[Openfontlibrary] Font formats accepted by OFLB

[Ed Trager]

Hi, Ben,

Someone has written a PHP class that handles ZIP/ GZIP / BZIP2
compression and decompression:

http://www.phpclasses.org/browse/package/945.html

I will take a look at this as soon as I get a chance and see what I think.

IMO, the best answer would be to seamlessly support all of the popular
compression formats automatically.  (This would possibly also include
7Zip which may not currently be in the above-mentioned class).

In a C++ program that I and a couple of colleagues wrote a couple of
years ago (unrelated to typography), we did exactly this kind of
thing.  That is, we automatically supported compressed file packages.
We read enough of the file headers to be able to detect which file
format needed to be read, and then called the appropriate
decompression library.  This solution was very nice -- we could throw
almost any sort of file at the program, and it just worked.


Best - Ed


On Sat, Oct 25, 2008 at 10:33 AM, Ben Weiner <ben at readingtype.org.uk> wrote:
> Hi,
>
> Ben Laenen wrote:
>> On Saturday 25 October 2008, Ben Weiner wrote:
>>
>>> Looks as though in the next OFLB site version we'll have to ask
>>> people to upload individual files. We'd certainly need to unpack
>>> archives if we allowed them, and ccHost currently cannot see inside
>>> tarballs, facts that together mean we're best avoiding them.
>>>
>>
>> I completely disagree with that, and it won't work anyway. You assume
>> that fonts are always one file, or everything could be pushed into one
>> file.
> Never ;-)
>
> Although that sounds a bit like an archive to me :-)
>>  If you look at DejaVu we have a *lot* more, like build files
>> (Makefile, some scripts to process the fonts when building etc), more
>> scripts that help in development, and other metadata files like
>> changelogs, readme, status files etc.
>>
>> Other projects have for example Xgridfit files for their hinting, or
>> other files that are used for building the fonts from source.
>>
> Aha! Source. Nobody's come back to me on that. I know humans can read
> .sfd files. What about the 'source' files used by other font-authoring
> applications? Do we accept these even thought they're not amenable to
> reuse except by people who also own that software?
>
> If we do decide to accept them, can someone provide a 'Hello
> Typographical World' example file for each? I  can do Macromedia
> Fontographer from my deep-stored Mac OS 8 box, but none of the others.
>> If ccHost cannot handle zipped files, then too bad.
> It can certainly handle them. What it doesn't do is make them usefully
> available in their unscrambled form. It's also not very deft with
> tarballs - although it'll accept them by default, you cannot find out
> what's inside them (without a plugin of some sort from the future, AFAIK).
>
> Incidentally, there is no reason why what I'm informally calling the
> 'typeface record' (the basic unit of ccHosting as applied to fonts, eg
> http://openfontlibrary.org/media/files/OSP/322) should not possess a mix
> of compiled fonts and zipped resources such as the source files and
> readmes. Anyone think that's a good idea?
>
> Cheers,
> Ben
> _______________________________________________
> Openfontlibrary mailing list
> Openfontlibrary at lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/openfontlibrary
>