////////////////////////////////////////////////////////////////////////////
 // Geometric   --> Tag {( "*" | "/" | "%" ) Tag}
 bool Eval::parseGeometric (
-  std::vector <std::pair <std::string, Lexer::Type> >& infix,
-  int &i) const
+  std::vector <std::pair <std::string, Lexer::Type>>& infix,
+  unsigned int &i) const
 {
   if (i < infix.size () &&
       parseTag (infix, i))
   {
     while (i < infix.size () &&
+           infix[i].second == Lexer::Type::op &&
            (infix[i].first == "*" ||
             infix[i].first == "/" ||
-            infix[i].first == "%") &&
-           infix[i].second == Lexer::typeOperator)
+            infix[i].first == "%"))
     {
       ++i;
       if (! parseTag (infix, i))
@@ -592,16 +609,16 @@ bool Eval::parseGeometric (
 ////////////////////////////////////////////////////////////////////////////////
 // Tag         --> Unary {( "_hastag_" | "_notag_" ) Unary}
 bool Eval::parseTag (
-  std::vector <std::pair <std::string, Lexer::Type> >& infix,
-  int &i) const
+  std::vector <std::pair <std::string, Lexer::Type>>& infix,
+  unsigned int &i) const
 {
   if (i < infix.size () &&
       parseUnary (infix, i))
   {
     while (i < infix.size () &&
+           infix[i].second == Lexer::Type::op &&
            (infix[i].first == "_hastag_" ||
-            infix[i].first == "_notag_") &&
-           infix[i].second == Lexer::typeOperator)
+            infix[i].first == "_notag_"))
     {
       ++i;
       if (! parseUnary (infix, i))
@@ -617,8 +634,8 @@ bool Eval::parseTag (
 ////////////////////////////////////////////////////////////////////////////////
 // Unary       --> [( "-" | "+" | "!" )] Exponent
 bool Eval::parseUnary (
-  std::vector <std::pair <std::string, Lexer::Type> >& infix,
-  int &i) const
+  std::vector <std::pair <std::string, Lexer::Type>>& infix,
+  unsigned int &i) const
 {
   if (i < infix.size ())
   {
@@ -644,15 +661,15 @@ bool Eval::parseUnary (
 ////////////////////////////////////////////////////////////////////////////////
 // Exponent    --> Primitive ["^" Primitive]
 bool Eval::parseExponent (
-  std::vector <std::pair <std::string, Lexer::Type> >& infix,
-  int &i) const
+  std::vector <std::pair <std::string, Lexer::Type>>& infix,
+  unsigned int& i) const
 {
   if (i < infix.size () &&
       parsePrimitive (infix, i))
   {
     while (i < infix.size () &&
-           infix[i].first == "^" &&
-           infix[i].second == Lexer::typeOperator)
+           infix[i].second == Lexer::Type::op &&
+           infix[i].first == "^")
     {
       ++i;
       if (! parsePrimitive (infix, i))
@@ -668,8 +685,8 @@ bool Eval::parseExponent (
 ////////////////////////////////////////////////////////////////////////////////
 // Primitive   --> "(" Logical ")" | Variant
 bool Eval::parsePrimitive (
-  std::vector <std::pair <std::string, Lexer::Type> >& infix,
-  int &i) const
+  std::vector <std::pair <std::string, Lexer::Type>>& infix,
+  unsigned int &i) const
 {
   if (i < infix.size ())
   {
@@ -706,7 +723,7 @@ bool Eval::parsePrimitive (
         ++i;
         return true;
       }
-      else if (infix[i].second != Lexer::typeOperator)
+      else if (infix[i].second != Lexer::Type::op)
       {
         ++i;
         return true;
@@ -750,32 +767,32 @@ bool Eval::parsePrimitive (
 //   Exit.
 //
 void Eval::infixToPostfix (
-  std::vector <std::pair <std::string, Lexer::Type> >& infix) const
+  std::vector <std::pair <std::string, Lexer::Type>>& infix) const
 {
   // Short circuit.
   if (infix.size () == 1)
     return;
 
   // Result.
-  std::vector <std::pair <std::string, Lexer::Type> > postfix;
+  std::vector <std::pair <std::string, Lexer::Type>> postfix;
 
   // Shunting yard.
-  std::vector <std::pair <std::string, Lexer::Type> > op_stack;
+  std::vector <std::pair <std::string, Lexer::Type>> op_stack;
 
   // Operator characteristics.
   char type;
-  int precedence;
+  unsigned int precedence;
   char associativity;
 
-  std::vector <std::pair <std::string, Lexer::Type> >::iterator token;
+  std::vector <std::pair <std::string, Lexer::Type>>::iterator token;
   for (token = infix.begin (); token != infix.end (); ++token)
   {
-    if (token->second == Lexer::typeOperator &&
+    if (token->second == Lexer::Type::op &&
         token->first == "(")
     {
       op_stack.push_back (*token);
     }
-    else if (token->second == Lexer::typeOperator &&
+    else if (token->second == Lexer::Type::op &&
              token->first == ")")
     {
       while (op_stack.size () &&
@@ -790,11 +807,11 @@ void Eval::infixToPostfix (
       else
         throw std::string ("Mismatched parentheses in expression");
     }
-    else if (token->second == Lexer::typeOperator &&
+    else if (token->second == Lexer::Type::op &&
              identifyOperator (token->first, type, precedence, associativity))
     {
       char type2;
-      int precedence2;
+      unsigned int precedence2;
       char associativity2;
       while (op_stack.size () > 0 &&
              identifyOperator (op_stack.back ().first, type2, precedence2, associativity2) &&
@@ -830,7 +847,7 @@ void Eval::infixToPostfix (
 bool Eval::identifyOperator (
   const std::string& op,
   char& type,
-  int& precedence,
+  unsigned int& precedence,
   char& associativity) const
 {
   for (unsigned int i = 0; i < NUM_OPERATORS; ++i)
@@ -849,22 +866,21 @@ bool Eval::identifyOperator (
 
 ////////////////////////////////////////////////////////////////////////////////
 std::string Eval::dump (
-  std::vector <std::pair <std::string, Lexer::Type> >& tokens) const
+  std::vector <std::pair <std::string, Lexer::Type>>& tokens) const
 {
   // Set up a color mapping.
   std::map <Lexer::Type, Color> color_map;
-  color_map[Lexer::typeNone]       = Color ("rgb000 on gray6");
-  color_map[Lexer::typeOperator]   = Color ("gray14 on gray6");
-  color_map[Lexer::typeNumber]     = Color ("rgb530 on gray6");
-  color_map[Lexer::typeHex]        = Color ("rgb303 on gray6");
-  color_map[Lexer::typeDecimal]    = Color ("rgb530 on gray6");
-  color_map[Lexer::typeString]     = Color ("rgb550 on gray6");
-  color_map[Lexer::typeIdentifier] = Color ("rgb035 on gray6");
-  color_map[Lexer::typeDate]       = Color ("rgb150 on gray6");
-  color_map[Lexer::typeDuration]   = Color ("rgb531 on gray6");
+  color_map[Lexer::Type::op]         = Color ("gray14 on gray6");
+  color_map[Lexer::Type::number]     = Color ("rgb530 on gray6");
+  color_map[Lexer::Type::hex]        = Color ("rgb303 on gray6");
+  color_map[Lexer::Type::string]     = Color ("rgb550 on gray6");
+  color_map[Lexer::Type::dom]        = Color ("rgb045 on gray6");
+  color_map[Lexer::Type::identifier] = Color ("rgb035 on gray6");
+  color_map[Lexer::Type::date]       = Color ("rgb150 on gray6");
+  color_map[Lexer::Type::duration]   = Color ("rgb531 on gray6");
 
   std::string output;
-  std::vector <std::pair <std::string, Lexer::Type> >::const_iterator i;
+  std::vector <std::pair <std::string, Lexer::Type>>::const_iterator i;
   for (i = tokens.begin (); i != tokens.end (); ++i)
   {
     if (i != tokens.begin ())
@@ -874,7 +890,7 @@ std::string Eval::dump (
     if (color_map[i->second].nontrivial ())
       c = color_map[i->second];
     else
-      c = color_map[Lexer::typeNone];
+      c = Color ("rgb000 on gray6");
 
     output += c.colorize (i->first);
   }

src/Eval.h

@@ -53,28 +53,28 @@ public:
   static void getBinaryOperators (std::vector <std::string>&);
 
 private:
-  void evaluatePostfixStack (const std::vector <std::pair <std::string, Lexer::Type> >&, Variant&) const;
-  void infixToPostfix (std::vector <std::pair <std::string, Lexer::Type> >&) const;
-  void infixParse (std::vector <std::pair <std::string, Lexer::Type> >&) const;
-  bool parseLogical (std::vector <std::pair <std::string, Lexer::Type> >&, int &) const;
-  bool parseRegex (std::vector <std::pair <std::string, Lexer::Type> >&, int &) const;
-  bool parseEquality (std::vector <std::pair <std::string, Lexer::Type> >&, int &) const;
-  bool parseComparative (std::vector <std::pair <std::string, Lexer::Type> >&, int &) const;
-  bool parseArithmetic (std::vector <std::pair <std::string, Lexer::Type> >&, int &) const;
-  bool parseGeometric (std::vector <std::pair <std::string, Lexer::Type> >&, int &) const;
-  bool parseTag (std::vector <std::pair <std::string, Lexer::Type> >&, int &) const;
-  bool parseUnary (std::vector <std::pair <std::string, Lexer::Type> >&, int &) const;
-  bool parseExponent (std::vector <std::pair <std::string, Lexer::Type> >&, int &) const;
-  bool parsePrimitive (std::vector <std::pair <std::string, Lexer::Type> >&, int &) const;
-  bool identifyOperator (const std::string&, char&, int&, char&) const;
+  void evaluatePostfixStack (const std::vector <std::pair <std::string, Lexer::Type>>&, Variant&) const;
+  void infixToPostfix (std::vector <std::pair <std::string, Lexer::Type>>&) const;
+  void infixParse (std::vector <std::pair <std::string, Lexer::Type>>&) const;
+  bool parseLogical (std::vector <std::pair <std::string, Lexer::Type>>&, unsigned int &) const;
+  bool parseRegex (std::vector <std::pair <std::string, Lexer::Type>>&, unsigned int &) const;
+  bool parseEquality (std::vector <std::pair <std::string, Lexer::Type>>&, unsigned int &) const;
+  bool parseComparative (std::vector <std::pair <std::string, Lexer::Type>>&, unsigned int &) const;
+  bool parseArithmetic (std::vector <std::pair <std::string, Lexer::Type>>&, unsigned int &) const;
+  bool parseGeometric (std::vector <std::pair <std::string, Lexer::Type>>&, unsigned int &) const;
+  bool parseTag (std::vector <std::pair <std::string, Lexer::Type>>&, unsigned int &) const;
+  bool parseUnary (std::vector <std::pair <std::string, Lexer::Type>>&, unsigned int &) const;
+  bool parseExponent (std::vector <std::pair <std::string, Lexer::Type>>&, unsigned int &) const;
+  bool parsePrimitive (std::vector <std::pair <std::string, Lexer::Type>>&, unsigned int &) const;
+  bool identifyOperator (const std::string&, char&, unsigned int&, char&) const;
 
-  std::string dump (std::vector <std::pair <std::string, Lexer::Type> >&) const;
+  std::string dump (std::vector <std::pair <std::string, Lexer::Type>>&) const;
 
 private:
   std::vector <bool (*)(const std::string&, Variant&)> _sources;
   bool _ambiguity;
   bool _debug;
-  std::vector <std::pair <std::string, Lexer::Type> > _compiled;
+  std::vector <std::pair <std::string, Lexer::Type>> _compiled;
 };
 
 

src/File.cpp

@@ -28,14 +28,10 @@
 #include <fstream>
 #include <sys/types.h>
 #include <sys/stat.h>
-#ifdef SOLARIS
-#include <fcntl.h> // for flock() replacement
-#include <string.h> // for memset()
-#else
-#include <sys/file.h>
-#endif
-#include <pwd.h>
+#include <stdio.h>
+#include <fcntl.h>
 #include <unistd.h>
+#include <pwd.h>
 #include <File.h>
 #include <text.h>
 #include <util.h>
@@ -151,6 +147,9 @@ void File::close ()
 {
   if (_fh)
   {
+    if (_locked)
+      unlock ();
+
     fclose (_fh);
     _fh = NULL;
     _h = -1;
@@ -161,38 +160,31 @@ void File::close ()
 ////////////////////////////////////////////////////////////////////////////////
 bool File::lock ()
 {
+  _locked = false;
   if (_fh && _h != -1)
   {
-    // Try three times before failing.
-    int retry = 0;
-    while (flock (_h, LOCK_NB | LOCK_EX) && ++retry <= 3)
-      ;
-
-    if (retry <= 3)
-    {
+                    // l_type   l_whence  l_start  l_len  l_pid
+    struct flock fl = {F_WRLCK, SEEK_SET, 0,       0,     0 };
+    fl.l_pid = getpid ();
+    if (fcntl (_h, F_SETLKW, &fl) == 0)
       _locked = true;
-      return true;
-    }
   }
 
-  _locked = false;
-  return false;
+  return _locked;
 }
 
 ////////////////////////////////////////////////////////////////////////////////
-bool File::waitForLock ()
+void File::unlock ()
 {
   if (_locked)
-    return true;
+  {
+                    // l_type   l_whence  l_start  l_len  l_pid
+    struct flock fl = {F_UNLCK, SEEK_SET, 0,       0,     0 };
+    fl.l_pid = getpid ();
 
