[
Date Prev][
Date Next][
Thread Prev][
Thread Next][
Date Index][
Thread Index]
[
List Home]
|
RE: [cdt-dev] Problems with indexing/searching
|
Hi Larry,
it was the wrong bug, https://bugs.eclipse.org/bugs/show_bug.cgi?id=147080
dealt with your actual problem and it was fixed.
In general the indexer still is far from perfect. With M4 we were able to
reduce the failure rate from 21% down to 8% (for indexing firefox 2.0).
Indexing is not the only thing that needs our attention we are trying to
improve lot's of other stuff (build, editor, outline-view, content-assist,
call- and type-hierarchy, ...).
I hope that we'll be able to adress more indexing problems in the 4.0
release cycle and you can help us here:
Before we can fix a problem we have to understand it, and with indexing
this is not an easy thing. As Dave pointed out, you can save us from
doing this part of the work by tracking down a problem such that it can
easily be reproduced with a small set of files. This way we have more
time to work on the actual fixes.
Markus.
-----Original Message-----
From: cdt-dev-bounces@xxxxxxxxxxx [mailto:cdt-dev-bounces@xxxxxxxxxxx] On Behalf Of Larry Streepy
Sent: Donnerstag, 25. Jänner 2007 18:40
To: CDT General developers list.
Subject: Re: [cdt-dev] Problems with indexing/searching
Hi Markus,
Sorry for the improper use of this channel. However, I did read through
that bug and the discussion seemed to morph from "the indexer takes too
long and gets wrong results" into "we need to add better tracing to the
indexer." The latter is what seems to have been done when the bug was
closed. The underlying issue (broken/long running indexer) seems to have
been left behind. Has it been moved to a new bug? If so, I'll use that
forum to comment further.
Thanks,
Larry.
Schorn, Markus wrote:
> Hi Larry,
> in general bugs are not discussed on this mailing list. You
> can find the discussion in bugzilla. The bug you are referring
> to has been fixed:
> https://bugs.eclipse.org/bugs/show_bug.cgi?id=155324
>
> Markus.
>
> -----Original Message-----
> From: cdt-dev-bounces@xxxxxxxxxxx [mailto:cdt-dev-bounces@xxxxxxxxxxx] On Behalf Of Larry Streepy
> Sent: Mittwoch, 24. Jänner 2007 21:39
> To: CDT General developers list.
> Subject: Re: [cdt-dev] Problems with indexing/searching
>
> I haven't seen any discussionon this for a while, so I'll ask: has this
> been addressed in the new 4.0 RC? It would really be nice to have a
> functional indexer to use on my code base. Right now, eclipse for C++ is
> not much better than vi because it has none of the navigational or
> reference searching available due to a busted indexer.
>
> Sorry to vent, but it's been very frustrating.
>
> If it's been addressed and I missed the message (I've been watching cdt-dev
> ever since I posted this), I apologize.
>
> Thanks,
> Larry.
>
> Sergey Prigogin wrote:
>> Exactly my experience. See bugzilla 155324
>> <https://bugs.eclipse.org/bugs/show_bug.cgi?id=155324>.
>>
>> -sergey
>>
>> On 10/12/06, *James Hawkins* <truiken@xxxxxxxxx
>> <mailto:truiken@xxxxxxxxx>> wrote:
>>
>> On 10/12/06, Larry Streepy <larry@xxxxxxxxxxxxxx
>> <mailto:larry@xxxxxxxxxxxxxx>> wrote:
>> > I've updated to the latest versions of the CDT and I'm still having
>> > trouble with the indexer.
>> >
>> > Here's some environmental data:
>> > Running on CentOs (Linux variant)
>> >
>> > Eclipse version 3.2.1
>> > CDT version 3.1.1.2006092708
>> >
>> > My project is big, but not massive (roughly 250,000 lines of C++
>> code).
>> >
>> > First, I just couldn't seem to get it to index my code properly
>> at all.
>> > The indexer would run for less than a minute and emit no
>> messages, so
>> > I thought things were just fine.
>> >
>> > Then I would try to do a navigate to a type (Navigate->Open Type) and
>> > I'd start typing in the name of *any* of the classes defined in
>> my code
>> > and none of them are present. The only things present seemed to be
>> > coming from some of my system includes, but nothing from my code
>> base.
>> > I tweaked a lot of settings and then told the indexer to run and
>> off it
>> > went (and has been going for more than an hour now).
