feat(ol_openedx_uai_content_customization): collapse two-CSV workflow into one

src/ol_openedx_uai_content_customization/CHANGELOG.rst

@@ -1,6 +1,16 @@
 Changelog
 =========
 
+[0.2.0] - 2026-07-07
+---------------------
+
+Changed
+~~~~~~~
+- ``generate_uai_course_versions`` now reads a **single** CSV file instead of
+  two.  The ``--edx-videos-csv`` argument has been removed.  Add an
+  ``edx_video_id`` column directly to the processed-videos CSV to provide the
+  Open edX video UUID for each row.
+
 [0.1.0] - 2026-05-15
 ---------------------
 

src/ol_openedx_uai_content_customization/README.rst

@@ -87,43 +87,39 @@ Usage
 Prerequisites
 ~~~~~~~~~~~~~
 
-You will need two CSV files:
-
-1. **Processed videos CSV** — produced by the video customization
-   workflow. Required columns:
-
-   - ``course_key`` — the Open edX course key of the **source course to
-     clone** (e.g. ``course-v1:UAI_SOURCE+UAI.2+1T2026``).  This course
-     **must already exist** in the CMS modulestore before the command runs.
-     The command validates all source keys up-front and aborts with an error
-     if any are missing.
-   - ``industry`` — one of: ``Healthcare``, ``Finance``, ``Energy``,
-     ``Original``
-   - ``duration`` — ``short`` or ``long``
-   - ``video_file_name`` — file name matching the ``name`` column in the edX videos CSV
-   - ``video_title`` — display name for the subsection/unit/video
-   - ``module_name`` — used to build the course display name
-   - ``course_intro`` — optional introduction content for the generated course.
-     If this value contains HTML tags, it is used as-is. If it is plain text,
-     the command HTML-escapes it and wraps it in ``<p>...</p>``.
-
-     Intro resolution precedence for each generated
-     ``(course_key, industry, duration)`` variant:
-
-     1. exact match on ``(course_key, industry, duration)``
-     2. fallback to ``(course_key, industry)`` (industry-only intro, reused for
-        both short and long)
-     3. fallback to ``(course_key, Original industry)`` (reused across all
-        industries and durations for that source course)
-
-     If no intro is resolved, no Introduction section is created for that
-     variant.
-
-2. **Open edX videos CSV** — exported from Studio / OVS after uploading
-   the customized videos. Required columns:
-
-   - ``name`` — video file name (matches ``video_file_name`` above)
-   - ``video_id`` — the Open edX UUID for the video
+You will need a single CSV file:
+
+**Processed videos CSV** — produced by the video customization
+workflow. Required columns:
+
+- ``course_key`` — the Open edX course key of the **source course to
+  clone** (e.g. ``course-v1:UAI_SOURCE+UAI.2+1T2026``).  This course
+  **must already exist** in the CMS modulestore before the command runs.
+  The command validates all source keys up-front and aborts with an error
+  if any are missing.
+- ``industry`` — one of: ``Healthcare``, ``Finance``, ``Energy``,
+  ``Original``
+- ``duration`` — ``short`` or ``long``
+- ``video_file_name`` — file name of the video (for reference/display)
+- ``video_title`` — display name for the subsection/unit/video
+- ``module_name`` — used to build the course display name
+- ``edx_video_id`` — the Open edX UUID for the video (exported from
+  Studio / OVS after uploading the customized video)
+- ``course_intro`` — optional introduction content for the generated course.
+  If this value contains HTML tags, it is used as-is. If it is plain text,
+  the command HTML-escapes it and wraps it in ``<p>...</p>``.
+
+  Intro resolution precedence for each generated
+  ``(course_key, industry, duration)`` variant:
+
+  1. exact match on ``(course_key, industry, duration)``
+  2. fallback to ``(course_key, industry)`` (industry-only intro, reused for
+     both short and long)
+  3. fallback to ``(course_key, Original industry)`` (reused across all
+     industries and durations for that source course)
+
+  If no intro is resolved, no Introduction section is created for that
+  variant.
 
 Running the Command
 ~~~~~~~~~~~~~~~~~~~
@@ -135,7 +131,6 @@ shell):
 
     python manage.py generate_uai_course_versions \
         --processed-videos-csv /path/to/processed_videos.csv \
-        --edx-videos-csv /path/to/edx_videos.csv \
         [--username studio_worker] \
         [--dry-run]
 
