Mercurial > dovecot > core-2.2

changeset 609:5470c0cb13a7 HEAD

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
We can support UTF-8 charset too without any translations.
author Timo Sirainen <tss@iki.fi>
date Wed, 13 Nov 2002 15:47:59 +0200
parents debb8468514e
children bc766660efc4
files src/lib-charset/Makefile.am src/lib-charset/charset-ascii.c src/lib-charset/charset-iconv.c src/lib-charset/charset-utf8.c
diffstat 4 files changed, 93 insertions(+), 81 deletions(-) [+]
line wrap: on
line diff
--- a/src/lib-charset/Makefile.am	Wed Nov 13 13:08:18 2002 +0200
+++ b/src/lib-charset/Makefile.am	Wed Nov 13 15:47:59 2002 +0200
@@ -4,8 +4,8 @@
 	-I$(top_srcdir)/src/lib
 
 libcharset_a_SOURCES = \
-	charset-ascii.c \
-	charset-iconv.c
+	charset-iconv.c \
+	charset-utf8.c
 
 noinst_HEADERS = \
 	charset-utf8.h
--- a/src/lib-charset/charset-ascii.c	Wed Nov 13 13:08:18 2002 +0200
+++ /dev/null	Thu Jan 01 00:00:00 1970 +0000
@@ -1,71 +0,0 @@
-/* Copyright (C) 2002 Timo Sirainen */
-
-#include "lib.h"
-#include "charset-utf8.h"
-
-#ifndef HAVE_ICONV_H
-
-#include <ctype.h>
-
-struct _CharsetTranslation {
-	int dummy;
-};
-
-static CharsetTranslation ascii_translation;
-
-CharsetTranslation *charset_to_utf8_begin(const char *charset,
-					  int *unknown_charset)
-{
-	if (unknown_charset != NULL)
-		*unknown_charset = FALSE;
-
-	if (strcasecmp(charset, "us-ascii") != 0 &&
-	    strcasecmp(charset, "ascii") != 0) {
-		/* no support for non-ascii charsets */
-		if (unknown_charset != NULL)
-			*unknown_charset = TRUE;
-		return NULL;
-	}
-
-	return &ascii_translation;
-}
-
-void charset_to_utf8_end(CharsetTranslation *t __attr_unused__)
-{
-}
-
-void charset_to_utf8_reset(CharsetTranslation *t __attr_unused__)
-{
-}
-
-int charset_to_ucase_utf8(CharsetTranslation *t __attr_unused__,
-			  const unsigned char **inbuf, size_t *insize,
-			  unsigned char *outbuf, size_t *outsize)
-{
-	size_t max_size, i;
-
-	max_size = I_MIN(*insize, *outsize);
-	for (i = 0; i < max_size; i++)
-		outbuf[i] = i_toupper((*inbuf)[i]);
-
-	*insize = 0;
-	*outsize = max_size;
-
-	return TRUE;
-}
-
-const char *
-charset_to_ucase_utf8_string(const char *charset, int *unknown_charset,
-			     const unsigned char *buf,
-			     size_t *size __attr_unused__)
-{
-	if (charset == NULL || strcasecmp(charset, "us-ascii") == 0 ||
-	    strcasecmp(charset, "ascii") == 0)
-		return str_ucase(t_strdup_noconst(buf));
-
-	if (unknown_charset != NULL)
-		*unknown_charset = TRUE;
-	return NULL;
-}
-
-#endif
--- a/src/lib-charset/charset-iconv.c	Wed Nov 13 13:08:18 2002 +0200
+++ b/src/lib-charset/charset-iconv.c	Wed Nov 13 15:47:59 2002 +0200
@@ -10,6 +10,7 @@
 
 struct _CharsetTranslation {
 	iconv_t cd;
+	int ascii;
 };
 
 CharsetTranslation *charset_to_utf8_begin(const char *charset,
@@ -17,16 +18,22 @@
 {
 	CharsetTranslation *t;
 	iconv_t cd;
+	int ascii;
 
 	if (unknown_charset != NULL)
 		*unknown_charset = FALSE;
 
 	if (strcasecmp(charset, "us-ascii") == 0 ||
 	    strcasecmp(charset, "ascii") == 0) {
-		/* no need to do any actual translation */
+		cd = NULL;
+		ascii = TRUE;
+	} else if (strcasecmp(charset, "UTF-8") == 0 ||
+		   strcasecmp(charset, "UTF8") == 0) {
 		cd = NULL;
+		ascii = FALSE;
 	} else {
-		cd = iconv_open("UTF8", charset);
+		ascii = FALSE;
+		cd = iconv_open("UTF-8", charset);
 		if (cd == (iconv_t)-1) {
 			if (unknown_charset != NULL)
 				*unknown_charset = TRUE;
@@ -36,6 +43,7 @@
 
 	t = i_new(CharsetTranslation, 1);
 	t->cd = cd;
+	t->ascii = ascii;
 	return t;
 }
 
