From asafgreenberg at gmail.com Thu Jan 3 07:09:10 2013 From: asafgreenberg at gmail.com (Asaf Greenberg) Date: Thu, 3 Jan 2013 17:09:10 +0200 Subject: [FriBidi] Is there a stable build (binaries) of fribidi for Windows ? In-Reply-To: <50E21308.6020709@behdad.org> References: <50E1A404.8050306@gmail.com> <50E21308.6020709@behdad.org> Message-ID: yes, the binaries are ok. thank you for uploading them. i'm using them once in a while. However, they present the same functionality as my old binaries. What i was looking for, specifically, was the ability to convert an HTML file from VISUAL hebrew to LOGICAL hebrew. indeed, each word in itself is converted well in fribidi. but as a whole, the sections of a sentence with html code should also change their order. e.g. (visual hebrew: very old style, seen well only with specific encoding, browser and backword fonts) zyx onml gfe dcba (after fribidi, as seen when ordered properly with . each section now looks good, but the sections are not in the right order) lmno xyz abcd efg (actually should've been: both text AND sections should be in this order) abcd efg lmno xyz in the last scenario, the OS/GUI toolkits knows to display it right, while on disk the text is saved letter by letter as they exist *logically *) so what i was hoping to find is a routine that will re-order the *sections of a sentence*. the logical hebrew's sentences are being broken with - HTML tags, - english text - numbers - certain punctuation characters So far i'm found a workaround to my specific task. And there are very few websites which are still written that way, I don't think there's a point in implementing this feature, if it doesn't exist yet. On Tue, Jan 1, 2013 at 12:34 AM, Behdad Esfahbod wrote: > I uploaded what I could build using mingw32 cross compiler here: > > http://fribidi.org/download/win/ > > Let me know if it's any good and I'll upload those with every new release. > > behdad > > On 12-12-31 08:41 AM, asaf greenberg wrote: > > Where can i find a stable build of fribidi for Windows ? > > I didn't find anything recent & working by myself. > > > > > > > > > > _______________________________________________ > > fribidi mailing list > > fribidi at lists.freedesktop.org > > http://lists.freedesktop.org/mailman/listinfo/fribidi > > > > -- > behdad > http://behdad.org/ > -------------- next part -------------- An HTML attachment was scrubbed... URL: From behdad at behdad.org Sun Jan 6 23:31:48 2013 From: behdad at behdad.org (Behdad Esfahbod) Date: Mon, 07 Jan 2013 01:31:48 -0600 Subject: [FriBidi] fribidi-vs-unicode, PyArabicShaping, PyByeDie, UAX#9 draft, etc Message-ID: <50EA79E4.5020102@behdad.org> Hi, Some of you may have noticed that (after about 8 years) I fixed a couple real bugs (though very insignificant) in the bidi algorithm in FriBidi. Those were discovered by running FriBidi against the Unicode BidiTest.txt test suite. The code for that is here: https://github.com/behdad/fribidi-vs-unicode A while back, I ported the Arabic shaping logic in FriBidi to Python. That can be found here: https://github.com/behdad/pyarabicshaping Finally, yesterday I wrote a Python reference bidi implementation from scratch. Note, it's a *reference* implementation. Very slow. It's 150 times slower than FriBidi. It does pass all tests and I'm fairly sure that it's bugfree now: https://github.com/behdad/pybyedie