diff options
author | Robert Brady <rwb197@ecs.soton.ac.uk> | 1999-08-02 20:36:55 +0000 |
---|---|---|
committer | Robert Brady <rbrady@src.gnome.org> | 1999-08-02 20:36:55 +0000 |
commit | 7e6f5f5e02cd8cb5c3977c708370f975729063af (patch) | |
tree | 137d039895e35af61045326482e85cbf771f3a70 | |
parent | 272b02eb016ac210da5ccbb3707e856e48a3ac32 (diff) | |
download | libxml2-USING_LIBUNICODE.tar.gz |
use libunicode to convert between latin1 and utf-8.USING_LIBUNICODE
1999-08-02 Robert Brady <rwb197@ecs.soton.ac.uk>
* encoding.c, Makefile.am: use libunicode to convert between latin1
and utf-8.
-rw-r--r-- | ChangeLog | 5 | ||||
-rw-r--r-- | Makefile.am | 11 | ||||
-rw-r--r-- | encoding.c | 165 |
3 files changed, 74 insertions, 107 deletions
@@ -1,3 +1,8 @@ +1999-08-02 Robert Brady <rwb197@ecs.soton.ac.uk> + + * encoding.c, Makefile.am: use libunicode to convert between latin1 + and utf-8. + Tue Jul 27 21:43:00 CEST 1999 Daniel Veillard <Daniel.Veillard@w3.org> * xpath.[ch] : improvements and debug of the XPath implementation diff --git a/Makefile.am b/Makefile.am index 08247e22..ce0d526f 100644 --- a/Makefile.am +++ b/Makefile.am @@ -2,11 +2,11 @@ SUBDIRS = doc -INCLUDES = -I@srcdir@ @CORBA_CFLAGS@ $(VERSION_FLAGS) +INCLUDES = -I@srcdir@ @CORBA_CFLAGS@ $(VERSION_FLAGS) `unicode-config --cflags` VERSION_FLAGS = -DLIBXML_VERSION=\"@LIBXML_VERSION@\" -noinst_PROGRAMS=tester testSAX testHTML testXPath +noinst_PROGRAMS=tester testSAX testHTML testXPath mytest bin_SCRIPTS=xml-config @@ -44,13 +44,18 @@ xmlinc_HEADERS = \ valid.h DEPS = $(top_builddir)/libxml.la -LDADDS = $(top_builddir)/libxml.la @Z_LIBS@ +LDADDS = $(top_builddir)/libxml.la @Z_LIBS@ `unicode-config --libs` tester_SOURCES=tester.c tester_LDFLAGS = tester_DEPENDENCIES = $(DEPS) tester_LDADD= $(LDADDS) +mytest_SOURCES=mytest.c +mytest_LDFLAGS = +mytest_DEPENDENCIES = $(DEPS) +mytest_LDADD= $(LDADDS) + testSAX_SOURCES=testSAX.c testSAX_LDFLAGS = testSAX_DEPENDENCIES = $(DEPS) @@ -22,8 +22,12 @@ #include <ctype.h> #include <string.h> #include <stdio.h> + +#include <unicode.h> + #include "encoding.h" + /* * From rfc2044: encoding of the Unicode values on UTF-8: * @@ -49,25 +53,20 @@ int isolat1ToUTF8(unsigned char* out, int outlen, unsigned char* in, int inlen) { - unsigned char* outstart= out; - unsigned char* outend= out+outlen; - unsigned char* inend= in+inlen; - unsigned char c; - - while (in < inend) { - c= *in++; - if (c < 0x80) { - if (out >= outend) return -1; - *out++ = c; - } - else { - if (out >= outend) return -1; - *out++ = 0xC0 | (c >> 6); - if (out >= outend) return -1; - *out++ = 0x80 | (0x3F & c); - } - } - return out-outstart; + iconv_t u; + size_t i; + + unsigned char *o = out; + + u = unicode_iconv_open("UTF-8", "ISO-8859-1"); + i = unicode_iconv(u, &in, &inlen, &out, &outlen); + + unicode_iconv_close(u); + + if (i==-1) + return i; + else + return out - o; } /** @@ -86,24 +85,20 @@ isolat1ToUTF8(unsigned char* out, int outlen, unsigned char* in, int inlen) int UTF8Toisolat1(unsigned char* out, int outlen, unsigned char* in, int inlen) { - unsigned char* outstart= out; - unsigned char* outend= out+outlen; - unsigned char* inend= in+inlen; - unsigned char c; - - while (in < inend) { - c= *in++; - if (c < 0x80) { - if (out >= outend) return -1; - *out++= c; - } - else if (((c & 0xFE) == 