Comments for Evan Teran's Blog http://blog.codef00.com Just some thoughts from a computer geek Thu, 05 Jan 2012 18:32:24 +0000 hourly 1 http://wordpress.org/?v=3.3.1 Comment on (Not so much) Fun with QSharedPointer by David Pinelo http://blog.codef00.com/2011/12/15/not-so-much-fun-with-qsharedpointer/comment-page-1/#comment-7839 David Pinelo Thu, 05 Jan 2012 18:32:24 +0000 http://blog.codef00.com/?p=174#comment-7839 Very interesting post. Sometimes it's hard to understand memory management when you see a lot of differents options (scope pointer, shared, shared data, parent-child...). A simple practical sample like you write made things clearer. Very interesting post. Sometimes it’s hard to understand memory management when you see a lot of differents options (scope pointer, shared, shared data, parent-child…). A simple practical sample like you write made things clearer.

]]> Comment on GMP error handling frustrates me by Evan Teran http://blog.codef00.com/2011/03/21/gmp-error-handling-frustrates-me/comment-page-1/#comment-7834 Evan Teran Mon, 05 Dec 2011 04:39:27 +0000 http://blog.codef00.com/?p=147#comment-7834 @gmp: that is fantastic information. I'll definitely look into rewriting the kcalc code to use the technique you mention. It definitely seems like a nice clean approach to memory allocation errors :-). @gmp: that is fantastic information. I’ll definitely look into rewriting the kcalc code to use the technique you mention. It definitely seems like a nice clean approach to memory allocation errors :-) .

]]> Comment on GMP error handling frustrates me by gmp http://blog.codef00.com/2011/03/21/gmp-error-handling-frustrates-me/comment-page-1/#comment-7833 gmp Tue, 29 Nov 2011 18:18:03 +0000 http://blog.codef00.com/?p=147#comment-7833 You can use alternate error handling in GMP. This is from Gmp Manual 5.0.4 chapter 14. So you can redifine a function for example to throw an exception, and it will be ok. By default GMP uses malloc, realloc and free for memory allocation, and if they fail GMP prints a message to the standard error output and terminates the program. Alternate functions can be specified, to allocate memory in a different way or to have a different error action on running out of memory. This feature is available in the Berkeley compatibility library (see Chapter 13 [BSD Compatible Functions], page 84) as well as the main GMP library. void mp_set_memory_functions ( void *(*alloc_func_ptr ) (size t), void *(*realloc_func_ptr ) (void *, size t, size t), void (*free_func_ptr ) (void *, size t)) [Function] Replace the current allocation functions from the arguments. If an argument is NULL, the corresponding default function is used. These functions will be used for all memory allocation done by GMP, apart from temporary space from alloca if that function is available and GMP is configured to use it. You can use alternate error handling in GMP. This is from Gmp Manual 5.0.4 chapter 14. So you can redifine a function for example to throw an exception, and it will be ok.

By default GMP uses malloc, realloc and free for memory allocation, and if they fail GMP prints a message to the standard error output and terminates the program.
Alternate functions can be specified, to allocate memory in a different way or to have a different error action on running out of memory.
This feature is available in the Berkeley compatibility library (see Chapter 13 [BSD Compatible Functions], page 84) as well as the main GMP library.
void mp_set_memory_functions (
void *(*alloc_func_ptr ) (size t),
void *(*realloc_func_ptr ) (void *, size t, size t),
void (*free_func_ptr ) (void *, size t))
[Function]
Replace the current allocation functions from the arguments. If an argument is NULL, the corresponding default function is used. These functions will be used for all memory allocation done by GMP, apart from temporary space from alloca if that function is available and GMP is configured to use it.

]]> Comment on It Lives! by NEO_2.0 http://blog.codef00.com/2008/12/05/it-lives/comment-page-1/#comment-7830 NEO_2.0 Wed, 24 Aug 2011 12:32:51 +0000 http://blog.codef00.com/?p=17#comment-7830 Hello, You have code the Best Debugger for Linux thank you! Greez NEO_2.0 Hello,

You have code the Best Debugger for Linux thank you!

Greez

NEO_2.0

]]> Comment on The Default Browser on Linux Debacle by “Default” Applications Obstacle to Desktop Linux Adoption | Ghodmode Development http://blog.codef00.com/2011/02/18/the-default-browser-on-linux-debacle/comment-page-1/#comment-7829 “Default” Applications Obstacle to Desktop Linux Adoption | Ghodmode Development Wed, 15 Jun 2011 09:23:27 +0000 http://blog.codef00.com/?p=84#comment-7829 [...] on those XML files, but I did stumble across a functional solution with some help from an article at Evan Teran’s Blog. Setting the default browser using the update-alternatives command (see [...] [...] on those XML files, but I did stumble across a functional solution with some help from an article at Evan Teran’s Blog. Setting the default browser using the update-alternatives command (see [...]