>> >
>> > I looked in the eclipse .log file and I'm seeing several
>> > StackOverflowError messages being emitted, so something is definitely
>> > off with the indexer. Here's a portion of one of the stack traces:
>> >
>> > java.lang.StackOverflowError
>> >
>> org.eclipse.cdt.internal.core.pdom.db.ShortString.compare(ShortString.java:202)
>> >
>> org.eclipse.cdt.internal.core.pdom.dom.PDOMNamedNode$NodeFinder.compare(PDOMNamedNode.java
>> :85)
>> > org.eclipse.cdt.internal.core.pdom.db.BTree.accept(BTree.java:226)
>> > org.eclipse.cdt.internal.core.pdom.db.BTree.accept(BTree.java:229)
>> > org.eclipse.cdt.internal.core.pdom.db.BTree.accept(BTree.java :229)
>> > org.eclipse.cdt.internal.core.pdom.db.BTree.accept(BTree.java:195)
>> >
>> org.eclipse.cdt.internal.core.pdom.dom.cpp.PDOMCPPLinkage.adaptBinding(PDOMCPPLinkage.java:292)
>> >
>> org.eclipse.cdt.internal.core.pdom.dom.cpp.PDOMCPPLinkage.resolveBinding
>> (PDOMCPPLinkage.java:316)
>> > org.eclipse.cdt.internal.core.pdom.PDOM.resolveBinding(PDOM.java:248)
>> >
>> org.eclipse.cdt.internal.core.dom.parser.cpp.CPPScope.getBinding(CPPScope.java:134)
>> > org.eclipse.cdt.internal.core.dom.parser.cpp.CPPSemantics.lookup
>> (CPPSemantics.java:1019)
>> >
>> org.eclipse.cdt.internal.core.dom.parser.cpp.CPPSemantics.resolveBinding(CPPSemantics.java:613)
>> >
>> org.eclipse.cdt.internal.core.dom.parser.cpp.CPPVisitor.createBinding(CPPVisitor.java
>> :164)
>> >
>> org.eclipse.cdt.internal.core.dom.parser.cpp.CPPASTName.resolveBinding(CPPASTName.java:52)
>> >
>> org.eclipse.cdt.internal.core.pdom.dom.cpp.PDOMCPPNamespace.getBinding(PDOMCPPNamespace.java:146)
>> > org.eclipse.cdt.internal.core.dom.parser.cpp.CPPScope.getBinding
>> (CPPScope.java:136)
>> >
>> org.eclipse.cdt.internal.core.dom.parser.cpp.CPPVisitor.createBinding(CPPVisitor.java:350)
>> >
>> org.eclipse.cdt.internal.core.dom.parser.cpp.CPPVisitor.createBinding(CPPVisitor.java:191)
>> >
>> org.eclipse.cdt.internal.core.dom.parser.cpp.CPPASTName.resolveBinding(CPPASTName.java:52)
>> >
>> org.eclipse.cdt.internal.core.pdom.dom.cpp.PDOMCPPNamespace.getBinding(PDOMCPPNamespace.java:146)
>> > org.eclipse.cdt.internal.core.dom.parser.cpp.CPPScope.getBinding
>> (CPPScope.java:136)
>> >
>> org.eclipse.cdt.internal.core.dom.parser.cpp.CPPVisitor.createBinding(CPPVisitor.java:350)
>> >
>> org.eclipse.cdt.internal.core.dom.parser.cpp.CPPVisitor.createBinding(CPPVisitor.java:191)
>> >
>> org.eclipse.cdt.internal.core.dom.parser.cpp.CPPASTName.resolveBinding(CPPASTName.java:52)
>> >
>> org.eclipse.cdt.internal.core.pdom.dom.cpp.PDOMCPPNamespace.getBinding(PDOMCPPNamespace.java:146)
>> > org.eclipse.cdt.internal.core.dom.parser.cpp.CPPScope.getBinding
>> (CPPScope.java:136)
>> >
>> org.eclipse.cdt.internal.core.dom.parser.cpp.CPPVisitor.createBinding(CPPVisitor.java:350)
>> >
>> org.eclipse.cdt.internal.core.dom.parser.cpp.CPPVisitor.createBinding(CPPVisitor.java:191)
>> >
>> org.eclipse.cdt.internal.core.dom.parser.cpp.CPPASTName.resolveBinding(CPPASTName.java:52)
>> > ...
