[debug-toolbar/display_source_code] add a pyramid view that display syntax highlighted python source file This will be used for tool building for easier debugging. Closes #17256791

[mod] allow to pass generic additional arguments to pygments HtmlFormatter

[debug/fix] ensure that not syntax highlighted code is a string Otherwise it could be in another unrendered form that makes marko complains. Closes #17256791

[debug-toolbar] add registry decisions debug panel Closes #17219866

Increase needed version of logilab-common to >= 1.5.0

[debug-toolbar] add registry panel Closes #17219866

[debug-toolbar] add sql panel Closes #17219873

[debug-toolbar] add cw general panel with controller Closes #17219897

[debug-toolbar] make linter happy

[pyramid/ctl/debugtoolbar] automatically include custom panels on -t

[debug-toolbar/rql] display sql queries generated by rql ones

[debug-toolbar/rql] add RQL panel Closes #17219673

[debug/emit/rql] add RQL debug channel and emit queries This is used for the RQL debug panel but can be used as a generic mechanism for several tools.

[debug/source_highlight] rename highlight to highlight_terminal

[debug/source_highlight] add highlight_html and generate_css for debugtool panels

[debug/rql] bind a uniq token per rql to trace its decomposition This is aimed to be used by the RQL debug panel (and the SQL one later on) but can also be used by other debugging tools.

[debug] add a new channel events mechanism for debugging

[ctl] allow to specific instance name for instance commands in $CW_INSTANCE Closes #17253994

[cubicweb/utils] add QueryCache.__contains__() to improve dict-like behavior. Closes #255426

[migrations/pdb] add to every failing migration operation a "p(db)" option Instead, the migration command will just crash without offering the possibility of the user to debug or continue the migration. Closes #17219772

[pyramid/ctl] add a new option to activate the debugtoolbar (-t) While this toolbar isn't yet very integrated with CW, it integrates an inline debugging shell like werkzeug for flask or django_extensions which greatly helps debugging on exception. It can already be manually activated by writting "pyramid.includes = pyramid_debugtoolbar" but it's hidden somewhere in the documentation and not very accessible and annoying to do. Closes #17219765

[changelog] add missing information to the 3.27 changelog

Merge 3.26

[debug] add ALL possibility to --dbglevel

[ux] display instance name in suggested upgrade command

[migration/pdb] display traceback instead of only the exception for easier debugging Closes #17219820

[migration/pdb] add option to use pdb.post_mortem if traceback is provided Post mortem is a mode where the pdb shell is opened **where** the exception as occured instead at the breakpoint for set_trace. This is way more useful for debugging for the user because is will have the full context of the error. Closes #17219827

[migration/shell] select ipdb if present on (d)ebug mode Closes #17219833

[mod] move get_pdb to utils.py

Add a migration function in order to relocate bfss Related to https://www.cubicweb.org/ticket/1903304

[pyramid/debug] on every request display request path and selected controller This will display lines like: 2019-06-12 16:26:57 - (cubicweb.pyramid.bwcompat) INFO: REQUEST [view] '/wiki/334037' selected controller <cubicweb.web.views.basecontrollers.ViewController object at 0x7f133802ae90> at /root/.virtualenvs/abreton/local/lib/python2.7/site-packages/cubicweb/web/views/basecontrollers.py:120 2019-06-12 16:26:59 - (cubicweb.pyramid.bwcompat) INFO: REQUEST [data] '/data/ea5eb3b6f89fba0ff4b671093c88d3dd/images/cartouche-mur.jpg' selected controller <cubicweb.web.views.staticcontrollers.DataController object at 0x7f1332a84d50> at /root/.virtualenvs/abreton/local/lib/python2.7/site-packages/cubicweb/web/views/staticcontrollers.py:179 The make debugging easier as this isn't done already. Most other web frameworks do something similar for the developper to be able to understand what is happening right now. Closes #17219741

[logging/debug] move transaction operations logs from self.debug to DBG_OPS There were a mixing between displaying transaction operations log on logger.debug and only when DBG_OPS was on, uniformize everything behind DBG_OPS. Closes #17229126

[debug] syntax highlight SQL and RQL debug output This prevent DBG_RQL and DBG_SQL output to looks like a gigantic blob of white text. Do this only if we detect that we are in a tty and that pygments is installed. Closes #17222885

[logging] send logs by default to stdout Following the move to have commands only working on foreground sending logs to a file by default doesn't make much more sens anymore so send them to stdout by default and add a new option to send them to a file if needed

[cubicweb-ctl] backport --dbglevel option of pyramid to all instance commands Closes #17219653

[cubicweb-ctl] backport --loglevel option of pyramid to all instance commands

[cubicweb-ctl/fix] correctly get exception traceback_ for pdb.post_mortem In python 3 the behavior of sys.exc_info had a very subtle change: - in python 2 you can call if whenever you want after a try/except statement and you'll get information about this last raise - ipython 3, once you get out of try/except, sys.exc_info is cleaned and you'll get (None, None, None) Hardened the test to avoid this error from happening again.

Document mercurial configuration for contributing * how to use master/review repositories * how to send the "Available at" command in patchbomb emails * how to add the "cubicweb" flag in patchbomb emails * how to use our "jenkins" mercurial extension Closes #17243538

Add documentation on workflow permission

Remove undefined function in documentation

Merge 3.26

[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

[clean] remove pyramid warning about looping task This warning isn't useful anymore.

[doc] add instruction to re-enable ckeditor

[fix] reference to underclared and unused variable kwargs

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

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.

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. :-/