At 11:40 PM 8/15/99 -0500, Shaw Terwilliger wrote:
>thomas wrote:
>> I found some type of file format SDK and downloaded
>> it -- however it didn't seem useful so I sent an e-mail to
>> corel's developer support. I'll let you know what they say.
>> In the mean time can can you point me to the abi import api?
>
>Be careful of the license you agree to when using vendor SDKs.
I don't know what's in that SDK -- does it include header files and tools,
or just the spec? -- and I haven't seen the license on it, so I can't
comment on it. I know Caolan sidestepped a lot of legal minefields in
obtaining information on the MS Word file formats, so he might be a good
source for advice on this topic.
However, I did find the following document (via Google) on Corel's site:
http://www.corel.com/partners_developers/ds/co32sdk/docs/ff/A_FRNTFF.htm
It's not an SDK, just a prose spec on the contents of the WP 7 file format,
and I couldn't find any indication that this spec was legally encumbered.
It's also linked from the following document, from which you should be able
to infer what the subset of WP 6 features are:
http://www.corel.com/partners_developers/ds/co32sdk/docs/changes7.htm
If the SDK license is a problem, then these documents might be a more
appropriate starting place for a "clean room" implementation. You'll have
to code everything from scratch, including the header files, but that's a
time-honored tradition in the Open Source world. :-)
>Either way, an import for Word Perfect would be great. As far
>as I know, as many as three other people have expressed interest
>in working on a Word Perfect filter. If anyone else on the list
>has been keeping track of those contact addresses, now would be
>a great time to pass them around again.
Have these been circulated in private, or should we dig through our email
archives and provide an answer publicly?
>If you're looking for information on AbiWord's importer and
>exporter interfaces, look at abi/src/wp/impexp. The base classes
>(from which specific importers and exporters derive) are there
>and are pretty self-explanatory. To start, I would recommend
>cloning an existing filter (say the "plain text" one) and using
>other filters as guidelines. Our native format is a good example
>in callback-driven I/O (using expat), the plain text importer shows
>a simple single pass at filling a document.
Was this enough information, or do you need another level of drill-down?
Paul
