Add unified API landing page and fix footer layout#1169
Merged
sbryngelson merged 6 commits intoMFlowCode:masterfrom Feb 19, 2026
Merged
Add unified API landing page and fix footer layout#1169sbryngelson merged 6 commits intoMFlowCode:masterfrom
sbryngelson merged 6 commits intoMFlowCode:masterfrom
Conversation
- Add docs/api/ Doxygen target with a landing page linking to Pre-Process, Simulation, and Post-Process API docs - Consolidate sidebar from 4 links to 2: "User Guide" and "API Documentation" (highlights for all API sub-projects) - Unify browser tab title to "MFC" across all Doxygen targets - Fix footer blank space caused by missing #nav-path element that navtree.js needs for content height calculation Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Contributor
|
CodeAnt AI is reviewing your PR. Thanks for using CodeAnt! 🎉We're free for open-source projects. if you're enjoying it, help us grow by sharing. Share on X · |
Contributor
📝 WalkthroughWalkthroughThis PR centralizes API documentation and adds automated architecture documentation: new CMake targets and Python generators produce an API landing page and an architecture.md generated from module metadata; header/footer/templates and linting are updated; per-target readmes are removed and module categories introduced. Changes
Sequence Diagram(s)sequenceDiagram
participant CMake as CMake
participant GenAPI as gen_api_landing.py
participant GenArch as gen_architecture.py
participant Doxygen as Doxygen
participant Lint as lint_docs.py
participant Output as docs/
CMake->>GenAPI: invoke gen_api_landing target
GenAPI-->>Output: write docs/api/readme.md
CMake->>GenArch: invoke gen_architecture target
GenArch-->>Output: write docs/documentation/architecture.md
CMake->>Doxygen: run api_doxygen (depends on gen_api_landing)
Doxygen-->>Output: generate API HTML under /api/
CMake->>Lint: run documentation lint (calls check_module_categories)
Lint-->>CMake: report missing module-category errors
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~25 minutes Possibly related PRs
Suggested labels
Suggested reviewers
Poem
🚥 Pre-merge checks | ✅ 2 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (2 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches
🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
codeant-ai
bot
added
the
size:M
Contributor
Nitpicks 🔍
|
![codeant-ai[bot]](https://avatars.githubusercontent.com/in/646884?s=60&v=4){kind=small}
Contributor
|
CodeAnt AI finished reviewing your PR. |
![cubic-dev-ai[bot]](https://avatars.githubusercontent.com/in/1082092?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Pull request overview
This PR improves the MFC documentation structure by creating a unified API landing page and consolidating navigation. The changes simplify the user experience by reducing the sidebar from 4 links to 2 (User Guide and API Documentation), while the new API landing page at docs/api/ provides organized access to all three sub-project APIs (Pre-Process, Simulation, Post-Process). Additionally, it fixes a footer layout issue caused by missing DOM elements that Doxygen's JavaScript expects for content height calculation.
Changes:
- Added unified API Documentation landing page that links to all three component APIs
- Consolidated sidebar navigation from 4 API links to 1 unified API Documentation link with smart highlighting
- Fixed footer blank space by adding required
#nav-pathelement and hiding it with CSS - Unified browser tab titles to "MFC" across all documentation targets
Reviewed changes
Copilot reviewed 6 out of 6 changed files in this pull request and generated no comments.
