[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [cdt-dev] indexing results difference between CDT 7.0.x and CDT 8.0

Hi Markus,

 

I will try create a simple example.

 

I would prefer to start with narrowing when the indexing regression occurred? For that I need to install CDT 8.0 M7 and other release candidates till CDT 8.0 final release. Unfortunately I am not able to find links to these installations L

 

Thanks,

Yevgeny

 

From: cdt-dev-bounces@xxxxxxxxxxx [mailto:cdt-dev-bounces@xxxxxxxxxxx] On Behalf Of Schorn, Markus
Sent: Monday, August 08, 2011 11:18 AM
To: CDT General developers list.
Subject: Re: [cdt-dev] indexing results difference between CDT 7.0.x and CDT 8.0

 

Hi Yevgeny,

Using line 75 of your source file, you should be able to construct an example that demonstrates the problem.

Could be https://bugs.eclipse.org/bugs/show_bug.cgi?id=347462

Markus.

 

From: cdt-dev-bounces@xxxxxxxxxxx [mailto:cdt-dev-bounces@xxxxxxxxxxx] On Behalf Of Yevgeny Shifrin
Sent: Sunday, August 07, 2011 10:58
To: CDT General developers list.
Subject: Re: [cdt-dev] indexing results difference between CDT 7.0.x and CDT 8.0
Importance: Low

 

Hi,

 

I've seen a case when call hierarchy is working fine from CDT 7.0.2 and not working from CDT 8.0.0.

 

Class B derives from class A. Call Hierarchy for class A constructor, shows Class B constructor in CDT 7.0.2 and not in CDT 8.0.0. I compared between the two Class B parser logs. In CDT 8.0.0 there is the following line ("Unresolved names" section): "Attempt to use symbol failed: A in file B.c++:75". Line 75 is exactly where constructor B calls constructor A. Any idea what that means?

 

Regarding my previous mail:

As I mentioned below, CDT 8.0 M7 had indexing statistics similar to CDT 7.0.2. How many milestones there are between CDT 8.0 M7 and CDT 8.0 official release. If there are not many, I could check them and narrow in which milestone this regression was introduced. Is there a link where I could find all CDT 8.0 milestones?

 

Thanks,

Yevgeny

 

From: cdt-dev-bounces@xxxxxxxxxxx [mailto:cdt-dev-bounces@xxxxxxxxxxx] On Behalf Of Yevgeny Shifrin
Sent: Thursday, August 04, 2011 9:56 AM
To: CDT General developers list.
Subject: Re: [cdt-dev] indexing results difference between CDT 7.0.x and CDT 8.0

 

Hi,

 

Thank you for your reply.

 

Is there a way to produce a full list of unresolved symbols/names? I've seen an option to provide a list of "Unresolved Includes" but it does not help.

 

BTW: As I mentioned below, CDT 8.0 M7 had indexing statistics similar to CDT 7.0.2. How many milestones there are between CDT 8.0 M7 and CDT 8.0 official release. If there are not many, I could check them and narrow in which milestone this regression was introduced. Is there a link where I could find all CDT 8.0 milestones?

 

Thanks,

Yevgeny

 

From: cdt-dev-bounces@xxxxxxxxxxx [mailto:cdt-dev-bounces@xxxxxxxxxxx] On Behalf Of Sergey Prigogin
Sent: Wednesday, August 03, 2011 7:43 PM
To: CDT General developers list.
Subject: Re: [cdt-dev] indexing results difference between CDT 7.0.x and CDT 8.0

 

Try to find examples of unresolved symbols and syntax errors that were not present in CDT 7.0.2, and then try to find patterns in them. Ideally you should be able to produce small reproducible examples and create bugs based on them.

 

-sergey

On Wed, Aug 3, 2011 at 5:10 AM, Yevgeny Shifrin <Yevgeny.Shifrin@xxxxxxxxxxx> wrote:

Hello,

 

I am working on big C++ code base. I see difference in indexing results between CDT 7.0.x and CDT 8.0 on exactly the same project with the same configurations.

 

CDT 7.0.2:

Indexed 'proj1' (several dozen K of sources and headers) in 2,602.16 sec: ~2M declarations; ~11M+17544 references; ~1K unresolved inclusions; ~1K syntax errors; ~80K unresolved names

 

CDT 8.0.0:

Indexed 'proj1' (several dozen K of sources and headers) in 3,428.65 sec: ~2M+35 declarations; ~11M references; ~1K unresolved inclusions; ~1K+25 syntax errors; ~200K unresolved names

 

I would like to understand the root cause for this difference. Is there a regression in CDT 8.0 behavior, or other reason? In CDT 8.0 M7 the statistics were similar to CDT 7.0.2. Is it possible that between CDT 8.0 M7 and CDT 8.0 release, change was submitted that causes this regression. From indexing functionality point of view (in the last month) I did not see regression in CDT 8.0 in compare to CDT 7.0.2. At least in one case CDT 8.0 was "smarter" than CDT 7.0.2.

 

I would appreciate any comments suggestion on this issue.

 

Thanks,

Yevgeny

 

 

 

 

This e-mail message is intended for the recipient only and contains information which is CONFIDENTIAL and which may be proprietary to ECI Telecom. If you have received this transmission in error, please inform us by e-mail, phone or fax, and then delete the original and all copies thereof.


_______________________________________________
cdt-dev mailing list
cdt-dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/cdt-dev

 

This e-mail message is intended for the recipient only and contains information which is CONFIDENTIAL and which may be proprietary to ECI Telecom. If you have received this transmission in error, please inform us by e-mail, phone or fax, and then delete the original and all copies thereof.

This e-mail message is intended for the recipient only and contains information which is CONFIDENTIAL and which may be proprietary to ECI Telecom. If you have received this transmission in error, please inform us by e-mail, phone or fax, and then delete the original and all copies thereof.

This e-mail message is intended for the recipient only and contains information which is CONFIDENTIAL and which may be proprietary to ECI Telecom. If you have received this transmission in error, please inform us by e-mail, phone or fax, and then delete the original and all copies thereof.