Closed tpluscode closed 1 year ago
Latest commit: 7350f75e9267cb07d80d829fa8050c60bcc589cb
The changes in this PR will be included in the next version bump.
Not sure what this means? Click here to learn what changesets are.
Click here if you're a maintainer who wants to add another changeset to this PR
Merging #1407 (b5d0984) into master (7884e0e) will decrease coverage by
3.26%
. The diff coverage is100.00%
.:exclamation: Current head b5d0984 differs from pull request most recent head 7350f75. Consider uploading reports for the commit 7350f75 to get more accurate results
@@ Coverage Diff @@
## master #1407 +/- ##
==========================================
- Coverage 80.77% 77.51% -3.26%
==========================================
Files 196 196
Lines 13448 13466 +18
Branches 741 721 -20
==========================================
- Hits 10862 10438 -424
- Misses 2578 3020 +442
Partials 8 8
Impacted Files | Coverage Δ | |
---|---|---|
apis/core/lib/domain/queries/dimension-mappings.ts | 100.00% <100.00%> (ø) |
... and 13 files with indirect coverage changes
Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.
This changes how new dimension mappings are saved to the database when "auto-mapping"
The current implementation performed an
INSERT
with a complex pattern and a long listVALUES
clause. Potentially thousands of items long.Since the full set of these
VALUES
is loaded to memory anyway, they are actually cross-referenced in memory and new items are streamed directly as quads using the graph protocol. This appears much more database-friendly. While Stardog does cope, local fuseki actually failed to insert when mapping a large dimension of municipalities (3k terms)