| Age | Commit message (Collapse) | Author |
|---|
|
|
|
As known from sbase, we want to manually flush stdout to see if all
data has been passed on.
|
|
Somehow, it broke the depth-value on libpng 1.4.*. Should be worth a
bug-report to the libpng-guys, but I can't be arsed right now.
This fucking library is a pain in the ass.
|
|
|
|
Thanks Dimitris, Hiltjo!
|
|
|
|
|
|
|
|
It took me quite a while to figure out that libpng already gives you
big endian values.
I also tweaked the library functions a bit more to really make sure
that all PNG-types (grayscale, palette, ...) are converted to RGBA
properly.
|
|
It always seemed odd to have a file-argument for 2ff, given all other
farbfeld conversion tools operate on stdin and stdout only.
Given I received numerous e-mails on this matter I've decided to change
the usage to lessen the confusion.
Maybe there's a way to do some magic with tee(1), but I didn't bother
just yet.
To ease the transition, the number of arguments are checked and a usage
printed if arguments are given.
|
|
thanks Quentin Rameau!
|
|
jpg-support is done.
|
|
I forgot to add Hiltjo to the 2015-committers.
|
|
|
|
|
|
Thanks z3bra for porting this!
Also change 2ff to use case instead of if-blocks.
|
|
This should also allow one to build the tools on both BSD and Linux
probably without modifying config.mk at all.
|
|
This is a wrapper for the different conversion utilities with
filetype detection using file(1) and an imagemagick fallback.
|
|
|
|
Use the arpa/inet.h conversion functions which are standardized
in POSIX.
|
|
Thanks Heiko ("rundstutzen") for reporting this!
|
|
No need for this when the tools don't accept arguments.
|
|
|
|
|
|
|
|
|
|
- Rename the format
- Change the format specification
- Drop old tools waiting to be fixed on a later date, just keep
fixed png for now
- Simplify other stuff
This is a direct consequence of my slcon2-talk on this topic.
At first I planned to have 64 bits per channel, but this is
overkill.
|
|
|
|
Thanks to Henrique Lengler <henriqueleng@opmbx.org> for the patch!
|
|
To simplify things just print that a gif error occured. The internal
error messages are not terribly descriptive and this approach saves on
extra preprocessor madness to work around it.
For DGifOpenFileHandle() we really can't do much as the prototype
was changed, just #if around that.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
The previous behaviour was correct. A width of 1900 pixels which is
0x00000640 in hex was stored on disk as:
00 00 06 40
That's BE order which is what we want.
Reading this uint32_t on an LE machine we get:
width = 0x40060000 and converting from BE to LE
we get the correct 0x640 value.
Reading on a BE machine we get the correct value
and the swap is a no-op.
|
|
Well, who would have thought fwrite already converts your integer to
NBO? It would've been bloody helpful to at least mention it in the
manual.
While at it, fix the wrong read-in-code and call it a day.
|
|
|
|
|
|
|
|
These tools are "one-shot", nobody cares about memory leaks while
the program is running.
Bring back -Os, no more false positives because there's no interaction
between gotos and setjmp/longjmp.
|
|
We could use the macros from stdint.h but this should be good
enough.
|
|
|
|
If we fail early, the call to free(png_row) could crash because
of an uninitialized (non-NULL) png_row.
|
|
|
|
-Os generates lots of false positives and it is not worth it much
for such a small set of tools.
|
|
Thanks to Robert Hülle for reporting this!
|
|
|
|
|
