Enhancement - related to, but not fixing bug #293

- Added new attribute modifiers 'word' and 'noword' which find the existence
  of whole words, or prove the non-existence of whole words.  If a task has
  the description "Pay the bill", then "description.word:the" will match, but
  "description.word:th" will not.  For partial word matches, there is still
  "description.contains:th".
- Added unit tests for the text.cpp functions.
- Added unit tests including the new modifiers in filters.
- Added unit tests to parse the new modifiers.
- Modified man page.
- Modified the Context::autoFilter processing to use the new modifiers for
  +tag and -tag filtering.
- Added a support email to an error message, while looking at the filter code.
- Added new modifiers to the help report.
- Modified a utf8.t unit test to include an alphanumeric tag, rather than a
  smiley face.

ChangeLog

@@ -10,6 +10,11 @@
  + Added feature #341 that makes explicit references to the task and taskrc
    man pages, both in the auto-generated .taskrc file and the version command
    output (thanks to Cory Donnelly).
+ + Added new attribute modifiers 'word' and 'noword' which find the existence
+   of whole words, or prove the non-existence of whole words.  If a task has
+   the description "Pay the bill", then "description.word:the" will match, but
+   "description.word:th" will not.  For partial word matches, there is still
+   "description.contains:th".
  + Fixed bug that showed a calendar for the year 2037 when 'task calendar due'
    was run, and there are no tasks with due dates.
  + Fixed bug #316 which caused the timesheet report to display an oddly sorted

doc/man/task.1

@@ -266,6 +266,10 @@ Attribute modifiers improve filters.  Supported modifiers are:
 .B  startswith (synonym left)
 .br
 .B  endswith   (synonym right)
+.br
+.B  word
+.br
+.B  noword
 .RE
 
 For example:

src/Att.cpp

@@ -77,6 +77,8 @@ static const char* modifierNames[] =
   "hasnt",
   "startswith", "left",
   "endswith",   "right",
+  "word",
+  "noword"
 };
 
 #define NUM_INTERNAL_NAMES   (sizeof (internalNames)   / sizeof (internalNames[0]))
@@ -375,7 +377,7 @@ bool Att::validNameValue (
 }
 
 ////////////////////////////////////////////////////////////////////////////////
-// TODO Obsolete
+// TODO Deprecated - remove.
 bool Att::validMod (const std::string& mod)
 {
   for (unsigned int i = 0; i < NUM_MODIFIER_NAMES; ++i)
@@ -412,7 +414,8 @@ std::string Att::type (const std::string& name) const
 std::string Att::modType (const std::string& name) const
 {
   if (name == "hasnt" ||
-      name == "isnt")
+      name == "isnt"  ||
+      name == "noword")
     return "negative";
 
   return "positive";
@@ -615,6 +618,35 @@ bool Att::match (const Att& other) const
     }
   }
 
+  // word = contains as a substring, with word boundaries.
+  else if (mMod == "word") // TODO i18n
+  {
+    // Fail if the substring is not found.
+    std::string::size_type sub = other.mValue.find (mValue);
+    if (sub == std::string::npos)
+      return false;
+
+    // Also fail if there is no word boundary at beginning and end.
+    if (!isWordStart (other.mValue, sub))
+      return false;
+
+    if (!isWordEnd (other.mValue, sub + mValue.length () - 1))
+      return false;
+  }
+
+  // noword = does not contain as a substring, with word boundaries.
+  else if (mMod == "noword") // TODO i18n
+  {
+    // Fail if the substring is not found.
+    std::string::size_type sub = other.mValue.find (mValue);
+    if (sub != std::string::npos &&
+        isWordStart (other.mValue, sub) &&
+        isWordEnd (other.mValue, sub + mValue.length () - 1))
+    {
+      return false;
+    }
+  }
+
   return true;
 }
 

src/Context.cpp

@@ -690,6 +690,9 @@ void Context::parse (
 }
 
 ////////////////////////////////////////////////////////////////////////////////
+// Note: The reason some of these are commented out is because the ::clear
+// method is not really "clear" but "clear_some".  Some members do not need to
+// be initialized.  That makes this method something of a misnomer.  So be it.
 void Context::clear ()
 {
 //  Config                    config;
@@ -759,6 +762,7 @@ void Context::autoFilter (Task& t, Filter& f)
     }
 
