remove remaining junk from bad merge

fix bad merge

[sync schema hooks] pylint fixes have broken hasattr test

[pylint] fix pylint detected errors and tweak it so that pylint -E will be much less verbose next time (+ update some copyrights on the way)

[doc] more doc about what's new in 3.14

[json] restore 2.5 compat

[views] implement json / jsonp export views (closes #1942658) Json export views are based on the same model as CSV export views. There are two distinct views : - *jsonexport* : direct conversion of the result set into json - *ejsonexport* : convert entities into json The JSONP parameter is named ``callback`` (same as on geonames, dbepdia and a lot of sites) An optional `_indent` request parameter can be passed to pretty print the results.

[form, entity] refactor '__linkto', now handled by the entity form, not the entity itself. Closes #1931543

[facet] add safety belt to rql path facet init (closes #1965481)

[cleanup] kill broken livetest module

[req cookie] fix name error

backport stable

[test] unicode name/value in cookie makes tests failing (SimpleCookie then suppose it's a dict...)

fix bad merge (unsaved buffers...)

backport stable

fix merge

Added tag cubicweb-debian-version-3.13.6-1 for changeset 8a8949ca5351

Added tag cubicweb-version-3.13.6 for changeset 17c007ad845a

backport oldstable

Added tag cubicweb-debian-version-3.12.10-1 for changeset 9dfd21fa0a8b

Added tag cubicweb-version-3.12.10 for changeset 074c848a3712

3.13.6

[web/facet] fix example selector in docstring

[migration doc] talk about changes in Entity.fetch_rql

[web request] fix stupid error that gone through the review

3.12.10

[skeleton] fix new class configuration in realdbtest

[web request] fix cookie 'expires' formating (closes #1953945) This was because cookie.expires wasn't processed in cw.etwist.http, though this code should had never existed, instead proper twisted method should be called. Also, move on the way to headers handling simplification and rewrite cw request.set_cookie / remove_cookie to directly use the Cookie class in cw.web.http_headers rather than going back and forth simple cookie <-> raw string <-> http_headers.Cookie IMO more on this should be done.

[web] properly redirect when user login on loggedout view (closes #1954050) This avoid getting a "you've no access to this view" message

[i18n] update

[schema, cleanup] one more case for add_etype_button

[i18n, facet] missing registration of <no relation> msgid

[test] sort to avoid random failure due to dict order

[facets] fix height computation (incorporates and/or element if present, use css value to compute likely maximum height, kill wdg_stack_size) (closes #1954077)

backport stable

[primary view] add 'verbatim' attribute view (closes #1947475) this patch has been applied to default accidentaly

[primary view] add 'verbatim' attribute view (closes #1947475)

backport stable

[doc] split API/book part of stuff that has been moved in primary view module docstring

[rqlrewrite] if inlined relation has to be moved to a subquery, take care of the object of the relation (closes #1945725) As the object of the relation is moved to the subquery, all variables/relations representing some of its attributes or inlined relations should be moved there as well. This avoid error such as "BadRQLQuery: variable AF should be selected by the subquery" on security insertion.

[views/facets] simple horizontal layout when there are a few facets only (closes #1921709)

[breadcrumbs] keep proper breadcrumbs when creating an entity linked to another by inspecting __linkto (closes #1947476)

[web test] secondary view isn't listed anymore by possible view since it use the class_deprecated metaclass

ensure the test for asource column won't crash on SQLServer (closes #1949621)

[doc] backport primary view documentation into the code, also document [URL]AttributeView

[primary sub views] use entity_call instead of cell_call

[doc] get back baseviews documentation into the code, enhance it and reorganize the module accordingly also properly deprecates SecondaryView as it should have been done a while ago.

typo

[reledit] honore 'rvid' for attribute relations (closes #1947471)

[view] fix TypeError when calling req.view('vid', entity=entity) (closes #1947474)

