Symbols browser issue of CC has been fixed for my Linux

[ollydbg]

What does pre-release means? Why don't you just call it "custom version"?

OK, I will change the tag name as soon as possible.

EDIT:
Done!

[eckard_klotz]

Hello Ollydbg.

Thanks for your effort.

Since I was unsure if your build contained the last changes for revision 12240 also I used the  C::B SVN head-revision together with the patch provided by Miguel Gimenez here https://sourceforge.net/p/codeblocks/tickets/1031/.

Sorry for not testing your build from November the 22nd.

However I was always following your discussion in this thread in the last weeks. Thus thank you to all who have shared their experiences here with us.

Stay well and healthy
                               Eckard Klotz.

[Miguel Gimenez]

The symbol browser patch is in trunk now, thank you to all who tested it.

[blauzahn]

compiling current trunk fails due to a syntax error in classbrowser.cpp:1188
within preprocessor block #ifndef CC_NO_COLLAPSE_ITEM

Code

void ClassBrowser:CollapseItem(CCTreeItem* item)

should be:

Code

void ClassBrowser::CollapseItem(CCTreeItem* item)

Thank you.

[MortenMacFly]

compiling current trunk fails due to a syntax error in classbrowser.cpp:1188
[...]

This is fixed in sVN-trunk but I wonder if CC_NO_COLLAPSE_ITEM is still needed... Maybe I should give it a shot...

[killerbot]

I can confirm builds are broken, I have it on 2 different systems (linux) (on the other hand the only 2 I tried )

Code

classbrowser.cpp: In member function ‘void ClassBrowser::CollapseItem(CCTreeItem*)’:
classbrowser.cpp:1188:33: error: no matching function for call to ‘ClassBrowser::GetId(CCTreeItem*&)’
     wxTreeItemId Id = GetId(item);
                                 ^
In file included from /usr/include/wx-3.0/wx/wx.h:38:0,
                 from /usr/include/wx-3.0/wx/wxprec.h:58,
                 from ./sdk_common.h:24,
                 from ./sdk_precomp.h:13,
                 from ./sdk.h:17:
/usr/include/wx-3.0/wx/window.h:241:16: note: candidate: wxWindowID wxWindowBase::GetId() const
     wxWindowID GetId() const { return m_windowId; }
                ^~~~~

[Miguel Gimenez]

Windows does not give this error, I will check on Linux.

EDIT: On Windows and Linux the CBP files have CC_NO_COLLAPSE_ITEM defined and the code is skipped. Looks like autotools does not define this macro.

I will fix this, but I think both build systems should use the same defines.

[oBFusCATed]

If anyone is interested in reliable CI builds: https://travis-ci.org/github/obfuscated/codeblocks_sf/builds (yes, they are still broken)

[Miguel Gimenez]

Fixed in ticket 1061, https://sourceforge.net/p/codeblocks/tickets/1061/

[killerbot]

Code

classbrowser.cpp: In member function ‘void ClassBrowser::CollapseItem(CCTreeItem*)’:
classbrowser.cpp:1188:33: error: no matching function for call to ‘ClassBrowser::GetId(CCTreeItem*&)’
     wxTreeItemId Id = GetId(item);
                                 ^
In file included from /usr/include/wx-3.0/wx/wx.h:38:0,
                 from /usr/include/wx-3.0/wx/wxprec.h:58,
                 from ./sdk_common.h:24,
                 from ./sdk_precomp.h:13,
                 from ./sdk.h:17:
/usr/include/wx-3.0/wx/window.h:241:16: note: candidate: wxWindowID wxWindowBase::GetId() const
     wxWindowID GetId() const { return m_windowId; }
                ^~~~~

Builds are broken for a few weeks now, should the code that broke it be reverted ?

[Miguel Gimenez]

The patch is there (1061), better fix than revert. This call to GetId() was a placeholder I later forgot about because this code is not compiled at all when using the CBP.

Other question is if the CC_NO_COLLAPSE_ITEM define must be removed from the CBP or added to makefile.am for coherency, but this is not urgent.

[killerbot]

I will apply the patch ,but the CC_NO_COLLAPSE_ITEM subject remains open.

[tigerbeard]

Yesterday I built trunc (12295) in Ubuntu18.04 and it compiled and runs fine (using wx3.0.4) and gcc 7.4.0.
The Symbol browser seems to be working fine.

Is there any impact from this issue with CC_NO_COLLAPSE_ITEM for normal use?

[Miguel Gimenez]

It is a minor optimization trading memory for speed, but hardly noticeable.

[tigerbeard]

It is a minor optimization trading memory for speed, but hardly noticeable.

Great relief 
Somehow I had the impression that this was a kind of blocking item to get the whole thing going.

I did not do the Windows build yet, but its really a major milestone to have a fully working symbol browser back. Your first patch was already a great step forward and I have been using that with a V17. Now even the few drawbacks of that one are gone now.

Thanks a lot for your great effort!