(CC:ed to authors)
I went through the file plugins and added magic patterns to those who didn't
register any, and corrected those already exisiting. The only case where I'm
not 100% certain is the MPEG magic number; the ones chosen are from my
/etc/magic file, and all mpegs I've looked at had eith
Hi,
> OK, I should have read the e-mail with the patch more carefully. This
> handling looks good.
So the patch actually has a good chance to get included. The probability
would be even better if you could upload it to ftp://ftp.gimp.org/incoming
as described in the README in that directory.
At
orrect file magic or the one that handles the correct extension ? I
> > would prefer the plug-in with the correct file magic.
> > Maybe the file magic of the faxg3 plug-in is too weak ?
>
> If I understood the patch correctly, only the extensions of magicless
> file-formats would
[ correct description of the patch ]
>The only problem I see with this patch are badly written file_plug_ins
>that don't provide a magic header but a bunch of extensions since the
>patch would significantly enhance the possibility that the wrong loader
>is used.
Grepping for gimp_register_load
xtension ? I
> would prefer the plug-in with the correct file magic.
> Maybe the file magic of the faxg3 plug-in is too weak ?
If I understood the patch correctly, only the extensions of magicless
file-formats would be probed before the magic header is used. That would
mean you couldn't eas
Mattias Engdegård wrote:
>
> >> Anyway, if the user finds that an application doesn't try to load a file
> >> whose name ends in .tga as a Targa file (or worse, attempts to load it as
> >> a Group 3 fax file), then the same user will assume that the application,
> >> or its programmers, or both,
>> Anyway, if the user finds that an application doesn't try to load a file
>> whose name ends in .tga as a Targa file (or worse, attempts to load it as
>> a Group 3 fax file), then the same user will assume that the application,
>> or its programmers, or both, are stupid beyond belief. And rightl
Mattias Engdegård wrote:
> Anyway, if the user finds that an application doesn't try to load a file
> whose name ends in .tga as a Targa file (or worse, attempts to load it as
> a Group 3 fax file), then the same user will assume that the application,
> or its programmers, or both, are stupid bey
>"New" Targa files have a magic string at the end of the file[1], plus
>a bunch of extra information. The Targa specification tells you how
>and where to expect it. It does says that for old files you are
>basically screwed :-(
I know that, but since a) the Gimp magic pattern language doesn't s
> Targa files have no magic header, and cannot be reliably identified that way.
[snip]
"New" Targa files have a magic string at the end of the file[1], plus
a bunch of extra information. The Targa specification tells you how
and where to expect it. It does says that for old files you are
basic
Targa files have no magic header, and cannot be reliably identified that way.
The TGA plugin tried to register some nonsense magic string which didn't work
anyway, and any attempt to load a TGA file without explicitly selecting the
load format will usually cause gimp to handle it as a group 3 fax
11 matches
Mail list logo