[sources management] nicer source management view and breadcrumbs (closes #1946329) * nice 'add host config button' * table displaying host configuration * display a source's config using <pre> or something like that that respects EOL * configure breadcrumbs for CWSourceSchemaConfig and CWSourceHostConfig

[cubicweb.js] log evalJSON errors (closes #1947473)

[views/table] parametrize table view co-classes (closes #1947472)

[cwproperties] refactor properties form to make it partly overrideable easily (closes #1936782) also fix some pitfalls on the way: * shouldn't capitalize labels, this has to be done in translation files as desired * form_row in CWProperties form didn't return subform as its parent class

[entity] rename Entity._rest_attr_info into cw_rest_attr_info (closes #1942759)

[entity] upgrade fetch_[unrelated_]order to benefit from changes introduced in 3.14 (closes #1942758) of rql generation parts of the ORM now based on rql syntax tree. This allows more powerful and flexible sort control by giving them the syntax tree instead of manipulating string. Also: * prefix new methods by 'cw_' * fix cases that currently crash in 3.14 due to the refactoring

[schema] drop CWSourceSchemaConfig[cw_for_source,cw_schema] unique constraint (closes #1946277) as they are not necessarily unique, eg when one want a configuration for both subject and object

[cleanup]

[jsoncontroller] deprecate the now unused js_set_cookie (closes #1944810)

[repo] ensure we don't restart a task while the repo is shutting down (closes #1942736) we need a safety belt in cw itself again looping task which use 'try / except' or doesn't need to access the db (which will trigger ShuttingDown exception)

[warnings] explicitly ask to show DeprecationWarning in debug mode (closes #1942760) as python 2.7 don't show DeprecationWarning by default

[testlib] make self.view() return raw output if no validator is used This will make it possible to parse and test view's output even if no validator is associated to the view's content type.

[tabs] do not use an ajax call to set cookie value, use jquery.cookie helper instead (closes #1944522)

[tabs] fix clic on tab that does nothing (closes #1942750) (there was an area around the text that would trigger the tab switch but not load the tab content)

[idownloadable] dtw and up copyright

[test] fix test broken by 7815:2a164a9cf81c

[entity doc] fix typo (rest_var instead of rest_attr) and cleanup doc

[exceptions] stop catching any exception in various places (closes #1942716)

[tableview] fix view selection when value is None (closes #1936781) * in case of None on final view, use explicitly asked view anyway * when etype is None, that means it's a non final entity type

[book] merge pyro chapter with faq entry with connexion snippet

i18n update

[schema view] fix label/name inversion leading to wrong css class / group url depending on user's language

[i18n] tweaks

[i18n] cleanup messages moved to localperms cube

[security view] slight changes to make things clearer

backport stable

[refactoring] introduce add_etype_button function to properly create button to add a new entity also properly use cw_create_url to generate the URL to add the entity

[user prefs] use same title as the action for consistency

[sources management] properly set __linkto, closes #1922970 when clicking on 'add mapping part' on an external source view, one expect to have source properly selected in form

some minor cleanups

[web test] fix test broken by new users/groups management views

[table view] allow to set None in 'headers', meaning the label should be fetched from the result set as done by default (closes #1931533) the point of this is to avoid having to define the whole headers list when only some of them have to be defined.

[user management] users and groups management now really lead to that (closes #1931478) ie, one tab for users, another for groups. By default the groups tab simply list existing groups (with the number of users they count), and propose to create a new group.

[css] use h* title properties from uiprops in cubicweb.old.css (closes #1931476) + minor cleanup of cubicweb.css

[entity fetchattrs] also fetch ambiguous rtypes even if we do not recurse on them (closes #1720823)

backport stable

[jquery] upgrade to jquery 1.6.4

[test] fix some tests broken since localperms cube extraction

refactor entity fetch_rql method to use a RQL syntax tree instead of RQL strings ; closes #1585650

[doc] add some documentation in cubicweb.web.action

[test] fix bad sed

[deprecation] fix unittest pending deprecation warnings on failIf/failUnless methods family

[test] fix test subsquently to localperms introduction

[deprecation] specify name/doc to deprecated else it's fetched from the object, which has then to be importable. Cleanup debian/control dependancy on the way (needs lgc 0.56.3 api)

[debian] fix init script dependencies (closes #1921729) Facility names starting with $ are reserved system facilities. Other scripts can be referred to using the facility name listed in Provides directly.