-  if (_fh && _h != -1)
-    if (flock (_h, LOCK_EX) == 0)
-    {
-      _locked = true;
-      return true;
-    }
-
-  return false;
+    fcntl (_h, F_SETLK, &fl);
+    _locked = false;
+  }
 }
 
 ////////////////////////////////////////////////////////////////////////////////

src/File.h

@@ -52,7 +52,7 @@ public:
   void close ();
 
   bool lock ();
-  bool waitForLock ();
+  void unlock ();
 
   void read (std::string&);
   void read (std::vector <std::string>&);

src/Filter.cpp

@@ -67,7 +67,7 @@ Filter::~Filter ()
 
 ////////////////////////////////////////////////////////////////////////////////
 // Take an input set of tasks and filter into a subset.
-void Filter::subset (const std::vector <Task>& input, std::vector <Task>& output)
+void Filter::subset (const std::vector <Task>& input, std::vector <Task>& output, bool applyContext /* = true */)
 {
   context.timer_filter.start ();
   _startCount = (int) input.size ();
@@ -75,7 +75,7 @@ void Filter::subset (const std::vector <Task>& input, std::vector <Task>& output
   if (context.config.getInteger ("debug.parser") >= 1)
     context.debug (context.cli.dump ("Filter::subset"));
 
-  std::string filterExpr = context.cli.getFilter ();
+  std::string filterExpr = context.cli.getFilter (applyContext);
   if (filterExpr.length ())
   {
     Eval eval;
@@ -111,7 +111,7 @@ void Filter::subset (const std::vector <Task>& input, std::vector <Task>& output
 
 ////////////////////////////////////////////////////////////////////////////////
 // Take the set of all tasks and filter into a subset.
-void Filter::subset (std::vector <Task>& output)
+void Filter::subset (std::vector <Task>& output, bool applyContext /* = true */)
 {
   context.timer_filter.start ();
 
@@ -119,7 +119,7 @@ void Filter::subset (std::vector <Task>& output)
     context.debug (context.cli.dump ("Filter::subset"));
 
   bool shortcut = false;
-  std::string filterExpr = context.cli.getFilter ();
+  std::string filterExpr = context.cli.getFilter (applyContext);
   if (filterExpr.length ())
   {
     context.timer_filter.stop ();

src/Filter.h

@@ -40,8 +40,8 @@ public:
   Filter ();
   ~Filter ();
 
-  void subset (const std::vector <Task>&, std::vector <Task>&);
-  void subset (std::vector <Task>&);
+  void subset (const std::vector <Task>&, std::vector <Task>&, bool applyContext = true);
+  void subset (std::vector <Task>&, bool applyContext = true);
   bool pendingOnly ();
   void safety ();
 

src/Hooks.cpp

@@ -38,8 +38,10 @@
 #include <Context.h>
 #include <JSON.h>
 #include <Hooks.h>
+#include <Timer.h>
 #include <text.h>
 #include <util.h>
+#include <i18n.h>
 
 extern Context context;
 
@@ -73,7 +75,17 @@ void Hooks::initialize ()
     {
       std::vector <std::string>::iterator i;
       for (i = _scripts.begin (); i != _scripts.end (); ++i)
-        context.debug ("Found hook script " + *i);
+      {
+        Path p (*i);
+        std::string name = p.name ();
+        if (name.substr (0, 6) == "on-add"    ||
+            name.substr (0, 9) == "on-modify" ||
+            name.substr (0, 9) == "on-launch" ||
+            name.substr (0, 7) == "on-exit")
+          context.debug ("Found hook script " + *i);
+        else
+          context.debug ("Found misnamed hook script " + *i);
+      }
     }
   }
   else if (_debug >= 1)
@@ -98,8 +110,7 @@ bool Hooks::enable (bool value)
 // - none
 //
 // Output:
-// - all emitted JSON lines are added/modified as tasks, if the exit code is
-//   zero, otherwise ignored.
+// - JSON not allowed.
 // - all emitted non-JSON lines are considered feedback or error messages
 //   depending on the status code.
 //
@@ -120,29 +131,28 @@ void Hooks::onLaunch ()
       std::vector <std::string> output;
       int status = callHookScript (*script, input, output);
 
-      std::vector <std::string>::iterator line;
-      for (line = output.begin (); line != output.end (); ++line)
-      {
-        if (isJSON (*line))
-        {
-          if (status == 0)
-          {
-            // Only 'add' is possible.
-            Task newTask (*line);
-            context.tdb2.add (newTask);
-          }
-        }
-        else
-        {
-          if (status == 0)
-            context.header (*line);
-          else
-            context.error (*line);
-        }
-      }
+      std::vector <std::string> outputJSON;
+      std::vector <std::string> outputFeedback;
+      separateOutput (output, outputJSON, outputFeedback);
+
+      assertNTasks (outputJSON, 0);
+
+      if (status == 0)
+      {
+        std::vector <std::string>::iterator message;
+        for (message = outputFeedback.begin (); message != outputFeedback.end (); ++message)
+          context.footnote (*message);
+      }
+      else
+      {
+        assertFeedback (outputFeedback);
+
+        std::vector <std::string>::iterator message;
+        for (message = outputFeedback.begin (); message != outputFeedback.end (); ++message)
+          context.error (*message);
 
-      if (status)
         throw 0;  // This is how hooks silently terminate processing.
+      }
     }
   }
 
@@ -157,7 +167,7 @@ void Hooks::onLaunch ()
 // - read-only line of JSON for each task added/modified
 //
 // Output:
-// - any emitted JSON is ignored
+// - all emitted JSON is ignored
 // - all emitted non-JSON lines are considered feedback or error messages
 //   depending on the status code.
 //
@@ -188,24 +198,28 @@ void Hooks::onExit ()
       std::vector <std::string> output;
       int status = callHookScript (*script, input, output);
 
-      std::vector <std::string>::iterator line;
-      for (line = output.begin (); line != output.end (); ++line)
-      {
-        if (isJSON (*line))
-        {
-          context.error ("JSON output ignored: {1}");
-        }
-        else
-        {
-          if (status == 0)
-            context.footnote (*line);
-          else
-            context.error (*line);
-        }
-      }
+      std::vector <std::string> outputJSON;
+      std::vector <std::string> outputFeedback;
+      separateOutput (output, outputJSON, outputFeedback);
+
+      assertNTasks (outputJSON, 0);
+
+      if (status == 0)
+      {
+        std::vector <std::string>::iterator message;
+        for (message = outputFeedback.begin (); message != outputFeedback.end (); ++message)
+          context.footnote (*message);
+      }
+      else
+      {
+        assertFeedback (outputFeedback);
+
+        std::vector <std::string>::iterator message;
+        for (message = outputFeedback.begin (); message != outputFeedback.end (); ++message)
+          context.error (*message);
 
-      if (status)
         throw 0;  // This is how hooks silently terminate processing.
+      }
     }
   }
 
@@ -219,14 +233,14 @@ void Hooks::onExit ()
 // - line of JSON for the task added
 //
 // Output:
-// - all emitted JSON lines are added/modified as tasks, if the exit code is
-//   zero, otherwise ignored.
+// - emitted JSON for the input task is added, if the exit code is zero,
+//   otherwise ignored.
 // - all emitted non-JSON lines are considered feedback or error messages
 //   depending on the status code.
 //
-void Hooks::onAdd (std::vector <Task>& tasks)
+void Hooks::onAdd (Task& task)
 {
-  if (! _enabled || tasks.size () < 1)
+  if (! _enabled)
     return;
 
   context.timer_hooks.start ();
@@ -234,9 +248,9 @@ void Hooks::onAdd (std::vector <Task>& tasks)
   std::vector <std::string> matchingScripts = scripts ("on-add");
   if (matchingScripts.size ())
   {
-    // Convert vector of tasks to a vector of strings.
+    // Convert task to a vector of strings.
     std::vector <std::string> input;
-    input.push_back (tasks[0].composeJSON ());
+    input.push_back (task.composeJSON ());
 
     // Call the hook scripts.
     std::vector <std::string>::iterator script;
@@ -245,33 +259,37 @@ void Hooks::onAdd (std::vector <Task>& tasks)
       std::vector <std::string> output;
       int status = callHookScript (*script, input, output);
 
-      input.clear ();
-      std::vector <std::string>::iterator line;
-      for (line = output.begin (); line != output.end (); ++line)
-      {
-        if (isJSON (*line))
-        {
-          if (status == 0)
-            input.push_back (*line);
-        }
-        else
-        {
-          if (status == 0)
-            context.footnote (*line);
-          else
-            context.error (*line);
-        }
-      }
+      std::vector <std::string> outputJSON;
+      std::vector <std::string> outputFeedback;
+      separateOutput (output, outputJSON, outputFeedback);
+
+      if (status == 0)
+      {
+        assertNTasks    (outputJSON, 1);
+        assertValidJSON (outputJSON);
+        assertSameTask  (outputJSON, task);
+
+        // Propagate forward to the next script.
+        input[0] = outputJSON[0];
+
+        std::vector <std::string>::iterator message;
+        for (message = outputFeedback.begin (); message != outputFeedback.end (); ++message)
+          context.footnote (*message);
+      }
+      else
+      {
+        assertFeedback (outputFeedback);
+
+        std::vector <std::string>::iterator message;
+        for (message = outputFeedback.begin (); message != outputFeedback.end (); ++message)
+          context.error (*message);
 
-      if (status)
         throw 0;  // This is how hooks silently terminate processing.
+      }
     }
 
-    // Transfer the modified task lines back to the original task list.
-    tasks.clear ();
-    std::vector <std::string>::iterator i;
-    for (i = input.begin (); i != input.end (); ++i)
-      tasks.push_back (Task (*i));
+    // Transfer the modified task back to the original task.
+    task = Task (input[0]);
   }
 
   context.timer_hooks.stop ();
@@ -285,14 +303,14 @@ void Hooks::onAdd (std::vector <Task>& tasks)
 // - line of JSON for the modified task, the diff being the modification
 //
 // Output:
-// - all emitted JSON lines are added/modified as tasks, if the exit code is
-//   zero, otherwise ignored.
+// - emitted JSON for the input task is saved, if the exit code is zero,
+//   otherwise ignored.
 // - all emitted non-JSON lines are considered feedback or error messages
 //   depending on the status code.
 //
-void Hooks::onModify (const Task& before, std::vector <Task>& tasks)
+void Hooks::onModify (const Task& before, Task& after)
 {
-  if (! _enabled || tasks.size () < 1)
+  if (! _enabled)
     return;
 
   context.timer_hooks.start ();
@@ -300,61 +318,48 @@ void Hooks::onModify (const Task& before, std::vector <Task>& tasks)
   std::vector <std::string> matchingScripts = scripts ("on-modify");
   if (matchingScripts.size ())
   {
-    // Prepare invariants.
-    std::string beforeJSON = before.composeJSON ();
-    std::string uuidPattern = "\"uuid\":\"" + before.get ("uuid") + "\"";
-
     // Convert vector of tasks to a vector of strings.
     std::vector <std::string> input;
-    input.push_back (beforeJSON);               // [0] original, never changes
-    input.push_back (tasks[0].composeJSON ());  // [1] original'
+    input.push_back (before.composeJSON ()); // [line 0] original, never changes
+    input.push_back (after.composeJSON ());  // [line 1] modified
 
     // Call the hook scripts.
     std::vector <std::string>::iterator script;
     for (script = matchingScripts.begin (); script != matchingScripts.end (); ++script)
     {
-      std::vector <std::string> firstTwoOnly;
-      firstTwoOnly.push_back (input[0]);
-      firstTwoOnly.push_back (input[1]);
-
       std::vector <std::string> output;
-      int status = callHookScript (*script, firstTwoOnly, output);
+      int status = callHookScript (*script, input, output);
 
-      // Start from scratch.
-      input[1] = "";                            // [1] placeholder for original'
+      std::vector <std::string> outputJSON;
+      std::vector <std::string> outputFeedback;
+      separateOutput (output, outputJSON, outputFeedback);
 
-      std::vector <std::string>::iterator line;
-      for (line = output.begin (); line != output.end (); ++line)
+      if (status == 0)
       {
-        if (isJSON (*line))
-        {
-          if (status == 0)
-          {
-            if (line->find (uuidPattern) != std::string::npos)
-              input[1] = *line;                 // [1] original'
-            else
-              input.push_back (*line);          // [n > 1] extras
-          }
-        }
-        else
-        {
-          if (status == 0)
-              context.footnote (*line);
-        else
-            context.error (*line);
-        }
-      }
+        assertNTasks    (outputJSON, 1);
+        assertValidJSON (outputJSON);
+        assertSameTask  (outputJSON, before);
+
+        // Propagate accepted changes forward to the next script.
+        input[1] = outputJSON[0];
+
+        std::vector <std::string>::iterator message;
+        for (message = outputFeedback.begin (); message != outputFeedback.end (); ++message)
+          context.footnote (*message);
+      }
+      else
+      {
+        assertFeedback (outputFeedback);
+
+        std::vector <std::string>::iterator message;
+        for (message = outputFeedback.begin (); message != outputFeedback.end (); ++message)
+          context.error (*message);
 
-      if (status)
         throw 0;  // This is how hooks silently terminate processing.
+      }
     }
 
