Re: [Pharo-dev] [Pharo-bugtracker] FogBugz (Case [Issue]2486) File System - Linux VM: Umlaute äöà etc in in file/directory names: UTF8 unicode problem?
This had been verified 6 times, is that normal ? On 08 Nov 2013, at 18:25, Pharo Issue Tracker <do-not-reply@pharo.fogbugz.com> wrote:
A FogBugz case was edited by Ulysse The Galactic Monkey From Outer Space.
Case ID: 2486 Title: Linux VM: Umlaute äöà etc in in file/directory names: UTF8 unicode problem? Status: Resolved (Fix Reviewed by the Monkey) Category: Bug Project: File System Area: 1. Pharo Image Priority: 5 - Fix If Time Milestone: Pharo3.0: 30/03/2014 Assigned To: Penelope, Ulysse's wife
URL: https://pharo.fogbugz.com/f/cases/2486
Changes: Added tag 'Validated in 30559'.
You are subscribed to this case. If you do not want to receive automatic notifications in the future, unsubscribe (https://pharo.fogbugz.com/default.asp?pre=preUnsubscribe&pg=pgEditBug&comman...) from this case.
_______________________________________________ Pharo-bugtracker mailing list Pharo-bugtracker@lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker
yes, you have re-validate for each new image version, since things might have changed in between ;) Note the tags on the left side: Validated in 30547, Validated in 30555, Validated in 30556, Validated in 30557, Validated in 30558, Validated in 30559 On 2013-11-08, at 18:26, Sven Van Caekenberghe <sven@stfx.eu> wrote:
This had been verified 6 times, is that normal ?
On 08 Nov 2013, at 18:25, Pharo Issue Tracker <do-not-reply@pharo.fogbugz.com> wrote:
A FogBugz case was edited by Ulysse The Galactic Monkey From Outer Space.
Case ID: 2486 Title: Linux VM: Umlaute äöà etc in in file/directory names: UTF8 unicode problem? Status: Resolved (Fix Reviewed by the Monkey) Category: Bug Project: File System Area: 1. Pharo Image Priority: 5 - Fix If Time Milestone: Pharo3.0: 30/03/2014 Assigned To: Penelope, Ulysse's wife
URL: https://pharo.fogbugz.com/f/cases/2486
Changes: Added tag 'Validated in 30559'.
You are subscribed to this case. If you do not want to receive automatic notifications in the future, unsubscribe (https://pharo.fogbugz.com/default.asp?pre=preUnsubscribe&pg=pgEditBug&comman...) from this case.
_______________________________________________ Pharo-bugtracker mailing list Pharo-bugtracker@lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker
On 08 Nov 2013, at 18:53, Camillo Bruni <camillobruni@gmail.com> wrote:
yes, you have re-validate for each new image version, since things might have changed in between ;)
Note the tags on the left side: Validated in 30547, Validated in 30555, Validated in 30556, Validated in 30557, Validated in 30558, Validated in 30559
I understand, but they should be integrated, right ? Instead of making noise in my mailbox ;-)
On 2013-11-08, at 18:26, Sven Van Caekenberghe <sven@stfx.eu> wrote:
This had been verified 6 times, is that normal ?
On 08 Nov 2013, at 18:25, Pharo Issue Tracker <do-not-reply@pharo.fogbugz.com> wrote:
A FogBugz case was edited by Ulysse The Galactic Monkey From Outer Space.
Case ID: 2486 Title: Linux VM: Umlaute äöà etc in in file/directory names: UTF8 unicode problem? Status: Resolved (Fix Reviewed by the Monkey) Category: Bug Project: File System Area: 1. Pharo Image Priority: 5 - Fix If Time Milestone: Pharo3.0: 30/03/2014 Assigned To: Penelope, Ulysse's wife
URL: https://pharo.fogbugz.com/f/cases/2486
Changes: Added tag 'Validated in 30559'.
You are subscribed to this case. If you do not want to receive automatic notifications in the future, unsubscribe (https://pharo.fogbugz.com/default.asp?pre=preUnsubscribe&pg=pgEditBug&comman...) from this case.
