fix(test): track pre-built indexes and surface pytest failures in CI

Author: arcangelo7

.github/workflows/run_tests.yml

@@ -37,6 +37,7 @@ jobs:
 
     - name: Run tests with coverage
       run: |
+        set -o pipefail
         uv run pytest | tee pytest-output.txt
         echo "COVERAGE=$(grep TOTAL pytest-output.txt | awk '{print $4}')" >> $GITHUB_ENV
 

.gitignore

@@ -28,10 +28,10 @@ var/
 
 # Tests
 test/virtuoso-test-data/
-test/virtuoso-meta-data/*
-!test/virtuoso-meta-data/meta_subset.nq
-test/qlever-index-data/*
-!test/qlever-index-data/index_subset.nt
+test/virtuoso-meta-data/database/virtuoso.lck
+test/virtuoso-meta-data/database/virtuoso.log
+test/virtuoso-meta-data/index_entities.txt
+test/qlever-index-data/oc-index-test.server-log.txt
 .coverage
 
 # uv

test/conftest.py

@@ -24,6 +24,7 @@
 TEST_DIR = os.path.dirname(os.path.abspath(__file__))
 QLEVER_DATA_DIR = os.path.join(TEST_DIR, "qlever-index-data")
 VIRTUOSO_DATA_DIR = os.path.join(TEST_DIR, "virtuoso-meta-data")
+VIRTUOSO_DB_DIR = os.path.join(VIRTUOSO_DATA_DIR, "database")
 
 
 def _wait_for_http(port: int, timeout: int = 60) -> None:
@@ -88,23 +89,13 @@ def virtuoso_endpoint():
             "-p", f"{VIRTUOSO_HTTP_PORT}:8890",
             "-p", f"{VIRTUOSO_ISQL_PORT}:1111",
             "-e", "DBA_PASSWORD=dba",
-            "-e", "VIRT_Parameters_DirsAllowed=/",
-            "-v", f"{VIRTUOSO_DATA_DIR}:/data",
+            "-v", f"{VIRTUOSO_DB_DIR}:/opt/virtuoso-opensource/database",
             VIRTUOSO_IMAGE,
         ],
         check=True,
         capture_output=True,
     )
     _wait_for_virtuoso(VIRTUOSO_CONTAINER)
-    subprocess.run(
-        [
-            "docker", "exec", VIRTUOSO_CONTAINER,
-            "isql", "1111", "dba", "dba",
-            "exec=ld_dir('/data', 'meta_subset.nq', ''); rdf_loader_run(); checkpoint;",
-        ],
-        check=True,
-        capture_output=True,
-    )
     yield f"http://127.0.0.1:{VIRTUOSO_HTTP_PORT}/sparql"
     subprocess.run(["docker", "stop", VIRTUOSO_CONTAINER], capture_output=True)
     subprocess.run(["docker", "rm", "-f", VIRTUOSO_CONTAINER], capture_output=True)

