Re: [Gimp-developer] [PATCH] Fix deprecated usage of gimp-free-select
Hi Martin! On Tue, Aug 23, 2011 at 1:44 AM, Martin Nordholts wrote: > There is already a patch that does this that no one have had time yet > (due to priorities) to commit: > https://bugzilla.gnome.org/show_bug.cgi?id=647834 Hummm... Next time I need to give a better look at the bug tracker :-) But good that there is already a patch that also fixes other deprecated calls. Best regards, Nelson ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
[Gimp-developer] [PATCH] Fix deprecated usage of gimp-free-select
Hi! Can somebody review and commit it, please? http://people.debian.org/~naoliv/misc/0001-Fix-deprecated-usage-of-gimp-free-select.txt Thank you! Best regards, Nelson ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
Re: [Gimp-developer] [PATCH] Fix some memleaks
Hi! On Wed, Aug 17, 2011 at 1:43 PM, Kevin Cozens wrote: > After you make the above mentioned changes, send a link to the updated patch > file. Updated patch available at http://people.debian.org/~naoliv/misc/0001-Fix-some-memleaks-2.txt Thank you! Best regards, Nelson ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
[Gimp-developer] [PATCH] Fix some memleaks
Hi! Can somebody verify and commit it, please? http://people.debian.org/~naoliv/misc/0001-Fix-some-memleaks.txt Thank you! Best regards, Nelson ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
Re: [Gimp-developer] [PATCH] Fix some errors and usage of deprecated procedures
Hi! On Tue, Feb 22, 2011 at 12:25 PM, Carol Spears wrote: > circuit.scm was repaired very recently (38 hours ago) in GIMP's git tree: > http://git.gnome.org/browse/gimp/commit/?id=b19645db4954a8fc8db83817dc7cd2e45e840efb > along with a few others. I saw it! :-) My patch only replaces the deprecated gimp-by-color-select by gimp-image-select-color (that isn't included in b19645d) > i recently opened a bug report with some patches for this same kind of > updates to different script-fus: > https://bugzilla.gnome.org/show_bug.cgi?id=642939 > if you would like to put your patches there, it might be easier for whoever > to get them from there when the time is right or whatever has to happen for > scripts to be updated. Included two patches there. Thank you! Best regards, Nelson ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
[Gimp-developer] [PATCH] Fix some errors and usage of deprecated procedures
Hi! While running the camouflage pattern plugin I saw an error: = Error while executing script-fu-camo-pattern: Error: ( : 1) Invalid number of arguments for gimp-image-insert-layer (expected 4 but received 3) = The attached patch fix this (and also fix the usage of some deprecated procedures). Thank you! Best regards, Nelson From 957448bd30ca612b0a291aec722576ac363105f7 Mon Sep 17 00:00:00 2001 From: Nelson A. de Oliveira Date: Tue, 22 Feb 2011 08:42:18 -0300 Subject: [PATCH] Fix some errors and usage of deprecated procedures --- plug-ins/script-fu/scripts/camo.scm| 11 +-- plug-ins/script-fu/scripts/circuit.scm | 10 +- plug-ins/script-fu/scripts/land.scm|6 +++--- 3 files changed, 9 insertions(+), 18 deletions(-) diff --git a/plug-ins/script-fu/scripts/camo.scm b/plug-ins/script-fu/scripts/camo.scm index 1b1faba..4c4311b 100644 --- a/plug-ins/script-fu/scripts/camo.scm +++ b/plug-ins/script-fu/scripts/camo.scm @@ -32,13 +32,13 @@ (gimp-context-push) -(gimp-image-insert-layer theImage baseLayer 0) +(gimp-image-insert-layer theImage baseLayer -1 0) (set! thickLayer (car (gimp-layer-new theImage theWidth theHeight RGBA-IMAGE "Camo Thick Layer" 100 NORMAL-MODE))) -(gimp-image-insert-layer theImage thickLayer 0) +(gimp-image-insert-layer theImage thickLayer -1 0) (set! thinLayer (car (gimp-layer-new theImage theWidth theHeight RGBA-IMAGE "Camo Thin Layer" 100 NORMAL-MODE))) -(gimp-image-insert-layer theImage thinLayer 0) +(gimp-image-insert-layer theImage thinLayer -1 0) (gimp-selection-all theImage) (gimp-context-set-background inColor1) @@ -54,8 +54,7 @@ (set! theBlur (- 16 inGrain)) (gimp-context-set-background inColor2) -(gimp-by-color-select thickLayer - '(0 0 0) 127 CHANNEL-OP-REPLACE TRUE FALSE 0 FALSE) +(gimp-image-select-color theImage CHANNEL-OP-REPLACE thickLayer '(0 0 0)) (gimp-edit-clear