-    // Transfer the modified task lines back to the original task list.
-    tasks.clear ();
-    std::vector <std::string>::iterator i;
-    for (i = input.begin (); i != input.end (); ++i)
-      if (i != input.begin ())
-        tasks.push_back (Task (*i));
+    after = Task (input[1]);
   }
 
   context.timer_hooks.stop ();
@@ -384,65 +389,162 @@ std::vector <std::string> Hooks::scripts (const std::string& event)
   return matching;
 }
 
+////////////////////////////////////////////////////////////////////////////////
+void Hooks::separateOutput (
+  const std::vector <std::string>& output,
+  std::vector <std::string>& json,
+  std::vector <std::string>& feedback) const
+{
+  std::vector <std::string>::const_iterator i;
+  for (i = output.begin (); i != output.end (); ++i)
+  {
+    if (isJSON (*i))
+      json.push_back (*i);
+    else
+      feedback.push_back (*i);
+  }
+}
+
 ////////////////////////////////////////////////////////////////////////////////
 bool Hooks::isJSON (const std::string& input) const
 {
-  // Does it even look like JSON?  {...}
-  if (input.length () > 2 &&
-      input[0] == '{'     &&
-      input[input.length () - 1] == '}')
+  return input.length ()            >  2   &&
+         input[0]                   == '{' &&
+         input[input.length () - 1] == '}';
+}
+
+////////////////////////////////////////////////////////////////////////////////
+void Hooks::assertValidJSON (const std::vector <std::string>& input) const
+{
+  std::vector <std::string>::const_iterator i;
+  for (i = input.begin (); i != input.end (); i++)
   {
+    if (i->length () < 3 ||
+        (*i)[0] != '{'   ||
+        (*i)[i->length () - 1] != '}')
+    {
+      context.error (STRING_HOOK_ERROR_OBJECT);
+      throw 0;
+    }
+
     try
     {
-      // The absolute minimum a task needs is:
-      bool foundDescription = false;
-
-      // Parse the whole thing.
-      json::value* root = json::parse (input);
-      if (root->type () == json::j_object)
+      json::value* root = json::parse (*i);
+      if (root->type () != json::j_object)
       {
-        json::object* root_obj = (json::object*)root;
-
-        // For each object element...
-        json_object_iter i;
-        for (i  = root_obj->_data.begin ();
-             i != root_obj->_data.end ();
-             ++i)
-        {
-          // If the attribute is a recognized column.
-          std::string type = Task::attributes[i->first];
-          if (type == "string" && i->first == "description")
-            foundDescription = true;
-        }
+        context.error (STRING_HOOK_ERROR_OBJECT);
+        throw 0;
       }
-      else
-        throw std::string ("Object expected.");
 
-      // It's JSON, but is it a task?
-      if (! foundDescription)
-        throw std::string ("Missing 'description' attribute, of type 'string'.");
+      if (((json::object*)root)->_data.find ("description") == ((json::object*)root)->_data.end ())
+      {
+        context.error (STRING_HOOK_ERROR_NODESC);
+        throw 0;
+      }
 
-      // Yep, looks like a JSON task.
-      return true;
+      if (((json::object*)root)->_data.find ("uuid") == ((json::object*)root)->_data.end ())
+      {
+        context.error (STRING_HOOK_ERROR_NOUUID);
+        throw 0;
+      }
     }
 
     catch (const std::string& e)
     {
-      if (_debug >= 1)
-        context.error ("Hook output looks like JSON, but is not a valid task.");
-
-      if (_debug >= 2)
-        context.error ("JSON " + e);
+      context.error (format (STRING_HOOK_ERROR_SYNTAX, *i));
+      if (_debug)
+        context.error (STRING_HOOK_ERROR_JSON + e);
+      throw 0;
     }
 
     catch (...)
     {
-      if (_debug >= 1)
-        context.error ("Hook output looks like JSON, but fails to parse.");
+      context.error (STRING_HOOK_ERROR_NOPARSE + *i);
+      throw 0;
     }
   }
+}
 
-  return false;
+////////////////////////////////////////////////////////////////////////////////
+void Hooks::assertNTasks (const std::vector <std::string>& input, unsigned int n) const
+{
+  if (input.size () != n)
+  {
+    context.error (format (STRING_HOOK_ERROR_BAD_NUM, n, (int) input.size ()));
+    throw 0;
+  }
+}
+
+////////////////////////////////////////////////////////////////////////////////
+void Hooks::assertSameTask (const std::vector <std::string>& input, const Task& task) const
+{
+  std::string uuid = task.get ("uuid");
+
+  std::vector <std::string>::const_iterator i;
+  for (i = input.begin (); i != input.end (); i++)
+  {
+    json::object* root_obj = (json::object*)json::parse (*i);
+
+    // If there is no UUID at all.
+    json_object_iter u = root_obj->_data.find ("uuid");
+    if (u == root_obj->_data.end ()          ||
+        u->second->type () != json::j_string)
+    {
+      context.error (format (STRING_HOOK_ERROR_SAME1, uuid));
+      throw 0;
+    }
+
+    json::string* up = (json::string*) u->second;
+    std::string json_uuid = json::decode (unquoteText (up->dump ()));
+    if (json_uuid != uuid)
+    {
+      context.error (format (STRING_HOOK_ERROR_SAME2, uuid, json_uuid));
+      throw 0;
+    }
+  }
+}
+
+////////////////////////////////////////////////////////////////////////////////
+void Hooks::assertFeedback (const std::vector <std::string>& input) const
+{
+  bool foundSomething = false;
+  std::vector <std::string>::const_iterator i;
+  for (i = input.begin (); i != input.end (); ++i)
+    if (nontrivial (*i))
+      foundSomething = true;
+
+  if (! foundSomething)
+  {
+    context.error (STRING_HOOK_ERROR_NOFEEDBACK);
+    throw 0;
+  }
+}
+
+////////////////////////////////////////////////////////////////////////////////
+std::vector <std::string>& Hooks::buildHookScriptArgs (std::vector <std::string>& args)
+{
+  Variant v;
+
+  // Hooks API version.
+  args.push_back ("api:2");
+
+  // Command line Taskwarrior was called with.
+  context.dom.get ("context.args", v);
+  args.push_back ("args:" + std::string (v));
+
+  // Command to be executed.
+  args.push_back ("command:" + context.cli.getCommand ());
+
+  // rc file used after applying all overrides.
+  args.push_back ("rc:" + context.rc_file._data);
+
+  // Directory containing *.data files.
+  args.push_back ("data:" + context.data_dir._data);
+
+  // Taskwarrior version, same as returned by "task --version"
+  args.push_back ("version:" + std::string(VERSION));
+
+  return args;
 }
 
 ////////////////////////////////////////////////////////////////////////////////
@@ -452,11 +554,11 @@ int Hooks::callHookScript (
   std::vector <std::string>& output)
 {
   if (_debug >= 1)
-    context.debug ("Hooks: Calling " + script);
+    context.debug ("Hook: Calling " + script);
 
   if (_debug >= 2)
   {
-    context.debug ("Hooks: input");
+    context.debug ("Hook: input");
     std::vector <std::string>::const_iterator i;
     for (i = input.begin (); i != input.end (); ++i)
       context.debug ("  " + *i);
@@ -467,21 +569,39 @@ int Hooks::callHookScript (
   for (i = input.begin (); i != input.end (); ++i)
     inputStr += *i + "\n";
 
-  std::string outputStr;
   std::vector <std::string> args;
-  int status = execute (script, args, inputStr, outputStr);
+  buildHookScriptArgs (args);
+  if (_debug >= 2)
+  {
+    context.debug ("Hooks: args");
+    for (auto arg: args)
+      context.debug ("  " + arg);
+  }
+
+  // Measure time for each hook if running in debug
+  int status;
+  std::string outputStr;
+  if (_debug >= 2)
+  {
+    Timer timer_per_hook("Hooks::execute (" + script + ")");
+    timer_per_hook.start();
+
+    status = execute (script, args, inputStr, outputStr);
+  }
+  else
+    status = execute (script, args, inputStr, outputStr);
 
   split (output, outputStr, '\n');
 
   if (_debug >= 2)
   {
-    context.debug ("Hooks: output");
+    context.debug ("Hook: output");
     std::vector <std::string>::iterator i;
     for (i = output.begin (); i != output.end (); ++i)
       if (*i != "")
         context.debug ("  " + *i);
 
-    context.debug (format ("Hooks: Completed with status {1}", status));
+    context.debug (format ("Hook: Completed with status {1}", status));
     context.debug (" "); // Blank line
   }
 

src/Hooks.h

@@ -44,14 +44,20 @@ public:
 
   void onLaunch ();
   void onExit ();
-  void onAdd (std::vector <Task>&);
-  void onModify (const Task&, std::vector <Task>&);
+  void onAdd (Task&);
+  void onModify (const Task&, Task&);
 
   std::vector <std::string> list ();
 
 private:
   std::vector <std::string> scripts (const std::string&);
+  void separateOutput (const std::vector <std::string>&, std::vector <std::string>&, std::vector <std::string>&) const;
   bool isJSON (const std::string&) const;
+  void assertValidJSON (const std::vector <std::string>&) const;
+  void assertNTasks (const std::vector <std::string>&, unsigned int) const;
+  void assertSameTask (const std::vector <std::string>&, const Task&) const;
+  void assertFeedback (const std::vector <std::string>&) const;
+  std::vector <std::string>& buildHookScriptArgs (std::vector <std::string>&);
   int callHookScript (const std::string&, const std::vector <std::string>&, std::vector <std::string>&);
 
 private:

src/ISO8601.cpp

@@ -25,6 +25,7 @@
 ////////////////////////////////////////////////////////////////////////////////
 
 #include <cmake.h>
+#include <Lexer.h>
 #include <ISO8601.h>
 
 ////////////////////////////////////////////////////////////////////////////////
@@ -196,7 +197,7 @@ bool ISO8601d::parse_date_time_ext (Nibbler& n)
       else if (parse_off_ext (n))
         ;
 
-      if (! isdigit (n.next ()))
+      if (! Lexer::isDigit (n.next ()))
         return true;
     }
 
@@ -228,16 +229,16 @@ bool ISO8601d::parse_date_time (Nibbler& n)
       if (n.skip ('Z'))
       {
         _utc = true;
-        if (!isdigit (n.next ()))
+        if (!Lexer::isDigit (n.next ()))
           return true;
       }
       else if (parse_off (n))
       {
-        if (!isdigit (n.next ()))
+        if (!Lexer::isDigit (n.next ()))
           return true;
       }
 
-      if (!isdigit (n.next ()))
+      if (!Lexer::isDigit (n.next ()))
         return true;
     }
 
@@ -277,13 +278,13 @@ bool ISO8601d::parse_date_ext (Nibbler& n)
       }
 
       _year = year;
-      if (!isdigit (n.next ()))
+      if (!Lexer::isDigit (n.next ()))
         return true;
     }
     else if (n.getDigit3 (_julian))
     {
       _year = year;
-      if (!isdigit (n.next ()))
+      if (!Lexer::isDigit (n.next ()))
         return true;
     }
     else if (n.getDigit2 (month) &&
@@ -293,7 +294,7 @@ bool ISO8601d::parse_date_ext (Nibbler& n)
       _year = year;
       _month = month;
       _day = day;
-      if (!isdigit (n.next ()))
+      if (!Lexer::isDigit (n.next ()))
         return true;
     }
   }
@@ -327,7 +328,7 @@ bool ISO8601d::parse_date (Nibbler& n, bool ambiguous)
           _weekday = day;
 
         _year = year;
-        if (!isdigit (n.next ()))
+        if (!Lexer::isDigit (n.next ()))
           return true;
       }
     }
@@ -336,7 +337,7 @@ bool ISO8601d::parse_date (Nibbler& n, bool ambiguous)
       if (n.getDigit2 (_month))
       {
         _year = year;
-        if (!isdigit (n.next ()))
+        if (!Lexer::isDigit (n.next ()))
           return true;
       }
     }
@@ -345,13 +346,13 @@ bool ISO8601d::parse_date (Nibbler& n, bool ambiguous)
       _year = year;
       _month = month / 100;
       _day = month % 100;
-      if (!isdigit (n.next ()))
+      if (!Lexer::isDigit (n.next ()))
         return true;
     }
     else if (ambiguous && n.getDigit3 (_julian))
     {
       _year = year;
-      if (!isdigit (n.next ()))
+      if (!Lexer::isDigit (n.next ()))
         return true;
     }
   }
@@ -382,7 +383,7 @@ bool ISO8601d::parse_off_ext (Nibbler& n)
           offset += mm * 60;
 
         _offset = (sign == "-") ? -offset : offset;
-        if (!isdigit (n.next ()))
+        if (!Lexer::isDigit (n.next ()))
           return true;
       }
     }
@@ -412,7 +413,7 @@ bool ISO8601d::parse_off (Nibbler& n)
           offset += mm * 60;
 
         _offset = (sign == "-") ? -offset : offset;
-        if (!isdigit (n.next ()))
+        if (!Lexer::isDigit (n.next ()))
           return true;
       }
     }
@@ -453,7 +454,7 @@ bool ISO8601d::parse_time_ext (Nibbler& n)
     if (_ambiguity)
     {
       _seconds = seconds;
-      if (!isdigit (n.next ()))
+      if (!Lexer::isDigit (n.next ()))
         return true;
     }
   }
@@ -487,7 +488,7 @@ bool ISO8601d::parse_time (Nibbler& n, bool ambiguous)
     }
 
     _seconds = seconds;
-    if (!isdigit (n.next ()))
+    if (!Lexer::isDigit (n.next ()))
       return true;
   }
 