@@ -145,9 +140,6 @@ Options
 ``--processed-videos-csv``
     Path to the processed video metadata CSV file. **Required.**
 
-``--edx-videos-csv``
-    Path to the Open edX videos CSV file. **Required.**
-
 ``--username``
     Username of the platform user under whose authority the courses are
     created. Defaults to ``studio_worker``.

src/ol_openedx_uai_content_customization/ol_openedx_uai_content_customization/constants.py

@@ -35,8 +35,7 @@
 CSV_COL_MODULE_NAME = "module_name"
 CSV_COL_COURSE_INTRO = "course_intro"
 
-CSV_COL_ASSET_NAME = "name"
-CSV_COL_ASSET_VIDEO_ID = "video_id"
+CSV_COL_EDX_VIDEO_ID = "edx_video_id"
 
 REQUIRED_CUSTOMIZED_CSV_COLS = [
     CSV_COL_COURSE_KEY,
@@ -46,9 +45,5 @@
     CSV_COL_VIDEO_TITLE,
     CSV_COL_MODULE_NAME,
     CSV_COL_COURSE_INTRO,
-]
-
-REQUIRED_ASSET_CSV_COLS = [
-    CSV_COL_ASSET_NAME,
-    CSV_COL_ASSET_VIDEO_ID,
+    CSV_COL_EDX_VIDEO_ID,
 ]

src/ol_openedx_uai_content_customization/ol_openedx_uai_content_customization/csv_utils.py

@@ -9,8 +9,6 @@
 from opaque_keys.edx.keys import CourseKey
 
 from ol_openedx_uai_content_customization.constants import (
-    CSV_COL_ASSET_NAME,
-    CSV_COL_ASSET_VIDEO_ID,
     CSV_COL_COURSE_INTRO,
     CSV_COL_COURSE_KEY,
     CSV_COL_DURATION,
@@ -58,21 +56,6 @@ def validate_csv_columns(fieldnames, required_cols, csv_label):
         raise ValueError(msg)
 
 
-def build_video_id_map(video_asset_rows):
-    """
-    Build a mapping of video file name → Open edX video ID.
-
-    Args:
-        video_asset_rows: Rows from the Open edX video asset CSV.
-
-    Returns:
-        dict mapping file name (e.g. "v004_h264.mp4") to video UUID string.
-    """
-    return {
-        row[CSV_COL_ASSET_NAME]: row[CSV_COL_ASSET_VIDEO_ID] for row in video_asset_rows
-    }
-
-
 def resolve_duration_code(duration_value):
     """
     Convert a duration cell value into a Short/Full code.

src/ol_openedx_uai_content_customization/ol_openedx_uai_content_customization/management/commands/generate_uai_course_versions.py

@@ -1,11 +1,10 @@
 """
 Management command: generate_uai_course_versions
 
-Reads two CSV files and uses Open edX modulestore APIs to build industry- and
-length-specific variants of UAI courses by cloning a base course:
+Reads a single CSV file and uses Open edX modulestore APIs to build industry-
+and length-specific variants of UAI courses by cloning a base course.
 
   - Processed videos CSV  (produced by the video-generation workflow)
-  - Open edX videos CSV (exported from Studio / OVS)
 
 For each unique (course_key, industry, duration) combination the command:
     1. Clones the source course (identified by the ``course_key`` CSV column)
@@ -22,7 +21,6 @@
 Usage:
     python manage.py generate_uai_course_versions \\
         --processed-videos-csv /path/to/processed_videos.csv \\
-        --edx-videos-csv /path/to/edx_videos.csv \\
         [--username studio_worker] \\
         [--dry-run]
 
