bpo-45847: Port codecs and unicodedata to PY_STDLIB_MOD (GH-29685)

Author: tiran

Modules/Setup.stdlib.in

@@ -56,6 +56,14 @@
 # _elementtree libexpat via CAPI hook in pyexpat.
 @MODULE__ELEMENTTREE_TRUE@_elementtree _elementtree.c
 
+@MODULE__CODECS_CN_TRUE@_codecs_cn cjkcodecs/_codecs_cn.c
+@MODULE__CODECS_HK_TRUE@_codecs_hk cjkcodecs/_codecs_hk.c
+@MODULE__CODECS_ISO2022_TRUE@_codecs_iso2022 cjkcodecs/_codecs_iso2022.c
+@MODULE__CODECS_JP_TRUE@_codecs_jp cjkcodecs/_codecs_jp.c
+@MODULE__CODECS_KR_TRUE@_codecs_kr cjkcodecs/_codecs_kr.c
+@MODULE__CODECS_TW_TRUE@_codecs_tw cjkcodecs/_codecs_tw.c
+@MODULE__MULTIBYTECODEC_TRUE@_multibytecodec cjkcodecs/multibytecodec.c
+@MODULE_UNICODEDATA_TRUE@unicodedata unicodedata.c
 
 ############################################################################
 # Modules with some UNIX dependencies

configure

@@ -654,6 +654,22 @@ MODULE__SHA1_FALSE
 MODULE__SHA1_TRUE
 MODULE__MD5_FALSE
 MODULE__MD5_TRUE
+MODULE_UNICODEDATA_FALSE
+MODULE_UNICODEDATA_TRUE
+MODULE__MULTIBYTECODEC_FALSE
+MODULE__MULTIBYTECODEC_TRUE
+MODULE__CODECS_TW_FALSE
+MODULE__CODECS_TW_TRUE
+MODULE__CODECS_KR_FALSE
+MODULE__CODECS_KR_TRUE
+MODULE__CODECS_JP_FALSE
+MODULE__CODECS_JP_TRUE
+MODULE__CODECS_ISO2022_FALSE
+MODULE__CODECS_ISO2022_TRUE
+MODULE__CODECS_HK_FALSE
+MODULE__CODECS_HK_TRUE
+MODULE__CODECS_CN_FALSE
+MODULE__CODECS_CN_TRUE
 MODULE__ELEMENTTREE_FALSE
 MODULE__ELEMENTTREE_TRUE
 MODULE_PYEXPAT_FALSE
@@ -19913,6 +19929,102 @@ fi
 $as_echo "$py_cv_module__elementtree" >&6; }
 
 
+     if true; then
+  MODULE__CODECS_CN_TRUE=
+  MODULE__CODECS_CN_FALSE='#'
+else
+  MODULE__CODECS_CN_TRUE='#'
+  MODULE__CODECS_CN_FALSE=
+fi
+
+
+
+
+
+     if true; then
+  MODULE__CODECS_HK_TRUE=
+  MODULE__CODECS_HK_FALSE='#'
+else
+  MODULE__CODECS_HK_TRUE='#'
+  MODULE__CODECS_HK_FALSE=
+fi
+
+
+
+
+
+     if true; then
+  MODULE__CODECS_ISO2022_TRUE=
+  MODULE__CODECS_ISO2022_FALSE='#'
+else
+  MODULE__CODECS_ISO2022_TRUE='#'
+  MODULE__CODECS_ISO2022_FALSE=
+fi
+
+
+
+
+
+     if true; then
+  MODULE__CODECS_JP_TRUE=
+  MODULE__CODECS_JP_FALSE='#'
+else
+  MODULE__CODECS_JP_TRUE='#'
+  MODULE__CODECS_JP_FALSE=
+fi
+
+
+
+
+
+     if true; then
+  MODULE__CODECS_KR_TRUE=
+  MODULE__CODECS_KR_FALSE='#'
+else
+  MODULE__CODECS_KR_TRUE='#'
+  MODULE__CODECS_KR_FALSE=
+fi
+
+
+
+
+
+     if true; then
+  MODULE__CODECS_TW_TRUE=
+  MODULE__CODECS_TW_FALSE='#'
+else
+  MODULE__CODECS_TW_TRUE='#'
+  MODULE__CODECS_TW_FALSE=
+fi
+
+
+
+
+
+     if true; then
+  MODULE__MULTIBYTECODEC_TRUE=
+  MODULE__MULTIBYTECODEC_FALSE='#'
+else
+  MODULE__MULTIBYTECODEC_TRUE='#'
+  MODULE__MULTIBYTECODEC_FALSE=
+fi
+
+
+
+
+
+     if true; then
+  MODULE_UNICODEDATA_TRUE=
+  MODULE_UNICODEDATA_FALSE='#'
+else
+  MODULE_UNICODEDATA_TRUE='#'
+  MODULE_UNICODEDATA_FALSE=
+fi
+
+
+
+
+
 
