The dashboard shell no longer inlines ~450 KB of scripts and CSS: two fixed routes, GET /config_tester/dashboard_bundle_js and _css (?v=<content hash>, weak ETag, 304, gzip, Cache-Control immutable when the hash matches, no-cache otherwise), serve the concatenated resources (pinned order, standalone last; rebuilt when a resource file changes, so F5 hot reload still works). The shell drops from ~527 KB to ~68 KB (12 KB gzipped) and repeat loads cost ~0 KB; the node iframe (srcdoc inherits the parent's base URL) uses the same routes, so ComfyUI's executed/history payloads shrink too. inline_assets=True (implied by inline_manifest, or USCG_INLINE_DASHBOARD_ASSETS=1) keeps a fully self-contained page. The v query is only compared, never used for lookup; no path parameters. Tests: 16 unit (order, page independence, inline == bundled bytes, hash rebuild, headers, v-is-never-a-path, node --check of globals + bundle), e2e (top-level + srcdoc iframe + inline + 500 path), CI assembled-page step covers globals + bundle as one script. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
9.0 KiB
Contributing
PRs are welcome and encouraged. This project is actively developed — if you have a bug fix, new feature, or quality improvement in mind, open an issue to discuss it first (for larger changes) or just submit a PR directly (for clear-cut fixes). The maintainer responds promptly.
Quick orientation
Before diving in, read these files in the repo root:
| File | What it covers |
|---|---|
README.md |
What the project does, how the three nodes connect, and user-facing features |
ProjectStructure.md |
Every file's purpose, all API routes (method, body, purpose), data shapes, communication patterns, dependency graph, and 36 development gotchas |
AI-CONTEXT.md |
Internal dev notes: architecture layers, critical codepaths, field naming rules, config data flow, common tasks |
Dev setup
Requirements:
- ComfyUI 0.12.3 or newer
- Python 3.10+ (whatever version your ComfyUI install uses)
Install:
ComfyUI/custom_nodes/
└── ComfyUI-Ultimate-Auto-Sampler-Config-Grid-Testing-Suite/ ← clone here
Clone into ComfyUI/custom_nodes/ and restart ComfyUI. The node will register itself automatically.
Python dependencies:
pip install -r requirements.txt
requirements.txt currently pulls in diffusers and piexif. ComfyUI itself provides PyTorch, numpy, and Pillow.
Running tests
# Python unit tests (ComfyUI + torch are stubbed): needs pytest, numpy, pillow
pip install -r tests/requirements-ci.txt
python -m pytest tests/ -q -ra --ignore=tests/e2e # 499 passed, 16 skipped without torch (measured 11/4/26)
# With real torch (e.g. ComfyUI's python), the 16 `requires_torch` tests run too.
# USCG_REQUIRE_TORCH=1 makes them fail instead of skip if torch is missing:
USCG_REQUIRE_TORCH=1 python -m pytest tests/ -q -ra --ignore=tests/e2e
# Dashboard JS unit tests (node:test, no browser)
node --test tests/js/
# Dashboard SPA browser smoke test (Playwright, Chromium, no ComfyUI needed)
pip install -r tests/e2e/requirements.txt
python -m playwright install chromium
USCG_REQUIRE_E2E=1 python -m pytest tests/e2e -q -ra --browser chromium
GitHub Actions (.github/workflows/tests.yml) runs four jobs on every push and PR:
| Job id | What it runs |
|---|---|
pytest-stubs |
the Python tests with ComfyUI and torch stubbed (asserts torch is NOT installed) |
pytest-torch |
the same tests with real CPU torch and USCG_REQUIRE_TORCH=1 |
js-syntax |
JS syntax checks (see below), the assembled Dashboard page, and node --test tests/js/ |
e2e-dashboard |
the Playwright smoke test (USCG_REQUIRE_E2E=1, Chromium) |
Tests live in tests/ and must not contain network or process-spawning code (no socket, http.server, subprocess, urllib): tests/ ships in the registry zip and is yara-scanned.
tests/test_docs_consistency.py keeps the docs honest: it fails when a route, event, source file or CI job exists in the repo but is missing from ProjectStructure.md / this file, when a README anchor link does not resolve, or when the TOC misses a section. Update the doc it names.
PRs that change Python logic should add tests where it's reasonable to do so.
