Mail Archives: djgpp/2001/07/11/12:30:07
> > > I think a much less painful way would be to edit the patch file, and
> > > manuall modify the file names to their original long names. If done
> > > carefully, this will produce a patch that will simply work with no fuss.
> >
> > I've modified some of the file names in the patch file, but with
> > each file comes more hunks that don't work.
>
> Sorry, I don't understand this sentence. Could you give an example?
I meant to say that when i ran patch with djgpp.diff, i got some hunks that were
rejected. When i modified the file names, I got even more hunks rejected (about
10-30 and only modified 3 filenames in the patch file). Some of the hunks are
applied to the file no problem, but for others it doesn't work.
As an example i changed the string "configur.lan" to "configure.lang" and some
of the hunks were applied, others could not be applied. Previously i'd had a
look at the erroneuos sections of code and tried to see what had gone wrong,
unfortunately the number of errors is rising and i'm unable to keep up.
The curious thing is that the files must be able to be built, otherwise there
would be no binary distribution. So by that logic, when i've modified the patch
file to accomodate long filenames, it should build even if some hunks cannot be
applied. Maybe I'm wrong, but I can't see any other explanation.
> > I don't even know what a hunk
>
> A hunk is a fragment of the patch file which describes a change to a
> contiguous range of source file lines. Each hunk begins with a header
> that states the affected lines' numbers, and then the actual changed
> lines.
Thanks for explaining that to me,
Chris Campbell
- Raw text -