]]> Comment on Combining Qt’s Signals and Slots with c++0x lamdas by Evan Teran http://blog.codef00.com/2011/03/27/combining-qts-signals-and-slots-with-c0x-lamdas/comment-page-1/#comment-7825 Evan Teran Tue, 24 May 2011 15:35:43 +0000 http://blog.codef00.com/?p=155#comment-7825 The object actually *does* get deleted (I've tested it by putting cout's in the destructor). The reason why is that the object is registers itself as a child of the sender. Note this part of my post: "We make the helper object a child of the some QObject (ideally the sender of the signal), so that it will be automatically deleted when the parent object gets destroyed." Granted, the object will live as long as the sender in this case. So you probably don't want to make 1000's of connections with this approach, but if it's a small number, there is no leak. The object actually *does* get deleted (I’ve tested it by putting cout’s in the destructor). The reason why is that the object is registers itself as a child of the sender. Note this part of my post:

“We make the helper object a child of the some QObject (ideally the sender of the signal), so that it will be automatically deleted when the parent object gets destroyed.”

Granted, the object will live as long as the sender in this case. So you probably don’t want to make 1000′s of connections with this approach, but if it’s a small number, there is no leak.

]]> Comment on Combining Qt’s Signals and Slots with c++0x lamdas by Essentia http://blog.codef00.com/2011/03/27/combining-qts-signals-and-slots-with-c0x-lamdas/comment-page-1/#comment-7824 Essentia Mon, 23 May 2011 17:55:31 +0000 http://blog.codef00.com/?p=155#comment-7824 This will almost certainly result in a memory leak, making this solution unattractive. This will almost certainly result in a memory leak, making this solution unattractive.

]]> Comment on The Default Browser on Linux Debacle by Evan Teran http://blog.codef00.com/2011/02/18/the-default-browser-on-linux-debacle/comment-page-1/#comment-7822 Evan Teran Sun, 24 Apr 2011 17:33:23 +0000 http://blog.codef00.com/?p=84#comment-7822 @Mike: I think that part of the issue is that I had build my system with both gnome and kde support (both kde and gnome in my gentoo USE flags). This seemed to be a bit a conflict where some programs would look to the gnome system, some would look to kde. Rebuilding my apps with just the kde USE flag made things a lot less confused. I'm not sure if I can still reproduce the issue after building it with both USE flags, but if I can, I'll open a bug. @Mike: I think that part of the issue is that I had build my system with both gnome and kde support (both kde and gnome in my gentoo USE flags).

This seemed to be a bit a conflict where some programs would look to the gnome system, some would look to kde. Rebuilding my apps with just the kde USE flag made things a lot less confused.

I’m not sure if I can still reproduce the issue after building it with both USE flags, but if I can, I’ll open a bug.

]]> Comment on The Default Browser on Linux Debacle by Mike http://blog.codef00.com/2011/02/18/the-default-browser-on-linux-debacle/comment-page-1/#comment-7820 Mike Sun, 24 Apr 2011 03:09:21 +0000 http://blog.codef00.com/?p=84#comment-7820 Hey there, just happened to run across this page. I'm the original author of xdg-settings, and it looks like you've found a bug. It's not supposed to do that. If you are so inclined, I'd be happy to try and figure out why it's not working for you, so that we can fix it to do what it's supposed to on systems like yours. You can open a bug at <a href="http://new.crbug.com" rel="nofollow">new.crbug.com</a> or just email me if you'd like. Hey there, just happened to run across this page.

I’m the original author of xdg-settings, and it looks like you’ve found a bug. It’s not supposed to do that. If you are so inclined, I’d be happy to try and figure out why it’s not working for you, so that we can fix it to do what it’s supposed to on systems like yours.

You can open a bug at new.crbug.com or just email me if you’d like.

]]> Comment on The Default Browser on Linux Debacle by Vladimir http://blog.codef00.com/2011/02/18/the-default-browser-on-linux-debacle/comment-page-1/#comment-7811 Vladimir Mon, 11 Apr 2011 21:32:43 +0000 http://blog.codef00.com/?p=84#comment-7811 update: Problem returned after system update... It seams that .desktop file was updated to its default state. Than I found out that problem is not there, but it is in "supposed-to-be-default-browser".desktop I found (also in MimeType line) part that said something like <code> /usr/lib/"not-default-browser"</code> removed it and problem is gone(again) update:
Problem returned after system update… It seams that .desktop file was updated to its default state. Than I found out that problem is not there, but it is in
“supposed-to-be-default-browser”.desktop I found (also in MimeType line) part that said something like /usr/lib/"not-default-browser"
removed it and problem is gone(again)

]]>