lib/util/charset: add back loading of charset modules

For autoconf builds these remain as modules, for waf builds they are
built into the charset library.

This is required to provide the CP850 charset when iconv is not available.

The charset modules static for the waf builds because with proper
shared libs, there isn't the same need for these to be in seperate .so
files.  The modules are also not produced if a system iconv is found,
except for developers, to allow testing of both code paths.

Andrew Bartlett

2 files changed, 64 insertions, 34 deletions

diff --git a/lib/util/charset/iconv.c b/lib/util/charset/iconv.c
index f63c4e6028..fd8b8ae8ee 100644
--- a/lib/util/charset/iconv.c
+++ b/lib/util/charset/iconv.c
@@ -244,7 +244,7 @@ static int smb_iconv_t_destructor(smb_iconv_t hwd)
 }
 
 _PUBLIC_ smb_iconv_t smb_iconv_open_ex(TALLOC_CTX *mem_ctx, const char *tocode, 
-			      const char *fromcode, bool native_iconv)
+			      const char *fromcode, bool allow_native_iconv)
 {
 	smb_iconv_t ret;
 	const struct charset_functions *from=NULL, *to=NULL;
@@ -268,6 +268,7 @@ _PUBLIC_ smb_iconv_t smb_iconv_open_ex(TALLOC_CTX *mem_ctx, const char *tocode,
 		return ret;
 	}
 
+	/* check if we have a builtin function for this conversion */
 	for (i=0;i<ARRAY_SIZE(builtin_functions);i++) {
 		if (strcasecmp(fromcode, builtin_functions[i].name) == 0) {
 			from = &builtin_functions[i];
@@ -277,43 +278,68 @@ _PUBLIC_ smb_iconv_t smb_iconv_open_ex(TALLOC_CTX *mem_ctx, const char *tocode,
 		}
 	}
 
-	if (from == NULL) {
-		for (from=charsets; from; from=from->next) {
-			if (strcasecmp(from->name, fromcode) == 0) break;
-		}
-	}
-
-	if (to == NULL) {
-		for (to=charsets; to; to=to->next) {
-			if (strcasecmp(to->name, tocode) == 0) break;
-		}
-	}
-
 #ifdef HAVE_NATIVE_ICONV
-	if ((!from || !to) && !native_iconv) {
-		goto failed;
-	}
-	if (!from) {
-		ret->pull = sys_iconv;
+	/* the from and to varaibles indicate a samba module or
+	 * internal conversion, ret->pull and ret->push are
+	 * initialised only in this block for iconv based
+	 * conversions */
+
+	if (allow_native_iconv && from == NULL) {
 		ret->cd_pull = iconv_open("UTF-16LE", fromcode);
 		if (ret->cd_pull == (iconv_t)-1)
 			ret->cd_pull = iconv_open("UCS-2LE", fromcode);
-		if (ret->cd_pull == (iconv_t)-1) goto failed;
+		if (ret->cd_pull != (iconv_t)-1) {
+			ret->pull = sys_iconv;
+		}
 	}
-
-	if (!to) {
-		ret->push = sys_iconv;
+	
+	if (allow_native_iconv && to == NULL) {
 		ret->cd_push = iconv_open(tocode, "UTF-16LE");
 		if (ret->cd_push == (iconv_t)-1)
 			ret->cd_push = iconv_open(tocode, "UCS-2LE");
-		if (ret->cd_push == (iconv_t)-1) goto failed;
+		if (ret->cd_push != (iconv_t)-1) {
+			ret->push = sys_iconv;
+		}
 	}
-#else
-	if (!from || !to) {
-		goto failed;
+#endif
+
+	/* If iconv was unable to provide the conversion, or if use of
+	 * it was disabled, and it wasn't a builtin charset, try a
+	 * module */
+	if (ret->pull == NULL && from == NULL) {
+		from = find_charset_functions(fromcode);
+	}
+	
+	if (ret->push == NULL && to == NULL) {
+		to = find_charset_functions(tocode);
+	}
+
+	/* In the WAF builds, all charset modules are linked in at compile
+	 * time, as we have shared libs.  Using run-time loading as well will
+	 * cause dependency loops.  For the autoconf build, try loading from a module */
+#ifndef _SAMBA_WAF_BUILD_
+	/* check if there is a module available that can do this conversion */
+	if (from == NULL && ret->pull == NULL && NT_STATUS_IS_OK(smb_probe_module("charset", fromcode))) {
+		if (!(from = find_charset_functions(fromcode))) {
+			DEBUG(0, ("Module %s doesn't provide charset %s!\n", fromcode, fromcode));
+		}
+	}
+	
+	if (to == NULL && ret->push == NULL && NT_STATUS_IS_OK(smb_probe_module("charset", tocode))) {
+		if (!(to = find_charset_functions(tocode))) {
+			DEBUG(0, ("Module %s doesn't provide charset %s!\n", tocode, tocode));
+		}
 	}
 #endif
 
+	if (ret->pull == NULL && from == NULL) {
+		goto failed;
+	}
+	
+	if (ret->push == NULL && to == NULL) {
+		goto failed;
+	}
+
 	/* check for conversion to/from ucs2 */
 	if (is_utf16(fromcode) && to) {
 		ret->direct = to->push;
diff --git a/lib/util/charset/wscript_build b/lib/util/charset/wscript_build
index 7a9918046d..d659c7a450 100644
--- a/lib/util/charset/wscript_build
+++ b/lib/util/charset/wscript_build
@@ -10,35 +10,39 @@ bld.SAMBA_SUBSYSTEM('charset',
                     deps='DYNCONFIG ICONV_WRAPPER',
                     public_deps='talloc')
 
+# In the WAF builds, all charset modules are linked in at compile
+# time, as we have shared libs.  Using run-time loading as well will
+# cause dependency loops
+
 bld.SAMBA_MODULE('charset_weird',
                  subsystem='charset',
                  source='weird.c',
-                 init_function='',
+                 init_function='charset_weird_init',
                  deps='samba-util',
-                 internal_module=bld.SAMBA3_IS_STATIC_MODULE('charset_weird'),
+                 internal_module=True,
                  enabled=bld.SAMBA3_IS_ENABLED_MODULE('charset_weird'))
 
 bld.SAMBA_MODULE('charset_CP850',
                  subsystem='charset',
                  source='CP850.c',
-                 init_function='',
+                 init_function='charset_CP850_init',
                  deps='samba-util',
-                 internal_module=bld.SAMBA3_IS_STATIC_MODULE('charset_CP850'),
+                 internal_module=True,
                  enabled=bld.SAMBA3_IS_ENABLED_MODULE('charset_CP850'))
 
 bld.SAMBA_MODULE('charset_CP437',
                  subsystem='charset',
                  source='CP437.c',
-                 init_function='',
+                 init_function='charset_CP437_init',
                  deps='samba-util',
-                 internal_module=bld.SAMBA3_IS_STATIC_MODULE('charset_CP437'),
+                 internal_module=True,
                  enabled=bld.SAMBA3_IS_ENABLED_MODULE('charset_CP437'))
 
 bld.SAMBA_MODULE('charset_macosxfs',
                  subsystem='charset',
                  source='charset_macosxfs.c',
-                 init_function='',
-                 internal_module=bld.SAMBA3_IS_STATIC_MODULE('charset_macosxfs'),
+                 init_function='charset_macosxfs_init',
+                 internal_module=True,
                  enabled=bld.SAMBA3_IS_ENABLED_MODULE('charset_macosxfs'))