Noe Gaumont <ngaumont@logilab.fr> [Fri, 11 Oct 2019 12:41:16 +0200] rev 12732
Remove undefined function in documentation
Philippe Pepiot <philippe.pepiot@logilab.fr> [Wed, 09 Oct 2019 11:44:22 +0200] rev 12731
Merge 3.26
Laurent Peuch <cortex@worlddomination.be> [Wed, 22 May 2019 01:16:55 +0200] rev 12730
[pyramid/ctl] pyramid command will always run in foreground from now on
Daemonization used to make sens in the past, but today "pyramid" command is
only used for dev and should never be used for deployement (you should use a
wsgi server for that instead) so daemonization doesn't fit any use case
anymore.
Closes #17232923
Laurent Peuch <cortex@worlddomination.be> [Thu, 22 Aug 2019 04:07:08 +0200] rev 12729
[clean] remove pyramid warning about looping task
This warning isn't useful anymore.
Laurent Peuch <cortex@worlddomination.be> [Thu, 20 Jun 2019 16:18:19 +0200] rev 12728
[doc] add instruction to re-enable ckeditor
Laurent Peuch <cortex@worlddomination.be> [Wed, 11 Sep 2019 18:31:08 +0200] rev 12727
[fix] reference to underclared and unused variable kwargs
Philippe Pepiot <philippe.pepiot@logilab.fr> [Wed, 11 Sep 2019 11:56:43 +0200] rev 12726
Avoid to update inlined relation column to NULL when deleting entities
When deleting entities, cubicweb run a rql DELETE on all relations to trigger hooks.
For an inlined relation this also mean set the column to NULL. This operation
may fail if there's additional constraints on the column.
Also this is a weird and useless behavior since deleting the entity row will by
definition delete the relation.
We still doesn't handle the case where both subject and object are going to be
deleted because rows need to be deleted in a particular order that cubicweb
doesn't handle.
Add a test checking UPDATE does not occur but hooks are correctly called.
Closes #17236690
Philippe Pepiot <philippe.pepiot@logilab.fr> [Wed, 11 Sep 2019 11:37:05 +0200] rev 12725
Drop useless ConstraintInsertionTC
This test does nothing and is here from the begining of repo history.
Now it raise an error "TypeError: Skipped expected string as 'msg' parameter, got 'SkipTest' instead. Perhaps you meant to use a mark?"
Just drop it.
Nicola Spanti <nicola.spanti@logilab.fr> [Tue, 27 Aug 2019 14:26:21 +0200] rev 12724
Backed out changeset 741dc2590f90
It was for cube s3storage, but this piece of information was
apparently not in the email (used for review), so this patch was
published in the wrong repository. :-/
Arthur Lutz <arthur.lutz@logilab.fr> [Fri, 23 Aug 2019 11:16:46 +0200] rev 12723
[.gitlab-ci] initial version (tox targets)
Denis Laxalde <denis.laxalde@logilab.fr> [Tue, 06 Aug 2019 16:43:46 +0200] rev 12722
Merge with 3.26
Denis Laxalde <denis.laxalde@logilab.fr> [Tue, 06 Aug 2019 16:36:21 +0200] rev 12721
Added tag debian/3.26.14-1, 3.26.14 for changeset 172f683a84f6
Denis Laxalde <denis.laxalde@logilab.fr> [Tue, 06 Aug 2019 15:43:59 +0200] rev 12720
[pkg] version 3.26.14
Denis Laxalde <denis.laxalde@logilab.fr> [Tue, 06 Aug 2019 14:26:17 +0200] rev 12719
[py3] Pass bytes as "msg" to smtplib.SMTP.sendmail()
When passing a unicode string to smtplib.SMTP.sendmail() as "msg"
argument, there is an implicit bytes encoding using "ascii" encoding in
python3. Of course this does not work if the string contains non-ASCII
characters. In fact, config's sendmails method intent to pass bytes to
smtplib.SMTP.sendmail() as it uses msg.as_string() method.
Unfortunately, in python3, this method returns a unicode string whereas
it returns a bytes string in python2; we thus fix this by calling
as_bytes() method on python3.
As there is no "as_bytes" method in python2, we need to handle python2
compatibility by hand and either call as_string() or as_bytes().
In testlib, where we mock smtplib.SMTP, we need to keep the "msg"
argument of Email class (defined in testlib as well) a unicode string.
Otherwise, it fails to be parsed by email.message_from_string() (from
stdlib) if it is bytes on python3.