@@ -504,7 +505,7 @@ bool ISO8601d::parse_time_utc_ext (Nibbler& n)
       n.skip ('Z'))
   {
     _utc = true;
-    if (!isdigit (n.next ()))
+    if (!Lexer::isDigit (n.next ()))
       return true;
   }
 
@@ -521,7 +522,7 @@ bool ISO8601d::parse_time_utc (Nibbler& n)
       n.skip ('Z'))
   {
     _utc = true;
-    if (!isdigit (n.next ()))
+    if (!Lexer::isDigit (n.next ()))
       return true;
   }
 
@@ -537,7 +538,7 @@ bool ISO8601d::parse_time_off_ext  (Nibbler& n)
   if (parse_time_ext (n) &&
       parse_off_ext (n))
   {
-    if (!isdigit (n.next ()))
+    if (!Lexer::isDigit (n.next ()))
       return true;
   }
 
@@ -553,7 +554,7 @@ bool ISO8601d::parse_time_off  (Nibbler& n)
   if (parse_time (n, true) &&
       parse_off (n))
   {
-    if (!isdigit (n.next ()))
+    if (!Lexer::isDigit (n.next ()))
       return true;
   }
 
@@ -702,7 +703,7 @@ void ISO8601d::resolve ()
     t.tm_min = (seconds % 3600) / 60;
     t.tm_sec = seconds % 60;
 
-  _value = utc ? timegm (&t) : timelocal (&t);
+  _value = utc ? timegm (&t) : mktime (&t);
 }
 
 ////////////////////////////////////////////////////////////////////////////////

src/Lexer.h

@@ -27,91 +27,86 @@
 #ifndef INCLUDED_LEXER
 #define INCLUDED_LEXER
 
-#include <vector>
 #include <string>
+#include <vector>
+#include <cstddef>
+
+// Lexer: A UTF8 lexical analyzer for every construct used on the Taskwarrior
+//        command line, with additional recognized types for disambiguation.
 
 class Lexer
 {
 public:
+  // These are overridable.
   static std::string dateFormat;
+  static bool isoEnabled;
 
-  enum Type
-  {
-    typeNone = 0,
-    typeString,
-    typeIdentifier,
-    typeIdentifierEscape,    // Intermediate
-    typeEscape,              // Intermediate
-    typeEscapeHex,           // Intermediate
-    typeEscapeUnicode,       // Intermediate
-    typeNumber,
-    typeDecimal,
-    typeExponentIndicator,   // Intermediate
-    typeExponent,            // Intermediate
-    typeHex,
-    typeOperator,
-    typeDate,
-    typeDuration,
-    typeTag,
-/*
-    Recognizing more types means that Lexer::*_split and Lexer::token approach
-    the ideal form, whereby the command line becomes just one string that is
-    lexed into tokens. Those tokens are then simply dissected by type..
-
-    typeUUID,
-    typePattern,
-    typeSubstitution,
-    typeNameValue,
-*/
-  };
+  enum class Type { uuid, number, hex,
+                    string,
+                    list, url, pair, separator,
+                    tag,
+                    path,
+                    substitution, pattern,
+                    op,
+                    dom, identifier, word,
+                    date, duration };
 
   Lexer (const std::string&);
-  virtual ~Lexer ();
-  Lexer (const Lexer&);            // Not implemented.
-  Lexer& operator= (const Lexer&); // Not implemented.
-  bool operator== (const Lexer&);  // Not implemented.
-  bool token (std::string&, Type&);
-  bool word (std::string&, Type&);
+  ~Lexer ();
   void ambiguity (bool);
+  bool token (std::string&, Lexer::Type&);
+  static std::vector <std::pair <std::string, Lexer::Type>> tokens (const std::string&);
+  static std::vector <std::string> split (const std::string&);
+  static std::string typeToString (Lexer::Type);
+  static bool isAllDigits (const std::string&);
+  static bool isOneWord (const std::string&);
 
-  static const std::string type_name (const Type&);
-  static bool is_ws (int);
-  static bool is_ident_start (int);
-  static bool is_ident (int);
-  static bool is_single_op (int);
-  static bool is_dec_digit (int);
-  static bool boundary (int, int);
-  static void word_split (std::vector <std::string>&, const std::string&);
-  static void token_split (std::vector <std::string>&, const std::string&);
-  static void token_split (std::vector <std::pair <std::string, Lexer::Type> >&, const std::string&);
+  // Static helpers.
+  static const std::string typeName (const Lexer::Type&);
+  static bool isWhitespace         (int);
+  static bool isAlpha              (int);
+  static bool isDigit              (int);
+  static bool isHexDigit           (int);
+  static bool isIdentifierStart    (int);
+  static bool isIdentifierNext     (int);
+  static bool isSingleCharOperator (int);
+  static bool isDoubleCharOperator (int, int, int);
+  static bool isTripleCharOperator (int, int, int, int);
+  static bool isBoundary           (int, int);
+  static bool isPunctuation        (int);
   static void dequote (std::string&);
 
-private:
-  bool is_date (std::string&);
-  bool is_duration (std::string&);
-  bool is_punct (int) const;
-  bool is_num (int) const;
-  bool is_triple_op (int, int, int) const;
-  bool is_double_op (int, int, int) const;
-  bool is_hex_digit (int) const;
-  int decode_escape (int) const;
-  int hex_to_int (int) const;
-  int hex_to_int (int, int) const;
-  int hex_to_int (int, int, int, int) const;
-  void shift ();
+  // Helpers.
+  bool isEOS () const;
+  int hexToInt (int) const;
+  int hexToInt (int, int) const;
+  int hexToInt (int, int, int, int) const;
+
+  // Classifiers.
+  bool isString       (std::string&, Lexer::Type&, int quote);
+  bool isDate         (std::string&, Lexer::Type&);
+  bool isDuration     (std::string&, Lexer::Type&);
+  bool isUUID         (std::string&, Lexer::Type&);
+  bool isNumber       (std::string&, Lexer::Type&);
+  bool isHexNumber    (std::string&, Lexer::Type&);
+  bool isSeparator    (std::string&, Lexer::Type&);
+  bool isList         (std::string&, Lexer::Type&);
+  bool isURL          (std::string&, Lexer::Type&);
+  bool isPair         (std::string&, Lexer::Type&);
+  bool isTag          (std::string&, Lexer::Type&);
+  bool isPath         (std::string&, Lexer::Type&);
+  bool isSubstitution (std::string&, Lexer::Type&);
+  bool isPattern      (std::string&, Lexer::Type&);
+  bool isOperator     (std::string&, Lexer::Type&);
+  bool isDOM          (std::string&, Lexer::Type&);
+  bool isIdentifier   (std::string&, Lexer::Type&);
+  bool isWord         (std::string&, Lexer::Type&);
 
 private:
-  const std::string _input;
-  std::string::size_type _i;
-  std::string::size_type _shift_counter;
-  int _n0;
-  int _n1;
-  int _n2;
-  int _n3;
-  bool _boundary01;
-  bool _boundary12;
-  bool _boundary23;
-  bool _ambiguity;
+  std::string _text;
+  std::size_t _cursor;
+  std::size_t _eos;
+  bool        _ambiguity;
 };
 
 #endif

src/Nibbler.cpp

@@ -27,9 +27,9 @@
 #include <cmake.h>
 #include <stdlib.h>
 #include <string.h>
-#include <ctype.h>
 #include <time.h>
 #include <inttypes.h>
+#include <Lexer.h>
 #include <Nibbler.h>
 #ifdef NIBBLER_FEATURE_DATE
 #include <Date.h>
@@ -284,7 +284,7 @@ bool Nibbler::getQuoted (
 bool Nibbler::getDigit (int& result)
 {
   if (_cursor < _length &&
-      isdigit (_input[_cursor]))
+      Lexer::isDigit (_input[_cursor]))
   {
     result = _input[_cursor++] - '0';
     return true;
@@ -300,12 +300,12 @@ bool Nibbler::getDigit6 (int& result)
   if (i < _length &&
       _length - i >= 6)
   {
-    if (isdigit (_input[i + 0]) &&
-        isdigit (_input[i + 1]) &&
-        isdigit (_input[i + 2]) &&
-        isdigit (_input[i + 3]) &&
-        isdigit (_input[i + 4]) &&
-        isdigit (_input[i + 5]))
+    if (Lexer::isDigit (_input[i + 0]) &&
+        Lexer::isDigit (_input[i + 1]) &&
+        Lexer::isDigit (_input[i + 2]) &&
+        Lexer::isDigit (_input[i + 3]) &&
+        Lexer::isDigit (_input[i + 4]) &&
+        Lexer::isDigit (_input[i + 5]))
     {
       result = strtoimax (_input.substr (_cursor, 6).c_str (), NULL, 10);
       _cursor += 6;
@@ -323,10 +323,10 @@ bool Nibbler::getDigit4 (int& result)
   if (i < _length &&
       _length - i >= 4)
   {
-    if (isdigit (_input[i + 0]) &&
-        isdigit (_input[i + 1]) &&
-        isdigit (_input[i + 2]) &&
-        isdigit (_input[i + 3]))
+    if (Lexer::isDigit (_input[i + 0]) &&
+        Lexer::isDigit (_input[i + 1]) &&
+        Lexer::isDigit (_input[i + 2]) &&
+        Lexer::isDigit (_input[i + 3]))
     {
       result = strtoimax (_input.substr (_cursor, 4).c_str (), NULL, 10);
       _cursor += 4;
@@ -344,9 +344,9 @@ bool Nibbler::getDigit3 (int& result)
   if (i < _length &&
       _length - i >= 3)
   {
-    if (isdigit (_input[i + 0]) &&
-        isdigit (_input[i + 1]) &&
-        isdigit (_input[i + 2]))
+    if (Lexer::isDigit (_input[i + 0]) &&
+        Lexer::isDigit (_input[i + 1]) &&
+        Lexer::isDigit (_input[i + 2]))
     {
       result = strtoimax (_input.substr (_cursor, 3).c_str (), NULL, 10);
       _cursor += 3;
@@ -364,8 +364,8 @@ bool Nibbler::getDigit2 (int& result)
   if (i < _length &&
       _length - i >= 2)
   {
-    if (isdigit (_input[i + 0]) &&
-        isdigit (_input[i + 1]))
+    if (Lexer::isDigit (_input[i + 0]) &&
+        Lexer::isDigit (_input[i + 1]))
     {
       result = strtoimax (_input.substr (_cursor, 2).c_str (), NULL, 10);
       _cursor += 2;
@@ -390,7 +390,7 @@ bool Nibbler::getInt (int& result)
   }
 
   // TODO Potential for use of find_first_not_of
-  while (i < _length && isdigit (_input[i]))
+  while (i < _length && Lexer::isDigit (_input[i]))
     ++i;
 
   if (i > _cursor)
@@ -403,39 +403,12 @@ bool Nibbler::getInt (int& result)
   return false;
 }
 
-////////////////////////////////////////////////////////////////////////////////
-bool Nibbler::getHex (int& result)
-{
-  std::string::size_type i = _cursor;
-
-  if (i < _length)
-  {
-    if (_input[i] == '-')
-      ++i;
-    else if (_input[i] == '+')
-      ++i;
-  }
-
-  // TODO Potential for use of find_first_not_of
-  while (i < _length && isxdigit (_input[i]))
-    ++i;
-
-  if (i > _cursor)
-  {
-    result = strtoimax (_input.substr (_cursor, i - _cursor).c_str (), NULL, 16);
-    _cursor = i;
-    return true;
-  }
-
-  return false;
-}
-
 ////////////////////////////////////////////////////////////////////////////////
 bool Nibbler::getUnsignedInt (int& result)
 {
   std::string::size_type i = _cursor;
   // TODO Potential for use of find_first_not_of
-  while (i < _length && isdigit (_input[i]))
+  while (i < _length && Lexer::isDigit (_input[i]))
     ++i;
 
   if (i > _cursor)
@@ -473,11 +446,11 @@ bool Nibbler::getNumber (std::string& result)
     ++i;
 
   // digit+
-  if (i < _length && isdigit (_input[i]))
+  if (i < _length && Lexer::isDigit (_input[i]))
   {
     ++i;
 
-    while (i < _length && isdigit (_input[i]))
+    while (i < _length && Lexer::isDigit (_input[i]))
       ++i;
 
     // ( . digit+ )?
@@ -485,7 +458,7 @@ bool Nibbler::getNumber (std::string& result)
     {
       ++i;
 
-      while (i < _length && isdigit (_input[i]))
+      while (i < _length && Lexer::isDigit (_input[i]))
         ++i;
     }
 
@@ -497,11 +470,11 @@ bool Nibbler::getNumber (std::string& result)
       if (i < _length && (_input[i] == '+' || _input[i] == '-'))
         ++i;
 
-      if (i < _length && isdigit (_input[i]))
+      if (i < _length && Lexer::isDigit (_input[i]))
       {
         ++i;
 
-        while (i < _length && isdigit (_input[i]))
+        while (i < _length && Lexer::isDigit (_input[i]))
           ++i;
 
         result = _input.substr (_cursor, i - _cursor);
@@ -555,11 +528,11 @@ bool Nibbler::getUnsignedNumber (double& result)
   std::string::size_type i = _cursor;
 
   // digit+
-  if (i < _length && isdigit (_input[i]))
+  if (i < _length && Lexer::isDigit (_input[i]))
   {
     ++i;
 
