Open AlexanderWells-diamond opened 1 year ago
Merging #110 (9c19bed) into master (dc4a63f) will decrease coverage by
0.13%
. The diff coverage is85.71%
.
@@ Coverage Diff @@
## master #110 +/- ##
==========================================
- Coverage 87.03% 86.90% -0.14%
==========================================
Files 14 14
Lines 972 985 +13
==========================================
+ Hits 846 856 +10
- Misses 126 129 +3
Impacted Files | Coverage Δ | |
---|---|---|
softioc/device.py | 94.67% <85.71%> (-0.93%) |
:arrow_down: |
Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.
15 files ± 0 15 suites ±0 30m 21s :stopwatch: + 2m 3s 253 tests + 6 246 :heavy_check_mark: + 5 7 :zzz: + 1 0 :x: ±0 3 795 runs +90 3 330 :heavy_check_mark: +65 465 :zzz: +25 0 :x: ±0
Results for commit 9c19bedb. ± Comparison against base commit dc4a63f8.
:recycle: This comment has been updated with latest results.
Thinking about it, in its current form this PR does mean you can no longer signal a boolOut/mbbOut record is in an invalid state by setting an out-of-range number to trigger the "ILLEGAL VALUE" error handling in EPICS. That's related to issue #53. I don't know if this is a significant concern.
Long, Bool and MBB record types are now protected from out of range values -
.set()
will throw an exception, andcaput()
will fail to store the new value. This means records can no longer be put into invalid values, or suffer from integer overflow/underflow.Also correct an oversight in
get_field_offsets
, to properly raise a Python exception when the record type cannot be found.Closes #66