Show a summary per file
| File | Description |
|---|---|
| docs/header.html | Updated navigation JavaScript to show 2 links instead of 4; added logic to highlight "API Documentation" when viewing any API sub-project |
| docs/gen_api_landing.py | Extended script to generate unified API landing page at docs/api/readme.md with descriptions and links to all three components |
| docs/footer.html | Added empty #nav-path div required by navtree.js for proper content height calculation |
| docs/custom.css | Added CSS rule to hide the empty nav-path element while preserving its layout function |
| docs/api/readme.md | New API landing page content with component descriptions and navigation links |
| CMakeLists.txt | Added 'api' target to documentation build system and made api_doxygen depend on gen_api_landing |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
🧹 Nitpick comments (2)
docs/gen_api_landing.py (1)
154-180: LGTM — unified landing page generation is clean and consistent with existing patterns.Minor note:
docs/api/readme.mdis both committed to the repo (new file in this PR) and regenerated by this script at build time. Because the content is derived solely from the staticTARGETSdict, the committed and generated versions will stay in sync as long asTARGETSis not changed without re-committing the output. IfTARGETSchanges in the future, the committed snapshot will silently become stale until the next build. Consider addingdocs/api/readme.md(and the analogous per-target files) to.gitignoreand treating them as pure build artifacts, or adding a CI step that diffs the committed file against the generated one.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@docs/gen_api_landing.py` around lines 154 - 180, The generated docs/api/readme.md (produced via the api_lines -> api_out.write_text flow in gen_api_landing.py from the TARGETS dict) is committed but should be treated as a build artifact; either add "docs/api/readme.md" (and per-target counterparts) to .gitignore to avoid committing the derived output, or add a CI check that runs the same generation (invoke the gen_api_landing.py logic that produces api_out from TARGETS) and fails the job if the generated content differs from the committed file, so the repo never drifts silently.docs/header.html (1)
76-90: LGTM — active-state detection correctly covers all four API path segments.The
apiPathslist is a manually maintained allowlist. If a new API sub-project is ever added (e.g.,syscheck), both this array and theGEN_DOCS/navigation in CMakeLists.txt would need to be updated in concert.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@docs/header.html` around lines 76 - 90, The active-state logic currently uses a hardcoded apiPaths array which must be updated manually; instead, remove the apiPaths constant and derive API path segments from the items array itself: when you detect items[i][1] === 'api', parse items[i][0] (the href string like '../<subproj>/index.html') to extract the subproject segment (e.g. '/subproj/') into a local variable or small array and use that to check if window.location.pathname.indexOf(thatSegment) !== -1 before setting a.className = 'active'; update the loop around items and the branch that sets a.className so all API subprojects added to items automatically get active-state detection without touching GEN_DOCS/CMakeLists.txt.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Nitpick comments:
In `@docs/gen_api_landing.py`:
- Around line 154-180: The generated docs/api/readme.md (produced via the
api_lines -> api_out.write_text flow in gen_api_landing.py from the TARGETS
dict) is committed but should be treated as a build artifact; either add
"docs/api/readme.md" (and per-target counterparts) to .gitignore to avoid
committing the derived output, or add a CI check that runs the same generation
(invoke the gen_api_landing.py logic that produces api_out from TARGETS) and
fails the job if the generated content differs from the committed file, so the
repo never drifts silently.
In `@docs/header.html`:
- Around line 76-90: The active-state logic currently uses a hardcoded apiPaths
array which must be updated manually; instead, remove the apiPaths constant and
derive API path segments from the items array itself: when you detect
items[i][1] === 'api', parse items[i][0] (the href string like
'../<subproj>/index.html') to extract the subproject segment (e.g. '/subproj/')
into a local variable or small array and use that to check if
window.location.pathname.indexOf(thatSegment) !== -1 before setting a.className
= 'active'; update the loop around items and the branch that sets a.className so
all API subprojects added to items automatically get active-state detection
without touching GEN_DOCS/CMakeLists.txt.
Contributor
There was a problem hiding this comment.
Actionable comments posted: 3
🧹 Nitpick comments (5)
docs/gen_architecture.py (3)
12-12:from __future__ import annotationsis unnecessary for Python 3.10+.
Path | Noneunion syntax (used on line 61) is natively supported in Python 3.10+ via PEP 604 without this import. Since the project targets Python 3.10+, this shim adds no value.♻️ Proposed fix
-from __future__ import annotations - import json🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@docs/gen_architecture.py` at line 12, Remove the unnecessary future import: delete the "from __future__ import annotations" statement at the top of docs/gen_architecture.py because the project targets Python 3.10+ and the union type syntax used elsewhere (e.g., the "Path | None" annotation on line 61) is natively supported; ensure no other code relies on postponed evaluation before committing.
