c++: Differentiate scanning logic from C

To enable raw string literals only in C++, add a flag indicating
the current source language is C++.
Suggested by Vaclav Slavik at:
<https://savannah.gnu.org/bugs/?43970>.

* gettext-tools/src/x-c.h (extract_cxx): New declaration.
(SCANNERS_C): Use extract_cxx for "C++".
* gettext-tools/src/x-c.c (cxx_extensions): New variable.
(phase5_get): Respect cxx_extensions and recognize raw string
literals only when it is set.
(extract_cxx): New function.
(extract_c, extract_objc): Reset cxx_extensions.

* gettext-tools/tests/xgettext-c-20: Adjust the source file name
from *.c to *.cc to enable handling of raw string literals.

diff --git a/gettext-tools/src/ChangeLog b/gettext-tools/src/ChangeLog
index fbfa43ffeb35f446c4a100db56829ffcce379374..bf50cb9304158a169feec0b714118d3f538b5c6d 100644 (file)
--- a/gettext-tools/src/ChangeLog
+++ b/gettext-tools/src/ChangeLog
@@ -1,3 +1,18 @@
+2015-01-13  Daiki Ueno  <ueno@gnu.org>
+
+       c++: Differentiate scanning logic from C
+       To enable raw string literals only in C++, add a flag indicating
+       the current source language is C++.
+       Suggested by Vaclav Slavik at:
+       <https://savannah.gnu.org/bugs/?43970>.
+       * x-c.h (extract_cxx): New declaration.
+       (SCANNERS_C): Use extract_cxx for "C++".
+       * x-c.c (cxx_extensions): New variable.
+       (phase5_get): Respect cxx_extensions and recognize raw string
+       literals only when it is set.
+       (extract_cxx): New function.
+       (extract_c, extract_objc): Reset cxx_extensions.
+
 2014-12-24  Daiki Ueno  <ueno@gnu.org>
 
        * gettext 0.19.4 released.

diff --git a/gettext-tools/src/x-c.c b/gettext-tools/src/x-c.c
index 001cfd91d6df4a94049038b345ca2e13c00faa8b..32b92d2ae01656c6232b03710d919ef2791f58de 100644 (file)
--- a/gettext-tools/src/x-c.c
+++ b/gettext-tools/src/x-c.c
@@ -829,6 +829,9 @@ phase4_ungetc (int c)
 /* True if ObjectiveC extensions are recognized.  */
 static bool objc_extensions;
 
+/* True if C++ extensions are recognized.  */
+static bool cxx_extensions;
+
 enum token_type_ty
 {
   token_type_character_constant,        /* 'x' */
@@ -1097,7 +1100,7 @@ phase5_get (token_ty *tp)
   int bufpos;
   int c;
   int last_was_backslash;
-  bool raw_expected = false;
+  bool raw_expected;
 
   if (phase5_pushback_length)
     {
@@ -1177,10 +1180,14 @@ phase5_get (token_ty *tp)
               continue;
 
             default:
-              /* Recognize C++ string literals prefixed by R, u8, u8R,
-                 u, uR, U, UR, L, or LR.  It is defined in ISO/IEC
-                 9899:2011 2.14.5.  Since gettext's argument is a byte
-                 sequence, we are only interested in u8, R, and u8R.  */
+              /* Recognize string literals prefixed by R, u8, u8R, u,
+                 uR, U, UR, L, or LR.  It is defined in the C standard
+                 ISO/IEC 9899:201x and the C++ standard ISO/IEC
+                 14882:2011.  The raw string literals prefixed by R,
+                 u8R, uR, UR, or LR are only valid in C++.
+
+                 Since gettext's argument is a byte sequence, we are
+                 only interested in u8, R, and u8R.  */
               if (c == '"')
                 {
                   bool is_prefix = false;
@@ -1188,37 +1195,61 @@ phase5_get (token_ty *tp)
                   switch (buffer[0])
                     {
                     case 'R':
-                      if (bufpos == 1)
-                        is_prefix = true;
+                      if (cxx_extensions && bufpos == 1)
+                        {
+                          is_prefix = true;
+                          raw_expected = true;
+                        }
                       break;
                     case 'u':
                       if (bufpos == 1)
-                        is_prefix = true;
+                        {
+                          is_prefix = true;
+                          raw_expected = false;
+                        }
                       else
                         switch (buffer[1])
                           {
                           case 'R':
-                            if (bufpos == 2)
-                              is_prefix = true;
+                            if (cxx_extensions && bufpos == 2)
+                              {
+                                is_prefix = true;
+                                raw_expected = true;
+                              }
                             break;
                           case '8':
-                            if (bufpos == 2
-                                || (bufpos == 3 && buffer[2] == 'R'))
-                              is_prefix = true;
+                            if (bufpos == 2)
+                              {
+                                is_prefix = true;
+                                raw_expected = false;
+                              }
+                            else if (cxx_extensions
+                                     && bufpos == 3 && buffer[2] == 'R')
+                              {
+                                is_prefix = true;
+                                raw_expected = true;
+                              }
                             break;
                           }
                       break;
                     case 'U':
                     case 'L':
-                      if (bufpos == 1
-                          || (bufpos == 2 && buffer[1] == 'R'))
-                        is_prefix = true;
+                      if (bufpos == 1)
+                        {
+                          is_prefix = true;
+                          raw_expected = false;
+                        }
+                      else if (cxx_extensions
+                               && bufpos == 2 && buffer[1] == 'R')
+                        {
+                          is_prefix = true;
+                          raw_expected = true;
+                        }
                       break;
                     }
 
