Age | Commit message (Collapse) | Author |
|
Just bringing in a commit from gerrit which was overlooked for too long
|
|
The query in get_taskbypolicy should limit results to tasks in
the public space. Public tasks that were created before spaces
could have space_phid set to NULL.
|
|
Mention phabtools.conf.example in README;
install hints for python-phabricator and python-mysqldb
Add dummy config value for update_lookback value (bugzilla_create.py dies without it).
Change-Id: I01aea0222c27ed0a91d94877e0627f394b9c7af0
|
|
refs T127766
NOTE: The points field name changed from `storypoints` to `points`
|
|
|
|
The project columns transaction type was migrated to a new type
name. See: https://secure.phabricator.com/D15634
Bug: T133994
Change-Id: I2d21336fd7e092c277d859170bb4637b1b8c3750
|
|
I migrated the Gather backlog board for Jon Katz, and now Growth,
so we have more username mappings. And tweaked my instructions.
Change-Id: I5d7136c2d6b75471e941dd86ba734decf13c6458
|
|
see https://github.com/phacility/phabricator/commit/32225d1dd0f8274ae9323650cbf20ea0571424de
Bug: T127524
Change-Id: Ib566ac738c3a2f942b019aa562fa7bf61abbeb30
|
|
Still only public tasks
Bug: T115936
Change-Id: I80862b65995448583a0faa24bc17f5d51c2ebf83
|
|
This changes things up for the public dump job
to more sanely work with mysql connections. Ideally this
is all redone and we ditch all of the migration logic
that was never meant to live this long. In practice at
least this does more sane things even though it all needs to be
scrapped and passing db con objects around in this way is odd.
Change-Id: Ib26179676c08996ea158c15f6d397b89987eeb2e
|
|
ref T104920
Change-Id: I3414422febe5886f09faa3f3f688cc652a92da46
|
|
Change-Id: I3665f0189765553af07e83e12471b0ce766c9f11
|
|
Change-Id: I8d7ac57cb9000c0d9e0130d2a737f296704b44fc
|
|
Change-Id: I74592323a28c1d2c6aa211c800cd643714d0222b
|
|
Change-Id: I57e0d863db1a4234f5e323caf62a7c6e407f0b61
|
|
ref T100846
Change-Id: I1d589356317d8e273d187c20b2f45f42b33f9fe3
|
|
Dumps to a json file for reporting and analytics that several
departments have requested. All security information is
derived from the security on the task, though project view policy
is also respected.
This is very specific as several columns in tables contain sensitive
or potentially sensitive information even on public tasks such as
mailKeys or hidden history.
Change-Id: I9e20d77a5642757c96f32f40914528b08104abbd
|
|
Upstream revamped their schema for subscribers
Change-Id: I3f920ac1d04c64ac252b1c4b6be508ead547b1e1
|
|
Change-Id: I3e0013fa355581079d4e72ef849f0e1114d13dd7
|
|
Point to bugrun.sh, say what scripts do.
Change-Id: I1973224676c25f6d2c1f0a1e0808af0678b734ef
|
|
Processes Trello board exported as JSON to import its cards as
Phabricator tasks. It imports name, description, members and commenters
(as task subscribers), labels (as text in description), and checklists
(as markup in description). Doesn't handle attachments, comments,
correct timestamps and authors, etc. Phabricator user "Trellimport"
makes all the changes in the present.
Similar to bugzilla_import.py, but does much less, runs in a single
shot, and only uses the phabricator conduit API.
Uses TrelloCard, TrelloDAO, and TrelloScrubber objects from a
"cburroughs"'s export_trello.py to figure_stuff_out().
Modifies cburroughs export_trello.py to
* handle changes in the Trello JSON format of idLabels, column list, and shortUrl
* write out idLabels, generate checklists in Description
* fake out its setup_logging()
* etc.
It uneasily combines two logging systems and options.
You have to first generate Trello name -> Phabricator PHID mappings, see
change Ia574d589. Detailed instructions to run this are in
trello_create.py, more info is in
https://www.mediawiki.org/wiki/Phabricator/Trello_to_Phabricator and
T821. spage is happy to help!
Bug: T821
Change-Id: Ia57b5c034bf58f8f3c502a2fa98ed779c135f1cd
|
|
This fine individual uploaded https://secure.phabricator.com/P1407 ,
an upcoming commit uses some of its code to parse information in the
JSON export of a Trello board.
Awesome dragon ASCII art.
Change-Id: Iec91bf3701af4e1c106338d3a82ff62c5396458d
|
|
A human being has to map e.g. 'spage1' in Trello to 'spage' on a
Phabricator host in a YAML data file.
Then the script trello_MakePHIDs.py can determine the mapping to
phabricator PHID. It writes out data/trello-scrub_<HOST>.yaml
Change-Id: Ia574d5896bd0ce73355d470b0bb5b0d337ac8f65
|
|
regarding operations project cleanup
Change-Id: I9273e8a31c2bfa5569739848cfdc83d86625c8cb
|
|
This script takes an old bz id and modifies
the ACL on the relevant Phab issue to allow
historical CC/Assignee/Author if they exist
in Phab now.
|
|
|
|
|
|
to refer/referd to tasks for RT
__This is not idempotent__
|
|
re key the xcomment dict
|
|
* Had to reconsider some unicode compromises
* Add bz reference from RT
* Rry to sanely represent literal stuffs and quoted
* Add bugzilla reference with links
* Comments missing metadata
* Attachments with weird names / formats
|
|
failures as actual failures and counts
other exclusions separately.
|
|
that happened to tasks from Security-* in bugzilla.
|
|
I was doing something clever to throttle a large backlog
over the long term, but I ended up just batch processsing.
This is much simpler to debug for days to come. Look at last
n seconds and fix new users.
|
|
* handles the ***'s literal case
* handles the private comment case
* handles the stripping of legacy attachment notice
|
|
This makes things slow but right. Adjusting CC's on a
ticket cannot be done in parallel without some race condition
risk of inaccuracy.
|
|
|
|
|
|
|
|
* close a few dangling sockets
* set perms on uploads for sec tickets
* do project checking via db not api
|
|
|
|
|
|
|
|
|
|
on author assignment
|
|
|
|
|
|
Change-Id: I63c88d1f0c8db5ba79b2aed5aee25a774fe07061
|
|
Change-Id: Ibe73f6de67b165c9b8992439a1dd7ab5d1ca756c
|
|
|
|
|