On 22 January 2011 21:07, Jose E. Marchesi <[email protected]> wrote: > You can find several unit tests for the currently implemented filters in > both torture/unit/base/stm/pdf-stm-read.c and > torture/unit/base/stm/pdf-stm-write.c. I guess that a similar approach > could be used to test the predictor filters.
The ones I looked at mostly encode a simple string, which can't be done for the predictor stuff because it's explicitly binary scanlines. (I knew nothing about the predictor filter before this afternoon - I surmise so far it's at the heart of embedding PNG in PDF?) Is the library advanced to the state that I can try to construct PDFs with authoring tools which stress this in various ways? I really just picked this up and ran with it since it didn't seem like a big task: I reworked the interface using some similar modules as guidelines and maintaining the same operational behaviour. It compiles, it successfully passes parameters to the module from the pdf-filter program. But I don't have any test data or anything of that nature, I was just working on a flyspray task. I'm reluctant to offer code I haven't even run though. I did port the decoder as well when I was bored this afternoon, but similar provisos apply. I guess if I had a test PDF I could round-trip it and demonstrate some functionality in both decoder and encoder modules. Ed