-    while (i < _length && isdigit (_input[i]))
+    while (i < _length && Lexer::isDigit (_input[i]))
       ++i;
 
     // ( . digit+ )?
@@ -567,7 +540,7 @@ bool Nibbler::getUnsignedNumber (double& result)
     {
       ++i;
 
-      while (i < _length && isdigit (_input[i]))
+      while (i < _length && Lexer::isDigit (_input[i]))
         ++i;
     }
 
@@ -579,11 +552,11 @@ bool Nibbler::getUnsignedNumber (double& result)
       if (i < _length && (_input[i] == '+' || _input[i] == '-'))
         ++i;
 
-      if (i < _length && isdigit (_input[i]))
+      if (i < _length && Lexer::isDigit (_input[i]))
       {
         ++i;
 
-        while (i < _length && isdigit (_input[i]))
+        while (i < _length && Lexer::isDigit (_input[i]))
           ++i;
 
         result = strtof (_input.substr (_cursor, i - _cursor).c_str (), NULL);
@@ -737,21 +710,21 @@ bool Nibbler::getDateISO (time_t& t)
   if (i < _length &&
       _length - i >= 16)
   {
-    if (isdigit (_input[i + 0]) &&
-        isdigit (_input[i + 1]) &&
-        isdigit (_input[i + 2]) &&
-        isdigit (_input[i + 3]) &&
-        isdigit (_input[i + 4]) &&
-        isdigit (_input[i + 5]) &&
-        isdigit (_input[i + 6]) &&
-        isdigit (_input[i + 7]) &&
+    if (Lexer::isDigit (_input[i + 0]) &&
+        Lexer::isDigit (_input[i + 1]) &&
+        Lexer::isDigit (_input[i + 2]) &&
+        Lexer::isDigit (_input[i + 3]) &&
+        Lexer::isDigit (_input[i + 4]) &&
+        Lexer::isDigit (_input[i + 5]) &&
+        Lexer::isDigit (_input[i + 6]) &&
+        Lexer::isDigit (_input[i + 7]) &&
         _input[i + 8] == 'T' &&
-        isdigit (_input[i + 9]) &&
-        isdigit (_input[i + 10]) &&
-        isdigit (_input[i + 11]) &&
-        isdigit (_input[i + 12]) &&
-        isdigit (_input[i + 13]) &&
-        isdigit (_input[i + 14]) &&
+        Lexer::isDigit (_input[i + 9]) &&
+        Lexer::isDigit (_input[i + 10]) &&
+        Lexer::isDigit (_input[i + 11]) &&
+        Lexer::isDigit (_input[i + 12]) &&
+        Lexer::isDigit (_input[i + 13]) &&
+        Lexer::isDigit (_input[i + 14]) &&
         _input[i + 15] == 'Z')
     {
       _cursor += 16;
@@ -817,7 +790,7 @@ bool Nibbler::parseDigits(std::string::size_type& i,
       // Check that 'f' of them are digits
       unsigned int g;
       for (g = 0; g < f; g++)
-        if (! isdigit (_input[i + g]))
+        if (! Lexer::isDigit (_input[i + g]))
             break;
       // Parse the integer when it is the case
       if (g == f)
@@ -908,9 +881,9 @@ bool Nibbler::getDate (const std::string& format, time_t& t)
     case 'a':
     case 'A':
       if (i + 3 <= _length          &&
-          ! isdigit (_input[i + 0]) &&
-          ! isdigit (_input[i + 1]) &&
-          ! isdigit (_input[i + 2]))
+          ! Lexer::isDigit (_input[i + 0]) &&
+          ! Lexer::isDigit (_input[i + 1]) &&
+          ! Lexer::isDigit (_input[i + 2]))
       {
         wday = Date::dayOfWeek (_input.substr (i, 3).c_str ());
         i += (format[f] == 'a') ? 3 : Date::dayName (wday).size ();
@@ -922,9 +895,9 @@ bool Nibbler::getDate (const std::string& format, time_t& t)
     case 'b':
     case 'B':
       if (i + 3 <= _length          &&
-          ! isdigit (_input[i + 0]) &&
-          ! isdigit (_input[i + 1]) &&
-          ! isdigit (_input[i + 2]))
+          ! Lexer::isDigit (_input[i + 0]) &&
+          ! Lexer::isDigit (_input[i + 1]) &&
+          ! Lexer::isDigit (_input[i + 2]))
       {
         if (month != -1)
           return false;
@@ -1030,14 +1003,14 @@ bool Nibbler::getName (std::string& result)
 
   if (i < _length)
   {
-    if (! isdigit (_input[i]) &&
+    if (! Lexer::isDigit (_input[i]) &&
         ! ispunct (_input[i]) &&
-        ! Lexer::is_ws (_input[i]))
+        ! Lexer::isWhitespace (_input[i]))
     {
       ++i;
       while (i < _length &&
              (_input[i] == '_' || ! ispunct (_input[i])) &&
-             ! Lexer::is_ws (_input[i]))
+             ! Lexer::isWhitespace (_input[i]))
       {
         ++i;
       }
@@ -1062,9 +1035,9 @@ bool Nibbler::getWord (std::string& result)
 
   if (i < _length)
   {
-    while (!isdigit (_input[i]) &&
-           !isPunctuation (_input[i]) &&
-           !Lexer::is_ws (_input[i]))
+    while (!Lexer::isDigit       (_input[i]) &&
+           !Lexer::isPunctuation (_input[i]) &&
+           !Lexer::isWhitespace  (_input[i]))
     {
       ++i;
     }
@@ -1254,21 +1227,6 @@ bool Nibbler::depleted ()
   return false;
 }
 
-////////////////////////////////////////////////////////////////////////////////
-// Override of ispunct, that considers #, $, _ and @ not to be punctuation.
-//
-// ispunct:      ! " # $ % & ' ( ) * + , - . / : ; < = > ? @ [ \ ] ^ _ ` { | } ~
-// Punctuation:  ! "     % & ' ( ) * + , - . / : ; < = > ?   [ \ ] ^ _ ` { | } ~
-// delta:            # $                                   @
-//
-bool Nibbler::isPunctuation (char c)
-{
-  if (c == '@' || c == '#' || c == '$' || c == '_')
-    return false;
-
-  return ispunct (c);
-}
-
 ////////////////////////////////////////////////////////////////////////////////
 std::string Nibbler::dump ()
 {

src/Nibbler.h

@@ -68,7 +68,6 @@ public:
   bool getDigit3 (int&);
   bool getDigit2 (int&);
   bool getInt (int&);
-  bool getHex (int&);
   bool getUnsignedInt (int&);
   bool getNumber (std::string&);
   bool getNumber (double&);
@@ -111,7 +110,6 @@ public:
 
   bool depleted ();
 
-  static bool isPunctuation (char);
   std::string dump ();
 
 private:

src/TDB2.cpp

@@ -102,7 +102,7 @@ bool TF2::get (int id, Task& task)
   // pending.data file, the task in question cannot appear earlier than line
   // (id - 1) in the file.  It can, however, appear significantly later because
   // it is not known how recent a GC operation was run.
-  for (int i = id - 1; i < _tasks.size (); ++i)
+  for (unsigned int i = id - 1; i < _tasks.size (); ++i)
   {
     if (_tasks[i].id == id)
     {
@@ -226,7 +226,7 @@ void TF2::commit ()
       if (_file.open ())
       {
         if (context.config.getBoolean ("locking"))
-          _file.waitForLock ();
+          _file.lock ();
 
         // Write out all the added tasks.
         std::vector <Task>::iterator task;
@@ -258,7 +258,7 @@ void TF2::commit ()
       if (_file.open ())
       {
         if (context.config.getBoolean ("locking"))
-          _file.waitForLock ();
+          _file.lock ();
 
         // Truncate the file and rewrite.
         _file.truncate ();
@@ -316,9 +316,15 @@ void TF2::load_tasks ()
       ++line_number;
       Task task (*i);
 
-      // Some tasks gets an ID.
+      // Some tasks get an ID.
       if (_has_ids)
-        task.id = context.tdb2.next_id ();
+      {
+        Task::status status = task.getStatus ();
+        // Completed / deleted tasks in pending.data get an ID if GC is off.
+        if (!context.run_gc ||
+            (status != Task::completed && status != Task::deleted))
+          task.id = context.tdb2.next_id ();
+      }
 
       _tasks.push_back (task);
 
@@ -352,7 +358,7 @@ void TF2::load_lines ()
   if (_file.open ())
   {
     if (context.config.getBoolean ("locking"))
-      _file.waitForLock ();
+      _file.lock ();
 
     _file.read (_lines);
     _file.close ();
@@ -462,9 +468,13 @@ void TF2::dependency_scan ()
         {
           if (right->get ("uuid") == *d)
           {
-            Task::status status = right->getStatus ();
-            if (status != Task::completed &&
-                status != Task::deleted)
+            // GC hasn't run yet, check both tasks for their current status
+            Task::status lstatus = left->getStatus ();
+            Task::status rstatus = right->getStatus ();
+            if (lstatus != Task::completed &&
+                lstatus != Task::deleted &&
+                rstatus != Task::completed &&
+                rstatus != Task::deleted)
             {
               left->is_blocked = true;
               right->is_blocking = true;
@@ -559,7 +569,9 @@ void TDB2::set_location (const std::string& location)
 void TDB2::add (Task& task, bool add_to_backlog /* = true */)
 {
   // Ensure the task is consistent, and provide defaults if necessary.
-  task.validate ();
+  // bool argument to validate() is "applyDefault". Pass add_to_backlog through
+  // in order to not apply defaults to synchronized tasks.
+  task.validate (add_to_backlog);
   std::string uuid = task.get ("uuid");
 
   // If the tasks are loaded, then verify that this uuid is not already in
@@ -567,32 +579,30 @@ void TDB2::add (Task& task, bool add_to_backlog /* = true */)
   if (!verifyUniqueUUID (uuid))
     throw format (STRING_TDB2_UUID_NOT_UNIQUE, uuid);
 
-  // Create a vector tasks, as hooks can cause them to multiply.
-  std::vector <Task> changes;
-  changes.push_back (task);
-  context.hooks.onAdd (changes);
+  // Only locally-added tasks trigger hooks.  This means that tasks introduced
+  // via 'sync' do not trigger hooks.
+  if (add_to_backlog)
+    context.hooks.onAdd (task);
 
-  update (uuid, task, add_to_backlog, changes);
+  update (uuid, task, add_to_backlog, true);
 }
 
 ////////////////////////////////////////////////////////////////////////////////
 void TDB2::modify (Task& task, bool add_to_backlog /* = true */)
 {
-  // Ensure the task is consistent, and provide defaults if necessary.
+  // Ensure the task is consistent.
   task.validate (false);
-  task.upgradeLegacyValues ();
   std::string uuid = task.get ("uuid");
 
   // Get the unmodified task as reference, so the hook can compare.
-  Task original;
-  get (uuid, original);
+  if (add_to_backlog)
+  {
+    Task original;
+    get (uuid, original);
+    context.hooks.onModify (original, task);
+  }
 
-  // Create a vector tasks, as hooks can cause them to multiply.
-  std::vector <Task> changes;
-  changes.push_back (task);
-  context.hooks.onModify (original, changes);
-
-  update (uuid, task, add_to_backlog, changes);
+  update (uuid, task, add_to_backlog);
 }
 
 ////////////////////////////////////////////////////////////////////////////////
@@ -600,61 +610,57 @@ void TDB2::update (
   const std::string& uuid,
   Task& task,
   const bool add_to_backlog,
-  std::vector <Task>& changes)
+  const bool addition /* = false */)
 {
-  std::vector <Task>::iterator i;
-  for (i = changes.begin (); i != changes.end (); ++i)
+  // Validate to add metadata.
+  task.validate (false);
+
+  // If the task already exists, it is a modification, else addition.
+  Task original;
+  if (not addition && get (task.get ("uuid"), original))
   {
-    // Validate to add metadata.
-    i->validate (false);
-
-    // If the task already exists, it is a modification, else addition.
-    Task original;
-    if (get (i->get ("uuid"), original))
-    {
-      // Update the task, wherever it is.
-      if (!pending.modify_task (*i))
-        completed.modify_task (*i);
-
-      // time <time>
-      // old <task>
-      // new <task>
-      // ---
-      undo.add_line ("time " + Date ().toEpochString () + "\n");
-      undo.add_line ("old " + original.composeF4 () + "\n");
-      undo.add_line ("new " + i->composeF4 () + "\n");
-      undo.add_line ("---\n");
-    }
-    else
-    {
-      // Re-validate to add default values.
-      i->validate ();
-
-      // Add new task to either pending or completed.
-      std::string status = i->get ("status");
-      if (status == "completed" ||
-          status == "deleted")
-        completed.add_task (*i);
-      else
-        pending.add_task (*i);
-
-      // Add undo data lines:
-      //   time <time>
-      //   new <task>
-      //   ---
-      undo.add_line ("time " + Date ().toEpochString () + "\n");
-      undo.add_line ("new " + i->composeF4 () + "\n");
-      undo.add_line ("---\n");
-    }
-
-    // Add task to backlog.
     if (add_to_backlog)
-      backlog.add_line (i->composeJSON () + "\n");
+    {
+      // All locally modified tasks are timestamped, implicitly overwriting any
+      // changes the user or hooks tried to apply to the "modified" attribute.
+      task.setAsNow ("modified");
+    }
 