test/qlever-index-data/oc-index-test.index-log.txt

@@ -0,0 +1,43 @@
+2026-03-08 15:08:39.865 - INFO: QLever index builder 6168b36, compiled on Fri Mar  6 04:34:10 UTC 2026 using git hash 6168b3
+2026-03-08 15:08:39.865 - INFO: Locale was not specified in settings file, default is en_US
+2026-03-08 15:08:39.865 - INFO: You specified "locale = en_US" and "ignore-punctuation = 0"
+2026-03-08 15:08:39.865 - INFO: By default, integers that cannot be represented by QLever will throw an exception
+2026-03-08 15:08:39.865 - WARN: Implicitly using the parallel parser for a single input file for reasons of backward compatibility; this is deprecated, please use the command-line option --parse-parallel or -p
+2026-03-08 15:08:39.865 - INFO: Processing triples from single input stream /dev/stdin (parallel = true) ...
+2026-03-08 15:08:39.868 - INFO: Parsing input triples and creating partial vocabularies, one per batch ...
+2026-03-08 15:08:39.930 - INFO: Triples parsed: 208 [average speed 0.0 M/s] 
+2026-03-08 15:08:39.939 - INFO: Number of triples created (including QLever-internal ones): 208 [may contain duplicates]
+2026-03-08 15:08:39.939 - INFO: Number of partial vocabularies created: 1
+2026-03-08 15:08:39.939 - INFO: Merging partial vocabularies ...
+2026-03-08 15:08:39.940 - INFO: Words merged: 117 [average speed 0.3 M/s] 
+2026-03-08 15:08:39.942 - INFO: Finished writing compressed internal vocabulary, size = 1288 B [uncompressed = 5662 B, ratio = 22%]
+2026-03-08 15:08:39.942 - INFO: Number of words in external vocabulary: 117
+2026-03-08 15:08:39.946 - INFO: Converting triples from local IDs to global IDs ...
+2026-03-08 15:08:39.947 - INFO: Triples converted: 208 [average speed 0.2 M/s] 
+2026-03-08 15:08:39.948 - INFO: Creating permutations SPO and SOP ...
+2026-03-08 15:08:39.949 - INFO: Number of inputs to `uniqueView`: 208
+2026-03-08 15:08:39.949 - INFO: Number of unique elements: 208
+2026-03-08 15:08:39.949 - INFO: Triples sorted: 208 [average speed 2.5 M/s] 
+2026-03-08 15:08:39.950 - INFO: Statistics for SPO: #relations = 52, #blocks = 1, #triples = 208
+2026-03-08 15:08:39.951 - INFO: Statistics for SOP: #relations = 52, #blocks = 1, #triples = 208
+2026-03-08 15:08:39.951 - INFO: Number of distinct patterns: 1
+2026-03-08 15:08:39.951 - INFO: Number of subjects with pattern: 52 [all]
+2026-03-08 15:08:39.951 - INFO: Total number of distinct subject-predicate pairs: 208
+2026-03-08 15:08:39.951 - INFO: Average number of predicates per subject: 4.0
+2026-03-08 15:08:39.951 - INFO: Average number of subjects per predicate: 52
+2026-03-08 15:08:39.951 - INFO: Creating permutations OSP and OPS ...
+2026-03-08 15:08:39.952 - INFO: Triples sorted: 208 [average speed 5.5 M/s] 
+2026-03-08 15:08:39.954 - INFO: Statistics for OSP: #relations = 56, #blocks = 1, #triples = 208
+2026-03-08 15:08:39.954 - INFO: Statistics for OPS: #relations = 56, #blocks = 1, #triples = 208
+2026-03-08 15:08:39.954 - INFO: Adding 52 triples to the POS and PSO permutation for the internal `ql:has-pattern` ...
+2026-03-08 15:08:39.955 - INFO: Creating permutations PSO and POS ...
+2026-03-08 15:08:39.956 - INFO: Number of inputs to `uniqueView`: 52
+2026-03-08 15:08:39.956 - INFO: Number of unique elements: 52
+2026-03-08 15:08:39.956 - INFO: Triples sorted: 52 [average speed 1.8 M/s] 
+2026-03-08 15:08:39.957 - INFO: Statistics for PSO: #relations = 1, #blocks = 1, #triples = 52
+2026-03-08 15:08:39.957 - INFO: Statistics for POS: #relations = 1, #blocks = 1, #triples = 52
+2026-03-08 15:08:39.957 - INFO: Creating permutations PSO and POS ...
+2026-03-08 15:08:39.958 - INFO: Triples sorted: 208 [average speed 6.3 M/s] 
+2026-03-08 15:08:39.958 - INFO: Statistics for PSO: #relations = 4, #blocks = 1, #triples = 208
+2026-03-08 15:08:39.958 - INFO: Statistics for POS: #relations = 4, #blocks = 1, #triples = 208
+2026-03-08 15:08:39.959 - INFO: Index build completed