New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Settle on 1 unicode name library and make it a requirement #1341
Comments
On April 25, 2014 11:12:25 PM Dave Crossland wrote:
You also need to consider the internal tables and files, like unicoderange.c For example, the current FontForge is synchronized with Unicode 6.3, but Ideally, FontForge should adapt to the different libuninameslist tables |
Some distros are not completely configured correctly for pkg-config. There is a school of thought that argues for using PKG_CHECK_MODULES, but that is not necessary. The standard way to check if a library exists and to link against it is to use AC_CHECK_LIB(), and also check for the developer header file(s) using AC_CHECK_HEADERS(). Fixed fontforge_create_pkg-config.m4 to only include libuninameslist as a dependecy ONLY if it exists AND user didn't say no to include it. This comes to a hard-stop if user wants libuninameslist, but it's not available. This also tests for version 0.3 and 0.4 of libuninameslist Libunicodenames was placed in '#' (commented) since it was going to be a fair bit of re-write to work-around the repeated AC_DEFINE in configure.ac Issue fontforge#1341 recommends sticking with 1 library rather than support both, and libunicodenames seems to be limited in function access and OS choices. Users that want to use libunicodenames can move the comment '#' to switch libraries for one or the other (see configure.ac) since the active code is still in place if you want to use the other library, or if there is another library in future, shouldn't be hard to modify for others: FONTFORGE_ARG_WITH_LIBUNINAMESLIST #FONTFORGE_ARG_WITH_LIBUNICODENAMES
This has been argued back and forth and the stance continues to be adjusted over time. |
It seems from #1322 that we get variant crash/no crash based on the availability of different unicode libraries. This is a poor situation to be in.
Therefore I propose that as part of @frank-trampe 's upcoming stable release, we make @JoesCat 's unicode library a mandatory dependency, so we don't get into this situation again.
The text was updated successfully, but these errors were encountered: