The 09 october 2006 build is out.

[killerbot]

Get quick announcements through the RSS feed http://www.codeblocks.org/nightly/CodeBlock_RSS.xml

A link to the unicode windows wxWidget dll for Code::Blocks : http://prdownload.berlios.de/codeblocks/wxmsw26u_gcc_cb_wx2.6.3p2.7z

For those who might need this one (when no MingW installed on your system) : the mingw10m.dll : http://prdownload.berlios.de/codeblocks/mingwm10.7z

For support of ansi builds, a link to the ansi windows wxWidget dll for Code::Blocks : http://prdownload.berlios.de/codeblocks/wxmsw26_gcc_cb_wx2.6.3p2.7z

The 09 October 2006 build is out.
  - Windows : http://prdownload.berlios.de/codeblocks/CB_20061009_rev3050_win32.7z
  - Linux :
         http://prdownload.berlios.de/codeblocks/CB_20061009_rev3050_Ubuntu6.06.deb
         http://prdownload.berlios.de/codeblocks/CB_20061009_rev3050_suse100+101.rpm


Resolved Fixed:

• Fixed minor bug in code-completion local block parsing (would miss the first line of the block, sometimes)

Regressions/Confirmed/Annoying/Common bugs:

• toolbar-images-not-changing-state (is a wx problem/Win XP problem)
• menu items with icon not correctly aligned (since wx263)

[NobbZ]

Sorry, but when I click the downloadlink on berlios-page, the I get a HTML-Document returned with the content "File doesnt exist" :-(

[stahta01]

Try http://prdownload.berlios.de/codeblocks/CB_20061009_rev3050_win32.7z

I can't try it because I have trouble with that server; I am on Dial-up and have trouble FTPing from Berlios, I use SVN instead.

[killerbot]

Sorry, but when I click the downloadlink on berlios-page, the I get a HTML-Document returned with the content "File doesnt exist" :-(

link is fixed, my mistake, sorry

[NobbZ]

Thanks, it is working now!

[lfm]

CB_20061009_rev3050_suse100+101.rpm intsalled only used 16M diskspace, but I installed (compile)  rev3050 from svn used 79M diskspace, why?

[killerbot]

? debug symbols stripped ?

[lfm]

? debug symbols stripped ?

How to strip  debug symbols?

my steps:

#./bootstrap
#./configure --prefix=/usr/local/codeblocks --enable-contrib
#make
#make install

[Ceniza]

There's a script called update that strips those symbols... but it only works when you compile it using the project file.

What I've used to compile it in release mode is to provide the flags for compiling and linking, like this:

# ./configure CFLAGS="-O3 -fomit-frame-pointer -pipe -ffast-math" CXXFLAGS="-O3 -fomit-frame-pointer -pipe -ffast-math -fpermissive" LDFLAGS="-s"

In fact I used some extra flags, but those should be enough.

[mandrav]

-fomit-frame-pointer

Last time I used this in linux with wxGTK, I got more segfaults than I could handle .

[BCCISProf]

I am getting the following error while editing a file in this build:

Assertion[cpMaxL=pdoc->length()] failed at sdk/wxscintilla/src/scintilla/scr/editor.c xx5977


Did not get this error on previous builds. Using Win XP SP2.

[Ceniza]

Last time I used this in linux with wxGTK, I got more segfaults than I could handle :wink:

It'd be safe to remove from those flags then :wink:

So far the only set of flags I've found to cause problems with some gtk+-1.2.x configure scripts are LDFLAGS="-Wl,-O1 -Wl,--enable-new-dtags -Wl,--sort-common -s -Wl,--as-needed -Wl,-z,combreloc" (haven't tried to find the guilty flags in there, even though -s is not one of those), but they worked just fine when I compiled wxGTK.

My whole Linux installation (Gentoo) is compiled using -fomit-frame-pointer, among other flags, (including wxGTK and Code::Blocks) and it hasn't caused any problems yet

[TheTuxKeeper]

Code Select Expand

#!/ bin/sh
binaries=`find /usr/local/codeblocks -type f \( -perm +111 -or -name "*.ko" \)`
# Strip ELF binaries
for f in $binaries
do
        file_output=`file "$f"`
        case $file_output in
            *ELF*", not stripped")
                ;;
            *ELF*", stripped")
                echo "WARNING: "`echo $f | sed -e "s,^$RPM_BUILD_ROOT/*,/,"`" is already stripped!"
                continue
                ;;
            *)
                continue
                ;;
        esac
        strip [--strip-all | --strip-unneeded]  "$f"
done

I took the /usr/lib/rpm/find-debuginfo.sh script of suse 10.0 (the one of 10.1 is different, don't know why) and removed unneeded things (object copying to .debug files and rpm specific things). I don't know whether --strip-all is a good idea or not, I think --strip-unneeded won't cause problems (I hope so :lol:).
Save this as a file, remove the space in the first line between "/" and "bash", replace [--strip-all | --strip-unneeded] with one of both options and run it as root with sh strip_codeblocks.sh. You can check whether te binaries have been stripped with for example file /usr/local/codeblocks/bin/codeblocks. A "stripped" should appear at the end
I havn't checked whether the script really works, but theoretically it should  :lol:

[lubos]

hello there
i noticed a bug in scrolling long(wide) files. it just stop scrolling at column 250 but there is 399 columns.when i use right mouse button it srolling normally.

i hope you understand :p
here is a small image:


can someone reply that?

[thomas]

hello there
i noticed a bug in scrolling long(wide) files. it just stop scrolling at column 250 but there is 399 columns.when i use right mouse button it srolling normally.

Scintilla feature, SciTE behaves the same way... if you need to scroll past 255, you need to place the cursor, sorry.

On a different note: Although this is truly not good behaviour from the editor's side, you should ask yourself if there is the possibility that something is wrong with your code, if you have more than 255 characters per line.
It is certainly debatable and hard to tell what is "normal" and what is not, but I guess most people will deem 60-120 characters as normal widths. Maybe up to 150, but that's about the maximum.
Anything that is a lot wider is quite unreadable, even on huge screens (and being required to scroll horizontally is generally considered very unergonomic).