                   if (is_prefix)
                     {
-                      raw_expected = buffer[bufpos - 1] == 'R';
                       bufpos = 0;
                       goto string;
                     }
@@ -2160,6 +2191,18 @@ extract_c (FILE *f,
            msgdomain_list_ty *mdlp)
 {
   objc_extensions = false;
+  cxx_extensions = false;
+  extract_whole_file (f, real_filename, logical_filename, flag_table, mdlp);
+}
+
+void
+extract_cxx (FILE *f,
+             const char *real_filename, const char *logical_filename,
+             flag_context_list_table_ty *flag_table,
+             msgdomain_list_ty *mdlp)
+{
+  objc_extensions = false;
+  cxx_extensions = true;
   extract_whole_file (f, real_filename, logical_filename, flag_table, mdlp);
 }
 
@@ -2170,5 +2213,6 @@ extract_objc (FILE *f,
               msgdomain_list_ty *mdlp)
 {
   objc_extensions = true;
+  cxx_extensions = false;
   extract_whole_file (f, real_filename, logical_filename, flag_table, mdlp);
 }

diff --git a/gettext-tools/src/x-c.h b/gettext-tools/src/x-c.h
index 64e4953ff35269e0b8d2195ad8fc36458a3269f3..9e819a0294a6cd1c7b6cfde3f3b63a8a35ad1eb7 100644 (file)
--- a/gettext-tools/src/x-c.h
+++ b/gettext-tools/src/x-c.h
@@ -45,7 +45,7 @@ extern "C" {
                         &flag_table_c,                                  \
                         &formatstring_c, NULL,                          \
                         &literalstring_c },                             \
-  { "C++",              extract_c,                                      \
+  { "C++",              extract_cxx,                                    \
                         &flag_table_c,                                  \
                         &formatstring_c, NULL,                          \
                         &literalstring_c },                             \
@@ -58,11 +58,16 @@ extern "C" {
                         &formatstring_gcc_internal, &formatstring_gfc_internal, \
                         &literalstring_c },                             \
 
-/* Scan a C/C++ file and add its translatable strings to mdlp.  */
+/* Scan a C file and add its translatable strings to mdlp.  */
 extern void extract_c (FILE *fp, const char *real_filename,
                        const char *logical_filename,
                        flag_context_list_table_ty *flag_table,
                        msgdomain_list_ty *mdlp);
+/* Scan a C++ file and add its translatable strings to mdlp.  */
+extern void extract_cxx (FILE *fp, const char *real_filename,
+                         const char *logical_filename,
+                         flag_context_list_table_ty *flag_table,
+                         msgdomain_list_ty *mdlp);
 /* Scan an ObjectiveC file and add its translatable strings to mdlp.  */
 extern void extract_objc (FILE *fp, const char *real_filename,
                           const char *logical_filename,

diff --git a/gettext-tools/tests/ChangeLog b/gettext-tools/tests/ChangeLog
index 47c51ff63c200055b2498d2b30db964a5f3387ad..8ab049aa5be309905abbc39ce87d29380e0c5b02 100644 (file)
--- a/gettext-tools/tests/ChangeLog
+++ b/gettext-tools/tests/ChangeLog
@@ -1,3 +1,8 @@
+2015-01-13  Daiki Ueno  <ueno@gnu.org>
+
+       * xgettext-c-20: Adjust the source file name from *.c to *.cc to
+       enable handling of raw string literals.
+
 2014-12-24  Daiki Ueno  <ueno@gnu.org>
 
        * gettext 0.19.4 released.

diff --git a/gettext-tools/tests/xgettext-c-20 b/gettext-tools/tests/xgettext-c-20
index 73e4442a0b264d25ebfec2092673964dc5e4ea02..2740962624281ce6bee36729df4bccf626104a0c 100755 (executable)
--- a/gettext-tools/tests/xgettext-c-20
+++ b/gettext-tools/tests/xgettext-c-20
@@ -5,21 +5,21 @@
 
 : ${XGETTEXT=xgettext}
 
-cat <<\EOF > err0.c
+cat <<\EOF > err0.cc
 gettext ("aaa(
 bbb
 )ccc");
 EOF
 
-(LANGUAGE= LC_ALL=C ${XGETTEXT} --no-location --no-wrap -o - err0.c 2>&1; exit) | grep 'unterminated string literal' || exit 1
+(LANGUAGE= LC_ALL=C ${XGETTEXT} --no-location --no-wrap -o - err0.cc 2>&1; exit) | grep 'unterminated string literal' || exit 1
 
-cat <<\EOF > err1.c
+cat <<\EOF > err1.cc
 gettext (R"aaa(
 bbb
 )ccc");
 EOF
 
-(LANGUAGE= LC_ALL=C ${XGETTEXT} --no-location --no-wrap -o - err1.c 2>&1; exit) | grep 'unterminated string literal' || exit 1
+(LANGUAGE= LC_ALL=C ${XGETTEXT} --no-location --no-wrap -o - err1.cc 2>&1; exit) | grep 'unterminated string literal' || exit 1
 
 cat <<\EOF > xg-c-20.cc
 #define X ""