oop found a problem

This commit is contained in:
John Cupitt 2016-07-09 18:01:46 +01:00
parent a9d1c654e2
commit 9d7667cbff
2 changed files with 9 additions and 15 deletions

22
TODO
View File

@ -7,6 +7,14 @@
try with convert, get a one-band png, no alpha at all try with convert, get a one-band png, no alpha at all
- try:
$ time vips magickload nipguide.pdf x.tif
VipsForeignLoad: images do not match
vips__region_start: start function failed for image nipguide.pdf
VipsRegion: images do not match in pixel size
- try loading one page of a PDF with the new magickload - try loading one page of a PDF with the new magickload
- try loading one page of a GIF with the new magickload - try loading one page of a GIF with the new magickload
@ -17,24 +25,10 @@
- try moving some more of the CLI tests to py - try moving some more of the CLI tests to py
- the gif tests in the suite sometimes fail with giflib5 because of an
uninitialized struct in giflib, see
https://sourceforge.net/p/giflib/bugs/94/
sadly ubuntu 16.04 only comes with giflib5, and giflib5 is currently broken
- I like the new int mask creator in reducev, can we use it in im_vips2imask()
as well?
what about making the int masks for the interpolators?
- try SEQ_UNBUFFERED on jpg source, get out of order error? - try SEQ_UNBUFFERED on jpg source, get out of order error?
- could load pdf thumbnails? - could load pdf thumbnails?
- still not happy about float->int mask conversion in im_vips2mask.c
- colour needs to split _build() into preprocess / process / postprocess - colour needs to split _build() into preprocess / process / postprocess
phases phases

View File

@ -32,8 +32,8 @@
*/ */
/* /*
#define DEBUG
*/ */
#define DEBUG
#ifdef HAVE_CONFIG_H #ifdef HAVE_CONFIG_H
#include <config.h> #include <config.h>