GitLab

[entity, metadata] huuum, use resolvable url as cwuri... And fix existing ones.

Closes #3390388

Get rid of extra space before `!' and `?'

Happens if a previous migration forgot a drop_attribute.
Closes #3471609

Added tag cubicweb-version-3.18.2, cubicweb-debian-version-3.18.2-1, cubicweb-centos-version-3.18.2-1 for changeset 6880674c1a26

Added tag cubicweb-version-3.17.12, cubicweb-debian-version-3.17.12-1, cubicweb-centos-version-3.17.12-1 for changeset 838d58a30f7e

--HG--
branch : stable

And fix a syntax error and wrong package name in the process.

--HG--
branch : stable

--HG--
branch : stable

--HG--
branch : stable

``Accept-Language`` should be overridable just as any other request header.
Default value is "nothing" since the application should be able the empty
header case.

--HG--
branch : stable

Also give proper name to subject (role, not target),
and kill unused variable.

--HG--
branch : stable

Closes jpl##49466

--HG--
branch : stable

--HG--
branch : stable

--HG--
branch : stable

--HG--
branch : stable

[web error] exception may not have a 'status' attribute, generating an AttributeError that hides the original error. Closes #3381670

--HG--
branch : stable

--HG--
branch : stable

The underlying design issue is not addressed here, only a cosmetic and
"we understand and document what's happening" note.

Related to #3400448.

--HG--
branch : stable

--HG--
branch : stable

assertEqual(msg, msg) had a hard time failing.

--HG--
branch : stable

This avoids a deprecation warning when following the documentation.

--HG--
branch : stable

If we're going to ignore errors, at least be loud about it.

sqlexec already wraps the sql execution in a try/except block, so we
were never going to catch any exceptions in the caller.  Check the
return value instead.  Closes #3459618.

Added tag cubicweb-version-3.18.1, cubicweb-debian-version-3.18.1-1, cubicweb-centos-version-3.18.1-1 for changeset 60322cb8636c

Fixes compat with jquery 1.10 (closes #3445689)

It tries to access a 'state' attribute on an entity without going
through the IWorkflowable adapter.

If the rdef is not final, yams will complain.

The addition of this attribute in 3.13.8 had default=False, which made
no sense, and was fixed in 3.13.9.  However no migration was done so
apps that ever saw the 3.13.8 code could still have '' as defaultval for
this attribute.

Make it clear that it is only an optimization.
Related to #3444095.

Fixes semantic change introduced in 96dba2efd16d where internal sessions
started to check attribute permissions.

Closes #3444095