Hello, just trying to do a sign-off following http://wiki.koha-community.org/wiki/Sign_off_on_patches It's ok that some test of prove t/ prove t/db_dependent prove xt/author fail even in a clean master? Regards, Bernardo -- Bernardo Gonzalez Kriegel bgkriegel@gmail.com
Bernardo, It's ok that some test of
prove t/ prove t/db_dependent prove xt/author
fail even in a clean master?
No unit tests should be failing on master. However, since they're passing on both jenkins and my server it is probably just a data issue, which means you can probably sign off if no additional tests fail after applying the patch, and hop on #koha when you have a chance so we can figure out why some of the tests are failing for you and fix it. Regards, Jared -- Jared Camins-Esakov Bibliographer, C & P Bibliography Services, LLC (phone) +1 (917) 727-3445 (e-mail) jcamins@cpbibliography.com (web) http://www.cpbibliography.com/
Hello, just trying to do a sign-off following http://wiki.koha-community.org/wiki/Sign_off_on_patches
It's ok that some test of
prove t/ prove t/db_dependent
Le 11/01/2013 18:08, Bernardo Gonzalez Kriegel a écrit : those one probably fail because you don't have the expected datas.
prove xt/author fail even in a clean master?
OTOH: signing-off is checking that the feature works fine. It's the QA team that will be in charge to check unit tests. (Of course, if a patch contains a unit test itself, you should not signoff until you've successfully run it ;-) ) HTH -- Paul POULAIN - BibLibre http://www.biblibre.com Free & Open Source Softwares for libraries Koha, Drupal, Piwik, Jasper
Paul, OTOH: signing-off is checking that the feature works fine. It's the QA
team that will be in charge to check unit tests. (Of course, if a patch contains a unit test itself, you should not signoff until you've successfully run it ;-) )
Signing off is both checking that a feature works and -- hopefully -- considering whether there are any bigger problems with it, such as regressions, to the extent that the developer is able to do this. Unit tests provide an easy way to test for regressions, which is why it's a good idea to run them -- if not all of them, at least the relevant ones -- while signing off. Obviously as RM I do not have the power to mandate that everyone must run unit tests before signing off, nor would I want to, but I will say that the more thorough the sign off, the more likely it is that I will spend time on a patch. Patches with only a perfunctory sign off are more likely to trigger my "one strike" rule (that's the rule that says the first problem I encounter with a patch, no matter how minor, can get the patch failed). The QA team should also be running unit tests, of course, but the point of the sign off/QA divide is that we want more eyes reviewing patches before they go in. If the developer who signs off is running unit tests, I can feel pretty confident that they didn't just add a "Signed-off-by" line before uploading a patch. Regards, Jared -- Jared Camins-Esakov Bibliographer, C & P Bibliography Services, LLC (phone) +1 (917) 727-3445 (e-mail) jcamins@cpbibliography.com (web) http://www.cpbibliography.com/
Paul and Jared, It seems that my 'clean' master was not that clean. I do a new clone, and the errors I found on master can be attributed to some missing sample data and a problem I have with Test::Pod::Spelling So now I know what to expect. Anyway, I was just trying to obey the rules. Thank you both, Bernardo -- Bernardo Gonzalez Kriegel bgkriegel@gmail.com On Fri, Jan 11, 2013 at 2:37 PM, Jared Camins-Esakov <jcamins@cpbibliography.com> wrote:
Paul,
OTOH: signing-off is checking that the feature works fine. It's the QA team that will be in charge to check unit tests. (Of course, if a patch contains a unit test itself, you should not signoff until you've successfully run it ;-) )
Signing off is both checking that a feature works and -- hopefully -- considering whether there are any bigger problems with it, such as regressions, to the extent that the developer is able to do this. Unit tests provide an easy way to test for regressions, which is why it's a good idea to run them -- if not all of them, at least the relevant ones -- while signing off. Obviously as RM I do not have the power to mandate that everyone must run unit tests before signing off, nor would I want to, but I will say that the more thorough the sign off, the more likely it is that I will spend time on a patch. Patches with only a perfunctory sign off are more likely to trigger my "one strike" rule (that's the rule that says the first problem I encounter with a patch, no matter how minor, can get the patch failed).
The QA team should also be running unit tests, of course, but the point of the sign off/QA divide is that we want more eyes reviewing patches before they go in. If the developer who signs off is running unit tests, I can feel pretty confident that they didn't just add a "Signed-off-by" line before uploading a patch.
Regards, Jared
-- Jared Camins-Esakov Bibliographer, C & P Bibliography Services, LLC (phone) +1 (917) 727-3445 (e-mail) jcamins@cpbibliography.com (web) http://www.cpbibliography.com/
_______________________________________________ Koha-devel mailing list Koha-devel@lists.koha-community.org http://lists.koha-community.org/cgi-bin/mailman/listinfo/koha-devel website : http://www.koha-community.org/ git : http://git.koha-community.org/ bugs : http://bugs.koha-community.org/
Hi, On Fri, Jan 11, 2013 at 9:37 AM, Jared Camins-Esakov < jcamins@cpbibliography.com> wrote:
Signing off is both checking that a feature works and -- hopefully -- considering whether there are any bigger problems with it, such as regressions, to the extent that the developer is able to do this. Unit tests provide an easy way to test for regressions, which is why it's a good idea to run them -- if not all of them, at least the relevant ones -- while signing off. Obviously as RM I do not have the power to mandate that everyone must run unit tests before signing off, nor would I want to, but I will say that the more thorough the sign off, the more likely it is that I will spend time on a patch. Patches with only a perfunctory sign off are more likely to trigger my "one strike" rule (that's the rule that says the first problem I encounter with a patch, no matter how minor, can get the patch failed).
