Merge pull request #1397: Fix failing CI tests (ancestral/translate)

nextstrain · Jan 25, 2024 · d83d671 · d83d671
2 parents cae562a + 0dd471b
commit d83d671
Show file tree

Hide file tree

Showing 6 changed files with 13 additions and 16 deletions.
diff --git a/tests/functional/ancestral/cram/case-sensitive.t b/tests/functional/ancestral/cram/case-sensitive.t
@@ -16,7 +16,7 @@ Change the _reference_ to lowercase
   $ python3 "$TESTDIR/../../../../scripts/diff_jsons.py" \
   >   "$TESTDIR/../data/simple-genome/nt_muts.ref-seq.json" \
   >   "nt_muts.ref-seq.json" \
-  >   --exclude-paths "root['meta']['updated']" 
+  >   --exclude-paths "root['generated_by']"
   {}
 
 
@@ -38,5 +38,5 @@ be lowecase which will be compared against the uppercase reference
   $ python3 "$TESTDIR/../../../../scripts/diff_jsons.py" \
   >   "$TESTDIR/../data/simple-genome/nt_muts.ref-seq.json" \
   >   "nt_muts.ref-seq.json" \
-  >   --exclude-paths "root['meta']['updated']" 
+  >   --exclude-paths "root['generated_by']"
   {}
diff --git a/tests/functional/ancestral/cram/general.t b/tests/functional/ancestral/cram/general.t
@@ -20,7 +20,7 @@ node-data JSON we diff against.
   $ python3 "$TESTDIR/../../../../scripts/diff_jsons.py" \
   >   "$TESTDIR/../data/simple-genome/nt_muts.ref-seq.json" \
   >   "nt_muts.ref-seq.json" \
-  >   --exclude-paths "root['meta']['updated']" 
+  >   --exclude-paths "root['generated_by']"
   {}
 
 Same as above but without providing a `--root-sequence`. The effect of this on behaviour is:
@@ -40,5 +40,5 @@ mutations (as there's nothing to compare the root node to)
   $ python3 "$TESTDIR/../../../../scripts/diff_jsons.py" \
   >   "$TESTDIR/../data/simple-genome/nt_muts.no-ref-seq.json" \
   >   "nt_muts.no-ref-seq.json" \
-  >   --exclude-paths "root['meta']['updated']" 
+  >   --exclude-paths "root['generated_by']"
   {}
diff --git a/tests/functional/ancestral/cram/vcf-multi-allele.t b/tests/functional/ancestral/cram/vcf-multi-allele.t
@@ -24,7 +24,7 @@ See <https://github.com/nextstrain/augur/issues/1380> for the bug this is testin
   $ python3 "$TESTDIR/../../../../scripts/diff_jsons.py" \
   >   "$DATA/nt_muts.ref-seq.json" \
   >   nt_muts.json \
-  >   --exclude-regex-paths "root\['nodes'\]\['.+'\]\['sequence'\]"
+  >   --exclude-regex-paths "root\['nodes'\]\['.+'\]\['sequence'\]" "root\['generated_by'\]"
   {'iterable_item_added': {"root['nodes']['sample_B']['muts'][0]": 'A30G'}}
 
   $ cat > expected.vcf <<EOF

diff --git a/tests/functional/ancestral/cram/vcf.t b/tests/functional/ancestral/cram/vcf.t
@@ -21,7 +21,7 @@ but it will have the reference sequence attached.
   $ python3 "$TESTDIR/../../../../scripts/diff_jsons.py" \
   >   "$DATA/nt_muts.ref-seq.json" \
   >   "nt_muts.vcf-input.ref-seq.json" \
-  >   --exclude-regex-paths "root\['nodes'\]\['.+'\]\['sequence'\]" "root['meta']['updated']"
+  >   --exclude-regex-paths "root\['nodes'\]\['.+'\]\['sequence'\]" "root\['generated_by'\]"
   {}
 
 Here's the same mutations as in $DATA/nt_muts.ref-seq.json,

diff --git a/tests/functional/ancestral/data/simple-genome/nt_muts.ref-seq.json b/tests/functional/ancestral/data/simple-genome/nt_muts.ref-seq.json
@@ -7,10 +7,6 @@
       "type": "source"
     }
   },
-  "generated_by": {
-    "program": "augur",
-    "version": "23.1.1"
-  },
   "mask": "00000000000000000000000000000000000000000000000000",
   "nodes": {
     "node_AB": {

diff --git a/tests/functional/translate/cram/root-mutations.t b/tests/functional/translate/cram/root-mutations.t
@@ -6,13 +6,14 @@ Setup
   $ export DATA="$TESTDIR/../data/simple-genome"
 
 This is the same as the "general.t" test, but we are modifying the input data
-such that the reference sequence contains "G" at pos 20 (1-based), and
-include a compensating mutation G20A on the root node.
-This results in the reference translation of gene1 to be MPCE* not MPCG*.
-(Note that the compensating nuc mutation doesn't actually need to be present
-in the JSON, `augur translate` just looks at the sequence attached to each node.)
+such that the reference sequence contains "G" at pos 20 (1-based), and include a
+compensating mutation G20A on the root node. (This manipulation would be much
+nicer using `jq` but it's not (yet) available in all nextstrain environments.)
+This results in the reference translation of gene1 to be MPCE* not MPCG*. (Note
+that the compensating nuc mutation doesn't actually need to be present in the
+JSON, `augur translate` just looks at the sequence attached to each node.)
 
-  $ sed '29s/^/        "G20A",\n/' "$ANC_DATA/nt_muts.ref-seq.json" | 
+  $ sed '24s/^/        "G20A",\n/' "$ANC_DATA/nt_muts.ref-seq.json" | 
   > sed 's/"nuc": "AAAAAAAAAATGCCCTGCGGG/"nuc": "AAAAAAAAAATGCCCTGCGAG/' > nt_muts.json
 
   $ ${AUGUR} translate \