Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [cdt-dev] [tcf-dev] Expression Working Sets (was: Preparing M6)

Well,

 

From a Platform point of view of course a feature that’s generally applicable (like Workingsets) is preferable over a point solution (like gdb enhanced expressions).

 

I have one customer who agreed to review the workingset feature in its current state and provide feedback; to you think that is still a good idea ? Or is there any other way in which I could help ?

 

I’d also like to hear Eugene’s opinion on the workingset feature, what is the easiest way in which one can test it ?

 

Thanks,

Martin

--

Martin Oberhuber, SMTS / Product Architect – Development Tools, Wind River

direct +43.662.457915.85  fax +43.662.457915.6

 

From: tcf-dev-bounces@xxxxxxxxxxx [mailto:tcf-dev-bounces@xxxxxxxxxxx] On Behalf Of Pawel Piech
Sent: Thursday, March 14, 2013 5:03 PM
To: tcf-dev@xxxxxxxxxxx; CDT General developers list.
Subject: Re: [tcf-dev] _expression_ Working Sets (was: Preparing M6)

 

(adding cdt-dev) to the discussion

Unfortunately I didn't get time to get the feature reviewed by another committer before hand, so I took a risk and committed it to give it some exposure for M6.  During testing, Mike found enough usability issues that he felt warranted pulling it out.  These issues can be addressed and since there's no APIs introduced by this feature, they can still be addressed and committed for M7.

However, we also had a lengthy discussion on the CDT call about this feature and we discussed the fact that there is considerable overlap between this working sets feature and the enhanced expressions feature that Marc added to CDT in Kepler.  Incidentally, the patch I wrote for CDT to better support working sets, is actually very applicable for improving workflow when using enhanced expressions.  So Abeer is checking with his product managers to determine how to proceed with this feature.  Of course, one down-side is that enhanced expressions are implemented by the gdb integration, so they wouldn't help TCF at all.

As far as JDT, Mike said that he doesn't use expressions so finds the working sets feature just a distraction.  I don't use expressions either for java and find that the display view is more applicable.

Cheers,
Pawel


On 03/14/2013 07:39 AM, Oberhuber, Martin wrote:

Hi Pawel,

 

Thanks for the information.

 

I had seen the comments about Grouping, but was not aware that it had been backed out entirely, and I’m very surprised to hear that.

I had thought that only the Pin&Clone was taken back.

 

Hadn’t the Java devs also found the feature useful ?

Did you find any stability or usability issues, or any regression ?

 

Thanks,

Martin

--

Martin Oberhuber, SMTS / Product Architect – Development Tools, Wind River

direct +43.662.457915.85  fax +43.662.457915.6

 

From: tcf-dev-bounces@xxxxxxxxxxx [mailto:tcf-dev-bounces@xxxxxxxxxxx] On Behalf Of Pawel Piech
Sent: Thursday, March 14, 2013 3:36 PM
To: TCF Development
Cc: TCF Development (tcf-dev@xxxxxxxxxxx)
Subject: Re: [tcf-dev] Preparing M6

 

Hi Martin. 

We decided to back out _expression_ working sets after initial test pass with m6 candidate.  Abeer who contributed the initial patch is also looking at using _expression_ groups implemented in cdt as an alternative. See bug comments for details. 

Cheers

Pawel


On Mar 14, 2013, at 6:50 AM, "Oberhuber, Martin" <Martin.Oberhuber@xxxxxxxxxxxxx> wrote:

Hi all,

 

Kepler M6 is coming up (Platform will freeze tomorrow, TCF as a +1 project on Monday March/18).

 

This will be our EclipseCon milestone. I plan to be doing some demos at the conference, and some people will want to download and try TCF after that; thus robustness of the code is important at this point, for increasing TCF community and adoption.

 

So I’d appreciate if you guys could do an extra test pass – especially on top of Platform Kepler and CDT 8.2/master – to make sure TCF is robust and solid on Monday. Please do especially look at any workflows that are enabled in plain Open Source – by order of priority, most important first :

 

-          TCF Linux native debug on localhost

-          TCF Linux ptrace remote debug, with connection setup + attach through Target Explorer

-          TE Terminals

-          Gdb remote launch and attach through Target Explorer

-          TCF RSE Integration (I could see this obsolete soon since we have Target Explorer).

 

Please do especially look at the new “Working Sets for _expression_ View” feature:

    https://bugs.eclipse.org/bugs/show_bug.cgi?id=372181

This has been merged into Platform M6 (any recent I-build), and Pawel implemented a small add-on to the CDT to enable dynamic tracking of working sets with the selected stack frame. I’d like to know whether that works on TCF Debug too (hopefully yes!).

 

At any rate, please refrain from any risky commits these couple days; use your time for test and fix as you can.

 

It’s going to be important and will hopefully pay back manifold.

 

Thanks !

 

Martin

--

Martin Oberhuber, SMTS / Product Architect – Development Tools, Wind River

direct +43.662.457915.85  fax +43.662.457915.6

 

_______________________________________________
tcf-dev mailing list
tcf-dev@xxxxxxxxxxx
http://dev.eclipse.org/mailman/listinfo/tcf-dev




_______________________________________________
tcf-dev mailing list
tcf-dev@xxxxxxxxxxx
http://dev.eclipse.org/mailman/listinfo/tcf-dev

 


Back to the top