The QA team should also be running unit tests, of course, but the point of the sign off/QA divide is that we want more eyes reviewing patches before they go in. If the developer who signs off is running unit tests, I can feel pretty confident that they didn't just add a "Signed-off-by" line before uploading a patch.
I cannot give this enough +1s. QA *must* be considered the responsibility of everybody who regularly submits patches to Koha, not just the QA team. I consider making sure that the tests pass if you submit a patch or if you sign off on one to be the *bare minimum* of what we should expect of ourselves. Regards, Galen -- Galen Charlton Manager of Implementation Equinox Software, Inc. / The Open Source Experts email: gmc@esilibrary.com direct: +1 770-709-5581 cell: +1 404-984-4366 skype: gmcharlt web: http://www.esilibrary.com/ Supporting Koha and Evergreen: http://koha-community.org & http://evergreen-ils.org
Many +1s here as well. Kind Regards, Chris On Fri, Jan 11, 2013 at 1:36 PM, Galen Charlton <gmc@esilibrary.com> wrote:
Hi,
On Fri, Jan 11, 2013 at 9:37 AM, Jared Camins-Esakov < jcamins@cpbibliography.com> wrote:
Signing off is both checking that a feature works and -- hopefully -- considering whether there are any bigger problems with it, such as regressions, to the extent that the developer is able to do this. Unit tests provide an easy way to test for regressions, which is why it's a good idea to run them -- if not all of them, at least the relevant ones -- while signing off. Obviously as RM I do not have the power to mandate that everyone must run unit tests before signing off, nor would I want to, but I will say that the more thorough the sign off, the more likely it is that I will spend time on a patch. Patches with only a perfunctory sign off are more likely to trigger my "one strike" rule (that's the rule that says the first problem I encounter with a patch, no matter how minor, can get the patch failed).
The QA team should also be running unit tests, of course, but the point of the sign off/QA divide is that we want more eyes reviewing patches before they go in. If the developer who signs off is running unit tests, I can feel pretty confident that they didn't just add a "Signed-off-by" line before uploading a patch.
I cannot give this enough +1s. QA *must* be considered the responsibility of everybody who regularly submits patches to Koha, not just the QA team. I consider making sure that the tests pass if you submit a patch or if you sign off on one to be the *bare minimum* of what we should expect of ourselves.
Regards,
Galen -- Galen Charlton Manager of Implementation Equinox Software, Inc. / The Open Source Experts email: gmc@esilibrary.com direct: +1 770-709-5581 cell: +1 404-984-4366 skype: gmcharlt web: http://www.esilibrary.com/ Supporting Koha and Evergreen: http://koha-community.org & http://evergreen-ils.org
_______________________________________________ Koha-devel mailing list Koha-devel@lists.koha-community.org http://lists.koha-community.org/cgi-bin/mailman/listinfo/koha-devel website : http://www.koha-community.org/ git : http://git.koha-community.org/ bugs : http://bugs.koha-community.org/
+1 :) -----Original Message----- From: koha-devel-bounces@lists.koha-community.org on behalf of Chris Nighswonger Sent: Fri 11.01.2013 19:45 To: Galen Charlton Cc: koha-devel@lists.koha-community.org Subject: Re: [Koha-devel] Failed unit test Many +1s here as well. Kind Regards, Chris On Fri, Jan 11, 2013 at 1:36 PM, Galen Charlton <gmc@esilibrary.com> wrote:
Hi,
On Fri, Jan 11, 2013 at 9:37 AM, Jared Camins-Esakov < jcamins@cpbibliography.com> wrote:
Signing off is both checking that a feature works and -- hopefully -- considering whether there are any bigger problems with it, such as regressions, to the extent that the developer is able to do this. Unit tests provide an easy way to test for regressions, which is why it's a good idea to run them -- if not all of them, at least the relevant ones -- while signing off. Obviously as RM I do not have the power to mandate that everyone must run unit tests before signing off, nor would I want to, but I will say that the more thorough the sign off, the more likely it is that I will spend time on a patch. Patches with only a perfunctory sign off are more likely to trigger my "one strike" rule (that's the rule that says the first problem I encounter with a patch, no matter how minor, can get the patch failed).
The QA team should also be running unit tests, of course, but the point of the sign off/QA divide is that we want more eyes reviewing patches before they go in. If the developer who signs off is running unit tests, I can feel pretty confident that they didn't just add a "Signed-off-by" line before uploading a patch.
I cannot give this enough +1s. QA *must* be considered the responsibility of everybody who regularly submits patches to Koha, not just the QA team. I consider making sure that the tests pass if you submit a patch or if you sign off on one to be the *bare minimum* of what we should expect of ourselves.
Regards,
Galen -- Galen Charlton Manager of Implementation Equinox Software, Inc. / The Open Source Experts email: gmc@esilibrary.com direct: +1 770-709-5581 cell: +1 404-984-4366 skype: gmcharlt web: http://www.esilibrary.com/ Supporting Koha and Evergreen: http://koha-community.org & http://evergreen-ils.org
_______________________________________________ Koha-devel mailing list Koha-devel@lists.koha-community.org http://lists.koha-community.org/cgi-bin/mailman/listinfo/koha-devel website : http://www.koha-community.org/ git : http://git.koha-community.org/ bugs : http://bugs.koha-community.org/
participants (6)
-
Bernardo Gonzalez Kriegel -
Chris Nighswonger -
Fischer, Katrin -
Galen Charlton -
Jared Camins-Esakov -
Paul Poulain