3.5.11

i18nupdate

must now properly use Binary for password (see 4180:42247d70105b). XXX Though this is somewhat backend related (eg store password using a binary type), this should be either well document or handled properly in the db helper.

TYPE_MAPPING expected to be a mapping from yams type to db type, and timestamp is the Datetime yams type

quick and dirty fix to avoid crash when running db-create command with an sqlite source

add es.po to default skel, closes #591625

closes #601987 1) sqlutils.restore_from_file have to use its confirm argument when a command fail, to propose to continue there (this can't be handled by the caller) 2) source.restore method hence needs to take this confirmation callback as argument 3) properly fix places where 'drop' was given instead of 'confirm'

missing one litle char

closes #607344 by simply appending '/' to base url if necessary instead of insulting the user

cleanup

test update. All cw tests OK

fix bad merge

make groupmap argument optional / sort for testing purpose

fix virtual relation detection: erschema is no more a relation type but a relation def in _synchronize_permissions

necessary to fix some security test, unclear why did I set perms below at some point...

import cleanup, check data file exists

should turn password into a Binary properly

should not simplify previous to preprocess

use ._cw instead of .req (reintroduced by merge of stable)

oops

backport stable

more extra_args cleanup

only give extra_args if there is something to be fed (breaks some db connectors otherwise)

merge

SqlServer: support single sign on / Windows credential authentication This is done through a new db-extra-arguments section in instance_dir/sources which is passed as a string to the extra_args named argument of lgc.db.get_connection. If this argument is Trusted_Connection, windows authentication is used instead of login/password. This change requires at least revision 5475ec3f6412 of logilab.common which will be included in logilab-common 0.46. debian/control updated to mention this new dependency.

SQL Server port: convert Password to Binary This conversion is required because pyodbc uses buffers for Binary fields and won't be able to cast a string/unicode value to the db VARBINARY column type otherwise.

[javascript] cleanup, remove tabs

[javascript] remove ajaxHtmlHead from dom response once it has been processed. This fixes #549138: treeview: folding is broken The new, systematic, pageid management (introduced by rev 37b376bb4088) made the treeview return : <ul class="treeview">...</ul> <div class="ajaxHtmlHead">...</div> which was then processed by getDomFromResponse and wrapped into a single div node : <div> <ul class="treeview">...</ul> <div class="ajaxHtmlHead">...</div> </div> In the treeview case, the node inserted into the tree was the wrapping <div> instead of the <ul>, causing the folding bug mentioned in the ticket.

msg isn't defined there, but we've to give traceback information to record error

all web tests OK

make new editcontroller works, based on a _cw_edited_fields hidden input

crecord introduced junk...

strip string by default

nicer StringField __init__

fix render (formerly form_render) prototype

process_posted yield field instead of field's name

proper field's role handling: may be 'subject' / 'object' *in case of entity's relation*, but None in other case. So you should correctly specify it to field_by_name

form_add_hidden -> add_hidden form_needs_multipart -> needs_multipart

substitute _render_attrs by values_and_attributes method, keeping bw compat

replace form_field_[encoding|vocabulary] methods on form by encoding|vocabylary(form) methods on fields

refactor vocabulary handling to avoid having to define methods on form objects, use .choices instead even for non final relations. Details: * deprecated support of <role>_<relation>_vocabulary methods on forms * new utility functions for relations vocabulary * most logic now on the RelationField class

cleanup

refactor form field value handling, to get a nicer api and an easier algorithm to get field's value Details: * new .typed_value / .display_value on fields * droped form_field_value on form * .value attribute of field instead of .initial * nicer field's __init__, allowing to give a lambda as value's value

missing imports

yams api update

drop form_field_name/form_field_id methods from form object, in favor of field.input_name(form) / field.dom_id(form)

NothingToEdit is not a ValidationError, simplify

fix translation of message to view newly created entity: must be done once language is set...

