opensource/galaxy
3
0
Fork 0
mirror of https://github.com/galaxyproject/galaxy.git synced 2026-09-21 13:50:20 +08:00
Code Issues Packages Projects Releases Wiki Activity
55,926 Commits 72 Branches 243 Tags
2dd0eed46b8beefdc650493948dbbcfe0eae01a4
Commit Graph
55926 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
Dannon Baker 2dd0eed46b Fix several 'to to' redundant word typos. 2021-10-21 12:28:45 -04:00
Marius van den Beek 31e6b32221 Merge pull request #12758 from dannon/fix_history_sort 
[21.09] Prefer model class when it exists in grid sorting
 2021-10-21 15:36:34 +02:00
Dannon Baker 29e0901794 Prefer model class when it exists (assuming it is more correct/refined logic) over column. This fixes history sort to work correctly in grids. 2021-10-21 07:37:54 -04:00
Marius van den Beek 0e25970adf Merge pull request #12745 from dannon/yaml-backport 
[21.09] Yaml datatype backport
 2021-10-19 20:39:45 +02:00
mvdbeek bbe3f25642 Move datatype test file to lib/galaxy/datatypes/test 2021-10-19 19:03:55 +02:00
John Chilton 6229eec62f Merge pull request #12721 from mvdbeek/fix_missing_hda_history_id 
[21.09] Fix missing history or hid for HDAs
 2021-10-19 10:53:45 -04:00
Marius van den Beek 84d86affa5 Merge pull request #12733 from bernt-matthias/topic/validator_needs_no_expression 
[21.09] Fix linter: Parameter validator needs no expression
 2021-10-19 15:50:10 +02:00
Alexander OSTROVSKY 7aa94c3291 re-add assert 2021-10-19 09:30:20 -04:00
Alexander OSTROVSKY ed303913c2 remove extra test lines 
python lint
 2021-10-19 09:30:20 -04:00
Alexander OSTROVSKY dd9e3fc373 added sniffer for yaml 
import order