JS syntax check
The Builder UI and Dashboard are plain JavaScript (no build step). CI runs these checks (the js-syntax job); run the same loops locally from the repo root:
# web/**/*.js are ES modules
for f in $(find web -name '*.js'); do
node --input-type=module --check < "$f" || { echo "FAIL $f"; break; }
done
# resources/*.js are plain scripts
for f in resources/*.js; do
node --check "$f" || { echo "FAIL $f"; break; }
done
The Dashboard files in resources/ share ONE script scope (see JS_FILES in html_generator.py; the page loads them as one bundle after a small globals script), so a duplicate top-level name only shows up in the assembled page. The js-syntax job therefore also builds the page with html_generator.get_html_template(..., inline_assets=True) and syntax-checks every inline script body, plus the globals script + bundle joined as one script (the snippet is in .github/workflows/tests.yml).
Architecture notes worth knowing before editing
The Builder UI is a DOM widget on the Python node. Its entire state is serialized to a single lora_config STRING widget as JSON. Python's generate_config() reads only that widget — all other INPUT_TYPES entries are vestigial.
State → configs_json transformation: All state-to-output transformation logic lives in state_to_configs_json (a @staticmethod on UltimateConfigBuilder in config_builder_node.py). Both runtime (generate_config) and the preview endpoint (POST /configbuilder/preview) call this single function. Do not add a JS-side equivalent transformer — that pattern existed historically (convertStateToConfigs) and was a recurring source of drift bugs; it was deliberately removed in favor of the single source of truth in Python.
When adding a new state field that should appear in configs_json:
- Add the field to
state_to_configs_jsoninconfig_builder_node.py. Make sure tests intests/test_state_to_configs_json.pycover the new field. - Update the Builder UI in
web/conf_builder/conf-builder-config-management.jsto read/write the field vianode.state.<field>. Also add the default in the initial state object inconf-builder-main.js(search forthis.state = {, around line 103), and add a migration backfill in the async init block below it (conf-builder-main.js, around line 484) so existing saved workflows pick up the default. - The preview panel will pick up the new field automatically — it just renders whatever
state_to_configs_jsonreturns.
Manifest field names — check before touching favorite/reject logic:
favorited(notfavorite) — used in__init__.pyroutesnote(notnotes) — per-item annotationfile— full/view?...URL, not a filenamerejected— required boolean for dashboard displayid— timestamp-based integer:int(time.time() * 100000) + random.randint(0, 1000)
Manifest items are inserted at index 0 (newest-first order).
ComfyUI V3 vs V1 nodes: Use the _call_node and _unwrap helpers when calling V3 nodes outside the executor pipeline. See ltx_video_generation.py for the established pattern.
Windows print safety: Two modules (config_builder_node.py, ltx_video_generation.py) shadow print with safe_print at module level (print = safe_print) to avoid colorama/stdout corruption that throws OSError [Errno 22] on long-running sessions. If you add new print-heavy code in those files, use the existing print = safe_print shadow at module top. Other modules use bare print() directly.
Distributed mode exists. Workers consume the already-rendered configs_json over the wire. Don't assume single-machine execution in generation orchestrator code.
ComfyUI Registry security constraints (violations get the version flagged and hidden from Manager):
- No outbound HTTP in this package (no
requests,urllib.request,http.client,socket,aiohttp.ClientSession): network code lives in the companion plugins; dev-only HTTP scripts go underdocs/tools/(gitignored) - No
subprocess,os.system,eval(), orexec() - No runtime
pip install - No custom file-serving endpoints — use ComfyUI's
/viewendpoint for images andWEB_DIRECTORYfor JS - All filesystem endpoints must validate paths with
_is_path_within()and sanitize user-supplied names
Dashboard JS changes require HTML regeneration. The resources/ files are inlined into a single HTML string by html_generator.py. After editing dashboard JS/CSS, you must re-run the sampler node or call POST /config_tester/get_session_html to see the changes. Browser refresh alone is not enough.
PR checklist
Before submitting:
python -m pytest tests/ -q --ignore=tests/e2epasses (andnode --test tests/js/)- JS syntax checks pass (the two loops above, plus the assembled-page check for
resources/changes) - New routes, events or source files are added to
ProjectStructure.md(tests/test_docs_consistency.pychecks this) - No regressions in existing features
- For new config fields:
state_to_configs_jsonupdated in Python, default state and migration backfill added inconf-builder-main.js, and UI read/write added inconf-builder-config-management.js - For user-facing features:
README.mdupdated - For Builder UI or Dashboard changes: smoke-tested in real ComfyUI (unit tests alone are not sufficient here)
- No new
import requests,subprocess,os.system,eval, orexeccalls (Registry security)
Where to discuss
- Bugs and feature requests: GitHub Issues
- Code changes: GitHub Pull Requests
The maintainer (Jason Hoku) actively monitors both.
License
This project is open source. The license is declared in pyproject.toml (license = {file = "LICENSE"}). A LICENSE file is referenced there but not yet present in the repo root — check the GitHub repository page for the current license terms.