PCRE2Project · NWilson · Feb 27, 2025 · Feb 26, 2025 · Feb 26, 2025
diff --git a/BUILD.bazel b/BUILD.bazel
@@ -20,9 +20,6 @@ copy_file(
     out = "src/pcre2_chartables.c",
 )
 
-# Removed src/pcre2_ucptables.c below because it is #included in
-# src/pcre2_tables.c. Also fixed typo: ckdint should be chkdint.
-# PH, 22-March-2023.
 cc_library(
     name = "pcre2",
     srcs = [
@@ -39,12 +36,15 @@ cc_library(
         "src/pcre2_extuni.c",
         "src/pcre2_find_bracket.c",
         "src/pcre2_jit_compile.c",
+        "src/pcre2_jit_match_inc.h",
+        "src/pcre2_jit_misc_inc.h",
         "src/pcre2_maketables.c",
         "src/pcre2_match.c",
         "src/pcre2_match_data.c",
         "src/pcre2_newline.c",
         "src/pcre2_ord2utf.c",
         "src/pcre2_pattern_info.c",
+        "src/pcre2_printint_inc.h",
         "src/pcre2_script_run.c",
         "src/pcre2_serialize.c",
         "src/pcre2_string_utils.c",
@@ -60,13 +60,11 @@ cc_library(
         "src/pcre2_internal.h",
         "src/pcre2_intmodedep.h",
         "src/pcre2_ucp.h",
+        "src/pcre2_ucptables_inc.h",
         "src/pcre2_util.h",
         ":config_h_generic",
     ],
     textual_hdrs = [
-        "src/pcre2_jit_match.c",
-        "src/pcre2_jit_misc.c",
-        "src/pcre2_ucptables.c",
     ],
     hdrs = [
         ":pcre2_h_generic",
@@ -119,7 +117,6 @@ cc_library(
     name = "pcre2test_dotc_headers",
     hdrs = [
         "src/pcre2_chkdint.c",
-        "src/pcre2_printint.c",
         "src/pcre2_tables.c",
         "src/pcre2_ucd.c",
         "src/pcre2_valid_utf.c",

diff --git a/Makefile.am b/Makefile.am
@@ -404,6 +404,8 @@ COMMON_SOURCES = \
   src/pcre2_intmodedep.h \
   src/pcre2_jit_char_inc.h \
   src/pcre2_jit_compile.c \
+  src/pcre2_jit_match_inc.h \
+  src/pcre2_jit_misc_inc.h \
   src/pcre2_jit_neon_inc.h \
   src/pcre2_jit_simd_inc.h \
   src/pcre2_maketables.c \
@@ -412,6 +414,7 @@ COMMON_SOURCES = \
   src/pcre2_newline.c \
   src/pcre2_ord2utf.c \
   src/pcre2_pattern_info.c \
+  src/pcre2_printint_inc.h \
   src/pcre2_script_run.c \
   src/pcre2_serialize.c \
   src/pcre2_string_utils.c \
@@ -421,14 +424,11 @@ COMMON_SOURCES = \
   src/pcre2_tables.c \
   src/pcre2_ucd.c \
   src/pcre2_ucp.h \
+  src/pcre2_ucptables_inc.h \
   src/pcre2_util.h \
   src/pcre2_valid_utf.c \
   src/pcre2_xclass.c
 
-# The pcre2_ucptables.c file is #included by pcre2_tables.c
-
-EXTRA_DIST += src/pcre2_ucptables.c
-
 if WITH_PCRE2_8
 lib_LTLIBRARIES += libpcre2-8.la
 libpcre2_8_la_SOURCES = \
@@ -520,12 +520,6 @@ EXTRA_DIST += \
   deps/sljit/sljit_src/allocator_src/sljitWXExecAllocatorPosix.c \
   deps/sljit/sljit_src/allocator_src/sljitWXExecAllocatorWindows.c
 
-# Some of the JIT sources are also in separate files that are #included.
-
-EXTRA_DIST += \
-  src/pcre2_jit_match.c \
-  src/pcre2_jit_misc.c
-
 if WITH_PCRE2_8
 libpcre2_8_la_LDFLAGS = $(EXTRA_LIBPCRE2_8_LDFLAGS)
 endif # WITH_PCRE2_8
@@ -681,12 +675,9 @@ pcre2_jit_test_LDADD += $(GCOV_LIBS)
 endif # WITH_GCOV
 endif # WITH_JIT
 