accord avec 'entité'

new catch_error context manager, nicer controller __init__ and new iter_and_commit(datakey) method

this bw compat doesn't work, drop it

don't try to write .row/.col, use new cw_row/cw_col attributes

etype_class('Any') should return bare 'Any' class

cubicweb security #36257: les urls /add/EntityName sont accessibles en anonyme

3.6 remove deprecation warning

Improve warning message

[mq]: fix_es_po_file_creation

[mq]: plural_display_name_fix

[mq]: doc

Add the search javascript to the html layout

improve error message

tweaks output

oops

clearer api: distinguish app_publish / ctrl_publish

override publisher's error handler to ease debugging

dataimport refactoring / improvments, keeping bw compat (for now)

look for cubicweb-ctl plugins in the ccplugin.py file, keeping bw compat on ecplugin.py

check entity has an eid before using absolute_url

force_session_key default to None in base form class...

fix mainvars detection bug with EXISTS (paren, actually)

call super class if choices specified to get std behaviour

merge

check name changed first

no 3.6 code in 3.5...

merge

make cubicweb-ctl db-init -d work with postgresql backend

merge

fix backout

merge

fix NameError

backout a75a6e770717, unmeetable condition and introduce a bug with boolean values

load site_cubicweb as a module instead of execing it, avoid reimport pbs

fix name error

quick and dirty fix trying to avoid rest directive conflicts when using sphinx (which seems to import the code)

merge

oops

merge

Passwords need to be considered as binary fields

TIMESTAMP column type has a special meaning for SQLServer, avoid using it for that driver (with help from logilab.common.adbh)

merge

use edited_attributes and uniformize attribute schema retreival

[mq]: fix_for_forms

fix deprecation warning

schema view fix

disable everything but metadata hooks while inserting schema

more advanced hooks control

fix 3.6 bug

[mq]: small_api_changes_for_3.6

yams api update

missing underscore leads to infinite recursion error

[mq]: fix_typo_and_forms

api update

yams api update

[i18n] use named substitutions for proper localization

remove redundant line

update sys.path at cubicweb startup

changed log severity from critical to info if a plugin cannot be loaded

fix non-xml_escaped url in box

nicer fix

avoid failure by simply not displaying this section for now"

rschema doesn't have may_have_permission

shouldn't be called from here

typos, api update

don't use setdefault with a tuple as value, a set is expected

api update

add hook selector

api update again

more api update

3.6 api update

3.6 api update

avoid infinite recursion error

test fixes

check a permission has been specified first...

api update

missing role argument

no more handled by the renderer

use implements instead of entity_implements

merge

cw 3.6 api update

some api update

deprecates entity_implements / but_etype selectors

fix initargs handling

fix guess_field to handle non internationalizable definition

perms are now held by the relation definition

password should now be named

__cmp__ should *NOT* return False when different

explain how to run as a windows service

be able to run cw instances as windows services

merge

oops, bad indent

backport stable

should not xml_escape rql

backport stable

properly escape arguments of user callbacks

new rset argument

api update

test update

fix infered rdef propagation

more api update

yams api update

move hook tests to hooks package tests

should use match_rtype, not entity_implements in such cases

3.5 api update

had been involontarly dropped

cw 3.6 api update

add a dummy and stupid dummy method for dummy form

cw 3.6 api update

card referenced later

give group mapping where needed

3.6 api update

fix sync_schema*: new syncprops argument to sync relation definition as well

Update generic RQL Based Constraints and RQLUniqueConstraint doc

fix class reference in old hook warning

merge

[mq]: typo

[mq]: fix_set_addition

Update generic RQL Based Constraints and RQLUniqueConstraint doc

fix class reference in old hook warning

missing groupmap propagation to rschema2rql

Update generic RQL Based Constraints and RQLUniqueConstraint doc

test api update

remove debug print, fix error message

missing mainvars

missing cache key

duh???

add regid when tracing selection

duh???

more api update

drop common subpackage

merge

move i18n / migration modules from cw.common to cw

merge

api update

remove old bw (pre 3.2) compat modules

use commit in migration script instead of checkpoint

deprecated one more custom add_entity in favor of req.create_entity

some cw 3.6 api update

yams api update

.id backward compat

backport stable branch

[migration] write migration instructions for permissions handling on relation definition

