released 1.3.3: improved -E, -G, -P GNU/BSD grep compatibility

README.md

@@ -1491,7 +1491,7 @@ Man page
 
 
 
-    ugrep 1.3.2                      July 25, 2019                        UGREP(1)
+    ugrep 1.3.3                      July 26, 2019                        UGREP(1)
 
 <a name="patterns"/>
 

man/ugrep.1

@@ -1,4 +1,4 @@
-.TH UGREP "1" "July 25, 2019" "ugrep 1.3.2" "User Commands"
+.TH UGREP "1" "July 26, 2019" "ugrep 1.3.3" "User Commands"
 .SH NAME
 \fBugrep\fR -- universal file pattern searcher
 .SH SYNOPSIS

patterns/README.md

@@ -1,24 +1,29 @@
+ugrep predefined patterns
+=========================
+
 This directory contains a collection of patterns that are helpful to search
 source code files.  Use ugrep option `-f` to specify one or more pattern files
 to use for searching these patterns in files.
 
 The list of patterns defined in this directory will expand over time.
 
-For example, to display all class defitions in C++ files in myproject directory:
+For example, to display class definitions in C++ files in the working directory
 
-    ugrep -R -o -tc++ -f c++/classes myproject
+    ugrep -r -o -tc++ -f c++/classes
 
 To display Java identifiers (Unicode) with the line numbers of the matches, but
 skipping all matches of identifiers in comments and strings:
 
-    ugrep -R -n -o -f java/names -f java/zap_comments -f java/zap_strings myproject
+    ugrep -r -n -o -f java/names -f java/zap_comments -f java/zap_strings
 
 Some patterns automatically enable ugrep option `-o` to match multiple lines.
 These pattern files start with `###-o` to enbale option `-o` that is required
 to match the pattern across multiple lines.  For example, strings and comments
 may span multiple lines, such as Python docstrings, requiring option `-o`.
 
-Patterns requiring Unicode matching are placed in Unicode mode with (?u:X),
+Empty lines and lines starting with a `#` are ignored.
+
+Patterns requiring Unicode matching are placed in Unicode mode with `(?u:X)`,
 just in case to prevent ugrep option -U from disabling them.
 
 We love your contributions to this effort! ❤️

patterns/php/README.md

@@ -6,5 +6,5 @@ PHP patterns
 - `names` matches identifiers (and keywords)
 - `strings` matches strings, auto-enables ugrep option -o to match multi-line strings
 - `zap_comments` removes comments from matches, recommend ugrep option -o
-- `zap_html` removes HTML content outside of <?php...?>, auto-enables ugrep option -o
+- `zap_html` removes HTML content outside of `<?php...?>` blocks, auto-enables ugrep option -o
 - `zap_strings` removes strings from matches, recommend ugrep option -o

src/ugrep.cpp

@@ -103,7 +103,7 @@ Optional libraries:
 #endif
 
 // ugrep version info
-#define UGREP_VERSION "1.3.2"
+#define UGREP_VERSION "1.3.3"
 
 // ugrep platform -- see configure.ac
 #if !defined(PLATFORM)
@@ -476,7 +476,7 @@ int main(int argc, char **argv)
             else if (strncmp(arg, "exclude-from=", 13) == 0)
               flag_exclude_from.emplace_back(arg + 13);
             else if (strcmp(arg, "extended-regexp") == 0)
-              ;
+              flag_basic_regexp = false;
             else if (strncmp(arg, "file=", 5) == 0)
               flag_file.emplace_back(arg + 5);
             else if (strncmp(arg, "file-extensions=", 16) == 0)
@@ -554,7 +554,7 @@ int main(int argc, char **argv)
             else if (strncmp(arg, "pager=", 6) == 0)
               flag_pager = arg + 6;
             else if (strcmp(arg, "perl-regexp") == 0)
-              flag_perl_regexp = true;
+              flag_basic_regexp = !(flag_perl_regexp = true);
             else if (strcmp(arg, "quiet") == 0 || strcmp(arg, "silent") == 0)
               flag_quiet = flag_no_messages = true;
             else if (strcmp(arg, "recursive") == 0)
@@ -650,6 +650,7 @@ int main(int argc, char **argv)
             break;
 
           case 'E':
+            flag_basic_regexp = false;
             break;
 
           case 'e':
@@ -774,6 +775,7 @@ int main(int argc, char **argv)
 
           case 'P':
             flag_perl_regexp = true;
+            flag_basic_regexp = false;
             break;
 
           case 'p':
@@ -894,7 +896,7 @@ int main(int argc, char **argv)
 #ifndef HAVE_LIBZ
   // -z: but we don't have libz
   if (flag_decompress)
-    help("option -z is disabled");
+    help("option -z is not available in this version of ugrep");
 #endif
 
   // -t list: list table of types
@@ -1394,6 +1396,9 @@ int main(int argc, char **argv)
 
     // enable --break
     flag_break = true;
+
+    // enable --line-buffered to flush output to the pager immediately
+    flag_line_buffered = true;
   }
 #endif
 
@@ -1468,16 +1473,18 @@ int main(int argc, char **argv)
         help("invalid --tabs=NUM value");
     }
 
-#ifdef HAVE_BOOST_REGEX
     if (flag_perl_regexp)
     {
+#ifdef HAVE_BOOST_REGEX
       // construct the NFA pattern matcher
       std::string pattern(reflex::BoostPerlMatcher::convert(regex, convert_flags));
       reflex::BoostPerlMatcher matcher(pattern, matcher_options.c_str());
       found = findinfiles(magic, matcher, infiles, encoding);
+#else
+      help("Option -P is not available in this version of ugrep");
+#endif
     }
     else
-#endif
     {
       // construct the DFA pattern matcher
       reflex::Pattern pattern(reflex::Matcher::convert(regex, convert_flags), "r");
@@ -1499,7 +1506,7 @@ int main(int argc, char **argv)
   {
     if (!flag_no_messages)
     {
-      std::cerr << "Boost regex error at position " << error.position() << " in " << regex << std::endl;
+      std::cerr << "Boost regex error in " << regex << std::endl;
       switch (error.code())
       {
         case boost::regex_constants::error_collate:
@@ -2774,9 +2781,12 @@ bool ugrep(reflex::AbstractMatcher& matcher, reflex::Input& input, const char *p
     ;
   }
 
-  // --break: add a line break
+  // --break: add a line break and flush
   if ((matches > 0 || flag_any_line) && flag_break)
+  {
     fputc('\n', out);
+    fflush(out);
+  }
 
   return matches > 0;
 }