-# Build the general pcre2test program. The file src/pcre2_printint.c is
-# #included by pcre2test as many times as needed, at different code unit
-# widths.
+# Build the general pcre2test program.
 
 bin_PROGRAMS += pcre2test
-EXTRA_DIST += src/pcre2_printint.c
 pcre2test_SOURCES = src/pcre2test.c
 pcre2test_CFLAGS = $(AM_CFLAGS)
 pcre2test_LDADD = $(LIBREADLINE)

diff --git a/NON-AUTOTOOLS-BUILD b/NON-AUTOTOOLS-BUILD
@@ -143,9 +143,7 @@ example.
      defined SUPPORT_JIT in src/config.h, because when JIT support is not
      configured, dummy functions are compiled. When JIT support IS configured,
      pcre2_jit_compile.c #includes other files from the sljit dependency,
-     all of whose names begin with "sljit". It also #includes
-     src/pcre2_jit_match.c and src/pcre2_jit_misc.c, so you should not compile
-     those yourself.
+     all of whose names begin with "sljit".
 
      Note also that the pcre2_fuzzsupport.c file contains special code that is
      useful to those who want to run fuzzing tests on the PCRE2 library. Unless

diff --git a/README b/README
@@ -860,10 +860,8 @@ The distribution should contain the files listed below.
   src/pcre2_extuni.c         )
   src/pcre2_find_bracket.c   )
   src/pcre2_jit_compile.c    )
-  src/pcre2_jit_match.c      ) sources for the functions in the library,
-  src/pcre2_jit_misc.c       )   and some internal functions that they use
-  src/pcre2_maketables.c     )
-  src/pcre2_match.c          )
+  src/pcre2_maketables.c     ) sources for the functions in the library,
+  src/pcre2_match.c          )   and some internal functions that they use
   src/pcre2_match_data.c     )
   src/pcre2_newline.c        )
   src/pcre2_ord2utf.c        )
@@ -876,11 +874,9 @@ The distribution should contain the files listed below.
   src/pcre2_substring.c      )
   src/pcre2_tables.c         )
   src/pcre2_ucd.c            )
-  src/pcre2_ucptables.c      )
   src/pcre2_valid_utf.c      )
   src/pcre2_xclass.c         )
 
-  src/pcre2_printint.c     debugging function that is used by pcre2test,
   src/pcre2_fuzzsupport.c  function for (optional) fuzzing support
 
   src/config.h.in          template for config.h, when built by "configure"
@@ -890,9 +886,13 @@ The distribution should contain the files listed below.
   src/pcre2_internal.h     header for internal use
   src/pcre2_intmodedep.h   a mode-specific internal header
   src/pcre2_jit_char_inc.h header used by JIT
+  src/pcre2_jit_match_inc.h header used by JIT
+  src/pcre2_jit_misc_inc.h header used by JIT
   src/pcre2_jit_neon_inc.h header used by JIT
   src/pcre2_jit_simd_inc.h header used by JIT
+  src/pcre2_printint_inc.h debugging function that is used by pcre2test
   src/pcre2_ucp.h          header for Unicode property handling
+  src/pcre2_ucptables_inc.h header with Unicode data tables
   src/pcre2_util.h         header for internal utils
 
   deps/sljit/sljit_src/*   source files for the JIT compiler

diff --git a/doc/html/NON-AUTOTOOLS-BUILD.txt b/doc/html/NON-AUTOTOOLS-BUILD.txt
@@ -143,9 +143,7 @@ example.
      defined SUPPORT_JIT in src/config.h, because when JIT support is not
      configured, dummy functions are compiled. When JIT support IS configured,
      pcre2_jit_compile.c #includes other files from the sljit dependency,
-     all of whose names begin with "sljit". It also #includes
-     src/pcre2_jit_match.c and src/pcre2_jit_misc.c, so you should not compile
-     those yourself.
+     all of whose names begin with "sljit".
 