19-19: Module-levelsrc_dirinitialization runs at import time.This is a side effect that makes the module hard to test and import as a library. The value could instead be resolved inside
main()and passed as a parameter togenerate_module_map()and_find_module_file().🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@docs/gen_architecture.py` at line 19, The module-level src_dir is evaluated at import time causing side effects; move resolution of Path(sys.argv[1]) into main() and pass the resulting src_dir explicitly into generate_module_map() and _find_module_file() (update their signatures to accept a src_dir parameter), removing the global src_dir variable so imports are side-effect free and tests can supply custom paths.
95-104: Two silent failure modes inmain().
- If
architecture.md.inis missing,template.read_text()raisesFileNotFoundErrorwith no user-friendly diagnostic.- If
<!-- MODULE_MAP -->is absent from the template,text.replace(...)is a silent no-op — the generatedarchitecture.mdwould have no module map without any warning.🛡️ Suggested defensive additions
def main() -> None: template = src_dir / "docs" / "documentation" / "architecture.md.in" output = src_dir / "docs" / "documentation" / "architecture.md" + if not template.exists(): + sys.exit(f"ERROR: template not found: {template}") + text = template.read_text(encoding="utf-8") module_map = generate_module_map() + if "<!-- MODULE_MAP -->" not in text: + sys.exit(f"ERROR: placeholder '<!-- MODULE_MAP -->' not found in {template}") text = text.replace("<!-- MODULE_MAP -->", module_map)🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@docs/gen_architecture.py` around lines 95 - 104, In main(), add defensive checks around reading and replacing the template: catch FileNotFoundError from template.read_text and log or print a clear diagnostic referencing the template variable and exit non‑zero; after reading, verify the marker "<!-- MODULE_MAP -->" exists in text before calling text.replace (if missing, log/raise a clear error instead of silently proceeding); only write output.write_text after a successful replacement (use generate_module_map() when replacing) and ensure any other unexpected exceptions are surfaced with a helpful message. This targets the main() function and the template.read_text / text.replace usage to avoid the two silent failure modes.toolchain/mfc/lint_docs.py (2)
847-859:catsstring is recomputed inside the per-error append, not precomputed.
", ".join(e["category"] for e in categories)is evaluated once per uncategorized module. Precompute it before the loop.♻️ Proposed fix
+ cats = ", ".join(e["category"] for e in categories) errors = [] for fpp in sorted(src_dir.rglob("m_*.[fF]*")): if fpp.suffix not in (".fpp", ".f90", ".F90"): continue name = fpp.stem if name not in categorized: rel = fpp.relative_to(repo_root) - cats = ", ".join(e["category"] for e in categories) errors.append(🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@toolchain/mfc/lint_docs.py` around lines 847 - 859, The code recomputes cats = ", ".join(e["category"] for e in categories) inside the loop for every uncategorized module; move that computation out of the loop and compute cats once before iterating over src_dir.rglob, then reference that precomputed cats inside the errors.append block (keep references to errors, fpp, categorized, categories, and repo_root as shown).
838-841: No validation ofmodule_categories.jsonstructure before accessingentry["modules"].If the JSON file is malformed (e.g., missing the
"modules"key), thejson.loadsorentry["modules"]access raises an opaqueJSONDecodeError/KeyErrorwith no actionable hint. The existing file-exists guard is good but doesn't cover structural problems.🛡️ Suggested minimal guard
- categories = json.loads(categories_file.read_text(encoding="utf-8")) - categorized = set() - for entry in categories: - categorized.update(entry["modules"]) + try: + categories = json.loads(categories_file.read_text(encoding="utf-8")) + categorized = {m for entry in categories for m in entry["modules"]} + except (json.JSONDecodeError, KeyError, TypeError) as exc: + return [f" docs/module_categories.json is malformed: {exc}"]🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@toolchain/mfc/lint_docs.py` around lines 838 - 841, The code assumes every entry in the loaded categories JSON has a "modules" key and that json.loads succeeds; wrap the json.loads call in a try/except to catch JSONDecodeError and raise/log a clear error mentioning "module_categories.json", then validate the parsed object before the loop: ensure `categories` is a list and for each `entry` (in the loop that updates `categorized`) verify `isinstance(entry, dict)` and that `"modules"` in entry and `isinstance(entry["modules"], (list, tuple))`; if any check fails, raise or log a ValueError/TypeError with a descriptive message referencing `categories` and the offending `entry` so the caller gets an actionable hint instead of a KeyError/opaque decode error.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@CMakeLists.txt`:
- Around line 851-867: Add an explicit ordering dependency so inject_page_dates
runs after gen_architecture: update the CMake targets to make the
inject_page_dates target (or its stamp/output) depend on gen_architecture (or
the "${CMAKE_CURRENT_BINARY_DIR}/gen-architecture.stamp") and/or add
add_dependencies(inject_page_dates gen_architecture) so documentation_doxygen
cannot run inject-dates.py in parallel with gen_architecture; reference the
existing gen_architecture target and the gen-architecture.stamp output when
adding the dependency.
