I’m creating a Unicode font, and I’m adding some Arabic characters. I just tried testing my font to see if it’s working the way I think it should, but I can’t get any program to use the characters.
The characters exist, and are mapped correctly, both in the Arabic block and the Arabic Presentations Block-B, and, for example, MS Word’s Insert Symbol array shows my characters, but when it inserts them into a document, it pulls them from some other font it seems to trust more to have those characters.
I believed that switching on the appropriate boxes in the Format/Settings/Ranges/Unicode Character Ranges/Edit… thingie would fix this problem, but I switched them all on, and nothing has changed, and the fact that every last box is labelled “Reserved for Unicode SubRange [sic]” concerns me…
Have you also calculated the Code Page Character Ranges on the Format, Settings, Ranges dialogue? That’s the only thing I can think of that you may have missed. Other than that, you may need to reinstall the font and reboot the computer.
I pressed Calculate first to see what it did. None of the boxes got checked. I switched them all on, and then pressed Calculate. It switched them all off again.
Screen capture, just so you understand what I mean:
Note that the scroll bar is all the way to the top. In previous versions of the software I’ve owned, these were labelled with Unicode code blocks. The help files have them labelled too.
It is more more likely you are modifying an existing font.
When the Content and Layout version (shown on the same dialog) of a font is set to “Version 0”, all Unicode ranges are reserved. So just increase the version to 1, 2 or 3 and then set the appropriate Unicode blocks.
I’m creating a Unicode font, and I’m adding some Arabic characters.
It is more more likely you are modifying an existing font.
I’m adding characters to one of my older fonts, created in an older version of your software. Does this matter? And did you mean to call me a liar in a public forum?
When the Content and Layout version (shown on the same dialog) of a font is set to “Version 0”, all Unicode ranges are reserved. So just increase the version to 1, 2 or 3 and then set the appropriate Unicode blocks.
Thank you. I’ll remember this – all of this – in the future.
Thanks for explaining, and no I did not call you a liar and I didn’t mean to call you a liar. I just wanted to be sure the problem wasn’t caused by a bug in the latest release of FontCreator.
Thank you. I’ll remember this – all of this – in the future.
There may still be a bug. I switched to Version 1 in the older font, calculated both Unicode Character Ranges and Code Page Ranges, and the characters still get ignored. I went in both and manually turned on everything in Unicode Ranges I had Arabic characters for, and everything in Code Page Ranges with the word “Arabic” in it. I also switched the Font Direction hint under Header to “Fully mixed directional glyphs”, since Arabic is read the other way. Still nothing. I added the first character from the main CJK block (code point $4E00), and it wasn’t recognized either.
I went for broke, started a new Unicode font, set the Units per em length to the same as the old font, wiped the glyph boxes past the first four, and added 500 new ones. I copied everything from the old font down to the mappings into those boxes, and did all the same steps as the preceding paragraph. (Arabic still was not recognized as being there with a checked box in Unicode Ranges, nor were any characters from the Arabic Presentations Block-B.) The Chinese character is now recognized and can be typed – without needing any of the CJK blocks checked under Unicode Page Ranges, or any reasonable ones checked under Code Page Ranges. The Arabic characters are the only holdouts.
I’ve run out of switches to twiddle. You have the whole story with no fabrications or omissions. Any ideas?
I believe previous versions of Windows did show Arabic characters from fonts made with FontCreator. I’ve just tried to make a font with Arabic characters, but it fails to show on my Windows Vista PC. I haven’t enabled Arabic language support. That might be related, I’m not sure yet.
Another reason why my test font failed might be related to the lack of OpenType Layout features, but again I’m not sure yet.
This post, just found on those forums contains some information about creating Arabic fonts. I’m not sure if it’s useful or not, but just in case: http://typophile.com/node/16288
I’ve tried copying my font to a Windows XP computer – still nothing.
I’m not convinced that Arabic typesetting requires Open Type. I suspect it’s better with it, but… Unicode allows for things like having the a and the composing acute accent next to each other, and Unicode replaces them with U+00E1 á, so why shouldn’t it do this with the Arabic characters? Also, my copy of Word 2003 was published before Open Type went live in 2005, so I don’t think that’s it. It won’t display my font, but will do the correct joined-up Arabic glyphs when I try copying characters one at a time to it.
Heck, Unicode itself allowed for Arabic characters back in 1991, so I doubt it requires the presence of OT to make them work.
I have asked questions at Typophile before. When I’ve gotten a response, it hasn’t been very helpful. I suspect I’m going to be told to move to Open Type.
I suspect at this point I’m just going to have to kludge it with some segment of the unassigned blocks, and typeset what I need manually.