@@ -48,6 +46,7 @@
     BLOCK_TYPE_SEQUENTIAL,
     BLOCK_TYPE_VERTICAL,
     BLOCK_TYPE_VIDEO,
+    CSV_COL_EDX_VIDEO_ID,
     CSV_COL_MODULE_NAME,
     CSV_COL_VIDEO_FILE_NAME,
     CSV_COL_VIDEO_TITLE,
@@ -56,13 +55,11 @@
     INTRO_SUBSECTION_DISPLAY_NAME,
     INTRO_UNIT_DISPLAY_NAME,
     LECTURES_SECTION_DISPLAY_NAME,
-    REQUIRED_ASSET_CSV_COLS,
     REQUIRED_CUSTOMIZED_CSV_COLS,
 )
 from ol_openedx_uai_content_customization.csv_utils import (
     build_course_intro_lookup,
     build_new_course_key,
-    build_video_id_map,
     group_videos_by_course,
     parse_csv,
     resolve_course_intro,
@@ -82,7 +79,7 @@ class Command(BaseCommand):
     """Generate industry/length-specific UAI courses using Open edX modulestore APIs."""
 
     help = (
-        "Generate industry and length-specific UAI courses by reading two CSV files "
+        "Generate industry and length-specific UAI courses by reading a CSV file "
         "and creating courses in the CMS modulestore."
     )
 
@@ -92,11 +89,6 @@ def add_arguments(self, parser):
             required=True,
             help="Path to the processed video metadata CSV file.",
         )
-        parser.add_argument(
-            "--edx-videos-csv",
-            required=True,
-            help="Path to the Open edX videos CSV file (exported from Studio/OVS).",
-        )
         parser.add_argument(
             "--username",
             default="studio_worker",
@@ -111,7 +103,6 @@ def add_arguments(self, parser):
 
     def handle(self, *args, **options):  # noqa: ARG002, PLR0915
         processed_videos_csv = options["processed_videos_csv"]
-        edx_videos_csv = options["edx_videos_csv"]
         username = options["username"]
         dry_run = options["dry_run"]
 
@@ -129,28 +120,17 @@ def handle(self, *args, **options):  # noqa: ARG002, PLR0915
         processed_video_rows, processed_video_fieldnames = parse_csv(
             processed_videos_csv
         )
-        edx_video_rows, edx_video_fieldnames = parse_csv(edx_videos_csv)
 
         try:
             validate_csv_columns(
                 processed_video_fieldnames,
                 REQUIRED_CUSTOMIZED_CSV_COLS,
                 "processed videos CSV",
             )
-            validate_csv_columns(
-                edx_video_fieldnames,
-                REQUIRED_ASSET_CSV_COLS,
-                "edX videos CSV",
-            )
         except ValueError as exc:
             raise CommandError(str(exc)) from exc
 
-        video_id_map = build_video_id_map(edx_video_rows)
-
-        self.stdout.write(
-            f"Loaded {len(processed_video_rows)} processed video rows "
-            f"and {len(edx_video_rows)} edX video rows."
-        )
+        self.stdout.write(f"Loaded {len(processed_video_rows)} processed video rows.")
 
         course_groups = group_videos_by_course(processed_video_rows)
         course_intro_lookup = build_course_intro_lookup(processed_video_rows)
@@ -196,8 +176,10 @@ def handle(self, *args, **options):  # noqa: ARG002, PLR0915
                 for video in videos:
                     vid_file = video[CSV_COL_VIDEO_FILE_NAME]
                     vid_title = video[CSV_COL_VIDEO_TITLE]
-                    mapped_id = video_id_map.get(vid_file, "<NOT FOUND>")
-                    self.stdout.write(f"    - {vid_title} ({vid_file} -> {mapped_id})")
+                    edx_video_id = video.get(CSV_COL_EDX_VIDEO_ID, "<NOT SET>")
+                    self.stdout.write(
+                        f"    - {vid_title} ({vid_file} -> {edx_video_id})"
+                    )
                 skipped += 1
                 continue
 
