Add guard so NuclearModelMap doesn't descend into subalgs #6
+3
−1
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
CAFMaker configuration chokes if this guard isn't enabled.
The culprit is in
NuclearModelMap::LoadConfig(), that checksNuclearModel@Pdg=...., as that callsGetConfig()and loops over all entries inGetItemMap(), which expands keys (such as the expected ones e.g.NuclearModel@Pdg=1000180400), treats them like algs, evaluates their config (in this case,SpectralFunc/ApproxElementsfor theAR25_20i_00_000tune), and appends keys from that config.This creates bogus keys like
NuclearModel@Pdg=1000180400/AllowElementMatch, which cause UB and segfault.A tag,
v3_06_02_sbn1p1, has been created pointing to commit 7e784bc, and published to SBN CVMFS.Production workflows using this tag have run successfully on the grid.
Summoning @sjgardiner (who setup most of this original magic 🪄) to review.
Ho Ho Ho! 🎅🌲❄️