-    // The original task may be further referenced by the caller.
-    if (i->get ("uuid") == uuid)
-      task = *i;
+    // Update the task, wherever it is.
+    if (!pending.modify_task (task))
+      completed.modify_task (task);
+
+    // time <time>
+    // old <task>
+    // new <task>
+    // ---
+    undo.add_line ("time " + Date ().toEpochString () + "\n");
+    undo.add_line ("old " + original.composeF4 () + "\n");
+    undo.add_line ("new " + task.composeF4 () + "\n");
+    undo.add_line ("---\n");
   }
+  else
+  {
+    // Add new task to either pending or completed.
+    std::string status = task.get ("status");
+    if (status == "completed" ||
+        status == "deleted")
+      completed.add_task (task);
+    else
+      pending.add_task (task);
+
+    // Add undo data lines:
+    //   time <time>
+    //   new <task>
+    //   ---
+    undo.add_line ("time " + Date ().toEpochString () + "\n");
+    undo.add_line ("new " + task.composeF4 () + "\n");
+    undo.add_line ("---\n");
+  }
+
+  // Add task to backlog.
+  if (add_to_backlog)
+    backlog.add_line (task.composeJSON () + "\n");
 }
 
 ////////////////////////////////////////////////////////////////////////////////
@@ -979,6 +985,9 @@ void TDB2::show_diff (
     view.add (Column::factory ("string", STRING_TDB2_UNDO_PRIOR));
     view.add (Column::factory ("string", STRING_TDB2_UNDO_CURRENT));
 
+    Color label (context.config.get ("color.label"));
+    view.colorHeader (label);
+
     Task after (current);
 
     if (prior != "")
@@ -1189,18 +1198,21 @@ int TDB2::gc ()
   // Allowed as an override, but not recommended.
   if (context.config.getBoolean ("gc"))
   {
-    std::vector <Task> pending_tasks   = pending.get_tasks ();
+    std::vector <Task> pending_tasks = pending.get_tasks ();
+
+    // TODO Thread.
     std::vector <Task> completed_tasks = completed.get_tasks ();
 
+    // TODO Assume pending < completed, therefore there is room here to process
+    //      data before joining with the completed.data thread.
+
     bool pending_changes = false;
     bool completed_changes = false;
-
     std::vector <Task> pending_tasks_after;
     std::vector <Task> completed_tasks_after;
 
     // Reduce unnecessary allocation/copies.
     pending_tasks_after.reserve (pending_tasks.size ());
-    completed_tasks_after.reserve (completed_tasks.size ());
 
     // Scan all pending tasks, looking for any that need to be relocated to
     // completed, or need to be 'woken'.
@@ -1237,6 +1249,11 @@ int TDB2::gc ()
       }
     }
 
+    // TODO Join completed.data thread.
+
+    // Reduce unnecessary allocation/copies.
+    completed_tasks_after.reserve (completed_tasks.size ());
+
     // Scan all completed tasks, looking for any that need to be relocated to
     // pending.
     for (task = completed_tasks.begin ();

src/TDB2.h

@@ -128,7 +128,7 @@ public:
 
 private:
   void gather_changes ();
-  void update (const std::string&, Task&, const bool, std::vector <Task>&);
+  void update (const std::string&, Task&, const bool, const bool addition = false);
   bool verifyUniqueUUID (const std::string&);
   void show_diff (const std::string&, const std::string&, const std::string&);
   void revert_undo (std::vector <std::string>&, std::string&, std::string&, std::string&, std::string&);

src/TLSClient.cpp

@@ -298,7 +298,6 @@ int TLSClient::verify_certificate () const
   // This verification function uses the trusted CAs in the credentials
   // structure. So you must have installed one or more CA certificates.
   unsigned int status = 0;
-
   const char* hostname = _host.c_str();
 #if GNUTLS_VERSION_NUMBER >= 0x030104
   if (_trust == TLSClient::ignore_hostname)
@@ -350,7 +349,7 @@ int TLSClient::verify_certificate () const
         if (_debug)
           std::cout << "c: ERROR x509 cert import. " << gnutls_strerror (ret) << "\n";
         gnutls_x509_crt_deinit(cert);
-        status = GNUTLS_E_CERTIFICATE_ERROR;
+        return GNUTLS_E_CERTIFICATE_ERROR;
       }
 
       if (gnutls_x509_crt_check_hostname (cert, hostname) == 0)
@@ -366,7 +365,7 @@ int TLSClient::verify_certificate () const
   }
 #endif
 
-#if GNUTLS_VERSION_NUMBER >= 0x030105
+#if GNUTLS_VERSION_NUMBER >= 0x030104
   gnutls_certificate_type_t type = gnutls_certificate_type_get (_session);
   gnutls_datum_t out;
   ret = gnutls_certificate_verification_status_print (status, type, &out, 0);
@@ -377,6 +376,8 @@ int TLSClient::verify_certificate () const
     return GNUTLS_E_CERTIFICATE_ERROR;
   }
 
+  if (_debug)
+    std::cout << "c: INFO " << out.data << "\n";
   gnutls_free (out.data);
 #endif
 

src/Task.cpp

@@ -65,14 +65,12 @@ static const float epsilon = 0.000001;
 #endif
 
 std::string Task::defaultProject  = "";
-std::string Task::defaultPriority = "";
 std::string Task::defaultDue      = "";
 bool Task::searchCaseSensitive    = true;
 bool Task::regex                  = false;
 std::map <std::string, std::string> Task::attributes;
 
 std::map <std::string, float> Task::coefficients;
-float Task::urgencyPriorityCoefficient    = 0.0;
 float Task::urgencyProjectCoefficient     = 0.0;
 float Task::urgencyActiveCoefficient      = 0.0;
 float Task::urgencyScheduledCoefficient   = 0.0;
@@ -87,6 +85,8 @@ float Task::urgencyBlockingCoefficient    = 0.0;
 float Task::urgencyAgeCoefficient         = 0.0;
 float Task::urgencyAgeMax                 = 0.0;
 
+std::map <std::string, std::vector <std::string>> Task::customOrder;
+
 static const std::string dummy ("");
 
 ////////////////////////////////////////////////////////////////////////////////
@@ -565,7 +565,6 @@ void Task::parse (const std::string& input)
               nl.getQuoted ('"', value))
           {
             legacyAttributeMap (name);
-            legacyValueMap (name, value);
 
             if (name.substr (0, 11) == "annotation_")
               ++annotation_count;
@@ -586,8 +585,6 @@ void Task::parse (const std::string& input)
       parseJSON (copy);
     else
       throw std::string (STRING_RECORD_NOT_FF4);
-
-    upgradeLegacyValues ();
   }
 
   catch (const std::string&)
@@ -636,8 +633,9 @@ void Task::parseJSON (const std::string& line)
         // Dates are converted from ISO to epoch.
         else if (type == "date")
         {
-          Date d (unquoteText (i->second->dump ()));
-          set (i->first, d.toEpochString ());
+          std::string text = unquoteText (i->second->dump ());
+          Date d (text);
+          set (i->first, text == "" ? "" : d.toEpochString ());
         }
 
         // Tags are an array of JSON strings.
@@ -718,8 +716,6 @@ void Task::parseJSON (const std::string& line)
         }
       }
     }
-
-    upgradeLegacyValues ();
   }
 }
 
@@ -804,6 +800,10 @@ std::string Task::composeJSON (bool decorate /*= false*/) const
     if (i->first.substr (0, 11) == "annotation_")
       continue;
 
+    // If value is an empty string, do not ever output it
+    if (i->second == "")
+        continue;
+
     if (attributes_written)
       out << ",";
 
@@ -815,16 +815,22 @@ std::string Task::composeJSON (bool decorate /*= false*/) const
     if (type == "date")
     {
       Date d (i->second);
-      if (i->first == "modification")
-        out << "\"modified\":\""
-            << d.toISO ()
-            << "\"";
-      else
-        out << "\""
-            << i->first
-            << "\":\""
-            << d.toISO ()
-            << "\"";
+      out << "\""
+          << (i->first == "modification" ? "modified" : i->first)
+          << "\":\""
+          // Date was deleted, do not export parsed empty string
+          << (i->second == "" ? "" : d.toISO ())
+          << "\"";
+
+      ++attributes_written;
+    }
+
+    else if (type == "numeric")
+    {
+      out << "\""
+          << i->first
+          << "\":"
+          << i->second;
 
       ++attributes_written;
     }
@@ -847,6 +853,8 @@ std::string Task::composeJSON (bool decorate /*= false*/) const
       }
 
       out << "]";
+
+      ++attributes_written;
     }
 
     // Everything else is a quoted value.
@@ -894,9 +902,8 @@ std::string Task::composeJSON (bool decorate /*= false*/) const
   // Include urgency.
   if (decorate)
     out << ","
-        << "\"urgency\":\""
-        << urgency_c ()
-        <<"\"";
+        << "\"urgency\":"
+        << urgency_c ();
 #endif
 
   out << "}";
@@ -1093,6 +1100,7 @@ int Task::getTagCount () const
 bool Task::hasTag (const std::string& tag) const
 {
   // Synthetic tags - dynamically generated, but do not occupy storage space.
+  // Note: This list must match that in CmdInfo::execute.
   if (tag == "BLOCKED")   return is_blocked;
   if (tag == "UNBLOCKED") return !is_blocked;
   if (tag == "BLOCKING")  return is_blocking;
@@ -1112,10 +1120,10 @@ bool Task::hasTag (const std::string& tag) const
   if (tag == "SCHEDULED") return has ("scheduled");
   if (tag == "CHILD")     return has ("parent");
   if (tag == "UNTIL")     return has ("until");
-  if (tag == "WAITING")   return has ("wait");
   if (tag == "ANNOTATED") return hasAnnotations ();
   if (tag == "TAGGED")    return has ("tags");
   if (tag == "PARENT")    return has ("mask");
+  if (tag == "WAITING")   return get ("status") == "waiting";
   if (tag == "PENDING")   return get ("status") == "pending";
   if (tag == "COMPLETED") return get ("status") == "completed";
   if (tag == "DELETED")   return get ("status") == "deleted";
@@ -1376,7 +1384,7 @@ void Task::validate (bool applyDefault /* = true */)
   if (!has ("modified") || get ("modified") == "")
     setAsNow ("modified");
 
-  if (applyDefault)
+  if (applyDefault && (! has ("parent") || get ("parent") == ""))
   {
     // Override with default.project, if not specified.
     if (Task::defaultProject != "" &&
@@ -1386,14 +1394,6 @@ void Task::validate (bool applyDefault /* = true */)
         set ("project", Task::defaultProject);
     }
 
-    // Override with default.priority, if not specified.
-    if (Task::defaultPriority != "" &&
-        ! has ("priority"))
-    {
-      if (context.columns["priority"]->validate (Task::defaultPriority))
-        set ("priority", Task::defaultPriority);
-    }
-
     // Override with default.due, if not specified.
     if (Task::defaultDue != "" &&
         ! has ("due"))
@@ -1479,17 +1479,6 @@ void Task::validate (bool applyDefault /* = true */)
         throw std::string (format (STRING_TASK_VALID_RECUR, value));
     }
   }
-
-  // Priorities must be valid.
-  if (has ("priority"))
-  {
-    std::string priority = get ("priority");
-    if (priority != "H" &&
-        priority != "M" &&
-        priority != "L" &&
-        priority != "")
-      throw format (STRING_TASK_VALID_PRIORITY, priority);
-  }
 }
 
 ////////////////////////////////////////////////////////////////////////////////
@@ -1647,7 +1636,6 @@ float Task::urgency_c () const
 {
   float value = 0.0;
 #ifdef PRODUCT_TASKWARRIOR
-  value += fabsf (Task::urgencyPriorityCoefficient)    > epsilon ? (urgency_priority ()    * Task::urgencyPriorityCoefficient)    : 0.0;
   value += fabsf (Task::urgencyProjectCoefficient)     > epsilon ? (urgency_project ()     * Task::urgencyProjectCoefficient)     : 0.0;
   value += fabsf (Task::urgencyActiveCoefficient)      > epsilon ? (urgency_active ()      * Task::urgencyActiveCoefficient)      : 0.0;
   value += fabsf (Task::urgencyScheduledCoefficient)   > epsilon ? (urgency_scheduled ()   * Task::urgencyScheduledCoefficient)   : 0.0;
@@ -1689,6 +1677,16 @@ float Task::urgency_c () const
           if (hasTag (tag))
             value += var->second;
         }
+
+        // urgency.user.keyword.<keyword>.coefficient
+        if (var->first.substr (13, 8) == "keyword." &&
+            (end = var->first.find (".coefficient")) != std::string::npos)
+        {
+          std::string keyword = var->first.substr (21, end - 21);
+
+          if (get ("description").find (keyword) != std::string::npos)
+            value += var->second;
+        }
       }
       else if (var->first.substr (0, 12) == "urgency.uda.")
       {
@@ -1734,18 +1732,6 @@ float Task::urgency ()
   return urgency_value;
 }
 
-////////////////////////////////////////////////////////////////////////////////
-float Task::urgency_priority () const
-{
-  const std::string& value = get_ref ("priority");
-
-       if (value == "H") return 1.0;
-  else if (value == "M") return 0.65;
-  else if (value == "L") return 0.3;
-
-  return 0.0;
-}
-
 ////////////////////////////////////////////////////////////////////////////////
 float Task::urgency_inherit () const
 {
@@ -1767,7 +1753,6 @@ float Task::urgency_inherit () const
     v += it->urgency_annotations ();
     v += it->urgency_due ();
     v += it->urgency_next ();
-    v += it->urgency_priority ();
     v += it->urgency_scheduled ();
     v += it->urgency_waiting ();
 
