Skip to content

test: verify #632 segfault fixed by nlp-engine #658 (engine 3.5.6) #31

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
ddehilster merged 1 commit into from
Jun 17, 2026
Merged

test: verify #632 segfault fixed by nlp-engine #658 (engine 3.5.6) #31

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion nlp-engine
Open in desktop
Submodule nlp-engine updated 4 files
+34 −0 .github/workflows/move-assets.yml
+4 −0 include/Api/lite/nlp_engine.h
+58 −40 lite/nlp_engine.cpp
+1 −1 nlp/main.cpp
35 changes: 35 additions & 0 deletions tests/test_engine.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -96,6 +96,41 @@ def test_links_parser(self):
self._run_analyzer("links")


class ReentrancyRegressionTest(TestCase):
"""nlp-engine #658 / #632 regression.

Re-entrant ``analyze()`` across working folders must reuse the loaded
analyzer instead of rebuilding and orphaning it, and switching working
folders (which tears down the prior engine) must not segfault on teardown.

Unlike the fixture-based suites above, these tests assert only that the
calls succeed and -- critically -- that the process survives teardown.
They deliberately do NOT compare against the output fixtures (which drift
with the engine and are regenerated separately), so a clean run here is an
unambiguous signal that the #632 SIGSEGV (exit 139) is gone.
"""

def test_repeated_analyze_same_process(self):
"""Several analyze() calls in one process exercise the re-entrant
addAna path that, pre-#658, logged 'Named analyzer already present'
and orphaned a fresh NLP. Uses the bundled default analyzer."""
for _ in range(3):
xml = NLPPlus.analyze("Hello world!")
self.assertTrue(xml)

def test_working_dir_switch_teardown(self):
"""The #632 repro: switch to a temp working folder, analyze, then let
TemporaryDirectory GC-clean it -- which tears down the prior engine.
Pre-#658 this SIGSEGV'd during teardown; post-fix it exits cleanly."""
with TemporaryDirectory(prefix="test-nlpplus") as tmp:
copytree(DATADIR.parent / "analyzers", Path(tmp) / "analyzers")
copytree(NLPPLUSDIR / "data", Path(tmp) / "data")
NLPPlus.set_working_folder(tmp)
text = read_file(DATADIR / "basic" / "text.txt")
results = NLPPlus.engine.analyze(text, "basic")
self.assertIsNotNone(results)


if __name__ == "__main__":
logging.basicConfig(level=logging.INFO)
main()
Loading