In `@docs/documentation/architecture.md`:
- Line 9: Update the source template that generates the architecture docs to add
a language specifier to each fenced code block: replace the opening fences for
the ASCII diagrams (e.g., the block that contains "pre_process ──> simulation
──> post_process" and the three other similarly formatted diagram/pseudo‑code
blocks) with ```text so the generated architecture.md uses ```text for those
fences; ensure all four untagged fenced blocks in the template are changed to
use the text specifier.
In `@docs/gen_architecture.py`:
- Around line 81-84: The Markdown table is emitted immediately after the section
heading in generate_module_map(), causing MD058 warnings; update the code that
appends the section header (where lines.append(f"### {cat}") is called) to
insert a blank line before the table (e.g., append an empty string or newline
into the lines list right after the heading) so the heading and the table are
separated by one blank line.
---
Duplicate comments:
In `@docs/documentation/architecture.md`:
- Line 100: The generated module map tables in
docs/documentation/architecture.md are missing surrounding blank lines (MD058);
update the docs/gen_architecture.py generate_module_map() function to emit an
empty line before and after each markdown table it produces (e.g., ensure the
code path that builds the table string for the "Module | Role" section inserts
'\n' or a blank-line entry both preceding and following the table output), and
apply the same blank-line handling for the other table positions referenced
(lines noted in the review) so every generated table is separated by blank
lines.
---
Nitpick comments:
In `@docs/gen_architecture.py`:
- Line 12: Remove the unnecessary future import: delete the "from __future__
import annotations" statement at the top of docs/gen_architecture.py because the
project targets Python 3.10+ and the union type syntax used elsewhere (e.g., the
"Path | None" annotation on line 61) is natively supported; ensure no other code
relies on postponed evaluation before committing.
- Line 19: The module-level src_dir is evaluated at import time causing side
effects; move resolution of Path(sys.argv[1]) into main() and pass the resulting
src_dir explicitly into generate_module_map() and _find_module_file() (update
their signatures to accept a src_dir parameter), removing the global src_dir
variable so imports are side-effect free and tests can supply custom paths.
- Around line 95-104: In main(), add defensive checks around reading and
replacing the template: catch FileNotFoundError from template.read_text and log
or print a clear diagnostic referencing the template variable and exit non‑zero;
after reading, verify the marker "<!-- MODULE_MAP -->" exists in text before
calling text.replace (if missing, log/raise a clear error instead of silently
proceeding); only write output.write_text after a successful replacement (use
generate_module_map() when replacing) and ensure any other unexpected exceptions
are surfaced with a helpful message. This targets the main() function and the
template.read_text / text.replace usage to avoid the two silent failure modes.
In `@toolchain/mfc/lint_docs.py`:
- Around line 847-859: The code recomputes cats = ", ".join(e["category"] for e
in categories) inside the loop for every uncategorized module; move that
computation out of the loop and compute cats once before iterating over
src_dir.rglob, then reference that precomputed cats inside the errors.append
block (keep references to errors, fpp, categorized, categories, and repo_root as
shown).