thickLayer) (gimp-selection-invert theImage) (gimp-edit-fill thickLayer BACKGROUND-FILL) @@ -66,7 +65,7 @@ (gimp-context-set-background inColor3) -(gimp-by-color-select thinLayer '(0 0 0) 127 CHANNEL-OP-REPLACE TRUE FALSE 0 FALSE) +(gimp-image-select-color theImage CHANNEL-OP-REPLACE thinLayer '(0 0 0)) (gimp-edit-clear thinLayer) (gimp-selection-invert theImage) (gimp-edit-fill thinLayer BACKGROUND-FILL) diff --git a/plug-ins/script-fu/scripts/circuit.scm b/plug-ins/script-fu/scripts/circuit.scm index 9ed5143..a30ca2b 100644 --- a/plug-ins/script-fu/scripts/circuit.scm +++ b/plug-ins/script-fu/scripts/circuit.scm @@ -112,15 +112,7 @@ (= remove-bg TRUE) (= separate-layer TRUE)) (begin - (gimp-by-color-select - active-layer - '(0 0 0) - 15 - 2 - TRUE - FALSE - 10 - FALSE) + (gimp-image-select-color image 2 active-layer '(0 0 0)) (gimp-edit-clear active-layer))) (if (= keep-selection FALSE) diff --git a/plug-ins/script-fu/scripts/land.scm b/plug-ins/script-fu/scripts/land.scm index ef48a77..cd9d8c1 100644 --- a/plug-ins/script-fu/scripts/land.scm +++ b/plug-ins/script-fu/scripts/land.scm @@ -38,19 +38,19 @@ ) (gimp-context-set-gradient gradient) (gimp-image-undo-disable img) - (gimp-image-add-layer img layer-one 0) + (gimp-image-insert-layer img layer-one -1 0) (plug-in-solid-noise RUN-NONINTERACTIVE img layer-one TRUE FALSE seed detail xscale yscale) (plug-in-c-astretch RUN-NONINTERACTIVE img layer-one) (set! layer-two (car (gimp-layer-copy layer-one TRUE))) - (gimp-image-add-layer img layer-two -1) + (gimp-image-insert-layer img layer-two -1 -1) (gimp-image-set-active-layer img layer-two) (plug-in-gradmap RUN-NONINTERACTIVE img layer-two) - (gimp-by-color-select layer-one '(190 190 190) 55 CHANNEL-OP-REPLACE FALSE FALSE 0 FALSE) + (gimp-image-select-color img CHANNEL-OP-REPLACE layer-one '(190 190 190)) (plug-in-bump-map RUN-NONINTERACTIVE img layer-two layer-one 135.0 35 landheight 0 0 0 0 TRUE FALSE 0) ;(plug-in-c-astretch RUN-NONINTERACTIVE img layer-two) -- 1.7.2.3 ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
Re: [Gimp-developer] Wrong tools contour
On Thu, Feb 3, 2011 at 2:32 PM, Alexia Death wrote: > On Thu, Feb 3, 2011 at 2:50 PM, Nelson A. de Oliveira > wrote: >> I don't know if it's a know issue, > Im sorry, but I find this question a bit annoying. It's a big issue > smack in the middle where you cant miss it. Of course its known. So users will have to always stay in doubt if one issue is known or not, because it annoys to ask? I did search the bug reports and didn't find anything (maybe I didn't search very well, but I did try). Also, there isn't a "known issues" list in gimp.org. How I am supposed to know what is known or not, besides asking? >> (using latest git version). > Git is a moving target. It is occasionally broken. If you want to > complain about something in it, its best done in IRC and after > building again to see if it has been fixed already. See that there is a difference between complaining and asking (even if I didn't use an interrogation, it indeed was a doubt): I didn't say that "Gimp sux! Brush contours are all broken and wrong". Also see that by latest git version, I really mean the latest version (7e6d2542f22f7168abc7160fc0960e405e2eb607 right now). So the problem is still there. And what is the difference/problem in asking in the mailing-list instead IRC? If I don't need to have a fast answer, my network has some kind of restriction or I don't know what IRC is, why do I have to ask on IRC? Nelson ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
[Gimp-developer] Square brushes
And hi one more time! Couldn't we have a square (and square fuzzy) brush by default on Gimp or it's out of the scope/vision/something else of the product? Thanks again! Best regards, Nelson ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
[Gimp-developer] Icon for dynamics
Hi! I don't know if it's related to the GTK theme or something else, but having a button with "?" on it, to choose the dynamics mode is somehow strange (see http://sites.google.com/site/naoliv/gimp/1.png). It looks something like a help button. If we look at the mode for pencil, for example (http://sites.google.com/site/naoliv/gimp/2.png), we can see that it's a drop-down menu (and not a "help" button). Couldn't we have the dynamics mode like this too? http://sites.google.com/site/naoliv/gimp/3.png seems much better and intuitive. Thank you! Best regards, Nelson ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
