Give each index a (reasonably) unique ID #984
Conversation
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## master #984 +/- ##
=======================================
Coverage 85.80% 85.81%
=======================================
Files 308 308
Lines 29261 29266 +5
Branches 3430 3430
=======================================
+ Hits 25108 25114 +6
- Misses 2787 2789 +2
+ Partials 1366 1363 -3 ☔ View full report in Codecov by Sentry. |
|
Kudos, SonarCloud Quality Gate passed!
|
hannahbast
force-pushed
the
index-id
branch
2 times, most recently
from
2d3d8b6 to
3b21d2d
Compare
hannahbast
force-pushed
the
index-id
branch
2 times, most recently
from
b71939e to
fc11809
Compare
hannahbast
force-pushed
the
index-id
branch
2 times, most recently
from
2872384 to
66d5893
Compare
|
Kudos, SonarCloud Quality Gate passed!
|
hannahbast
force-pushed
the
index-id
branch
from
de346a8 to
fd77641
Compare
hannahbast
force-pushed
the
index-id
branch
from
fd77641 to
7938778
Compare
|
Kudos, SonarCloud Quality Gate passed!
|
|
Kudos, SonarCloud Quality Gate passed!
|
The ID is a string that can be obtained via the API with cmd=get-index-id. It is currently just a concatenation of the index name and several statistics on the number of triples (which is reasonably but not perfectly unique). This feature is useful for applications like https://github.com/ad-freiburg/qlever-petrimaps, which have an internal cache that depends on the index. When the index changes, the application should be able to notice this an clear its cache. With an index ID it can.
|
|
@joka921 This has been lying around for some time now (and has already been in active use for month). It's simplistic but good enough IMO. I just rebased it to the current master and added a simple test. Please have a quick look and let me know whether you want anything changed before we merge this? |
hannahbast
changed the title
The index ID is currently just a concatenation of the index name, and the number of triples, subjects, predicates, and objects. This makes it unlikely (but not impossible) that two different indexes have the same ID. The ID can be obtained via the API with
cmd=get-index-id.This feature is useful for applications like https://github.com/ad-freiburg/qlever-petrimaps, which have an internal cache that depends on the index. When the index changes, the application should be able to notice that, and with the index ID, it now can.