     // The mechanism for filtering on tags is +/-<tag>.
+    // Do not handle here - see below.
     else if (att->second.name () == "tags")
     {
     }
@@ -776,17 +780,22 @@ void Context::autoFilter (Task& t, Filter& f)
     }
   }
 
+  // This is now a correct implementation of a filter on the presence or absence
+  // of a tag.  The prior code provided the illusion of leftmost partial tag
+  // matches, but was really using the 'contains' and 'nocontains' attribute
+  // modifiers.  See bug #293.
+
   // Include tagAdditions.
   foreach (tag, tagAdditions)
   {
-    f.push_back (Att ("tags", "has", *tag));
+    f.push_back (Att ("tags", "word", *tag));
     debug ("auto filter: +" + *tag);
   }
 
   // Include tagRemovals.
   foreach (tag, tagRemovals)
   {
-    f.push_back (Att ("tags", "hasnt", *tag));
+    f.push_back (Att ("tags", "noword", *tag));
     debug ("auto filter: -" + *tag);
   }
 }

src/Filter.cpp

@@ -125,7 +125,7 @@ void Filter::applySequence (std::vector<Task>& all, Sequence& sequence)
     std::vector <int> right;
     listDiff (filteredSequence, (std::vector <int>&)sequence, left, right);
     if (left.size ())
-      throw std::string ("Sequence filtering error - please report this error");
+      throw std::string ("Sequence filtering error - please report this error to support@taskwarrior.org");
 
     if (right.size ())
     {

src/report.cpp

@@ -260,6 +260,8 @@ int longUsage (std::string &outs)
       << "  hasnt"                                                            << "\n"
       << "  startswith (synonym left)"                                        << "\n"
       << "  endswith   (synonym right)"                                       << "\n"
+      << "  word"                                                             << "\n"
+      << "  noword"                                                           << "\n"
       <<                                                                         "\n"
       << "  For example:"                                                     << "\n"
       << "    task list due.before:eom priority.not:L"                        << "\n"

src/tests/att.t.cpp

@@ -33,7 +33,7 @@ Context context;
 ////////////////////////////////////////////////////////////////////////////////
 int main (int argc, char** argv)
 {
-  UnitTest t (95);
+  UnitTest t (97);
 
   Att a;
   t.notok (a.valid ("name"),            "Att::valid name         -> fail");
@@ -146,6 +146,14 @@ int main (int argc, char** argv)
   try {a6.mod ("endswith");} catch (...) {good = false;}
   t.ok (good, "Att::mod (endswith)");
 
+  good = true;
+  try {a6.mod ("word");} catch (...) {good = false;}
+  t.ok (good, "Att::mod (word)");
+
+  good = true;
+  try {a6.mod ("noword");} catch (...) {good = false;}
+  t.ok (good, "Att::mod (noword)");
+
   good = true;
   try {a6.mod ("fartwizzle");} catch (...) {good = false;}
   t.notok (good, "Att::mod (fartwizzle)");

src/tests/filt.t.cpp

@@ -34,7 +34,7 @@ Context context;
 ////////////////////////////////////////////////////////////////////////////////
 int main (int argc, char** argv)
 {
-  UnitTest test (14);
+  UnitTest test (20);
 
   // Create a filter consisting of two Att criteria.
   Filter f;
@@ -66,6 +66,7 @@ int main (int argc, char** argv)
   // Modifiers.
   Task mods;
   mods.set ("name", "value");
+  mods.set ("description", "hello, world.");
 
   Att a ("name", "is", "value");
   f.clear ();
@@ -124,6 +125,42 @@ int main (int argc, char** argv)
 "below"
 */
 