@@ -208,7 +190,6 @@ def handle(self, *args, **options):  # noqa: ARG002, PLR0915
                     display_name,
                     intro_by_course_key[new_course_key],
                     videos,
-                    video_id_map,
                     user,
                 )
                 created += 1
@@ -273,7 +254,6 @@ def _create_course(  # noqa: PLR0913
         display_name,
         course_intro,
         videos,
-        video_id_map,
         user,
     ):
         """
@@ -353,7 +333,7 @@ def _create_course(  # noqa: PLR0913
             for video in videos:
                 title = video[CSV_COL_VIDEO_TITLE]
                 vid_file = video[CSV_COL_VIDEO_FILE_NAME]
-                edx_video_id = video_id_map.get(vid_file)
+                edx_video_id = video.get(CSV_COL_EDX_VIDEO_ID)
 
                 if not edx_video_id:
                     log.warning(

src/ol_openedx_uai_content_customization/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "ol-openedx-uai-content-customization"
-version = "0.1.2"
+version = "0.2.0"
 description = "An Open edX plugin to generate industry/length-specific UAI custom courses from video CSV data."
 authors = [
   {name = "MIT Office of Digital Learning"}

src/ol_openedx_uai_content_customization/tests/test_csv_utils.py

@@ -4,7 +4,6 @@
 from ol_openedx_uai_content_customization.csv_utils import (
     build_course_intro_lookup,
     build_new_course_key,
-    build_video_id_map,
     group_videos_by_course,
     normalize_course_intro,
     parse_csv,
@@ -16,27 +15,29 @@
 
 def test_parse_csv_returns_list_of_dicts(tmp_path):
     """Each row in the CSV is returned as a dict keyed by column header."""
-    csv_text = "name,video_id\nv004_h264.mp4,abc-123\nv005_h264.mp4,def-456\n"
-    csv_file = tmp_path / "assets.csv"
+    csv_text = (
+        "video_file_name,edx_video_id\nv004_h264.mp4,abc-123\nv005_h264.mp4,def-456\n"
+    )
+    csv_file = tmp_path / "videos.csv"
     csv_file.write_text(csv_text)
 
     rows, fieldnames = parse_csv(str(csv_file))
 
     assert len(rows) == 2  # noqa: PLR2004
-    assert rows[0]["name"] == "v004_h264.mp4"
-    assert rows[0]["video_id"] == "abc-123"
-    assert fieldnames == ["name", "video_id"]
+    assert rows[0]["video_file_name"] == "v004_h264.mp4"
+    assert rows[0]["edx_video_id"] == "abc-123"
+    assert fieldnames == ["video_file_name", "edx_video_id"]
 
 
 def test_parse_csv_empty_file(tmp_path):
     """A CSV with only a header row returns an empty row list but keeps the headers."""
     csv_file = tmp_path / "empty.csv"
-    csv_file.write_text("name,video_id\n")
+    csv_file.write_text("video_file_name,edx_video_id\n")
 
     rows, fieldnames = parse_csv(str(csv_file))
 
     assert rows == []
-    assert fieldnames == ["name", "video_id"]
+    assert fieldnames == ["video_file_name", "edx_video_id"]
 
 
 @pytest.mark.parametrize(
@@ -58,27 +59,6 @@ def test_validate_csv_columns(fieldnames, required, should_raise):
     validate_csv_columns(fieldnames, required, "test CSV")
 
 
-@pytest.mark.parametrize(
-    ("rows", "expected"),
-    [
-        (
-            [
-                {"name": "v004_h264.mp4", "video_id": "abc-123"},
-                {"name": "v005_h264.mp4", "video_id": "def-456"},
-            ],
-            {
-                "v004_h264.mp4": "abc-123",
-                "v005_h264.mp4": "def-456",
-            },
-        ),
-        ([], {}),
-    ],
-)
-def test_build_video_id_map(rows, expected):
-    """Map video asset rows to edX video IDs."""
-    assert build_video_id_map(rows) == expected
-
-
 @pytest.mark.parametrize(
     ("duration_value", "expected"),
     [
@@ -164,6 +144,7 @@ def _make_row(course_key, industry, duration, video_file="v001.mp4", title="Titl
         "video_title": title,
         "module_name": "Module 2",
         "course_intro": "",
+        "edx_video_id": "aaaaaaaa-0000-0000-0000-aaaaaaaaaaaa",
     }