[server] portable dump format supports now ZIP64 extensions by default (closes #1912535) ZIP files can use the ZIP64 extensions when the zipfile is larger than 2 GB. zipfile will raise an exception when the ZIP file would require ZIP64 extensions. Note: ZIP64 extensions are disabled by default in stdlib because the default zip and unzip commands on Unix (the InfoZIP utilities) don’t support these extensions.

[repo] use smarter query to check asource presence on startup (closes #1922099)

[js] closes #1558870 : update to jquery 1.6 (1.6.3)

[cubification] moved CWPermission handling into a localperms cube. closes #1914011 keeping bw compat provided the cube is available (else user should be warned he has to install it).

backout 7781:e95cfd5eca61

backout 7780:a1d5365fefc1

[js] closes #1558870 : update to jquery 1.6 (1.6.3)

[cubification] moved CWPermission handling into a localperms cube. closes #1914011 keeping bw compat provided the cube is available (else user should be warned he has to install it).

[debug] use repr() in some place where you'll have a chance to get an unicode error if output encoding isn't properly detected

[i18n fr] fix typo and enhance translation

[user management] disable http cache on user management view (not really a startup view)

backport stable

Added tag cubicweb-debian-version-3.13.5-1 for changeset 2e9dd7d94555

Added tag cubicweb-version-3.13.5 for changeset 20d9c550c57e

3.13.5

i18nupdate

[multi-sources] closes #1918120: drop behaviour leading to wrong result as demonstrated by the ticket The formerly implemented behaviour doesn't seems to be useful, nor to have a chance to be correct. At least no test case demonstrate the opposite, so I've dropped it.

[session] enforce coherency of login/passwd couple values when no anonymous user (closes: #1910849) None may be returned for both if anonymous connection is not allowed or if an empty login is used in configuration Raise ConfigurationError is anonymous login is non-ascii (convention)

[session] take care of non-ascii characters in login and session id (closes: #1910849) We ensure a valid <str> conversion of the session id by using unormalize() at initialization.

closes #1914824: fix retrieval of _andor input, test and fix generated rql

[debian] build the book with no auto-compression of book files (closes: #1906581) Book content paths (html and rest) are now excluded from Debian compression step. This will fix a documentation search issue on the website (doc.cubicweb.org) Documentation build is now made optional since it can break for old distributions and we don't want to block a new release of Cubicweb because of documentation issues.

[views/navigation] use the proper icons instead of ascii art (closes #1624943)

devtools: Fix cleanup of temporary database (closes #1908656) The stored database path for clean up was not absolute. We now make the ``db-name`` absolute at database manade initialization. before storing the name for cleanup.

Drop privileges properly (closes #1910561) Use new function from logilab.common.daemon to switch to non-privileged uid and gid.

drop unittest deprecation warning

[web] add option to make resources-concat optional (implements #1910615)

backport stable

oops, forgotten to qref enhancements made to the reledit doc addition

[book] section about how to disable reledit feature/view

[skel] drop unused import

[hooks control] test and fix session.reset_hooks_mode_categories, closes #1908680

[form fields doc] explain when field.role should be None

Added tag cubicweb-version-3.10.x for changeset a62f24e1497e

merged back oldstable fix to stable

work around cursor.rowcount not returning anything useful before 1st fetch with mssql (closes #1910869) use select count(*) from table to get the number of rows

[devctl] do not crash in the (rare, odd) case where mod.__file__ is None (closes #1908664)

