From c9908d97e3ab4e4d0cd87435c52aea88a5a4abc0 Mon Sep 17 00:00:00 2001 From: Matthias Melcher Date: Tue, 18 Apr 2006 13:07:42 +0000 Subject: THIS FIX CONTAINS TWO MORE FILES THAT MUST BE COMPILED. I would like to ask the maintainers of the build environments to please add these files to the setup: src/fl_encoding_latin1.cxx src/fl_encoding_mac_roman.cxx I ADDED SOME DOCUMENTATION THAT NEEDS TO BE FIXED. Beeing not a native English speaker, I have a hard time writing documentation. Would someone please update my babbeling in documentation/drawing.html? Thanks. This commit fixes some very basic problems with OS X's code page in preparation for the compose-character keyboard fix. It also fixes issues with MS Windows and X11 not rendering the characters in the Western (Latin-1) set between 0x80 and 0x9F. In the original ISO font, they were unused, but are now assigned to international characters like the Euro currency sign. This patch also tries to fix one basic flaw with FLTK 1 and font encoding. I will not put much more work into this because FLTK 1.2 and FLTK 2 fix the problem entirely by using UTF-8 instead of 8-bit "C"-style strings. All these changes are only meaningful for foreign language users or users of special characters like the Euro, the Degree or the Permille symbol. A short explanation follows. Max OS X uses a different code page than X11 and Win32. This means that all characters above 0x7f have an entirely different meaning. If your source code contains international characters, your text will appear different if you change to another OS. This patch provides two functions that convert text with international characters from the character set of the source code into the local character set. Two more functions are provided to convert them back. The functions are fl_latin1_to_local (source is in Win32 or X11), fl_mac_roman_to_local (source was written on OS X) and the corresponding fl_local_to_latin1 and fl_local_to_mac_roman, which are very useful if yoou want to store strings with intl. characters that will be moved between systems. All this is assuming a "Western" code page as it is common in the Americas and most of Europe. User of other languages will have to use FLTK 2. git-svn-id: file:///fltk/svn/fltk/branches/branch-1.1@4975 ea41ed52-d2ee-0310-a9c1-e6b18d33e121 --- src/fl_encoding_latin1.cxx | 136 +++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 136 insertions(+) create mode 100644 src/fl_encoding_latin1.cxx (limited to 'src/fl_encoding_latin1.cxx') diff --git a/src/fl_encoding_latin1.cxx b/src/fl_encoding_latin1.cxx new file mode 100644 index 000000000..a0115199d --- /dev/null +++ b/src/fl_encoding_latin1.cxx @@ -0,0 +1,136 @@ +// +// "$Id: $" +// +// Convert MSWindows-1252 (Latin-1) encoded text to the local encoding. +// +// Copyright 1998-2006 by Bill Spitzak and others. +// +// This library is free software; you can redistribute it and/or +// modify it under the terms of the GNU Library General Public +// License as published by the Free Software Foundation; either +// version 2 of the License, or (at your option) any later version. +// +// This library is distributed in the hope that it will be useful, +// but WITHOUT ANY WARRANTY; without even the implied warranty of +// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU +// Library General Public License for more details. +// +// You should have received a copy of the GNU Library General Public +// License along with this library; if not, write to the Free Software +// Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 +// USA. +// +// Please report all bugs and problems on the following page: +// +// http://www.fltk.org/str.php +// + +#include +#include +#include "flstring.h" + +// These function assume a western code page. If you need to support +// scripts that are not part of this code page, you might want to +// take a look at FLTK2, which uses utf8 for text encoding. +// +// By keeping these conversion tables in their own module, they will not +// be statically linked (by a smart linker) unless actually used. +// +// On MS-Windows, nothing need to be converted. We simply return the +// original pointer. +// +// Most X11 implementations seem to default to Latin-1 as a code since it +// is a superset of ISO 8859-1, the original wetsern codepage on X11. +// +// Apple's OS X however renders text in MacRoman for western settings. The +// lookup tables below will convert all common character codes and replace +// unknown characters with an upsidedown question mark. + +// This table converts MSWindows-1252/Latin 1 into MacRoman encoding +static uchar latin2roman[256] = { +0xdb, 0xc0, 0xe2, 0xc4, 0xe3, 0xc9, 0xa0, 0xe0, 0xf6, 0xe4, 0xc0, 0xdc, 0xce, 0xc0, 0xc0, 0xc0, +0xc0, 0xd4, 0xd5, 0xd2, 0xd3, 0xa5, 0xd0, 0xd1, 0xf7, 0xaa, 0xc0, 0xdd, 0xcf, 0xc0, 0xc0, 0xd9, +0xca, 0xc1, 0xa2, 0xa3, 0xc0, 0xb4, 0xc0, 0xa4, 0xac, 0xa9, 0xbb, 0xc7, 0xc2, 0xc0, 0xa8, 0xf8, +0xa1, 0xb1, 0xc0, 0xc0, 0xab, 0xb5, 0xa6, 0xe1, 0xfc, 0xc0, 0xbc, 0xc8, 0xc0, 0xc0, 0xc0, 0xc0, +0xcb, 0xe7, 0xe5, 0xcc, 0x80, 0x81, 0xae, 0x82, 0xe9, 0x83, 0xe6, 0xe8, 0xed, 0xea, 0xeb, 0xec, +0xc0, 0x84, 0xf1, 0xee, 0xef, 0xcd, 0x85, 0xc0, 0xaf, 0xf4, 0xf2, 0xf3, 0x86, 0xc0, 0xc0, 0xa7, +0x88, 0x87, 0x89, 0x8b, 0x8a, 0x8c, 0xbe, 0x8d, 0x8f, 0x8e, 0x90, 0x91, 0x93, 0x92, 0x94, 0x95, +0xc0, 0x96, 0x98, 0x97, 0x99, 0x9b, 0x9a, 0xd6, 0xbf, 0x9d, 0x9c, 0x9e, 0x9f, 0xc0, 0xc0, 0xd8 +}; + +// This table converts MacRoman into MSWindows-1252/Latin 1 +static uchar roman2latin[256] = { +0xc4, 0xc5, 0xc7, 0xc9, 0xd1, 0xd6, 0xdc, 0xe1, 0xe0, 0xe2, 0xe4, 0xe3, 0xe5, 0xe7, 0xe9, 0xe8, +0xea, 0xeb, 0xed, 0xec, 0xee, 0xef, 0xf1, 0xf3, 0xf2, 0xf4, 0xf6, 0xf5, 0xfa, 0xf9, 0xfb, 0xfc, +0x86, 0xb0, 0xa2, 0xa3, 0xa7, 0x95, 0xb6, 0xdf, 0xae, 0xa9, 0x99, 0xb4, 0xa8, 0xbf, 0xc6, 0xd8, +0xbf, 0xb1, 0xbf, 0xbf, 0xa5, 0xb5, 0xbf, 0xbf, 0xbf, 0xbf, 0xbf, 0xaa, 0xba, 0xbf, 0xe6, 0xf8, +0xbf, 0xa1, 0xac, 0xbf, 0x83, 0xbf, 0xbf, 0xab, 0xbb, 0x85, 0xa0, 0xc0, 0xc3, 0xd5, 0x8c, 0x9c, +0x96, 0x97, 0x93, 0x94, 0x91, 0x92, 0xf7, 0xbf, 0xff, 0x9f, 0xbf, 0x80, 0x8b, 0x9b, 0xbf, 0xbf, +0x87, 0xb7, 0x82, 0x84, 0x89, 0xc2, 0xca, 0xc1, 0xcb, 0xc8, 0xcd, 0xce, 0xcf, 0xcc, 0xd3, 0xd4, +0xbf, 0xd2, 0xda, 0xdb, 0xd9, 0xbf, 0x88, 0x98, 0xaf, 0xbf, 0xbf, 0xbf, 0xb8, 0xbf, 0xbf, 0xbf +}; + +static char *buf = 0; +static int n_buf = 0; + +#ifdef __APPLE__ + +const char *fl_latin1_to_local(const char *t, int n) +{ + if (n==-1) n = strlen(t); + if (n<=n_buf) { + n_buf = (n + 257) & 0x7fffff00; + if (buf) free(buf); + buf = (char*)malloc(n_buf); + } + const uchar *src = (const uchar*)t; + uchar *dst = (uchar*)buf; + for ( ; n>0; n--) { + uchar c = *src; + if (c>127) + *dst = latin2roman[c-128]; + else + *dst = c; + } + //*dst = 0; // this would be wrong! + return buf; +} + +const char *fl_local_to_latin1(const char *t, int n) +{ + if (n==-1) n = strlen(t); + if (n<=n_buf) { + n_buf = (n + 257) & 0x7fffff00; + if (buf) free(buf); + buf = (char*)malloc(n_buf); + } + const uchar *src = (const uchar*)t; + uchar *dst = (uchar*)buf; + for ( ; n>0; n--) { + uchar c = *src++; + if (c>127) + *dst++ = roman2latin[c-128]; + else + *dst++ = c; + } + //*dst = 0; // this would be wrong + return buf; +} + +#else + +const char *fl_latin1_to_local(const char *t, int) +{ + return t; +} + +const char *fl_local_to_latin1(const char *t, int) +{ + return t; +} + +#endif + +// +// End of "$Id: $". +// -- cgit v1.2.3