@@ -1948,17 +1933,13 @@ void Task::modify (modType type, bool text_required /* = false */)
         }
         else
         {
-          // Some columns are not modifiable.
-          if (name == "uuid"  ||
-              name == "id"    ||
-              name == "mask"  ||
-              name == "imask" ||
-              name == "parent")
-            throw format (STRING_INVALID_MOD, name, value);
-
           // Get the column info.
           Column* column = context.columns[name];
 
+          // Some columns are not modifiable.
+          if (! column->modifiable ())
+            throw format (STRING_INVALID_MOD, name, value);
+
           // Dependencies are specified as IDs.
           if (name == "depends")
           {
@@ -2068,9 +2049,6 @@ void Task::modify (modType type, bool text_required /* = false */)
             if (v.type () == Variant::type_string)
               throw format (STRING_UDA_NUMERIC, v.get_string ());
 
-            v.cast (Variant::type_real);
-            v.cast (Variant::type_string);
-
             set (name, v);
             ++modCount;
           }
@@ -2209,78 +2187,3 @@ void Task::modify (modType type, bool text_required /* = false */)
 }
 
 ////////////////////////////////////////////////////////////////////////////////
-void Task::upgradeLegacyValues ()
-{
-  // 2.4.0 Update recurrence values.
-  if (has ("recur"))
-  {
-    std::string value = get ("recur");
-    if (value != "")
-    {
-      std::string new_value = value;
-      upgradeLegacyValue (new_value);
-
-      if (new_value != value)
-      {
-        set ("recur", new_value);
-        context.debug (format ("Legacy upgrade: recur {1} --> {2}", value, new_value));
-      }
-    }
-  }
-
-  // 2.4.0 Update UDA duration values.
-  Config::const_iterator name;
-  for (name = context.config.begin (); name != context.config.end (); ++name)
-  {
-    if (name->first.substr (0, 4) == "uda." &&
-        name->first.find (".type") != std::string::npos)
-    {
-      if (name->second == "duration")
-      {
-        std::string::size_type period = name->first.find ('.', 4);
-        if (period != std::string::npos)
-        {
-          std::string uda = name->first.substr (4, period - 4);
-          std::string value = get (uda);
-          std::string new_value = value;
-          upgradeLegacyValue (new_value);
-
-          if (new_value != value)
-          {
-            set ("recur", new_value);
-            context.debug (format ("Legacy upgrade: UDA {1}, {2} --> {3}", uda, value, new_value));
-          }
-        }
-      }
-    }
-  }
-}
-
-////////////////////////////////////////////////////////////////////////////////
-void Task::upgradeLegacyValue (std::string& value)
-{
-  std::string::size_type len = value.length ();
-  std::string::size_type p;
-
-       if (value == "-")                                                    value = "0s";
-  else if ((p = value.find ("hr"))    != std::string::npos && p == len - 2) value = value.substr (0, p) + "h";
-  else if ((p = value.find ("hrs"))   != std::string::npos && p == len - 3) value = value.substr (0, p) + "h";
-  else if ((p = value.find ("mins"))  != std::string::npos && p == len - 4) value = value.substr (0, p) + "min";
-  else if ((p = value.find ("mnths")) != std::string::npos && p == len - 5) value = value.substr (0, p) + "mo";
-  else if ((p = value.find ("mos"))   != std::string::npos && p == len - 3) value = value.substr (0, p) + "mo";
-  else if ((p = value.find ("mth"))   != std::string::npos && p == len - 3) value = value.substr (0, p) + "mo";
-  else if ((p = value.find ("mths"))  != std::string::npos && p == len - 4) value = value.substr (0, p) + "mo";
-  else if ((p = value.find ("qrtrs")) != std::string::npos && p == len - 5) value = value.substr (0, p) + "q";
-  else if ((p = value.find ("qtr"))   != std::string::npos && p == len - 3) value = value.substr (0, p) + "q";
-  else if ((p = value.find ("qtrs"))  != std::string::npos && p == len - 4) value = value.substr (0, p) + "q";
-  else if ((p = value.find ("sec"))   != std::string::npos && p == len - 3) value = value.substr (0, p) + "s";
-  else if ((p = value.find ("secs"))  != std::string::npos && p == len - 4) value = value.substr (0, p) + "s";
-  else if ((p = value.find ("wk"))    != std::string::npos && p == len - 2) value = value.substr (0, p) + "w";
-  else if ((p = value.find ("wks"))   != std::string::npos && p == len - 3) value = value.substr (0, p) + "w";
-  else if ((p = value.find ("yr"))    != std::string::npos && p == len - 2) value = value.substr (0, p) + "y";
-  else if ((p = value.find ("yrs"))   != std::string::npos && p == len - 3) value = value.substr (0, p) + "y";
-
-  // It is not an error to have a non-legacy value.
-}
-
-////////////////////////////////////////////////////////////////////////////////

src/Task.h

@@ -37,13 +37,12 @@ class Task : public std::map <std::string, std::string>
 {
 public:
   static std::string defaultProject;
-  static std::string defaultPriority;
   static std::string defaultDue;
   static bool searchCaseSensitive;
   static bool regex;
   static std::map <std::string, std::string> attributes;  // name -> type
   static std::map <std::string, float> coefficients;
-  static float urgencyPriorityCoefficient;
+  static std::map <std::string, std::vector <std::string>> customOrder;
   static float urgencyProjectCoefficient;
   static float urgencyActiveCoefficient;
   static float urgencyScheduledCoefficient;
@@ -154,7 +153,6 @@ public:
 
   enum modType {modReplace, modPrepend, modAppend, modAnnotate};
   void modify (modType, bool text_required = false);
-  void upgradeLegacyValues ();
 
 private:
   int determineVersion (const std::string&);
@@ -163,10 +161,8 @@ private:
   void validate_before (const std::string&, const std::string&);
   const std::string encode (const std::string&) const;
   const std::string decode (const std::string&) const;
-  void upgradeLegacyValue (std::string&);
 
 public:
-  float urgency_priority () const;
   float urgency_project () const;
   float urgency_active () const;
   float urgency_scheduled () const;

src/Variant.cpp

@@ -26,6 +26,7 @@
 
 #include <cmake.h>
 #include <sstream>
+#include <algorithm>
 #include <stdio.h>
 #include <math.h>
 #include <stdlib.h>
@@ -41,6 +42,7 @@
 std::string Variant::dateFormat = "";
 bool Variant::searchCaseSensitive = true;
 bool Variant::searchUsingRegex = true;
+bool Variant::isoEnabled = true;
 
 ////////////////////////////////////////////////////////////////////////////////
 Variant::Variant ()
@@ -134,7 +136,7 @@ Variant::Variant (const char* value)
 }
 
 ////////////////////////////////////////////////////////////////////////////////
-Variant::Variant (const time_t value, const enum type new_type /*=type_date*/)
+Variant::Variant (const time_t value, const enum type new_type)
 : _type (new_type)
 , _bool (false)
 , _integer (0)
@@ -165,7 +167,7 @@ void Variant::source (const std::string& input)
 }
 
 ////////////////////////////////////////////////////////////////////////////////
-std::string Variant::source () const
+const std::string& Variant::source () const
 {
   return _source;
 }
@@ -313,19 +315,25 @@ bool Variant::operator< (const Variant& other) const
       return left._string < right._string;
 
     case type_string:
-      if (left.source () == "priority" || right.source () == "priority")
       {
-             if (left._string != "H" && right._string == "H") return true;
-        else if (left._string == "L" && right._string == "M") return true;
-        else if (left._string == ""  && right._string != "")  return true;
-        else                                                  return false;
-      }
-      else
-      {
-        if (left.trivial () || right.trivial ())
+        if (left._string == right._string)
           return false;
 
-        return left._string < right._string;
+        auto order = Task::customOrder.find (left.source ());
+        if (order != Task::customOrder.end ())
+        {
+          // Guaranteed to be found, because of ColUDA::validate ().
+          auto posLeft  = std::find (order->second.begin (), order->second.end (), left._string);
+          auto posRight = std::find (order->second.begin (), order->second.end (), right._string);
+          return posLeft < posRight;
+        }
+        else
+        {
+          if (left.trivial () || right.trivial ())
+            return false;
+
+          return left._string < right._string;
+        }
       }
 
     case type_date:
@@ -458,20 +466,25 @@ bool Variant::operator<= (const Variant& other) const
       return left._string <= right._string;
 
     case type_string:
-      if (left.source () == "priority" || right.source () == "priority")
       {
-             if (left._string        == right._string       ) return true;
-        else if (                       right._string == "H") return true;
-        else if (left._string == "L" && right._string == "M") return true;
-        else if (left._string == ""                         ) return true;
-        else                                                  return false;
-      }
-      else
-      {
-        if (left.trivial () || right.trivial ())
-          return false;
+        if (left._string == right._string)
+          return true;
 
-        return left._string <= right._string;
+        auto order = Task::customOrder.find (left.source ());
+        if (order != Task::customOrder.end ())
+        {
+          // Guaranteed to be found, because of ColUDA::validate ().
+          auto posLeft  = std::find (order->second.begin (), order->second.end (), left._string);
+          auto posRight = std::find (order->second.begin (), order->second.end (), right._string);
+          return posLeft <= posRight;
+        }
+        else
+        {
+          if (left.trivial () || right.trivial ())
+            return false;
+
+          return left._string <= right._string;
+        }
       }
 
     case type_date:
@@ -604,20 +617,27 @@ bool Variant::operator> (const Variant& other) const
       return left._string > right._string;
 
     case type_string:
-      if (left.source () == "priority" || right.source () == "priority")
       {
-             if (left._string == "H" && right._string != "H") return true;
-        else if (left._string == "M" && right._string == "L") return true;
-        else if (left._string != ""  && right._string == "")  return true;
-        else                                                  return false;
-      }
-      else
-      {
-        if (left.trivial () || right.trivial ())
+        if (left._string == right._string)
           return false;
 
-        return left._string> right._string;
+        auto order = Task::customOrder.find (left.source ());
+        if (order != Task::customOrder.end ())
+        {
+          // Guaranteed to be found, because of ColUDA::validate ().
+          auto posLeft  = std::find (order->second.begin (), order->second.end (), left._string);
+          auto posRight = std::find (order->second.begin (), order->second.end (), right._string);
+          return posLeft > posRight;
+        }
+        else
+        {
+          if (left.trivial () || right.trivial ())
+            return false;
+
+          return left._string > right._string;
+        }
       }
+
     case type_date:
       if (left.trivial () || right.trivial ())
         return false;
@@ -748,20 +768,25 @@ bool Variant::operator>= (const Variant& other) const
       return left._string >= right._string;
 
     case type_string:
-      if (left.source () == "priority" || right.source () == "priority")
       {
-             if (left._string        == right._string       ) return true;
-        else if (left._string == "H"                        ) return true;
-        else if (left._string == "M" && right._string == "L") return true;
-        else if (                       right._string == "" ) return true;
-        else                                                  return false;
-      }
-      else
-      {
-        if (left.trivial () || right.trivial ())
-          return false;
+        if (left._string == right._string)
+          return true;
 
-        return left._string >= right._string;
+        auto order = Task::customOrder.find (left.source ());
+        if (order != Task::customOrder.end ())
+        {
+          // Guaranteed to be found, because of ColUDA::validate ().
+          auto posLeft  = std::find (order->second.begin (), order->second.end (), left._string);
+          auto posRight = std::find (order->second.begin (), order->second.end (), right._string);
+          return posLeft >= posRight;
+        }
+        else
+        {
+          if (left.trivial () || right.trivial ())
+            return false;
+
+          return left._string >= right._string;
+        }
       }
 
     case type_date:
@@ -1963,6 +1988,8 @@ Variant::operator std::string () const
   case type_unknown:
     throw std::string (STRING_VARIANT_RENDER_UNK);
   }
+
+  return "";
 }
 
 ////////////////////////////////////////////////////////////////////////////////
@@ -2065,7 +2092,8 @@ void Variant::cast (const enum type new_type)
 
         ISO8601d iso;
         std::string::size_type pos = 0;
-        if (iso.parse (_string, pos) &&
+        if (isoEnabled               &&
+            iso.parse (_string, pos) &&
             pos == _string.length ())
         {
           _date = (time_t) iso;
@@ -2170,7 +2198,7 @@ double Variant::get_real () const
 }
 
 ////////////////////////////////////////////////////////////////////////////////
-std::string Variant::get_string () const
+const std::string& Variant::get_string () const
 {
   return _string;
 }

src/Variant.h

@@ -27,6 +27,7 @@
 #ifndef INCLUDED_VARIANT
 #define INCLUDED_VARIANT
 
+#include <map>
 #include <string>
 #include <time.h>
 #include <Task.h>
@@ -37,6 +38,7 @@ public:
   static std::string dateFormat;
   static bool searchCaseSensitive;
   static bool searchUsingRegex;
+  static bool isoEnabled;
 
   enum type {type_unknown, type_boolean, type_integer, type_real, type_string,
              type_date, type_duration};
@@ -48,11 +50,11 @@ public:
   Variant (const double);
   Variant (const std::string&);
   Variant (const char*);
-  Variant (const time_t, const enum type new_type = type_date);
+  Variant (const time_t, const enum type);
   ~Variant ();
 
   void source (const std::string&);
-  std::string source () const;
+  const std::string& source () const;
 
   Variant& operator= (const Variant&);
 