[Gimp-developer] Wrong tools contour
Hi! I don't know if it's a know issue, but it seems that the contour from pencil, paintbrush and rubber are being wrongly drawn on the screen (using latest git version). A video is available at http://sites.google.com/site/naoliv/gimp demonstrating this issue. See that when using the pencil, the "round" contour isn't symmetric as it should be. For the paintbrush and the rubber the contour keeps changing with every move. It's a little annoying and makes us think "Why it's drawing outside the contour?" or "Why it's not drawing where it's supposed to?" Thank you! Best regards, Nelson ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
Re: [Gimp-developer] scanfs without field width limits making Gimp crash
Hi!
On Mon, Jan 24, 2011 at 9:41 AM, Simon Budig wrote:
> Ah sorry, should have mentioned that. The bug report is older than your
> mail to the list. We had a report on the %s conversion earlier which is
> what the patch attached to the bug attempts to fix.
Here (also from your patch):
snprintf (fmt_str, sizeof (fmt_str), "%%%lds %%%lds %%%lds %%%lds",
sizeof (colorstr_r) - 1, sizeof (colorstr_g) - 1,
sizeof (colorstr_b) - 1, sizeof (colorstr_a) - 1);
sscanf (ptr, fmt_str, colorstr_r, colorstr_g, colorstr_b, colorstr_a);
It will protects against the overflow, but there is a chance to get
wrong data (if the first string is also very big). For example, with
this ugly example code that I think that is similar to the one from
your patch:
#include
int main()
{
char str1[16];
char str2[16];
char str3[16];
char str4[16];
char fmt[128];
char buf[] =
"
bbb ccc ddd";
snprintf(fmt, sizeof(fmt), "%%%lds %%%lds %%%lds %%%lds",
sizeof(str1) - 1, sizeof(str2) - 1, sizeof(str3) - 1, sizeof(str4) -
1);
sscanf(buf, fmt, str1, str2, str3, str4);
printf("*%s* *%s* *%s* *%s*", str1, str2, str3, str4);
return 0;
}
See that we have one big string first and all the four vars (wrongly)
were used by it:
*aaa* *aaa* *aaa* *aaa*
Right?
Best regards,
Nelson
___
Gimp-developer mailing list
Gimp-developer@lists.XCF.Berkeley.EDU
https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
Re: [Gimp-developer] scanfs without field width limits making Gimp crash
Hi! On Mon, Jan 24, 2011 at 8:26 AM, Simon Budig wrote: > For Gimp itself there is a bug report on this issue at > https://bugzilla.gnome.org/show_bug.cgi?id=639203 > > I guess I'll commit the patch attached to the bugreport soon unless > someone has a better suggestion. But here, for example (from your patch): snprintf (fmt_str, sizeof (fmt_str), "%%d %%d %%%lds", sizeof (endbuf) - 1); if (sscanf (line, fmt_str, &t->majtype, &t->type, end) != 3) Won't it still be affected by a very large integer (like the example that I sent on my initial message) at the first or second position in the file? I get this when trying to load an example with the the big number: = Plug-in crashed: "sphere-designer" (/usr/lib/gimp/2.0/plug-ins/sphere-designer) The dying plug-in may have messed up GIMP's internal state. You may want to save your images and restart GIMP to be on the safe side. = But I can't say how bad or ignorable it is. Best regards, Nelson ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
Re: [Gimp-developer] scanfs without field width limits making Gimp crash
Hi! On Mon, Jan 24, 2011 at 1:11 AM, Christopher Curtis wrote: > On Sat, Jan 22, 2011 at 2:04 AM, Nelson A. de Oliveira > wrote: > >> To make it crash: >> perl -e 'print "5"x210' | ./a.out > > I believe the unbounded '%s' is a legitimate bug, but is the '%i' > assertion true? > > The example it gives doesn't crash when I run it. Instead scanf > returns ERANGE and (oddly) sets 'a' to -1. This may be Linux specific > behavior though. Both on Linux and FreeBSD I get a segmentation fault with the example code. I can't test on other platforms however. Best regards, Nelson ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
[Gimp-developer] scanfs without field width limits making Gimp crash
Hi!