[server] don't try to set perms twice on the same relation (e.g. symetric relations)

merge

fix class reference in old hook warning

more api renaming/fixing update

return relation property using rdef.get

should return item type not relation

api renaming update

fix bad merge

should delete state once replacement has been done

dont crash if directory doesn't exist

fix jsonization of interval

added replace_state method to ease workflow migration. backported from forge as a reminder, not really tested yet

backport stable into default

Added tag cubicweb-debian-version-3.5.10-1 for changeset dfe2f245248c

Added tag cubicweb-version-3.5.10 for changeset 9b52725d8c53

merge

3.5.10

quick & dirty fix for auto completion widget

update i18n doc

Pasting blog about facet in the facet section of the doc book.

hide en apidoc in .hgignore

héhé, get the same result with a slightly enhanced condition and all this code removed

unused, cleanup

fix sql generation bug with neged inlined relation where the object is invariant and subject is only referenced by the relation

add postcommit_event on Operations

add some notes

i18nupdate

nicer error message on rql constraints

implements 'parent' method on workflow entity types

this invalid constraint now fails with new constraints checking

this constraint is actually unusable in yams schema for cw

unused since a long time now

finish yesterday work on rql constraints: * Fix inheritance pb: there are some places where we filter constraints according to the class hierarchy, so rql unique constraint should'nt be neither a RQLContraint nor a RQLVocabularyConstraint subclass. Added test in unittest_schema reflecting this. * So now we have to get explicitly RQLUniqueConstraint where desired (eg in server/hooks.py) * Update i18ncubicweb command to include constraint types in generated pot file (this should have been there for a while...) * Update unittest_schemaserial which has been broken when serializing format for rql constraints has changed

synchronize rdef where unique constraints have been updated/added

oops, should have been updated

add a fourth item to 'view box' defintion, dispctrl, so that we can later globally sort all boxes instead of having view boxes before component boxes. 'view' boxes order is configured through uicfg.primaryview_display_ctrl, 'component' boxes order through the cwproperty system.

slightly adapt title and hide boxes when view is called as a non primary view

allow to override is_primary result using an argument given to object constructor (have to be given to 'initargs' argument of .render like initargs={'is_primary': False}

typo

new generic urlattr view

update _render_attribute / _render_relation prototype to gain more control on label through uicfg (for attributes and relations). Keep bw compat for crm/iliane

when fields are displayed in a table and label should'nt be displayed, set colspan to 2 on the field value cell

fix format string

cleanup

fix workflow consistency: check state/transition name uniqness in the wf when state/transition name is modified ; use the new msg argument of rql constraints to have a nicer error message on failure

nicer __str__ and __repr__ methods on rql constraints

refactor so that we can specify an extra argument to constraints which will be used as error message

nicer mainvars/expression handling when initializing rql constraints

merge

fix RQLUniqueConstraint behaviour by using a DISTINCT query and allowing to specify variables that should be used in selection. See usage on state_of / transition_of relations.

don't fail on error while render a notification mail

improve error message for redundant initial workflow states

fix name error while debugging

oops, this is an object relation

don't use html/dtd validator as default validator for non html views. Closes #550162

support x.set_relations(a_relation=None) to delete a_relation relations from this entity

properly fail if user has entered wrong connection information for the system database, closes #549136

[forms] deprecate form_render(**kwargs) in favor of render(formvalues=None, rendervalues=None, renderer=None) to fix a pitfall in form_render prototype: there is no way to distinguish between arguments that should be given to build_context from arguments that should be given to the renderer. This was problematic for instance with inlined form where a 'title' argument is given for the renderer, but is also used as default value for field whose name is 'title'...

wfhistory isn't anymore selectable if the entity has no record in its history

fix name error if ValueError is raised

merge

fix rset.column_types docstring to reflect reality

make wfhistory non selectable for entities without record in its workflow history

when one is adding an inline entity for a relation of a single card, the 'add a new xxx' link disappears. If the user then cancel the addition, we have to make this link appears back. This is done by giving add new link id to removeInlineForm.

Added tag cubicweb-debian-version-3.5.9-1 for changeset d7f2d32340fb