Python: Add test for string encoding dataset check

tausbn · tausbn · commit d01593e57126 · 2024-10-21T12:08:46.000Z
Note that this test checks that the current setup creates dataset check
violations. A later commit will fix this (and flip the negation in the
test).
diff --git a/python/extractor/cli-integration-test/string-encoding/repo_dir/test.py b/python/extractor/cli-integration-test/string-encoding/repo_dir/test.py
@@ -0,0 +1,2 @@
+"\uD800"
+"?"
diff --git a/python/extractor/cli-integration-test/string-encoding/test.sh b/python/extractor/cli-integration-test/string-encoding/test.sh
@@ -0,0 +1,19 @@
+#!/bin/bash
+
+set -Eeuo pipefail # see https://vaneyckt.io/posts/safer_bash_scripts_with_set_euxo_pipefail/
+
+set -x
+
+CODEQL=${CODEQL:-codeql}
+
+SCRIPTDIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" >/dev/null 2>&1 && pwd )"
+cd "$SCRIPTDIR"
+
+rm -rf db
+
+$CODEQL database create db --language python --source-root repo_dir/
+
+# Note the negation in front -- it witnesses the fact that currently the dataset check FAILS.
+! $CODEQL dataset check db/db-python
+
+echo "Test successfully completed."
