ISO-8859-1 support

    - Try to detect genuine ISO-8859-1 encoding

6 files changed, 118 insertions, 8 deletions

diff --git a/VERSION b/VERSION
index 69258b7..1a03094 100644
--- a/VERSION
+++ b/VERSION
@@ -1 +1 @@
-0.1.9CVS
+0.1.9
diff --git a/autogen.sh b/autogen.sh
index d23378a..82a102f 100755
--- a/autogen.sh
+++ b/autogen.sh
@@ -1,8 +1,71 @@
-#!/bin/bash
+#!/bin/sh
 
+PROG=`basename $0`
+
+KEYFILE=src/librcd.c
+
+# Some OS's have multiple versions (autoconf259, etc.) and don't have an 
+# autoconf binary
+
+AUTOCONF=`which autoconf`
+if test x"${AUTOCONF}" != x -a -f ${AUTOCONF}
+then
+    AUTOCONF=autoconf
+    AUTOMAKE=automake
+    ACLOCAL=aclocal
+    LIBTOOLIZE=libtoolize
+    AUTOHEADER=autoheader
+else
+    FINDPATH=`echo ${PATH}|sed -e 's,:, ,g'` 
+    AUTOCONF=`find ${FINDPATH} -name "autoconf*"|sort -r|head -1`
+    AUTOMAKE=`find ${FINDPATH} -name "automake*"|sort -r|head -1`
+    ACLOCAL=`find ${FINDPATH} -name "aclocal*"|sort -r|head -1`
+    LIBTOOLIZE=`find ${FINDPATH} -name "libtoolize*"|sort -r|head -1`
+    AUTOHEADER=`find /usr/bin /usr/local/bin -name "autoheader*"|sort -r|head -1`
+    echo "$0: autoconf: using ${AUTOCONF}"
+    echo "$0: automake: using ${AUTOMAKE}"
+    echo "$0: aclocal: using ${ACLOCAL}"
+    echo "$0: libtoolize: using ${LIBTOOLIZE}"
+    echo "$0: autoheader: using ${AUTOHEADER}"
+fi
+
+GETTEXTIZE_FLAGS=--no-changelog
+AUTOPOINT_FLAGS=
+LIBTOOLIZE_FLAGS=--copy
+
+# Some OS's require /usr/local/share/aclocal
+
+if test ! -d /usr/local/share/aclocal
+then
+  ACLOCAL_FLAGS=''
+else
+  ACLOCAL_FLAGS='-I /usr/local/share/aclocal'
+fi
+AUTOHEADER_FLAGS=-Wall
+AUTOMAKE_FLAGS='--add-missing --copy -Wall'
+AUTOCONF_FLAGS=-Wno-obsolete
+
+die()
+{
+    err=$?
+    echo "$PROG: exited by previous error(s), return code was $err" >&2
+    exit 1
+}
+
+if [ ! -f $KEYFILE ]
+then
+    echo "$PROG: key-file \`$KEYFILE' not found, exiting." >&2
+    echo "$PROG: hint: you should run $PROG from top-level sourse directory." >&2
+    exit 1
+fi
+
+# gettextize ${GETTEXTIZE_FLAGS}
+# autopoint ${AUTOPOINT_FLAGS}
+# xml-i18n-toolize       || die
+# intltoolize            || die
+${LIBTOOLIZE} ${LIBTOOLIZE_FLAGS} --force || die
+${ACLOCAL} ${ACLOCAL_FLAGS}        || die
+${AUTOHEADER} ${AUTOHEADER_FLAGS}  || die
+${AUTOMAKE} ${AUTOMAKE_FLAGS}      || die
+${AUTOCONF} ${AUTOCONF_FLAGS}      || die
 
-rm -f configure
-aclocal
-libtoolize --copy --force
-automake --copy --add-missing
-autoconf
diff --git a/configure.in b/configure.in
index 1ad8cd6..15d197f 100644
--- a/configure.in
+++ b/configure.in
@@ -1,5 +1,6 @@
 dnl Process this file with autoconf to produce a configure script.
 AC_INIT(src/librcd.c)
+AC_CONFIG_HEADERS(config.h)
 
