Working with TrueType fonts

Anon

Hi everyone, just wondering if anyone has worked with PHP and TrueType fonts... I have found plenty of documentation on writing out images with TrueType fonts, but I cant find a damn thing on anything else to do with them.

Basically, I'd like to use PHP to first grab the information out of a font such as it's name, family, char set, etc. I know this info is embedded in the font and accessible... just can't figure out how to get to it.

I don't think I could use the FreeType API's to do this could I?

Any help would be greatly appreciated, no matter how small.

Anon

Hi there!

I recently wrote a PHP script to parse TTF files to look for additional information about the font, including:
Copyright, Family, Filename, Style, Identity, Full Name, Version, PS Name, Trademark information.
By parsing the files directly, you can extract all kinds of stuff, down to the individual vector information that determins how the characters look. You could even write your own font editor ...

I used the TrueType Font File documentation available by Apple at

http://fonts.apple.com/TTRefMan/RM06/Chap6.html

and by Microsoft at

http://www.microsoft.com/typography/tt/tt.htm

Hope this helps

Xtian

Anon

Thanks Xtian- you have put me on the right path.

I have never worked with 'raw' data files like these before, so it's all a bit daunting.

One thing that confuses me is that there are mention of various tables within the ttf file that hold the information; but I cant find any information on the actual process of reading these tables out... do you know of any resources on this? Or even better (I know this is a big ask) do you have any sample code I could look at?

Thanks again!

Anon

I'll see if I can get my code into a more presentable (and commented) form, so I can share it. As it is it's still got some things hardcoded, that just won't work anywhere else ...
In the meantime, here's the way it works:
The TrueType Font files are organised in "tables", which can contain data of varying sizes. The locations of the tables in the file are determined by their offsets from a predefined location (either the beginning of the file, or the start of a section).
If this sounds confusing, you should see my code!
;o)
Xtian

Anon

OK Xtian, now that you've suceeded in completely crushing my spirit... 😉

No seriously, I'd love to see some of your code no matter how messy. I'm mainly interested in how you are interpreting the binary objects and deriving any sort of sense from that raw data.

Thanks a million!