Re: [Gimp-print-devel] canon-unprint

[<sh...@al...>]

> * Added parse_canon() with calls to page_update()
> * Added rle_decode() since canon rather knows the compressed buffer size than
> the uncompressed
> * Added switching from parse_escp2() to parse_canon() via environment variabl
> e

I spent about 30 seconds scanning your code.  What I saw looked pretty good.

> UNPRINT:
>    "UNPRINT=canon ./unprint" will use parse_canon()

Ok, although I'd like to remove this feature eventually.

Unfortunately, there's no really simple way to do that.  What we want is
for each routine to be given a file pointer and be able to read that file
from the beginning, yet we need to read some of the file to determine what
the file type is.  There are only two ways to do this that I can think of.
One is to create a pipe internally, fork the process, read the data in
one process, determine the file type, and write the data to the pipe which
then gets read and parsed by the appropriate child process.

The other way is to read to a buffer, and then only read from the file
if the buffer is empty.

I think the second method would require less in the way of computer resources,
although I think the first is stylistically cleaner.

I haven't decided which is the better way to go.

> The output produced from canon-printfiles looks extremely poor - there are
> some weird artefacts. Eric, could you please have a look at it? Maybe I've go
> t
> something terribly wrong?

I just compiled it.  I'll have a look.

Eric