+  a = Att ("description", "word", "hello");
+  f.clear ();
+  f.push_back (a);
+  test.ok (f.pass (mods), "description:hello, world. -> description.word:hello = match");
+  // TODO test inverse.
+
+  a = Att ("description", "word", "world");
+  f.clear ();
+  f.push_back (a);
+  test.ok (f.pass (mods), "description:hello, world. -> description.word:world = match");
+  // TODO test inverse.
+
+  a = Att ("description", "word", "pig");
+  f.clear ();
+  f.push_back (a);
+  test.notok (f.pass (mods), "description:hello, world. -> description.word:pig = no match");
+  // TODO test inverse.
+
+  a = Att ("description", "noword", "hello");
+  f.clear ();
+  f.push_back (a);
+  test.notok (f.pass (mods), "description:hello, world. -> description.noword:hello = no match");
+  // TODO test inverse.
+
+  a = Att ("description", "noword", "world");
+  f.clear ();
+  f.push_back (a);
+  test.notok (f.pass (mods), "description:hello, world. -> description.noword:world = no match");
+  // TODO test inverse.
+
+  a = Att ("description", "noword", "pig");
+  f.clear ();
+  f.push_back (a);
+  test.ok (f.pass (mods), "description:hello, world. -> description.noword:pig = match");
+  // TODO test inverse.
+
   return 0;
 }
 

src/tests/text.t.cpp

@@ -34,7 +34,7 @@ Context context;
 ////////////////////////////////////////////////////////////////////////////////
 int main (int argc, char** argv)
 {
-  UnitTest t (117);
+  UnitTest t (147);
 
   // void wrapText (std::vector <std::string>& lines, const std::string& text, const int width)
   std::string text = "This is a test of the line wrapping code.";
@@ -267,6 +267,44 @@ int main (int argc, char** argv)
   t.notok (noVerticalSpace ("a\rb"),              "noVerticalSpace 'a\\rb' -> false");
   t.notok (noVerticalSpace ("a\fb"),              "noVerticalSpace 'a\\fb' -> false");
 
+  text = "Hello, world.";
+  //      0123456789012
+  //      s   e  s   e
+
+  // bool isWordStart (const std::string&, std::string::size_type);
+  t.notok (isWordStart ("", 0),                   "isWordStart (\"\", 0) -> false");
+  t.ok    (isWordStart ("foo", 0),                "isWordStart (\"foo\", 0) -> true");
+  t.ok    (isWordStart (text, 0),                 "isWordStart (\"Hello, world.\", 0) -> true");
+  t.notok (isWordStart (text, 1),                 "isWordStart (\"Hello, world.\", 1) -> false");
+  t.notok (isWordStart (text, 2),                 "isWordStart (\"Hello, world.\", 2) -> false");
+  t.notok (isWordStart (text, 3),                 "isWordStart (\"Hello, world.\", 3) -> false");
+  t.notok (isWordStart (text, 4),                 "isWordStart (\"Hello, world.\", 4) -> false");
+  t.notok (isWordStart (text, 5),                 "isWordStart (\"Hello, world.\", 5) -> false");
+  t.notok (isWordStart (text, 6),                 "isWordStart (\"Hello, world.\", 6) -> false");
+  t.ok    (isWordStart (text, 7),                 "isWordStart (\"Hello, world.\", 7) -> true");
+  t.notok (isWordStart (text, 8),                 "isWordStart (\"Hello, world.\", 8) -> false");
+  t.notok (isWordStart (text, 9),                 "isWordStart (\"Hello, world.\", 9) -> false");
+  t.notok (isWordStart (text, 10),                "isWordStart (\"Hello, world.\", 10) -> false");
+  t.notok (isWordStart (text, 11),                "isWordStart (\"Hello, world.\", 11) -> false");
+  t.notok (isWordStart (text, 12),                "isWordStart (\"Hello, world.\", 12) -> false");
+
+  // bool isWordEnd (const std::string&, std::string::size_type);
+  t.notok (isWordEnd ("", 0),                     "isWordEnd (\"\", 0) -> false");
+  t.ok    (isWordEnd ("foo", 2),                  "isWordEnd (\"foo\", 2) -> true");
+  t.notok (isWordEnd (text, 0),                   "isWordEnd (\"Hello, world.\", 0) -> false");
+  t.notok (isWordEnd (text, 1),                   "isWordEnd (\"Hello, world.\", 1) -> false");
+  t.notok (isWordEnd (text, 2),                   "isWordEnd (\"Hello, world.\", 2) -> false");
+  t.notok (isWordEnd (text, 3),                   "isWordEnd (\"Hello, world.\", 3) -> false");
+  t.ok    (isWordEnd (text, 4),                   "isWordEnd (\"Hello, world.\", 4) -> true");
+  t.notok (isWordEnd (text, 5),                   "isWordEnd (\"Hello, world.\", 5) -> false");
+  t.notok (isWordEnd (text, 6),                   "isWordEnd (\"Hello, world.\", 6) -> false");
+  t.notok (isWordEnd (text, 7),                   "isWordEnd (\"Hello, world.\", 7) -> false");
+  t.notok (isWordEnd (text, 8),                   "isWordEnd (\"Hello, world.\", 8) -> false");
+  t.notok (isWordEnd (text, 9),                   "isWordEnd (\"Hello, world.\", 9) -> false");
+  t.notok (isWordEnd (text, 10),                  "isWordEnd (\"Hello, world.\", 10) -> false");
+  t.ok    (isWordEnd (text, 11),                  "isWordEnd (\"Hello, world.\", 11) -> true");
+  t.notok (isWordEnd (text, 12),                  "isWordEnd (\"Hello, world.\", 12) -> false");
+
   return 0;
 }
 