[doc/book] fix ref to trustedauth cube

site schema should be named data model schema

backport stable

Added tag cubicweb-debian-version-3.13.4-1 for changeset 52f26475d764

Added tag cubicweb-version-3.13.4 for changeset 223ecf0620b6

3.13.4

[.po generation] closes #1902315: c i18ncube doesn't consider relation definition description

closes #1901436: view/modify action in table don't work

[css] closes #1880740: update cubicweb.reset.css file (meyerweb.com)

[web test] fix test needed after change of primary_email definition

[doc] fix ReST in README file (make pypi happier)

[web] fix user management view selector

[base schema] do similar changes as use_email for primary_email

[book, rql] update rql langage chapter with new operators and outer join capabilities

[web] closes #1892494: emailaddress and anonymity notice though mangle-address is weak since address won't be mangled when fetched through rql

[3.13 migration] closes #1896724: some db backend (eg sqlserver) doesn't support ADD COLUMN

[source] update system table and internal structures on source renaming. Closes #1896721

[rql2sql] closes #1892473: enhance handling of optional inlined relation * some outer join aren't supported while they could * consider newly available 'optcomparisons' st information on variables (rql 0.30)

[schema] use definition for `use_email` relation instead of type declaration (closes: #1894929) This will be more adaptable for other existent `use_relastion` as the one in the cube Person.

[web] rql directive breaks if explicit vid and empty rset (closes #1893433)

[datafeed] properly encode/decode external uri

[querier] we should not insert None but skip the line when some unknown eid is detected (closes #1892491)

[rql2sql] closes #1892471: support for unary expression introduced in rql 0.30

[deprecation] __message hidden generate deprecation warning

[server] host mapping is better named url mapping (closes #1892461)

[packaging] should depend on yams 0.33

[packaging] should depend on yams 0.33

[rql2sql test] sqlserver use 0/1 for booleans, not FALSE/TRUE

[repo] Stop setting hashmode on schema At best it was a no-op, at worst it created race conditions.

[devtool] Make the sqlite database used for test execution process specific (closes #1230401) * Database cache are not impacted. * This allow multiple process to run test with the same template at the same time The db-name property return an unaltered value to for database cache mechanism.

backport stable

Added tag cubicweb-debian-version-3.13.3-1 for changeset fb48c55cb802

Added tag cubicweb-version-3.13.3 for changeset 0e82e7e5a34f

[rql2sql] take care to boolean_value returning an int (eg sqlite/mysql)

[repo test] remove debug

[repo ms] commit/rollback session if it has been created in the method

3.13.3

i18nupdate

[sql] more test and fix boolean handling

backport oldstable

[repo test] Fix test broken by BFSSTestable changeset bd56a29acaa8 broke several fragile test.

[sql] fix querying of boolean value w/ sqlite

backport oldstable

[datafeed] for datafeed source, we don't want commit in extid2eid but explicitly handled by the source. Also, we should use 'safe' internal session

[sql gen] handle optional on comparison node (eg HAVING expression) and on rhs of final relation. Closes #1859609

[session] safe internal sessions don't deactivate integrity hooks

[sql test] lgdb > 1.6 use int for bool with sqlite backend

[rql2sql] don't introduce substitution for Int and Boolean constants. This is needed for COUNT(DISTINCT) support

[css] syntax fix in cubicweb.css

[entities] skip_copy_for should be considered for object relations too (closes #1857474)

[datafeed] case insensitive etype. Closes #1877014

[datafeed] closes #1875322: empty tag means None value

[datafeed] properly take care to cwsource=system in imported xml. Closes #1877017

[datafeed / fti] rather control a 'complete' parameter than setting empty attribute values

[repo api] deprecates source.extid2eid, only the system source should implement it, and one should directly use the repo api instead of this confusing and error-prone short-cut. Deprecates eid2ext as well for consistency.

backport stable

backport oldstable

[bfss] Fix update of BFSS attribute to None (close #1875289) The current code assume attr value is alway a valide Binary value. But when the attr is optional, the value might be None. The change handle this case.