@@ -60,10 +68,10 @@
 	size_t outleft, max_size, i;
 
 	if (t->cd == NULL) {
-		/* ascii - just copy it to outbuf uppercased */
+		/* no translation needed - just copy it to outbuf uppercased */
 		max_size = I_MIN(*insize, *outsize);
 		for (i = 0; i < max_size; i++)
-			outbuf[i] = i_toupper((*inbuf)[i]);
+			outbuf[i] = i_toupper((*inbuf)[i]); /* FIXME: utf8 */
 		*insize = 0;
 		*outsize = max_size;
 		return TRUE;
@@ -86,7 +94,7 @@
 
 	max_size = *outsize;
 	for (i = 0; i < max_size; i++)
-		outbuf[i] = i_toupper(outbuf[i]);
+		outbuf[i] = i_toupper(outbuf[i]); /* FIXME: utf8 */
 
 	return TRUE;
 }
@@ -103,7 +111,7 @@
 	    strcasecmp(charset, "ascii") == 0)
 		return str_ucase(t_strdup_noconst(buf));
 
-	cd = iconv_open("UTF8", charset);
+	cd = iconv_open("UTF-8", charset);
 	if (cd == (iconv_t)-1) {
 		if (unknown_charset != NULL)
 			*unknown_charset = TRUE;
@@ -139,8 +147,7 @@
 	*outpos++ = '\0';
 	t_buffer_alloc(*size + 1);
 
-	/* FIXME: this works only for ASCII */
-	str_ucase(outbuf);
+	str_ucase(outbuf); /* FIXME: utf8 */
 
 	iconv_close(cd);
 	return outbuf;
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/src/lib-charset/charset-utf8.c	Wed Nov 13 15:47:59 2002 +0200
@@ -0,0 +1,76 @@
+/* Copyright (C) 2002 Timo Sirainen */
+
+#include "lib.h"
+#include "charset-utf8.h"
+
+#ifndef HAVE_ICONV_H
+
+#include <ctype.h>
+
+struct _CharsetTranslation {
+	int dummy;
+};
+
+static CharsetTranslation ascii_translation, utf8_translation;
+
+CharsetTranslation *charset_to_utf8_begin(const char *charset,
+					  int *unknown_charset)
+{
+	if (unknown_charset != NULL)
+		*unknown_charset = FALSE;
+
+	if (strcasecmp(charset, "us-ascii") == 0 ||
+	    strcasecmp(charset, "ascii") == 0)
+		return &ascii_translation;
+
+	if (strcasecmp(charset, "UTF-8") == 0 ||
+	    strcasecmp(charset, "UTF8") == 0)
+		return &utf8_translation;
+
+	/* no support for charsets that need translation */
+	if (unknown_charset != NULL)
+		*unknown_charset = TRUE;
+	return NULL;
+}
+
+void charset_to_utf8_end(CharsetTranslation *t __attr_unused__)
+{
+}
+
+void charset_to_utf8_reset(CharsetTranslation *t __attr_unused__)
+{
+}
+
+int charset_to_ucase_utf8(CharsetTranslation *t __attr_unused__,
+			  const unsigned char **inbuf, size_t *insize,
+			  unsigned char *outbuf, size_t *outsize)
+{
+	size_t max_size, i;
+
+	max_size = I_MIN(*insize, *outsize);
+	for (i = 0; i < max_size; i++)
+		outbuf[i] = i_toupper((*inbuf)[i]); /* FIXME: utf8 */
+
+	*insize = 0;
+	*outsize = max_size;
+
+	return TRUE;
+}
+
+const char *
+charset_to_ucase_utf8_string(const char *charset, int *unknown_charset,
+			     const unsigned char *buf,
+			     size_t *size __attr_unused__)
+{
+	if (charset == NULL || strcasecmp(charset, "us-ascii") == 0 ||
+	    strcasecmp(charset, "ascii") == 0 ||
+	    strcasecmp(charset, "UTF-8") == 0 ||
+	    strcasecmp(charset, "UTF8") == 0)
+		return str_ucase(t_strdup_noconst(buf)); /* FIXME: utf8 */
+
+	if (unknown_charset != NULL)
+		*unknown_charset = TRUE;
+	return NULL;
+}
+
+#endif