Closed d-lamb closed 6 years ago
Hi
Indeed the issue is microscopic, so if you can ignore it please do. I tried to reproduce it on my virtual machine, but it just works, also with Apple Color Emoji font. I got it from https://github.com/potyt/fonts/raw/master/macfonts/Apple%20Color%20Emoji/Apple%20Color%20Emoji.ttf because my MacOS didn't have it, but it might be the wrong one. I wonder if you could end me your font file?
Dmitry
Sure, will send you a link via email.
Also, just in case could you run this in your Prima build directory and send me the output?
perl -Iblib/arch -Iblib/lib t/Objects/Fonts.t --debug=F 'Apple Color Emoji'
Hi, this font does all sort of crazy things if I load it on my linux. I guess fontconfig just cannot do much with it. So I just ignore it explicitly in tests. Not the best practice, but let's give fontconfig a chance , probably they fix it later
/dk
Thanks for checking on this. With 8d24a73 all tests pass.
Unrelated to that commit, there are warning messages from t/Object/Fonts.t
of the form:
X Error: BadValue (integer parameter out of range for operation), request: 45(X_OpenFont), after unix/apc_font.c:1822
X Error: BadValue (integer parameter out of range for operation), request: 45(X_OpenFont), after unix/apc_font.c:1161
and
Wide character in print at /opt/local/lib/perl5/vendor_perl/5.26/Test2/Formatter/TAP.pm line 113.
Wide character in print at /opt/local/lib/perl5/vendor_perl/5.26/Test2/Formatter/TAP.pm line 144.
But they are just noisy and do not appear to be causing any problems.
Hi,
The first one I have no control over - I've seen this before and that happens when XQuartz tries to load an already installed font that it has trouble reading.
The second one looks like coming from a malformed font name, which unicode name or something similar. If you could trace the font and send me the font file I might fix it.
/dk
Almost certainly a unicode font name issue. The 14 (I think) fonts that cause that TAP.pm warning are:
ديوان كوفي فارسي ديوان ثلث النيل التاريخ منى صنعاء مِصحفي ذهبي وسيم فرح مِصحفي بغداد نسخ بيروت
I'll leave it up to you whether and how to fix that!
Wow, that's impressive :) Do you think you can send me one of those?
Have similar problem with a couple of Japanese fonts slipped in with don't know what application. Unfortunately, can't trace down to the file name of the font (Prima's font structure doesn't seem to include that information). But the problem could easily be reproduced with the following line of code:
diag "Font: 苹方-繁";
Guess that
diag "А що за дідко тут коїться?";
would server the purpose too.
I have recently had similar issue on my hands with Test2 framework. Simply include 'use Test2::Plugin::UTF8' in a test file. I have tried it on my side with the recent 'git pull' – it works:
ok 4383 # skip height ok 4384 # skip size ok 4385 # skip direction ok 4386 # skip width / 苹方-繁 ok 4387 # skip width / 苹方-繁 ok 4388 # skip style ok 4389 - text wrap 苹方-繁
3 трав. 2018 р. о 13:45 Dmitry Karasik notifications@github.com написав(ла):
Wow, that's impressive :) Do you think you can send me one of those?
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/dk/Prima/issues/56#issuecomment-386378802, or mute the thread https://github.com/notifications/unsubscribe-auth/ABg3MDVIZ3BOQ8tTnbqbTq6V1dqflwJvks5tu0I2gaJpZM4TtMS-.
Best regards, Vadim Belman
Hi Dmitry,
I am unable to install Prima from CPAN due a test failure. This does not appear to be a serious failure that should prevent me from force-installing Prima. But it does seem to be the source of most of the Darwin install failures that are reported for recent Perl versions on CPAN Testers. The test failure is of t/Object/Fonts.t for the "Apple Color Emoji" font, and it looks like the following. I'm not too familiar with Prima, but am happy to help debug. I am running perl 5.26.2 on MacOS 10.12.6 (Sierra).
perl -V
follows the test output.