      Note also that the pcre2_fuzzsupport.c file contains special code that is
      useful to those who want to run fuzzing tests on the PCRE2 library. Unless

diff --git a/doc/html/README.txt b/doc/html/README.txt
@@ -860,10 +860,8 @@ The distribution should contain the files listed below.
   src/pcre2_extuni.c         )
   src/pcre2_find_bracket.c   )
   src/pcre2_jit_compile.c    )
-  src/pcre2_jit_match.c      ) sources for the functions in the library,
-  src/pcre2_jit_misc.c       )   and some internal functions that they use
-  src/pcre2_maketables.c     )
-  src/pcre2_match.c          )
+  src/pcre2_maketables.c     ) sources for the functions in the library,
+  src/pcre2_match.c          )   and some internal functions that they use
   src/pcre2_match_data.c     )
   src/pcre2_newline.c        )
   src/pcre2_ord2utf.c        )
@@ -876,11 +874,9 @@ The distribution should contain the files listed below.
   src/pcre2_substring.c      )
   src/pcre2_tables.c         )
   src/pcre2_ucd.c            )
-  src/pcre2_ucptables.c      )
   src/pcre2_valid_utf.c      )
   src/pcre2_xclass.c         )
 
-  src/pcre2_printint.c     debugging function that is used by pcre2test,
   src/pcre2_fuzzsupport.c  function for (optional) fuzzing support
 
   src/config.h.in          template for config.h, when built by "configure"
@@ -890,9 +886,13 @@ The distribution should contain the files listed below.
   src/pcre2_internal.h     header for internal use
   src/pcre2_intmodedep.h   a mode-specific internal header
   src/pcre2_jit_char_inc.h header used by JIT
+  src/pcre2_jit_match_inc.h header used by JIT
+  src/pcre2_jit_misc_inc.h header used by JIT
   src/pcre2_jit_neon_inc.h header used by JIT
   src/pcre2_jit_simd_inc.h header used by JIT
+  src/pcre2_printint_inc.h debugging function that is used by pcre2test
   src/pcre2_ucp.h          header for Unicode property handling
+  src/pcre2_ucptables_inc.h header with Unicode data tables
   src/pcre2_util.h         header for internal utils
 
   deps/sljit/sljit_src/*   source files for the JIT compiler

diff --git a/maint/.gitignore b/maint/.gitignore
@@ -2,7 +2,7 @@ ucptest
 utf8
 
 pcre2_ucp.h
-pcre2_ucptables.c
+pcre2_ucptables_inc.h
 pcre2_ucd.c
 
 testinput

diff --git a/maint/GenerateUcpTables.py b/maint/GenerateUcpTables.py
@@ -3,7 +3,7 @@
 #                   PCRE2 UNICODE PROPERTY SUPPORT
 #                   ------------------------------
 
-# This script generates the pcre2_ucptables.c file, which contains tables for
+# This script generates the pcre2_ucptables_inc.h file, which contains tables for
 # recognizing Unicode property names. It is #included by pcre2_tables.c. In
 # order to reduce the number of relocations when loading the PCRE2 library, the
 # names are held as a single large string, with offsets in the table. This is
@@ -60,7 +60,7 @@
 # Open the output file (no return on failure). This call also writes standard
 # header boilerplate.
 
-f = open_output("pcre2_ucptables.c")
+f = open_output("pcre2_ucptables_inc.h")
 
 # The list in bidi_classes contains just the Unicode classes such as AN, LRE,
 # etc., along with comments. We need to add "bidi" in front of each value, in
@@ -195,7 +195,7 @@ def stdnames(x):
 
 #endif /* SUPPORT_UNICODE */
 
-/* End of pcre2_ucptables.c */
+/* End of pcre2_ucptables_inc.h */
 """)
 
 f.close

diff --git a/maint/README b/maint/README
@@ -56,7 +56,7 @@ GenerateUcpHeader.py
   Unicode property values.
 
 GenerateUcpTables.py
-  A Python script that generates the file pcre2_ucptables.c from
+  A Python script that generates the file pcre2_ucptables_inc.h from
   GenerateCommon.py and Unicode data files. The generated file contains tables
   for looking up Unicode property names.
 
@@ -145,17 +145,17 @@ of supported scripts, and the command to do so is part of the documentation.
 You can give an output file name as an argument to the following scripts, but
 by default:
 
-GenerateUcd.py        creates pcre2_ucd.c        )
-GenerateUcpHeader.py  creates pcre2_ucp.h        ) in the current directory
-GenerateUcpTables.py  creates pcre2_ucptables.c  )
+GenerateUcd.py        creates pcre2_ucd.c            )
+GenerateUcpHeader.py  creates pcre2_ucp.h            ) in the current directory
+GenerateUcpTables.py  creates pcre2_ucptables_inc.h  )
 
 These files can be compared against the existing versions in the src directory
 to check on any changes before replacing the old files, but you can also
 generate directly into the final location by running:
 
 ./GenerateUcd.py       ../src/pcre2_ucd.c
 ./GenerateUcpHeader.py ../src/pcre2_ucp.h
-./GenerateUcpTables.py ../src/pcre2_ucptables.c
+./GenerateUcpTables.py ../src/pcre2_ucptables_inc.h
 
 Once the .c and .h files are in the ../src directory, the ucptest program can
 be compiled and used to check that the new tables work properly. The data files
@@ -176,7 +176,7 @@ In summary:
 ```
 ./GenerateUcd.py       ../src/pcre2_ucd.c
 ./GenerateUcpHeader.py ../src/pcre2_ucp.h
-./GenerateUcpTables.py ../src/pcre2_ucptables.c
+./GenerateUcpTables.py ../src/pcre2_ucptables_inc.h
 ./GenerateTest.py
 mv testinput ../testdata/testinput27
 mv testoutput ../testdata/testoutput27

diff --git a/maint/UpdateAlways b/maint/UpdateAlways
@@ -282,8 +282,8 @@ c_files=(
   src/pcre2_intmodedep.h
   src/pcre2_jit_char_inc.h
   src/pcre2_jit_compile.c
-  src/pcre2_jit_match.c
-  src/pcre2_jit_misc.c
+  src/pcre2_jit_match_inc.h
+  src/pcre2_jit_misc_inc.h
   src/pcre2_jit_neon_inc.h
   src/pcre2_jit_simd_inc.h
   src/pcre2_jit_test.c
@@ -293,7 +293,7 @@ c_files=(
   src/pcre2_newline.c
   src/pcre2_ord2utf.c
   src/pcre2_pattern_info.c
-  src/pcre2_printint.c
+  src/pcre2_printint_inc.h
   src/pcre2_script_run.c
   src/pcre2_serialize.c
   src/pcre2_string_utils.c
@@ -303,7 +303,7 @@ c_files=(
   src/pcre2_tables.c
   src/pcre2_ucd.c
   src/pcre2_ucp.h
-  src/pcre2_ucptables.c
+  src/pcre2_ucptables_inc.h
   src/pcre2_util.h
   src/pcre2_valid_utf.c
   src/pcre2_xclass.c

diff --git a/maint/manifest-tarball b/maint/manifest-tarball
@@ -327,8 +327,8 @@ drwxr-xr-x tarball-dir/pcre2-SNAPSHOT/src
 -rw-r--r-- tarball-dir/pcre2-SNAPSHOT/src/pcre2_intmodedep.h
 -rw-r--r-- tarball-dir/pcre2-SNAPSHOT/src/pcre2_jit_char_inc.h
 -rw-r--r-- tarball-dir/pcre2-SNAPSHOT/src/pcre2_jit_compile.c
--rw-r--r-- tarball-dir/pcre2-SNAPSHOT/src/pcre2_jit_match.c
--rw-r--r-- tarball-dir/pcre2-SNAPSHOT/src/pcre2_jit_misc.c
+-rw-r--r-- tarball-dir/pcre2-SNAPSHOT/src/pcre2_jit_match_inc.h
+-rw-r--r-- tarball-dir/pcre2-SNAPSHOT/src/pcre2_jit_misc_inc.h
 -rw-r--r-- tarball-dir/pcre2-SNAPSHOT/src/pcre2_jit_neon_inc.h
 -rw-r--r-- tarball-dir/pcre2-SNAPSHOT/src/pcre2_jit_simd_inc.h
 -rw-r--r-- tarball-dir/pcre2-SNAPSHOT/src/pcre2_jit_test.c
@@ -338,7 +338,7 @@ drwxr-xr-x tarball-dir/pcre2-SNAPSHOT/src
 -rw-r--r-- tarball-dir/pcre2-SNAPSHOT/src/pcre2_newline.c
 -rw-r--r-- tarball-dir/pcre2-SNAPSHOT/src/pcre2_ord2utf.c
 -rw-r--r-- tarball-dir/pcre2-SNAPSHOT/src/pcre2_pattern_info.c
--rw-r--r-- tarball-dir/pcre2-SNAPSHOT/src/pcre2_printint.c
+-rw-r--r-- tarball-dir/pcre2-SNAPSHOT/src/pcre2_printint_inc.h
 -rw-r--r-- tarball-dir/pcre2-SNAPSHOT/src/pcre2_script_run.c
 -rw-r--r-- tarball-dir/pcre2-SNAPSHOT/src/pcre2_serialize.c
 -rw-r--r-- tarball-dir/pcre2-SNAPSHOT/src/pcre2_string_utils.c
@@ -348,7 +348,7 @@ drwxr-xr-x tarball-dir/pcre2-SNAPSHOT/src
 -rw-r--r-- tarball-dir/pcre2-SNAPSHOT/src/pcre2_tables.c
 -rw-r--r-- tarball-dir/pcre2-SNAPSHOT/src/pcre2_ucd.c