Finally, there's a new draft of UAX#9 out that adds "bidi isolates". It's a major update to the algorithm. The current draft is at: http://www.unicode.org/reports/tr9/tr9-28.html#I1 I'm implementing that in the 6.3 branch of pybyedie. Eventually I hope it will make it in FriBidi too, though the N0 rule is hard to implement given FriBidi's API. Cheers, -- behdad http://behdad.org/ From johnbrown_105 at yahoo.com Thu Jan 10 08:37:14 2013 From: johnbrown_105 at yahoo.com (John Brown) Date: Thu, 10 Jan 2013 08:37:14 -0800 (PST) Subject: [FriBidi] Building charsets library on Windows Message-ID: <1357835834.36667.YahooMailNeo@web141105.mail.bf1.yahoo.com> Hello All, I am using the git version. These are my observations. 1) Currently, ./charset/Makefile is executed before ./lib/Makefile. On Windows this will not work because the linker will complain about undefined symbols (functions from libfribidi). 2) I am able to reverse this order by editing ./Makefile: - change SUBDIRS = gen.tab charset lib?bin doc test ? to SUBDIRS = gen.tab lib charset bin doc test, and - comment out am__append_4 and am__append_5 in ./lib/Makefile # am__append_4 = $(top_builddir)/charset/libfribidi-char-sets.la # am__append_5 = $(top_builddir)/charset/libfribidi-char-sets.la? ? 3) I also noticed that in ./charset/Makefile libfribidi_char_sets_la_LDFLAGS = $(am__append_2) where am__append_2 = -static, which I changed to -no-undefined ? 4) Finally, in ./charset/Makefile, I changed ? libfribidi_char_sets_la_LIBADD = $(GLIB_LIBS) to libfribidi_char_sets_la_LIBADD = ../lib/.libs/libfribidi.la $(GLIB_LIBS) because the charsets library depends on libfribidi. ? 5) I even #defined FRIBIDI_ENTRY when running configure: ? $ CFLAGS='-DFRIBIDI_ENTRY=__declspec$dllexport$' ./configure \ --enable-static --enable-shared --with-glib=no 6) By the way, even though I specified --with-glib=no, ./configure initially failed because it did not recognise PKG_CHECK_MODULES, which is in pkg.m4, which is not in the default location on my system. ? The relevant section of configure.ac seems to be: ? # --with[out]-glib AC_ARG_WITH(glib, ????????????? AC_HELP_STRING([--with-glib=@<:@no/auto/yes@:>@], ???????????????????????????? [use Glib @<:@default=auto@:>@])) GLIB_PACKAGE=glib-2.0 GLIB_MINVERSION=2.4 GLIB_LIBS= GLIB_CFLAGS= if test x$with_glib = xyes; then ??????? PKG_CHECK_MODULES(GLIB,$GLIB_PACKAGE >= $GLIB_MINVERSION) ??????? FRIBIDI_USE_GLIB=1 else if test x$with_glib = xno; then ??????? FRIBIDI_USE_GLIB=0 else ??????? PKG_CHECK_MODULES(GLIB,$GLIB_PACKAGE >= $GLIB_MINVERSION, ??????????????? FRIBIDI_USE_GLIB=1, ??????????????? FRIBIDI_USE_GLIB=0) fi fi ? In my opinion, --with-glib=no should have caused the check for Glib to be skipped. ? Despite all of the above, libtool still refuses to build the shared library. Can you say why? ? Regards, John Brown. From rogerdpack2 at gmail.com Thu Jan 10 16:32:40 2013 From: rogerdpack2 at gmail.com (Roger Pack) Date: Thu, 10 Jan 2013 17:32:40 -0700 Subject: [FriBidi] wrong date? Message-ID: hello, I noticed http://www.fribidi.org/ lists 0.19.5 as "aug 27" but it's listed as dec. in http://www.fribidi.org/download/ is this expected? Thank you. -r From behdad at behdad.org Thu Jan 10 16:38:04 2013 From: behdad at behdad.org (Behdad Esfahbod) Date: Thu, 10 Jan 2013 18:38:04 -0600 Subject: [FriBidi] wrong date? In-Reply-To: References: Message-ID: <50EF5EEC.3010301@behdad.org> Fixed. Thanks. b On 13-01-10 06:32 PM, Roger Pack wrote: > hello, I noticed http://www.fribidi.org/ lists 0.19.5 as "aug 27" but > it's listed as dec. in http://www.fribidi.org/download/ is this > expected? > Thank you. > -r > _______________________________________________ > fribidi mailing list > fribidi at lists.freedesktop.org > http://lists.freedesktop.org/mailman/listinfo/fribidi > -- behdad http://behdad.org/ From lrn1986 at gmail.com Sun Jan 20 09:46:25 2013 From: lrn1986 at gmail.com (LRN) Date: Sun, 20 Jan 2013 21:46:25 +0400 Subject: [FriBidi] Improper use of PKG_CHECK_MODULES Message-ID: <50FC2D71.9060908@gmail.com> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 First call to PKG_CHECK_MODULES must be unconditional. If it isn't, PKG_PROG_PKG_CONFIG should be called first. Otherwise subsequent PKG_CHECK_MODULES calls (after first PKG_CHECK_MODULES that was skipped) will not be able to find pkg-config. I'm hitting this when i configure without explicit --with-glib -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.11 (MingW32) Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQEcBAEBAgAGBQJQ/C1vAAoJEOs4Jb6SI2Cw0JcIALsI4GZRmhZF7qsmwYUASGRa LDX0sHNlcMRtgtmC9p+FCGpGHibNwXobh21EmrHVz78e8Jj8rEPvYdTaJvcIOOGf g2Ypwt/yjH3agU7cSKrD/2gmiFzLSIdS05//fzqWF97CWsQrkb2lFtGxnHqynUYr lmLKhGoO8PReFPGJ/0IzW5UL9DjHZclqwWfh+LBrPpHx6s8inio3m7MZZN66anq+ Ll3ILKQsMnHURWhiyX+WCRCg9uIHetGthCmHL+2DC1EDTZIhfTiyNhS6afkjNlmf Px++FpzZTP2sUfe25JOObdxjPe3Sg8Y35A45hu2dniryJMNcJR8o0rhmx325Wc8= =gcxW -----END PGP SIGNATURE----- From andrey.nautilus at yandex.ru Thu Jan 24 06:33:03 2013 From: andrey.nautilus at yandex.ru (Nautilus Andrey) Date: Thu, 24 Jan 2013 18:33:03 +0400 Subject: [FriBidi] Possible misprint in fribidi-bidi.h Message-ID: <719991359037983@web17e.yandex.ru> Hello. I noticed, that in fribidi-bidi.h lines 46 and 71 are equal (in latest version, 0.19.5). According to surrounding code, line 46 should look like #define fribidi_get_par_direction FRIBIDI_NAMESPACE(get_par_direction) instead of #define fribidi_get_par_embedding_levels FRIBIDI_NAMESPACE(get_par_embedding_levels) This misprint doesn't break anything for now, but in future, if FRIBIDI_NAMESPACE is defined to something else, it will break the api. Andrey. From behdad at behdad.org Tue Jan 29 20:05:16 2013 From: behdad at behdad.org (Behdad Esfahbod) Date: Tue, 29 Jan 2013 23:05:16 -0500 Subject: [FriBidi] Possible misprint in fribidi-bidi.h In-Reply-To: <719991359037983@web17e.yandex.ru> References: <719991359037983@web17e.yandex.ru> Message-ID: <51089BFC.4000806@behdad.org> Fixed. Thanks. On 13-01-24 09:33 AM, Nautilus Andrey wrote: > Hello. > > I noticed, that in fribidi-bidi.h lines 46 and 71 are equal (in latest version, 0.19.5). According to surrounding code, line 46 should look like > > #define fribidi_get_par_direction FRIBIDI_NAMESPACE(get_par_direction) > > instead of > > #define fribidi_get_par_embedding_levels FRIBIDI_NAMESPACE(get_par_embedding_levels) > > This misprint doesn't break anything for now, but in future, if FRIBIDI_NAMESPACE is defined to something else, it will break the api. > > Andrey. > _______________________________________________ > fribidi mailing list > fribidi at lists.freedesktop.org > http://lists.freedesktop.org/mailman/listinfo/fribidi > -- behdad http://behdad.org/