@@ -98,12 +100,12 @@ public:
   int type ();
   bool trivial () const;
 
-  bool        get_bool () const;
-  int         get_integer () const;
-  double      get_real () const;
-  std::string get_string () const;
-  time_t      get_date () const;
-  time_t      get_duration () const;
+  bool               get_bool () const;
+  int                get_integer () const;
+  double             get_real () const;
+  const std::string& get_string () const;
+  time_t             get_date () const;
+  time_t             get_duration () const;
 
 private:
   enum type   _type;

src/ViewTask.cpp

@@ -114,6 +114,7 @@ std::string ViewTask::render (std::vector <Task>& data, std::vector <int>& seque
 {
   context.timer_render.start ();
 
+  bool const obfuscate           = context.config.getBoolean ("obfuscate");
   bool const print_empty_columns = context.config.getBoolean ("print.empty.columns");
   std::vector <Column*> nonempty_columns;
   std::vector <bool> nonempty_sort;
@@ -143,6 +144,11 @@ std::string ViewTask::render (std::vector <Task>& data, std::vector <int>& seque
 
       if (min   > global_min)   global_min   = min;
       if (ideal > global_ideal) global_ideal = ideal;
+
+      // If a fixed-width column was just measured, there is no point repeating
+      // the measurement for all tasks.
+      if (_columns[i]->is_fixed_width ())
+        break;
     }
 
     if (print_empty_columns || global_min != 0)
@@ -237,7 +243,7 @@ std::string ViewTask::render (std::vector <Task>& data, std::vector <int>& seque
 
   // Compose column headers.
   unsigned int max_lines = 0;
-  std::vector <std::vector <std::string> > headers;
+  std::vector <std::vector <std::string>> headers;
   for (unsigned int c = 0; c < _columns.size (); ++c)
   {
     headers.push_back (std::vector <std::string> ());
@@ -290,7 +296,7 @@ std::string ViewTask::render (std::vector <Task>& data, std::vector <int>& seque
 
   // Compose, render columns, in sequence.
   _rows = 0;
-  std::vector <std::vector <std::string> > cells;
+  std::vector <std::vector <std::string>> cells;
   std::vector <int>::iterator s;
   for (unsigned int s = 0; s < sequence.size (); ++s)
   {
@@ -316,6 +322,11 @@ std::string ViewTask::render (std::vector <Task>& data, std::vector <int>& seque
 
       if (cells[c].size () > max_lines)
         max_lines = cells[c].size ();
+
+      if (obfuscate)
+        if (_columns[c]->type () == "string")
+          for (unsigned int line = 0; line < cells[c].size (); ++line)
+            cells[c][line] = obfuscateText (cells[c][line]);
     }
 
     // Listing breaks are simply blank lines inserted when a column value

src/ViewText.cpp

@@ -114,6 +114,8 @@ void ViewText::set (int row, int col, Color color)
 ////////////////////////////////////////////////////////////////////////////////
 std::string ViewText::render ()
 {
+  bool const obfuscate = context.config.getBoolean ("obfuscate");
+
   // Determine minimal, ideal column widths.
   std::vector <int> minimal;
   std::vector <int> ideal;
@@ -132,6 +134,11 @@ std::string ViewText::render ()
 
       if (min   > global_min)   global_min = min;
       if (ideal > global_ideal) global_ideal = ideal;
+
+      // If a fixed-width column was just measured, there is no point repeating
+      // the measurement for all tasks.
+      if (_columns[col]->is_fixed_width ())
+        break;
     }
 
     minimal.push_back (global_min);
@@ -181,7 +188,7 @@ std::string ViewText::render ()
 
   // Compose column headers.
   unsigned int max_lines = 0;
-  std::vector <std::vector <std::string> > headers;
+  std::vector <std::vector <std::string>> headers;
   for (unsigned int c = 0; c < _columns.size (); ++c)
   {
     headers.push_back (std::vector <std::string> ());
@@ -233,7 +240,7 @@ std::string ViewText::render ()
 
   // Compose, render columns, in sequence.
   _rows = 0;
-  std::vector <std::vector <std::string> > cells;
+  std::vector <std::vector <std::string>> cells;
   for (unsigned int row = 0; row < _data.size (); ++row)
   {
     max_lines = 0;
@@ -260,6 +267,11 @@ std::string ViewText::render ()
 
       if (cells[col].size () > max_lines)
         max_lines = cells[col].size ();
+
+      if (obfuscate)
+        if (_columns[col]->type () == "string")
+          for (unsigned int line = 0; line < cells[col].size (); ++line)
+            cells[col][line] = obfuscateText (cells[col][line]);
     }
 
     for (unsigned int i = 0; i < max_lines; ++i)

src/ViewText.h

@@ -68,24 +68,24 @@ public:
   std::string render ();
 
 private:
-  std::vector <std::vector <std::string> > _data;
-  std::vector <std::vector <Color> >       _color;
-  std::vector <Column*>                    _columns;
-  int                                      _width;
-  int                                      _left_margin;
-  Color                                    _header;
-  Color                                    _odd;
-  Color                                    _even;
-  int                                      _intra_padding;
-  Color                                    _intra_odd;
-  Color                                    _intra_even;
-  int                                      _extra_padding;
-  Color                                    _extra_odd;
-  Color                                    _extra_even;
-  int                                      _truncate_lines;
-  int                                      _truncate_rows;
-  int                                      _lines;
-  int                                      _rows;
+  std::vector <std::vector <std::string>> _data;
+  std::vector <std::vector <Color>>       _color;
+  std::vector <Column*>                   _columns;
+  int                                     _width;
+  int                                     _left_margin;
+  Color                                   _header;
+  Color                                   _odd;
+  Color                                   _even;
+  int                                     _intra_padding;
+  Color                                   _intra_odd;
+  Color                                   _intra_even;
+  int                                     _extra_padding;
+  Color                                   _extra_odd;
+  Color                                   _extra_even;
+  int                                     _truncate_lines;
+  int                                     _truncate_rows;
+  int                                     _lines;
+  int                                     _rows;
 };
 
 #endif

src/args.cpp

@@ -135,7 +135,6 @@ int main (int argc, const char** argv)
     cli.entity ("attribute",  "mask");
     cli.entity ("attribute",  "modified");
     cli.entity ("attribute",  "parent");
-    cli.entity ("attribute",  "priority");
     cli.entity ("attribute",  "project");
     cli.entity ("attribute",  "recur");
     cli.entity ("attribute",  "scheduled");

src/calc.cpp

@@ -57,7 +57,7 @@ int main (int argc, char** argv)
   try
   {
     bool infix     = true;
-    bool ambiguous = true;
+    bool ambiguous = false;
 
     // Add a source for constants.
     Eval e;
@@ -77,8 +77,8 @@ int main (int argc, char** argv)
                   << "  -d|--debug        Debug mode\n"
                   << "  -i|--infix        Infix expression (default)\n"
                   << "  -p|--postfix      Postfix expression\n"
-                  << "  -a|--ambiguous    Choose dates over numbers when ambiguous (default)\n"
-                  << "  -n|--noambiguous  Choose numbers over dates when ambiguous\n"
+                  << "  -a|--ambiguous    Choose dates over numbers when ambiguous\n"
+                  << "  -n|--noambiguous  Choose numbers over dates when ambiguous (default)\n"
                   << "\n";
         exit (1);
       }

src/columns/CMakeLists.txt

@@ -17,7 +17,6 @@ set (columns_SRCS Column.cpp Column.h
                   ColMask.cpp ColMask.h
                   ColModified.cpp ColModified.h
                   ColParent.cpp ColParent.h
-                  ColPriority.cpp ColPriority.h
                   ColProject.cpp ColProject.h
                   ColRecur.cpp ColRecur.h
                   ColScheduled.cpp ColScheduled.h

src/columns/ColDate.cpp

@@ -207,8 +207,6 @@ void ColumnDate::render (
           color.colorize (
             rightJustify (
               Duration (date - now).format (), width)));
-      else
-        lines.push_back (rightJustify ("", width));
     }
   }
 }

src/columns/ColDate.h

@@ -37,7 +37,7 @@ class ColumnDate : public Column
 {
 public:
   ColumnDate ();
-  ~ColumnDate ();
+  virtual ~ColumnDate ();
 
   virtual bool validate (std::string&);
   virtual void measure (Task&, unsigned int&, unsigned int&);

src/columns/ColDepends.cpp

@@ -83,8 +83,15 @@ void ColumnDepends::measure (Task& task, unsigned int& minimum, unsigned int& ma
   std::vector <Task> blocking;
   dependencyGetBlocking (task, blocking);
 
-       if (_style == "indicator") minimum = maximum = utf8_width (context.config.get ("dependency.indicator"));
-  else if (_style == "count")     minimum = maximum = 2 + format ((int) blocking.size ()).length ();
+  if (_style == "indicator")
+  {
+    minimum = maximum = utf8_width (context.config.get ("dependency.indicator"));
+    _fixed_width = true;
+  }
+  else if (_style == "count")
+  {
+    minimum = maximum = 2 + format ((int) blocking.size ()).length ();
+  }
   else if (_style == "default" ||
            _style == "list")
   {

src/columns/ColIMask.cpp

@@ -62,11 +62,16 @@ bool ColumnIMask::validate (std::string& value)
 // Set the minimum and maximum widths for the value.
 void ColumnIMask::measure (Task& task, unsigned int& minimum, unsigned int& maximum)
 {
-  minimum = maximum = task.get ("imask").length ();
+  minimum = maximum = 0;
 
-  if (_style != "default" &&
-      _style != "number")
-    throw format (STRING_COLUMN_BAD_FORMAT, _name, _style);
+  if (task.has (_name))
+  {
+    minimum = maximum = task.get ("imask").length ();
+
+    if (_style != "default" &&
+        _style != "number")
+      throw format (STRING_COLUMN_BAD_FORMAT, _name, _style);
+  }
 }
 
 ////////////////////////////////////////////////////////////////////////////////
@@ -76,7 +81,8 @@ void ColumnIMask::render (
   int width,
   Color& color)
 {
-  lines.push_back (color.colorize (rightJustify (task.get ("imask"), width)));
+  if (task.has (_name))
+    lines.push_back (color.colorize (rightJustify (task.get ("imask"), width)));
 }
 
 ////////////////////////////////////////////////////////////////////////////////

src/columns/ColMask.cpp

@@ -62,10 +62,14 @@ bool ColumnMask::validate (std::string& value)
 // Set the minimum and maximum widths for the value.
 void ColumnMask::measure (Task& task, unsigned int& minimum, unsigned int& maximum)
 {
-  minimum = maximum = task.get ("mask").length ();
+  minimum = maximum = 0;
+  if (task.has (_name))
+  {
+    minimum = maximum = task.get ("mask").length ();
 
-  if (_style != "default")
-    throw format (STRING_COLUMN_BAD_FORMAT, _name, _style);
+    if (_style != "default")
+      throw format (STRING_COLUMN_BAD_FORMAT, _name, _style);
+  }
 }
 
 ////////////////////////////////////////////////////////////////////////////////
@@ -75,7 +79,8 @@ void ColumnMask::render (
   int width,
   Color& color)
 {
-  lines.push_back (color.colorize (task.get ("mask")));
+  if (task.has (_name))
+    lines.push_back (color.colorize (task.get ("mask")));
 }
 
 ////////////////////////////////////////////////////////////////////////////////

src/columns/ColParent.cpp

@@ -62,12 +62,17 @@ bool ColumnParent::validate (std::string& value)
 
 ////////////////////////////////////////////////////////////////////////////////
 // Set the minimum and maximum widths for the value.
-void ColumnParent::measure (Task&, unsigned int& minimum, unsigned int& maximum)
+void ColumnParent::measure (Task& task, unsigned int& minimum, unsigned int& maximum)
 {
-       if (_style == "default" || _style == "long") minimum = maximum = 36;
-  else if (_style == "short")                       minimum = maximum = 8;
-  else
-    throw format (STRING_COLUMN_BAD_FORMAT, _name, _style);
+  minimum = maximum = 0;
+
+  if (task.has (_name))
+  {
+         if (_style == "default" || _style == "long") minimum = maximum = 36;
+    else if (_style == "short")                       minimum = maximum = 8;
+    else
+      throw format (STRING_COLUMN_BAD_FORMAT, _name, _style);
+  }
 }
 
 ////////////////////////////////////////////////////////////////////////////////
@@ -77,20 +82,23 @@ void ColumnParent::render (
   int width,
   Color& color)
 {
-  // f30cb9c3-3fc0-483f-bfb2-3bf134f00694  default
-  //                             34f00694  short
-  if (_style == "default" ||
-      _style == "long")
+  if (task.has (_name))
   {
-    lines.push_back (color.colorize (leftJustify (task.get (_name), width)));
-  }
+    // f30cb9c3-3fc0-483f-bfb2-3bf134f00694  default
+    //                             34f00694  short
+    if (_style == "default" ||
+        _style == "long")
+    {
+      lines.push_back (color.colorize (leftJustify (task.get (_name), width)));
+    }
 
-  else if (_style == "short")
-  {
-    if (task.has (_name))
-      lines.push_back (color.colorize (leftJustify (task.get (_name).substr (28), width)));
-    else
-      lines.push_back (color.colorize (leftJustify ("", width)));
+    else if (_style == "short")
+    {
+      if (task.has (_name))
+        lines.push_back (color.colorize (leftJustify (task.get (_name).substr (28), width)));
+      else
+        lines.push_back (color.colorize (leftJustify ("", width)));
+    }
   }
 }