Fix TMDb episode ID handling for appended responses

[Teagan42]

Summary

• compute TMDb episode identifiers when missing from append_to_response payloads
• propagate show identifiers into episode fetch helpers and normalise ids as strings in the shared TMDb episode model
• bump the project version metadata to 2.0.2

Why

• appended TMDb episode payloads omit the id field, which previously caused validation failures when parsing responses

Affects

• loader enrichment episode fetching logic
• shared TMDb episode Pydantic model and related tests
• project version metadata and lockfile

Testing

• uv run ruff check .
• uv run pytest

Documentation

• not needed

---

https://chatgpt.com/codex/tasks/task_e_68e4c17efcd48328bafb1234ee7bc2eb

[github-actions]

Coverage Report

File	Stmts	Miss	Cover	Missing
mcp_plex/common
types.py	166	1	99%	168
validation.py	28	2	93%	37–38
mcp_plex/loader
__init__.py	374	17	95%	84–91, 165, 255–260, 662, 777–779, 781, 867–886
__main__.py	4	4	0%	2–8
imdb_cache.py	43	2	95%	59, 63
mcp_plex/loader/pipeline
channels.py	64	2	97%	134–135
enrichment.py	520	75	86%	240, 263–269, 322, 327, 358, 360, 367, 371, 475, 496, 516, 524–526, 529–531, 534–536, 544–549, 557, 611–612, 636, 696–697, 713, 742, 749, 751, 755, 807, 825, 837, 866, 872, 875–883, 909, 912–914, 931–933, 953–956, 960, 974–980, 1002, 1004, 1007–1009
ingestion.py	155	13	92%	81, 130–150, 185, 222, 238, 331, 337, 357
orchestrator.py	95	5	95%	63, 141, 170, 193–194
persistence.py	130	9	93%	122, 164–165, 175, 186, 190–192, 245
mcp_plex/server
__init__.py	351	28	92%	55–56, 136–137, 185, 202, 240, 261–264, 278–281, 289, 363, 366, 404, 422–423, 461, 491, 509, 515, 533, 632–634, 638
__main__.py	4	4	0%	3–8
cli.py	62	1	98%	73
config.py	82	12	85%	58, 61–69, 76, 82, 102, 116
media.py	85	4	95%	62, 84–85, 95
mcp_plex/server/tools
media_library.py	289	11	96%	477, 760, 783–791
TOTAL	2653	190	93%

Tests	Skipped	Failures	Errors	Time
157	0 💤	0 ❌	0 🔥	1m 28s ⏱️

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting

• @codex fix this CI failure
• @codex address that feedback

[chatgpt-codex-connector]

Update TMDb ID payload schema for string episode IDs

Normalising TMDBEpisode.id to a string means episode payloads can now contain values such as "1/season/2/episode/3", but the loader still creates the data.tmdb.id Qdrant index as an integer and the server casts TMDb identifiers to int before querying. Ingesting an episode after this change will either fail when the integer index receives a non‑integer string or make lookups by TMDb ID impossible because the stored payload no longer matches the integer filter. The index and query logic should be adjusted to accept string IDs (or the model should remain numeric) so episode data stays accessible.

Useful? React with 👍 / 👎.