 -rw-r--r-- tarball-dir/pcre2-SNAPSHOT/src/pcre2_ucp.h
--rw-r--r-- tarball-dir/pcre2-SNAPSHOT/src/pcre2_ucptables.c
+-rw-r--r-- tarball-dir/pcre2-SNAPSHOT/src/pcre2_ucptables_inc.h
 -rw-r--r-- tarball-dir/pcre2-SNAPSHOT/src/pcre2_util.h
 -rw-r--r-- tarball-dir/pcre2-SNAPSHOT/src/pcre2_valid_utf.c
 -rw-r--r-- tarball-dir/pcre2-SNAPSHOT/src/pcre2_xclass.c

diff --git a/src/pcre2_compile.c b/src/pcre2_compile.c
@@ -59,7 +59,7 @@ POSSIBILITY OF SUCH DAMAGE.
 #define CHAR_OUTPUT_HEX(c)  (c)
 #define CHAR_INPUT(c)       (c)
 #define CHAR_INPUT_HEX(c)   (c)
-#include "pcre2_printint.c"
+#include "pcre2_printint_inc.h"
 #undef PRINTABLE
 #undef CHAR_OUTPUT
 #undef CHAR_OUTPUT_HEX

diff --git a/src/pcre2_internal.h b/src/pcre2_internal.h
@@ -1949,7 +1949,7 @@ pcre2_dfa_match.c that must be updated. */
 
 /* This macro defines textual names for all the opcodes. These are used only
 for debugging, and some of them are only partial names. The macro is referenced
-only in pcre2_printint.c, which fills out the full names in many cases (and in
+only in pcre2_printint_inc.h, which fills out the full names in many cases (and in
 some cases doesn't actually use these names at all). */
 
 #define OP_NAME_LIST \

diff --git a/src/pcre2_intmodedep.h b/src/pcre2_intmodedep.h
@@ -47,7 +47,7 @@ to have access to the hidden structures at all supported widths.
 
 Some of the mode-dependent macros are required at different widths for
 different parts of the pcre2test code (in particular, the included
-pcre2_printint.c file). We undefine them here so that they can be re-defined for
+pcre2_printint_inc.h file). We undefine them here so that they can be re-defined for
 multiple inclusions. Not all of these are used in pcre2test, but it's easier
 just to undefine them all. */
 

diff --git a/src/pcre2_jit_compile.c b/src/pcre2_jit_compile.c
@@ -14129,7 +14129,7 @@ return 0;
 
 #define INCLUDED_FROM_PCRE2_JIT_COMPILE
 
-#include "pcre2_jit_match.c"
-#include "pcre2_jit_misc.c"
+#include "pcre2_jit_match_inc.h"
+#include "pcre2_jit_misc_inc.h"
 
 /* End of pcre2_jit_compile.c */
diff --git a/src/pcre2_jit_match.c → src/pcre2_jit_match_inc.h b/src/pcre2_jit_match.c → src/pcre2_jit_match_inc.h
@@ -197,4 +197,4 @@ return match_data->rc;
 #endif  /* SUPPORT_JIT */
 }
 
-/* End of pcre2_jit_match.c */
+/* End of pcre2_jit_match_inc.h */
diff --git a/src/pcre2_jit_misc.c → src/pcre2_jit_misc_inc.h b/src/pcre2_jit_misc.c → src/pcre2_jit_misc_inc.h
@@ -231,4 +231,4 @@ return executable_sizes[0] + executable_sizes[1] + executable_sizes[2];
 #endif
 }
 
-/* End of pcre2_jit_misc.c */
+/* End of pcre2_jit_misc_inc.h */
diff --git a/src/pcre2_printint.c → src/pcre2_printint_inc.h b/src/pcre2_printint.c → src/pcre2_printint_inc.h
@@ -1125,4 +1125,4 @@ for(;;)
   }
 }
 
-/* End of pcre2_printint.c */
+/* End of pcre2_printint_inc.h */
-Original file line number
+Diff line change
@@ Expand Up @@
     #endif
     }
-    /* End of pcre2_jit_misc.c */
+    /* End of pcre2_jit_misc_inc.h */