Closed daemontus closed 3 years ago
Merging #7 into master will decrease coverage by
0.07%
. The diff coverage is96.92%
.
@@ Coverage Diff @@
## master #7 +/- ##
==========================================
- Coverage 85.00% 84.93% -0.08%
==========================================
Files 28 29 +1
Lines 1761 1859 +98
==========================================
+ Hits 1497 1579 +82
- Misses 264 280 +16
Impacted Files | Coverage Δ | |
---|---|---|
src/impl_variable_id.rs | 0.00% <ø> (ø) |
|
src/sbml/mod.rs | 73.24% <96.92%> (+2.41%) |
:arrow_up: |
src/impl_fn_update.rs | 87.50% <0.00%> (-6.25%) |
:arrow_down: |
src/impl_regulatory_graph.rs | 84.90% <0.00%> (-1.89%) |
:arrow_down: |
src/bdd_params/impl_witness_generator.rs | 81.08% <0.00%> (-1.81%) |
:arrow_down: |
src/bdd_params/impl_bdd_parameter_encoder.rs | 85.45% <0.00%> (-0.61%) |
:arrow_down: |
src/display_regulatory_graph.rs | 95.00% <0.00%> (-0.46%) |
:arrow_down: |
src/sbml/export.rs | 94.78% <0.00%> (-0.26%) |
:arrow_down: |
src/async_graph/impl_async_graph.rs | 88.00% <0.00%> (-0.24%) |
:arrow_down: |
... and 6 more |
Continue to review full report at Codecov.
Legend - Click here to learn more
Δ = absolute <relative> (impact)
,ø = not affected
,? = missing data
Powered by Codecov. Last update e1a10e1...f11a06f. Read the comment docs.
SBML qualitative species can have an
id
and aname
attribute, whereid
is used in the model file when referencing the species, butname
is the actual value displayed to the user. In a lot of modelsid == name
, but this is not always the case.For now, we used
id
as the variable name in the imported model, but this causes unnecessary confusion since models are sometimes imported with seemingly random or unrelated species names. For example, Ginsim SBML seems to be ok, but anything from Cellcollective is problematic.The problem with using SBML names directly is that they have a very wide range of allowed characters (including
(
or?
). We thus have to perform a normalisation, replacing every invalid character with_
so that the model can be imported successfully.TL;DR, this pull request:
Note that this is closely related to #1, but does not solve it completely because there are still other known problematic SBML models. However, I've tested it with a custom version of aeon backend and it seems to work reasonable on models from cellcollective.org now (previously, only numbers were shown instead of names).