>> >
>> >
>> > There are two basic forms of stack overflow that have been occuring,
>> > here's a sample of the other:
>> >
>> > java.lang.StackOverflowError
>> >
>> org.eclipse.cdt.internal.core.dom.parser.ASTNode.getTranslationUnit(ASTNode.java:108)
>> >
>> org.eclipse.cdt.internal.core.dom.parser.ASTNode.getTranslationUnit(ASTNode.java
>> :108)
>> >
>> org.eclipse.cdt.internal.core.dom.parser.ASTNode.getTranslationUnit(ASTNode.java:108)
>> >
>> org.eclipse.cdt.internal.core.dom.parser.ASTNode.getTranslationUnit(ASTNode.java:108)
>> > org.eclipse.cdt.internal.core.dom.parser.cpp.CPPScope.getBinding
>> (CPPScope.java:128)
>> >
>> org.eclipse.cdt.internal.core.dom.parser.cpp.CPPClassScope.getBinding(CPPClassScope.java:238)
>> >
>> org.eclipse.cdt.internal.core.dom.parser.cpp.CPPSemantics.lookup(CPPSemantics.java:1019)
>> >
>> org.eclipse.cdt.internal.core.dom.parser.cpp.CPPSemantics.postResolution(CPPSemantics.java:665)
>> >
>> org.eclipse.cdt.internal.core.dom.parser.cpp.CPPSemantics.resolveBinding(CPPSemantics.java:629)
>> >
>> org.eclipse.cdt.internal.core.dom.parser.cpp.CPPVisitor.createBinding
>> (CPPVisitor.java:297)
>> >
>> org.eclipse.cdt.internal.core.dom.parser.cpp.CPPVisitor.createBinding(CPPVisitor.java:191)
>> >
>> org.eclipse.cdt.internal.core.dom.parser.cpp.CPPASTName.resolveBinding(CPPASTName.java:52)
>> >
>> org.eclipse.cdt.internal.core.pdom.dom.cpp.PDOMCPPNamespace.getBinding(PDOMCPPNamespace.java:146)
>> >
>> org.eclipse.cdt.internal.core.dom.parser.cpp.CPPVisitor.createBinding(CPPVisitor.java:350)
>> >
>> org.eclipse.cdt.internal.core.dom.parser.cpp.CPPVisitor.createBinding
>> (CPPVisitor.java:191)
>> >
>> org.eclipse.cdt.internal.core.dom.parser.cpp.CPPASTName.resolveBinding(CPPASTName.java:52)
>> >
>> org.eclipse.cdt.internal.core.pdom.dom.cpp.PDOMCPPNamespace.getBinding(PDOMCPPNamespace.java
>> :146)
>> >
>> org.eclipse.cdt.internal.core.dom.parser.cpp.CPPVisitor.createBinding(CPPVisitor.java:350)
>> >
>> org.eclipse.cdt.internal.core.dom.parser.cpp.CPPVisitor.createBinding(CPPVisitor.java:191)
>> >
>> org.eclipse.cdt.internal.core.dom.parser.cpp.CPPASTName.resolveBinding
>> (CPPASTName.java:52)
>> > ...
>> >
>> >
>> > I have increased the heap size and thread stack size considerably
>> (with
>> > -vmargs -Xmx2048M -Xss512M) so I don't think it's mu configuration.
>> >
>> > Any ideas on what might be wrong? Has this behavior been seen
>> before?
>> > I did search the archives and other blogs, but I didn't find anything
>> > directly relevant.
>> >
>>
>> I run into this problem when trying to index the source of the Wine
>> code base. The indexer either doesn't run at all, runs but stops
>> because of stack overflow, or it runs to completion (or so it says)
>> but only recognizes some (a very limited amount) of my information.
>>
>> --
>> James Hawkins
>> _______________________________________________
>> cdt-dev mailing list
>> cdt-dev@xxxxxxxxxxx <mailto:cdt-dev@xxxxxxxxxxx>
>> https://dev.eclipse.org/mailman/listinfo/cdt-dev
>>
>>
>>
>> ------------------------------------------------------------------------
>>
>> _______________________________________________
>> cdt-dev mailing list
>> cdt-dev@xxxxxxxxxxx
>> https://dev.eclipse.org/mailman/listinfo/cdt-dev
> _______________________________________________
> cdt-dev mailing list
> cdt-dev@xxxxxxxxxxx
> https://dev.eclipse.org/mailman/listinfo/cdt-dev
> _______________________________________________
> cdt-dev mailing list
> cdt-dev@xxxxxxxxxxx
> https://dev.eclipse.org/mailman/listinfo/cdt-dev
_______________________________________________
cdt-dev mailing list
cdt-dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/cdt-dev