While testing gimp with cppcheck I saw a lot of warnings caused by the
usage of scanf and fscanf without specifying a width limit.
One example:
=
[./app/gegl/gimpcurvesconfig.c:392]: (warning) scanf without field
width limits can crash with huge input data. To fix this error message
add a field width specifier:
%s => %20s
%i => %3i
Sample program that can crash:
#include
int main()
{
int a;
scanf("%i", &a);
return 0;
}
To make it crash:
perl -e 'print "5"x210' | ./a.out
=
Indeed it's possible to make gimp crash by using a curve file with
such big value; you can test by trying to import
http://people.debian.org/~naoliv/misc/gimp/curve.cur
A full list of fscanf/scanf warnings is available at
http://people.debian.org/~naoliv/misc/gimp/scanf.txt
Thank you!
Best regards,
Nelson
___
Gimp-developer mailing list
Gimp-developer@lists.XCF.Berkeley.EDU
https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
[Gimp-developer] [PATCH] Fix for memory and file descriptor leaks
Hi! Can somebody review two patches that (hopefully) fix some memory and file descriptor leaks, please? The first one is basically g_free()s and fclose()s: http://people.debian.org/~naoliv/misc/gimp/patches/0001-Fix-memory-and-descriptor-leaks.patch For the second one, instead including a fclose(fp) for every block with a "return FALSE;" in it I just moved it below (since it was not being used before). http://people.debian.org/~naoliv/misc/gimp/patches/0002-Move-file-opening-block-to-after-some-tests.patch Is there something wrong with them, please? Thank you! Best regards, Nelson ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
Re: [Gimp-developer] GIMP 2.8 schedule
Hi! Can donations (money) to GIMP somehow speed up its development, please? Thank you! Best regards, Nelson ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
Re: [Gimp-developer] Fixes for memory leaks
Hi again! I gave another look on the possible mem leaks and created another patch: http://people.debian.org/~naoliv/misc/0001-Fix-memory-leakages.patch Can somebody review them, please? There is both http://people.debian.org/~naoliv/misc/0001-Fix-memleaks-in-plug-ins.patch http://people.debian.org/~naoliv/misc/0001-Fix-memory-leakages.patch Thank you! Best regards, Nelson ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
Re: [Gimp-developer] Fixes for memory leaks
Hi! On Wed, Nov 18, 2009 at 4:23 AM, Martin Nordholts wrote: > Please generate the patch as a commit according to these instructions: > http://gimp.org/bugs/howtos/submit-patch.html Patch available at http://people.debian.org/~naoliv/misc/0001-Fix-memleaks-in-plug-ins.patch >> A run on the full source code of gimp (version is the latest one from >> git), also gives some messages: >> http://people.debian.org/~naoliv/misc/gimp-ccpcheck.txt > > Interesting. Most leaks are for tests or plug-ins though which are less > crucial, but there are four leaks in the core. There are also possible memleaks: http://people.debian.org/~naoliv/misc/gimp-possible-memleaks.txt Best regards, Nelsob ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
Re: [Gimp-developer] Scope of variables
Hi again! On Wed, Nov 18, 2009 at 12:23 AM, Nelson A. de Oliveira wrote: > For example, i is not used inside only one if here in > app/core/gimpimage-convert.c (and thus has been moved to where it's > only necessary): I need to sleep. Read it as: "For example, i is used inside only one if here in app/core/gimpimage-convert.c" ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
[Gimp-developer] Scope of variables
Hi!
Is it interesting to limit the scope of the variables to where they
are only used?