 PACKAGE=librcd
 LIBRCD_VERSION_MAJOR=`cat VERSION | sed -e s/CVS// | cut -d . -f 1 | sed -e s/^$/0/`
@@ -17,6 +18,14 @@ AC_SUBST(LIBRCD_VERSION_INFO)
 
 AM_INIT_AUTOMAKE($PACKAGE, $VERSION)
 
+AC_ARG_ENABLE( latin,
+    [  --disable-latin		disable detection of ISO-8859-1],,
+    disable_latin="yes")
+
+if test "x$disable_latin" == "xyes"; then
+    AC_DEFINE(DETECT_LATIN,1,[Defines if ISO-8859-1 detection is enabled])
+fi
+
 AC_PROG_CC
 AC_PROG_INSTALL
 AM_PROG_LIBTOOL
diff --git a/release b/release
index fa578bf..b51819c 100755
--- a/release
+++ b/release
@@ -49,4 +49,6 @@ cp -a * $RDIR/$rname
     fi
     tar cjf $rname.tar.bz2 $rname
     rm -rf $rname
+
+    scp $rname.tar.bz2 csa@dside.dyndns.org:/home/csa/public_html/files/rusxmms/
 )
diff --git a/src/librcd.c b/src/librcd.c
index 55eefc0..36986cc 100644
--- a/src/librcd.c
+++ b/src/librcd.c
@@ -1,5 +1,7 @@
 #include <stdio.h>
 
+#include "../config.h"
+
 #define _LIBRCD_C
 #include "librcd.h"
 
@@ -252,6 +254,36 @@ static int check_utf8(const unsigned char *buf, int len) {
     return res;
 }
 
+/* In russian language we will have whole word consisting of >127 characters,
+with latin languages there is in every word besides umlauts should exist at
+least one standard latin character with code < 127. */
+static int check_latin(const unsigned char *buf, int len) {
+    long i;
+    int word = 0;
+    int latin = 0;
+    
+    for (i=0;i<len;i++) {
+	if (buf[i]<128) {
+	    if (((buf[i]>='a')&&(buf[i]<='z'))||((buf[i]>='A')&&(buf[i]<='Z'))) {
+		    // Latin character inside a word, so it isn't cyrillic word
+		latin++;
+	    } else {
+		    // Treating as a word separator.
+		if (word > 0) {
+		    if (!latin) return 0;
+		    if ((word/latin)>4) return 0;
+		}
+
+		word = 0;
+		latin = 0;
+	    }
+	} else {
+		// Could be cyrillic word
+	    if (word>=0) word++;
+	}
+    }
+    return 1;
+}
 
 
 rcd_russian_charset rcdGetRussianCharset(const char *buf,int len) {
@@ -259,6 +291,9 @@ rcd_russian_charset rcdGetRussianCharset(const char *buf,int len) {
 
     l = len?len:strlen(buf);
     if (check_utf8(buf,l)>1) return RUSSIAN_CHARSET_UTF8;
+#ifdef DETECT_LATIN
+    if (check_latin(buf,l)) return RUSSIAN_CHARSET_LATIN;
+#endif /* DETECT_LATIN */
     return is_win_charset2(buf,l);
 }
 
diff --git a/src/librcd.h b/src/librcd.h
index 56db6c8..6fc3281 100644
--- a/src/librcd.h
+++ b/src/librcd.h
@@ -9,7 +9,8 @@ enum rcd_russian_charset_t {
     RUSSIAN_CHARSET_WIN = 0,
     RUSSIAN_CHARSET_KOI,
     RUSSIAN_CHARSET_UTF8,
-    RUSSIAN_CHARSET_ALT
+    RUSSIAN_CHARSET_ALT,
+    RUSSIAN_CHARSET_LATIN
 };
 typedef enum rcd_russian_charset_t rcd_russian_charset;