   { $as_echo "$as_me:${as_lineno-$LINENO}: checking for stdlib extension module _md5" >&5
 $as_echo_n "checking for stdlib extension module _md5... " >&6; }
@@ -20669,6 +20781,38 @@ if test -z "${MODULE__ELEMENTTREE_TRUE}" && test -z "${MODULE__ELEMENTTREE_FALSE
   as_fn_error $? "conditional \"MODULE__ELEMENTTREE\" was never defined.
 Usually this means the macro was only invoked conditionally." "$LINENO" 5
 fi
+if test -z "${MODULE__CODECS_CN_TRUE}" && test -z "${MODULE__CODECS_CN_FALSE}"; then
+  as_fn_error $? "conditional \"MODULE__CODECS_CN\" was never defined.
+Usually this means the macro was only invoked conditionally." "$LINENO" 5
+fi
+if test -z "${MODULE__CODECS_HK_TRUE}" && test -z "${MODULE__CODECS_HK_FALSE}"; then
+  as_fn_error $? "conditional \"MODULE__CODECS_HK\" was never defined.
+Usually this means the macro was only invoked conditionally." "$LINENO" 5
+fi
+if test -z "${MODULE__CODECS_ISO2022_TRUE}" && test -z "${MODULE__CODECS_ISO2022_FALSE}"; then
+  as_fn_error $? "conditional \"MODULE__CODECS_ISO2022\" was never defined.
+Usually this means the macro was only invoked conditionally." "$LINENO" 5
+fi
+if test -z "${MODULE__CODECS_JP_TRUE}" && test -z "${MODULE__CODECS_JP_FALSE}"; then
+  as_fn_error $? "conditional \"MODULE__CODECS_JP\" was never defined.
+Usually this means the macro was only invoked conditionally." "$LINENO" 5
+fi
+if test -z "${MODULE__CODECS_KR_TRUE}" && test -z "${MODULE__CODECS_KR_FALSE}"; then
+  as_fn_error $? "conditional \"MODULE__CODECS_KR\" was never defined.
+Usually this means the macro was only invoked conditionally." "$LINENO" 5
+fi
+if test -z "${MODULE__CODECS_TW_TRUE}" && test -z "${MODULE__CODECS_TW_FALSE}"; then
+  as_fn_error $? "conditional \"MODULE__CODECS_TW\" was never defined.
+Usually this means the macro was only invoked conditionally." "$LINENO" 5
+fi
+if test -z "${MODULE__MULTIBYTECODEC_TRUE}" && test -z "${MODULE__MULTIBYTECODEC_FALSE}"; then
+  as_fn_error $? "conditional \"MODULE__MULTIBYTECODEC\" was never defined.
+Usually this means the macro was only invoked conditionally." "$LINENO" 5
+fi
+if test -z "${MODULE_UNICODEDATA_TRUE}" && test -z "${MODULE_UNICODEDATA_FALSE}"; then
+  as_fn_error $? "conditional \"MODULE_UNICODEDATA\" was never defined.
+Usually this means the macro was only invoked conditionally." "$LINENO" 5
+fi
 if test -z "${MODULE__MD5_TRUE}" && test -z "${MODULE__MD5_FALSE}"; then
   as_fn_error $? "conditional \"MODULE__MD5\" was never defined.
 Usually this means the macro was only invoked conditionally." "$LINENO" 5

configure.ac

@@ -6074,6 +6074,14 @@ PY_STDLIB_MOD([termios], [], [test "$ac_cv_header_termios_h" = yes])
 dnl _elementtree loads libexpat via CAPI hook in pyexpat
 PY_STDLIB_MOD([pyexpat], [], [], [$LIBEXPAT_CFLAGS], [$LIBEXPAT_LDFLAGS])
 PY_STDLIB_MOD([_elementtree], [], [], [$LIBEXPAT_CFLAGS], [])
+PY_STDLIB_MOD_SIMPLE([_codecs_cn])
+PY_STDLIB_MOD_SIMPLE([_codecs_hk])
+PY_STDLIB_MOD_SIMPLE([_codecs_iso2022])
+PY_STDLIB_MOD_SIMPLE([_codecs_jp])
+PY_STDLIB_MOD_SIMPLE([_codecs_kr])
+PY_STDLIB_MOD_SIMPLE([_codecs_tw])
+PY_STDLIB_MOD_SIMPLE([_multibytecodec])
+PY_STDLIB_MOD_SIMPLE([unicodedata])
 
 dnl By default we always compile these even when OpenSSL is available
 dnl (issue #14693). The modules are small.

setup.py

@@ -1022,7 +1022,7 @@ def detect_simple_extensions(self):
         # profiler (_lsprof is for cProfile.py)
         self.add(Extension('_lsprof', ['_lsprof.c', 'rotatingtree.c']))
         # static Unicode character database
-        self.add(Extension('unicodedata', ['unicodedata.c']))
+        self.addext(Extension('unicodedata', ['unicodedata.c']))
         # _opcode module
         self.add(Extension('_opcode', ['_opcode.c']))
         # asyncio speedups
@@ -1432,11 +1432,12 @@ def detect_expat_elementtree(self):
 
     def detect_multibytecodecs(self):
         # Hye-Shik Chang's CJKCodecs modules.
-        self.add(Extension('_multibytecodec',
-                           ['cjkcodecs/multibytecodec.c']))
+        self.addext(Extension('_multibytecodec',
+                              ['cjkcodecs/multibytecodec.c']))
         for loc in ('kr', 'jp', 'cn', 'tw', 'hk', 'iso2022'):
-            self.add(Extension('_codecs_%s' % loc,
-                               ['cjkcodecs/_codecs_%s.c' % loc]))
+            self.addext(Extension(
+                f'_codecs_{loc}', [f'cjkcodecs/_codecs_{loc}.c']
+            ))
 
     def detect_multiprocessing(self):
         # Richard Oudkerk's multiprocessing module