For example, i is not used inside only one if here in
app/core/gimpimage-convert.c (and thus has been moved to where it's
only necessary):
--- app/core/gimpimage-convert.c.old2009-11-18 00:18:55.0 -0200
+++ app/core/gimpimage-convert.c2009-11-18 00:19:34.0 -0200
@@ -835,8 +835,6 @@
/* Convert to indexed? Build histogram if necessary. */
if (new_type == GIMP_INDEXED)
{
- gint i;
-
/* fprintf(stderr, " TO INDEXED(%d) ", num_cols); */
/* don't dither if the input is grayscale and we are simply
@@ -897,6 +895,7 @@
! needs_quantize &&
palette_type == GIMP_MAKE_PALETTE)
{
+ gint i;
/* If this is an RGB image, and the user wanted a custom-built
* generated palette, and this image has no more colours than
* the user asked for, we don't need the first pass (quantization).
Are these kind of minor changes interesting for Gimp, please?
Thank you!
Best regards,
Nelson
___
Gimp-developer mailing list
Gimp-developer@lists.XCF.Berkeley.EDU
https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
[Gimp-developer] Fixes for memory leaks
Hi! While giving a look at bug #595605 (from Gimp Bugzilla), I have remembered about cppcheck. I have created a patch available at http://people.debian.org/~naoliv/misc/gimp-mememleak.diff fixing (I hope) some leakages inside the plug-ins dir. Can somebody review it, please? There are some errors remaining (that or seemed to be unnecessary to fix or I couldn't find an easy way to fix): [./common/file-compressor.c:775]: (error) Memory leak: filename_copy [./common/file-xmc.c:888]: (error) Memory leak: positions [./common/mail.c:610]: (error) Memory leak: filename_copy [./common/plugin-browser.c:329]: (error) Memory leak: tmp_ptr [./common/plugin-browser.c:355]: (error) Memory leak: leaf_ptr [./file-psd/psd-util.c:200]: (error) Memory leak: pascal_str [./flame/libifs.c:1234]: (error) Memory leak: points [./file-xjt/xjt.c:3138]: (error) Memory leak: l_file_buff [./gfig/gfig-star.c:222]: (error) Memory leak: line_pnts [./gfig/gfig-star.c:222]: (error) Memory leak: min_max A run on the full source code of gimp (version is the latest one from git), also gives some messages: http://people.debian.org/~naoliv/misc/gimp-ccpcheck.txt Thank you! Best regards, Nelson ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
Re: [Gimp-developer] Improved brush editing interface mock-up
Hi! On Sat, Jul 25, 2009 at 8:19 AM, SHIRAKAWA Akira wrote: > New interface and improvements mock-up link: > http://i31.tinypic.com/2qjd55k.png This is something that I, as an user, would like to have. It also seems to save some screen space when editing on a notebook that can do only 1024x768 too :-) Best regards, Nelson ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
[Gimp-developer] RFC and some doubts about crop guide overlays
Hi! I would like to have some comments/suggestions/criticisms/etc about the quality and the implementation of the following crop guide overlays. This is related with http://bugzilla.gnome.org/show_bug.cgi?id=575911 Please, consider them as proof of concept; also, I am only including the relevant part of the code. First the easy one: a rule of fifths (more info about the rule of fifths at http://markhancock.blogspot.com/2006/03/address-basic-composition.html) Code is available at http://naoliv.googlepages.com/gimp-fifhts.txt Can it be implemented in a better way, optimized, etc? Now about the other overlays, like this golden triangle overlay: http://naoliv.googlepages.com/gimp-triangle.txt This is the code to represent http://pendery.org/files/other%20images/Techniques%20-%20Lightroom/Triangle.png (that is the pink lines from http://www.corel.com/img/content/community/tips/px/2007-03d/Divine_Proportion_Final_1.jpg). Can I use "phi" as I am using now or can I define it like SQRT5 is already defined? Also, it will be necessary to flip it vertically and horizontally (like it's possible to see at http://www.corel.com/img/content/community/tips/px/2007-03d/Divine_Proportion_Final_0.jpg). Any hints how can I do this, please? Is it possible to draw lines with different colors? (like http://www.corel.com/img/content/community/tips/px/2007-03d/Divine_Proportion_Final_1.jpg) Thank you very much! Best regards, Nelson ___ Gimp-developer mailing list Gimp-developer@lists.XCF.Berkeley.EDU https://lists.XCF.Berkeley.EDU/mailman/listinfo/gimp-developer