_______________________________________________ Pharo-bugtracker mailing list Pharo-bugtracker@lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker
On 08 Nov 2013, at 19:16, Sven Van Caekenberghe <sven@stfx.eu> wrote:
On 08 Nov 2013, at 18:53, Camillo Bruni <camillobruni@gmail.com> wrote:
yes, you have re-validate for each new image version, since things might have changed in between ;)
Note the tags on the left side: Validated in 30547, Validated in 30555, Validated in 30556, Validated in 30557, Validated in 30558, Validated in 30559
I understand, but they should be integrated, right ? Instead of making noise in my mailbox ;-)
maybe this could be seen as a hint to have a look at the change and write a short not âYes, this is goodâ. Marcus
On 08 Nov 2013, at 19:39, Marcus Denker <marcus.denker@inria.fr> wrote:
On 08 Nov 2013, at 19:16, Sven Van Caekenberghe <sven@stfx.eu> wrote:
On 08 Nov 2013, at 18:53, Camillo Bruni <camillobruni@gmail.com> wrote:
yes, you have re-validate for each new image version, since things might have changed in between ;)
Note the tags on the left side: Validated in 30547, Validated in 30555, Validated in 30556, Validated in 30557, Validated in 30558, Validated in 30559
I understand, but they should be integrated, right ? Instead of making noise in my mailbox ;-)
maybe this could be seen as a hint to have a look at the change and write a short not âYes, this is goodâ.
Marcus
But so many issues are integrated without a second vote of approval, so why is it needed here ?
On 09 Nov 2013, at 10:35, Sven Van Caekenberghe <sven@stfx.eu> wrote:
On 08 Nov 2013, at 19:39, Marcus Denker <marcus.denker@inria.fr> wrote:
On 08 Nov 2013, at 19:16, Sven Van Caekenberghe <sven@stfx.eu> wrote:
On 08 Nov 2013, at 18:53, Camillo Bruni <camillobruni@gmail.com> wrote:
yes, you have re-validate for each new image version, since things might have changed in between ;)
Note the tags on the left side: Validated in 30547, Validated in 30555, Validated in 30556, Validated in 30557, Validated in 30558, Validated in 30559
I understand, but they should be integrated, right ? Instead of making noise in my mailbox ;-)
maybe this could be seen as a hint to have a look at the change and write a short not âYes, this is goodâ.
Marcus
But so many issues are integrated without a second vote of approval, so why is it needed here ?
because I just can not review everything. I just canât. Marcus
yes, you have re-validate for each new image version, since things might have changed in between ;)
Note the tags on the left side: Validated in 30547, Validated in 30555, Validated in 30556, Validated in 30557, Validated in 30558, Validated in 30559
I understand, but they should be integrated, right ? Instead of making noise in my mailbox ;-)
maybe this could be seen as a hint to have a look at the change and write a short not âYes, this is goodâ.
Marcus
But so many issues are integrated without a second vote of approval, so why is it needed here ?
When we are doing a sprint on certain topics it is important to get issues integrated fast because if they break something unexpected the guy that proposed the fix can work on it. Often sprinters are just occasionally working on Pharo so this is not like we can tell them: ok wait one week and be prepared that next week you will have to prepare to allocate some time. Now running the validation several is good. with a dash boards we could spot these guys and integrate.
That would seem to be a lot of additional work for the monkey, which might not be a problem if the set of Cases is small or CPU resources are high.
Perhaps the process could be amended so the monkey operates in two stages. 1. In taking the status from "Fix review needed" to "Fix Reviewed by the Monkey" - the monkey runs once and does not run again. 2. When a human comes along and changes the status to "Fix to include", the monkey runs again, and also for each image update until integrated. The status would not change from "Fix to include".
Optionally Step 2 could also... 2a. In consideration of how several fixes are usually integrated in one image step, the monkey could consecutively load several Fixes on top of the previous image step, to check for incompatibilities between Fixes, and advise the human integrator of a known good order of integration; OR... 2b. A successful monkey test of "Fix to Include" automatically kicks off a CI run on that Fix. So that you end up with one Fix per image step. Probably this doubles up on the CI test run, but should ensure the CI never goes red; OR... 2c. A combination of 2a & 2b. The monkey loads several "Fix to Include" Fixes until there is a problem, and copies all the good ones to kick off a CI run.
So I know that would require some effort, but just food for thought. I'd be interested in helping do this.
The noise is some side-effect of the current somewhat limited situation. In the ideal case the monkey would just attach a new tag for each new image version if the validation report did not change. But for that you need to have your own complete service which keeps track of the previous validations. As you can imagine this will take some time and effort ;) For now I think we stick with the current approach which is more noisy but safer. Everybody is of course invited to improve the situation: http://smalltalkhub.com/#!/~Pharo/ci https://ci.inria.fr/pharo-contribution/job/CI/ http://smalltalkhub.com/#!/~PharoExtras/ImageWorker https://ci.inria.fr/pharo-contribution/job/ImageWorker/
participants (5)
-
btc＠openinworld.com -
Camillo Bruni -
Marcus Denker -
Stéphane Ducasse -
Sven Van Caekenberghe