[skeleton/pytestconf] provide an acceptable solution for windows, closes #1875488

[facets] make sure has_text facet data is posted correctly (closes #1857405) Posted data was associated to facet id (e.g. has-text-123098dasdf898adssdd) rather than facet name (i.e. has-text).

[rql] fix min/max args definition of [TEXT_]LIMIT_SIZE registered procedures (long outstanding bug discovered after recent fix in lgdb)

[schema] rephrase relation definition since constraint doesn't apply to CWRelation

[datafeed] closes #1873616 (user's url corruption when etype is not recognized

[schema] restrict source mapping 'cw_schema' relation to non final relation, we currently have no use for attributes

[c-c db-check] add some sql queries to fix cw_source/is/is_instance relations according to data in the entities table

[c-c db-check] missing comma ending print statement to get expected output

[c-c db-check] disable write security as well so we can delete stuff when needed (e.g. orphan TrInfo)

[c-c db-check] having no owner is allowed, don't try to fix that

[metadata, ldap] make backport of ldap user actually working

[c-c db-check] drop some deprecation warning

[index page] use 'New xxx' generated labels, 'add a xxx' labels are only generated according to inlined form configuration and with an associated context

[doc packaging] prune _sources containing generated text file, include js else search doesn't work

[boxes] 3.13.2 regression: should specify dom id else we may generate bad xhtml

[facets] fix missing 'warnings' import and missing self reference on 'filtered_variable'

stable is now 3.13

oldstable is now 3.12

Added tag cubicweb-debian-version-3.13.2-1 for changeset 9ad5411199e0

Added tag cubicweb-version-3.13.2 for changeset f9227b9d6183

3.13.2

[rql2sql] closes #1832859: fake HAVING terms w/ EXISTS terms there are cases where we want to use the HAVING trick on terms that should belong to an EXISTS subquery. Currently , this causes bad sql. We should fix that until we've a proper rql grammar avoiding the need for this trick. Notice at some point, we may want **actuall** HAVING clauses and GROUPBY used in EXISTS subquery, but that's another story.

backport stable

[boxes] closes #1832844: translate views category in possible views boxes

[cwctl] set min/max args for reset admin password command. Closes #1832986

[web] Image optimisation using trimage (pngcrush inside)

[preferences js] jquery 1.6 needs quote when querying attribute value

fix test broken by previous commit

[session] add session id to closed connection errors

unsaved change...

[facets] closes #1827464: facet box always appear by implementing proper selectors

[uilib] fix bad argument order, introduced by previous commit

[uilib] reimplement printable_value function using a dictionary so one can easily change how to base type is displayed everywhere in the application. Closes #1827465

closes #1826543: error on some ajax_replace_url call

Added tag cubicweb-debian-version-3.13.1-1 for changeset cc0578049cbe

Added tag cubicweb-version-3.13.1 for changeset 7d84317ef185

3.13.1

i18nupdate

backport stable

Added tag cubicweb-debian-version-3.12.9-1 for changeset 92464e39134c

Added tag cubicweb-version-3.12.9 for changeset cb2990aaa63c

3.12.9

[repo, ms] fix planning of some queries where variable 'non-invarianess' should be considered, leading for instance to junk in entities table on source deletion

[repo] closes #1821172: fix source deletion

[ajax, IE] rewrite some code to avoid duplicated evaluation with IE

[repo, hooks] small task optimization: don't start cleanup_old_transactions when undo support is not activated, and retrieve a session to update a source only when some source to update is found

[http] closes #1816411, restore expires_header on static files in non-debug mode

[native source] fix bug introduced in 7645:cb2459aaa7d7: fetchone() may return None

[source] closes #1816412: try to reconnect to dbms if connection is reset in eid_type_source()

[twisted] add an option to configure twisted's threadpool size Twisted's threadpool size defaults to 10, which means we virtually try to serve up to 10 simulateneous requests but connection-pool size defaults to 4 on CW-server side. That means that under heavy load, chances are high that some HTTP queries end up with a 500 error code / connection-pool exhausted.

[book, form] include complete example of self-posted form with custom field/widget + minor other changes