src/tests/utf8.t

@@ -65,8 +65,8 @@ qx{../task rc:utf8.rc add project:Çirçös utf8 in project};
 $output = qx{../task rc:utf8.rc ls project:Çirçös};
 like ($output, qr/Çirçös.+utf8 in project/, 'utf8 in project works');
 
-qx{../task rc:utf8.rc add utf8 in tag +☺};
-$output = qx{../task rc:utf8.rc ls +☺};
+qx{../task rc:utf8.rc add utf8 in tag +Zwölf};
+$output = qx{../task rc:utf8.rc ls +Zwölf};
 like ($output, qr/utf8 in tag/, 'utf8 in tag works');
 
 # Cleanup.

src/text.cpp

@@ -420,3 +420,45 @@ bool noVerticalSpace (const std::string& input)
 }
 
 ////////////////////////////////////////////////////////////////////////////////
+//          Input: hello, world
+// Result for pos: y......y....
+bool isWordStart (const std::string& input, std::string::size_type pos)
+{
+  // Short circuit: no input means no word start.
+  if (input.length () == 0)
+    return false;
+
+  // If pos is the first alphanumeric character of the string.
+  if (pos == 0 && isalnum (input[pos]))
+    return true;
+
+  // If pos is not the first alphanumeric character, but there is a preceding
+  // non-alphanumeric character.
+  if (pos > 0 && isalnum (input[pos]) && !isalnum (input[pos - 1]))
+    return true;
+
+  return false;
+}
+
+////////////////////////////////////////////////////////////////////////////////
+//          Input: hello, world
+// Result for pos: ....y......y
+bool isWordEnd (const std::string& input, std::string::size_type pos)
+{
+  // Short circuit: no input means no word start.
+  if (input.length () == 0)
+    return false;
+
+  // If pos is the last alphanumeric character of the string.
+  if (pos == input.length () - 1 && isalnum (input[pos]))
+    return true;
+
+  // If pos is not the last alphanumeric character, but there is a following
+  // non-alphanumeric character.
+  if (pos < input.length () - 1 && isalnum (input[pos]) && !isalnum (input[pos + 1]))
+    return true;
+
+  return false;
+}
+
+////////////////////////////////////////////////////////////////////////////////

src/text.h

@@ -52,6 +52,8 @@ void guess (const std::string&, std::vector<std::string>&, std::string&);
 bool digitsOnly (const std::string&);
 bool noSpaces (const std::string&);
 bool noVerticalSpace (const std::string&);
+bool isWordStart (const std::string&, std::string::size_type);
+bool isWordEnd (const std::string&, std::string::size_type);
 
 #endif
 ////////////////////////////////////////////////////////////////////////////////