Better sniff for big files, removed extraneous comments

Better test for bigger files

lint

safe load, yaml error

typo
 2021-10-19 09:30:20 -04:00
Alexander OSTROVSKY cf9298b762 linted 2021-10-19 09:30:20 -04:00
Alexander OSTROVSKY e4d821e4f1 Added yaml support 2021-10-19 09:30:20 -04:00
Marius van den Beek b5d7c95b1c Merge pull request #12726 from dannon/chart-download-overhaul 
[21.09] Charts downloads fixes and enhancements.
 2021-10-19 14:57:42 +02:00
Marius van den Beek 2f0ab3f154 Merge pull request #12713 from bernt-matthias/topic/pretext 
[21.09] Add file_ext property to Pretext and others datatypes
 2021-10-19 14:54:22 +02:00
Marius van den Beek 99857fd39f Merge pull request #12737 from dannon/fix-chrome-scrolling 
[21.09] Fix chrome scrolling on insert in markdown editor
 2021-10-19 14:21:03 +02:00
Matthias Bernt 2c28cc85d6 fix length validator message 2021-10-19 14:18:32 +02:00
Matthias Bernt 990e60fc99 fix linting for metadata validators 2021-10-19 13:14:30 +02:00
Matthias Bernt fa4655ac53 fix for options without text 2021-10-19 13:03:54 +02:00
Matthias Bernt 3a2718470d fix linter error 2021-10-19 12:36:32 +02:00
Matthias Bernt a898771e2b add more missing file_ext 2021-10-19 12:36:32 +02:00
Matthias Bernt a9099c2586 add file_ext property to Pretext 
otherwise sniffing does not work
 2021-10-19 12:36:32 +02:00
mvdbeek fb0a1c0e6d Fix virtual scroller missing item updates 
I don't fully understand what is going on, but here's my current
diagnosis of why this is ncessary:
Rhe contents/near response contains the max_hid header, and the scroller
uses that to determine if we're updating at the top of the scroller,
[here](https://github.com/mvdbeek/galaxy/blob/90c797bc485e238d9bfd0bd55b1b7c5f978037a7/client/src/components/History/providers/HistoryContentProvider/contentPayload.js#L153.)
For some reason `updatesAtTop` is false when the test fails because both
`response.maxHid` and `lastResponse.maxHid` are at 2. If I change > to >=
the test consistently passes in 10 tries, while the original fails in 5
out of 10 tries. I guess pairwise only starts emitting once there are
actually 2 responses so if the initial response contains both items (hid
1 and hid2) I guess it'll fail.
 2021-10-18 20:43:41 +02:00
Dannon Baker 4dadfff6bc Bugfix -- markdown editor scrolling in chrome for dialog-based elements (which change focus) 
Need to set selectionEnd prior to focus() call to get this to work
right.
 2021-10-18 13:01:47 -04:00
Matthias Bernt 2849c81373 fix 2021-10-18 13:28:04 +02:00
Matthias Bernt f5ca471c8e more fixes to validator linters 
and test of most correct combinations
 2021-10-18 13:23:34 +02:00
Matthias Bernt ec60b94e92 linter: regex validator does not require expression 2021-10-18 12:40:52 +02:00
mvdbeek 90c797bc48 Flush just once when copying workflow inputs 2021-10-18 11:42:42 +02:00
mvdbeek 44c3d8fee0 Do flush when remapping job 2021-10-18 11:42:42 +02:00
mvdbeek 4352b8c1ac Add to history before flushing (and moving to storage) 2021-10-18 11:42:42 +02:00
mvdbeek 06370c31fa Skip flushing for tags in more places 2021-10-18 11:42:42 +02:00
mvdbeek b2454dcafe Use TagHandlerSession in ModelPersistenceContext 
Eliminates an unnecessary flush
 2021-10-18 11:42:41 +02:00
mvdbeek f0092c9926 Use TagHandlerSession to avoid flushes 2021-10-18 11:42:41 +02:00
mvdbeek 0f685ab65c Now that we don't flush when adding hids we can do it earlier 2021-10-18 11:42:41 +02:00
mvdbeek ee6ec264c4 Flush just once per import 
That might mean we accumulate a lot of items in memory, but if we don't
do this we need to assign hids for every flush (which is possible).
 2021-10-18 11:42:41 +02:00
mvdbeek f63dc3b9df Do not explicitly begin a new transaction when getting next_hid 
There is no transactional state, so it seems unncessary. Not starting an
explicit transaction means sqlalchemy won't flush the current session,
which means we're not committing a HDA/HDCA to the database that doesn't
have a hid yet.
 2021-10-18 11:42:41 +02:00
mvdbeek cfdd6e9dfb Also raise eception if item has no hid 2021-10-18 11:42:41 +02:00
mvdbeek 6c15c1334c Fix copy action 2021-10-18 11:42:41 +02:00
mvdbeek 00a274c26a Explicitly pass history into DatasetCollectionManager.create 2021-10-18 11:42:40 +02:00
mvdbeek 53e3cf56c3 Fix hda_manager.copy so it doesn't flush datasets without a history_id 2021-10-18 11:42:40 +02:00
mvdbeek bf7cc9072c Always set item.history to self 
What I've observed is that for a job with a immediate post job action
(e.g setting a dataset tag within https://github.com/galaxyproject/galaxy/blob/202bff420761c63fc30ac933630d72dd40496248/lib/galaxy_test/api/test_workflows.py#L3605)
something is resetting hda.history_id. I have no idea how this happens.
Possibly https://github.com/galaxyproject/galaxy/blob/f66520e4a6f6b71516c48861f473c1b430ad967f/lib/galaxy/tools/actions/__init__.py#L557
fetches a new HDA from a different sqlalchemy session and modifies it ??
 2021-10-18 11:42:40 +02:00
mvdbeek b4befd02b5 Raise exception for if we're saving a historyless HDA 2021-10-18 11:42:39 +02:00
mvdbeek ea40edf1b7 Merge branch 'release_21.05' into release_21.09 2021-10-18 11:08:01 +02:00
Marius van den Beek 6fcdc5c44d Merge pull request #12731 from mvdbeek/fix_bad_merge 
[21.05] Fix bad merge
 2021-10-18 11:06:21 +02:00
mvdbeek f162322b35 Fix bad merge 
we've changed the import
 2021-10-18 10:16:17 +02:00
mvdbeek 9c4b6305b9 Merge branch 'release_21.05' into release_21.09 2021-10-18 08:29:12 +02:00
mvdbeek 2a4cd49083 Merge branch 'release_21.01' into release_21.05 2021-10-18 08:28:04 +02:00
mvdbeek 14b6579eb1 Merge branch 'release_20.09' into release_21.01 2021-10-18 08:23:32 +02:00
mvdbeek b60e5ea76d Merge branch 'release_20.05' into release_20.09 2021-10-18 08:22:04 +02:00
David López aee63e5aba Merge pull request #12728 from jmchilton/show_page_pdf_fixes 
[21.09] Various show page PDF bug fixes.
 2021-10-17 20:03:36 +02:00
John Chilton 5cd46c3d5e Various show page PDF bug fixes. 
- In FastAPI mode GET /api/pages/{id}.pdf was being masked by /api/pages/{id} and the tested 400 was an invalid ID error.
- trans.response was being used in the service layer and was unavailable under FastAPI - moved setting response type back up into the controller.
- API tests should skip if the Galaxy configuration is invalid - it wasn't doing this.

While I was in there I also tweaked the exception handling a bit to indicate that this is not a USER FORBIDDEN (403) operation so much as an optional SERVER configuration NOT IMPLEMENTED (501).
 2021-10-16 15:00:59 -04:00