User forums > Nightly builds
The 10 October 2009 build (5859) is out.
Jenna:
--- Quote from: jens on October 13, 2009, 09:14:07 am ---To all with large projects and long timeouts, please try to disable the symbols browser, not the whole codecompletion and post here whether it changes anything.
Check "Settings -> Editor... -> Code-completion and symbols browser -> Symbols browser -> Disable symbols browser"
--- End quote ---
Symbols-browser should now (svn r5872) should now be much faster.
Please test (if possible) and report back any issues.
ollydbg:
--- Quote from: jens on October 21, 2009, 01:55:05 pm ---
--- Quote from: jens on October 13, 2009, 09:14:07 am ---To all with large projects and long timeouts, please try to disable the symbols browser, not the whole codecompletion and post here whether it changes anything.
Check "Settings -> Editor... -> Code-completion and symbols browser -> Symbols browser -> Disable symbols browser"
--- End quote ---
Symbols-browser should now (svn r5872) should now be much faster.
Please test (if possible) and report back any issues.
--- End quote ---
Symbols browser runs faster!!!!!
Thanks.
But I still have more questions: see the screenshot below:
Q1, it seems in vector.tcc, the function has correctly parsed, butin the symbols browser, why there is only one member under std::vector?
Q2, this question maybe is not related to symbols-browser, it seems there is a namespace "std" in namespace "std". :(
ollydbg:
Also, when I open the CC debug dialog, There are many members of std::vector. see the screen shot below:
This is more weird.
Jenna:
--- Quote from: ollydbg on October 21, 2009, 02:36:20 pm ---Q1, it seems in vector.tcc, the function has correctly parsed, butin the symbols browser, why there is only one member under std::vector?
--- End quote ---
This has a simple cause.
The previous code has looked for double-entries while building the tree, the last parsed entry remained, all others have been deleted.
I do something similar, with one difference: the last entry after sorting remains.
I thought about not searching for doubles, but decided to do so.
The parser tells us there are several places where vector (in your example) is declared, even if that can not be (at least not in the same namespace).
In my opinion that is a parsers issue, because the parser dos not handle the gcc include-files correctly (due to defines, macros and whatever).
I tend to remove the RemoveDoubles-function, because it's not the clasbrowserbuilderthread's purpose to (try to) correct parser-issues.
BCCISProf:
Question: The last time the Mac OS X version has been updated was last March. Is there anyone who could possibly prepare a Mac OS X version of a recent nightly?
Thank you
Navigation
[0] Message Index
[#] Next page
[*] Previous page
Go to full version