0xC2) && in<inend) { - if (out >= outend) return -1; - *out++= ((c & 0x03) << 6) | (*in++ & 0x3F); - } - else return -2; - } - return out-outstart; + iconv_t u; + size_t i; + + unsigned char *o = out; + + u = unicode_iconv_open("ISO-8859-1", "UTF-8"); + i = unicode_iconv(u, &in, &inlen, &out, &outlen); + + unicode_iconv_close(u); + + if (i==-1) + return i; + else + return out - o; } /** @@ -120,38 +115,20 @@ UTF8Toisolat1(unsigned char* out, int outlen, unsigned char* in, int inlen) int UTF16ToUTF8(unsigned char* out, int outlen, unsigned short* in, int inlen) { - unsigned char* outstart= out; - unsigned char* outend= out+outlen; - unsigned short* inend= in+inlen; - unsigned int c, d; - int bits; - - while (in < inend) { - c= *in++; - if ((c & 0xFC00) == 0xD800) { /* surrogates */ - if ((in<inend) && (((d=*in++) & 0xFC00) == 0xDC00)) { - c &= 0x03FF; - c <<= 10; - c |= d & 0x03FF; - c += 0x10000; - } - else return -1; - } - - /* assertion: c is a single UTF-4 value */ - - if (out >= outend) return -1; - if (c < 0x80) { *out++= c; bits= -6; } - else if (c < 0x800) { *out++= (c >> 6) | 0xC0; bits= 0; } - else if (c < 0x10000) { *out++= (c >> 12) | 0xE0; bits= 6; } - else { *out++= (c >> 18) | 0xF0; bits= 12; } - - for ( ; bits < 0; bits-= 6) { - if (out >= outend) return -1; - *out++= (c >> bits) & 0x3F; - } - } - return out-outstart; + iconv_t u; + size_t i; + + unsigned char *o = out; + + u = unicode_iconv_open("UTF-8", "UTF-16"); + i = unicode_iconv(u, &in, &inlen, &out, &outlen); + + unicode_iconv_close(u); + + if (i==-1) + return i; + else + return out - o; } /** @@ -170,40 +147,20 @@ UTF16ToUTF8(unsigned char* out, int outlen, unsigned short* in, int inlen) int UTF8ToUTF16(unsigned short* out, int outlen, unsigned char* in, int inlen) { - unsigned short* outstart= out; - unsigned short* outend= out+outlen; - unsigned char* inend= in+inlen; - unsigned int c, d, trailing; - - while (in < inend) { - d= *in++; - if (d < 0x80) { c= d; trailing= 0; } - else if (d < 0xC0) return -2; /* trailing byte in leading position */ - else if (d < 0xE0) { c= d & 0x1F; trailing= 1; } - else if (d < 0xF0) { c= d & 0x0F; trailing= 2; } - else if (d < 0xF8) { c= d & 0x07; trailing= 3; } - else return -2; /* no chance for this in UTF-16 */ - - for ( ; trailing; trailing--) { - if ((in >= inend) || (((d= *in++) & 0xC0) != 0x80)) return -1; - c <<= 6; - c |= d & 0x3F; - } - - /* assertion: c is a single UTF-4 value */ - if (c < 0x10000) { - if (out >= outend) return -1; - *out++ = c; - } - else if (c < 0x110000) { - if (out+1 >= outend) return -1; - c -= 0x10000; - *out++ = 0xD800 | (c >> 10); - *out++ = 0xDC00 | (c & 0x03FF); - } - else return -1; - } - return out-outstart; + iconv_t u; + size_t i; + + unsigned short *o = out; + + u = unicode_iconv_open("UTF-16", "UTF-8"); + i = unicode_iconv(u, &in, &inlen, &out, &outlen); + + unicode_iconv_close(u); + + if (i==-1) + return i; + else + return out - o; } @@ -247,7 +204,7 @@ xmlDetectCharEncoding(const unsigned char* in) /** * xmlParseCharEncoding: - * @name: the encoding name as parsed, in UTF-8 format (ASCCI actually) + * @name: the encoding name as parsed, in UTF-8 format (ASCII actually) * * Conpare the string to the known encoding schemes already known. Note * that the comparison is case insensitive accordingly to the section |