Merge pull request #356 from bbrowning/chunking_testing

Consolidate test sample documents into one subdir
instructlab · Nov 11, 2024 · 9173237 · 9173237
2 parents 119d309 + 93fae83
commit 9173237
Show file tree

Hide file tree

Showing 3 changed files with 4 additions and 3 deletions.
diff --git a/tests/functional/test_chunkers.py b/tests/functional/test_chunkers.py
@@ -5,14 +5,15 @@
 # First Party
 from instructlab.sdg.utils.chunkers import DocumentChunker
 
-TEST_DATA_DIR = os.path.join(os.path.dirname(__file__), "testdata")
+TEST_DATA_DIR = os.path.join(os.path.dirname(__file__), "..", "testdata")
 
 
 def test_chunk_pdf(tmp_path):
+    pdf_path = Path(os.path.join(TEST_DATA_DIR, "sample_documents", "phoenix.pdf"))
     leaf_node = [
         {
             "documents": ["Lorem ipsum"],
-            "filepaths": [Path(os.path.join(TEST_DATA_DIR, "phoenix.pdf"))],
+            "filepaths": [pdf_path],
             "taxonomy_path": "knowledge",
         }
     ]
@@ -33,7 +34,7 @@ def test_chunk_pdf(tmp_path):
 
 
 def test_chunk_md(tmp_path):
-    markdown_path = Path(os.path.join(TEST_DATA_DIR, "phoenix.md"))
+    markdown_path = Path(os.path.join(TEST_DATA_DIR, "sample_documents", "phoenix.md"))
     leaf_node = [
         {
             "documents": [markdown_path.read_text(encoding="utf-8")],

diff --git a/tests/functional/testdata/phoenix.md → tests/testdata/sample_documents/phoenix.md b/tests/functional/testdata/phoenix.md → tests/testdata/sample_documents/phoenix.md
diff --git a/tests/functional/testdata/phoenix.pdf → tests/testdata/sample_documents/phoenix.pdf b/tests/functional/testdata/phoenix.pdf → tests/testdata/sample_documents/phoenix.pdf