- Around line 838-841: The code assumes every entry in the loaded categories
JSON has a "modules" key and that json.loads succeeds; wrap the json.loads call
in a try/except to catch JSONDecodeError and raise/log a clear error mentioning
"module_categories.json", then validate the parsed object before the loop:
ensure `categories` is a list and for each `entry` (in the loop that updates
`categorized`) verify `isinstance(entry, dict)` and that `"modules"` in entry
and `isinstance(entry["modules"], (list, tuple))`; if any check fails, raise or
log a ValueError/TypeError with a descriptive message referencing `categories`
and the offending `entry` so the caller gets an actionable hint instead of a
KeyError/opaque decode error.
- Add docs/documentation/architecture.md.in template with hand-written sections (pipeline, data structures, simulation loop, MPI, extending) - Add docs/gen_architecture.py to generate module map from source briefs - Add docs/module_categories.json as single source of truth for groupings - Add linter check: warns when a module is missing from categories JSON - Update contributing guide with step to register new modules - Link architecture page from User Guide sidebar Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
sbryngelson
force-pushed
the
docs-api-landing
branch
from
827a94e to
3d0d843
Compare
New entry on the landing site gallery with thumbnail from the YouTube video (https://youtube.com/watch?v=zDJoe0NYZsQ), run on a single A100 on Phoenix in under 1 hour. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Contributor
There was a problem hiding this comment.
Actionable comments posted: 2
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
toolchain/mfc/lint_docs.py (1)
373-375:⚠️ Potential issue | 🔴 CriticalAdd
"architecture"to the hardcodedpage_idsset — this is the root cause of both pipeline failures.
architecture.mdis gitignored (generated at build time), sodocs/documentation/*.mdglobbing during CI never collects the{#architecture}anchor. All other auto-generated pages (parameters,case_constraints, etc.) are pre-seeded in this set for exactly this reason.architecturemust follow the same pattern.Fix
- page_ids = {"citelist", "parameters", "case_constraints", "physics_constraints", "examples", "cli-reference"} + page_ids = {"citelist", "parameters", "case_constraints", "physics_constraints", "examples", "cli-reference", "architecture"}🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@toolchain/mfc/lint_docs.py` around lines 373 - 375, Add the missing "architecture" entry to the hardcoded page_ids set so generated/ignored architecture.md is treated like other auto-generated pages; update the page_ids declaration (the set assigned to variable page_ids in lint_docs.py) to include "architecture" alongside "citelist", "parameters", "case_constraints", "physics_constraints", "examples", and "cli-reference" so the subsequent doc_dir.glob("*.md") processing/logic recognizes the generated architecture page.
🧹 Nitpick comments (1)
toolchain/mfc/lint_docs.py (1)
847-859:check_module_categoriesmay emit duplicate errors for the same module name across multiplesrc/subdirectories.Several module names (e.g.,
m_global_parameters,m_checker) exist independently undersrc/pre_process/,src/simulation/, andsrc/post_process/. If any such module is missing frommodule_categories.json, the current code reports one error per file path rather than one error per unique module name. Consider deduplicating onnamebefore reporting, and computing thecatsstring once outside the loop.Suggested refactor
+ cats = ", ".join(e["category"] for e in categories) errors = [] - for fpp in sorted(src_dir.rglob("m_*.[fF]*")): - if fpp.suffix not in (".fpp", ".f90", ".F90"): - continue - name = fpp.stem - if name not in categorized: - rel = fpp.relative_to(repo_root) - cats = ", ".join(e["category"] for e in categories) - errors.append( - f" {rel}: module {name} is not in docs/module_categories.json.\n" - f" Fix: open docs/module_categories.json and add \"{name}\" to one of: {cats}.\n" - f" This ensures it appears on the Code Architecture page." - ) + missing = set() + for fpp in sorted(src_dir.rglob("m_*.[fF]*")): + if fpp.suffix not in (".fpp", ".f90", ".F90"): + continue + name = fpp.stem + if name not in categorized: + missing.add(name) + for name in sorted(missing): + errors.append( + f" module {name} is not in docs/module_categories.json." + f" Fix: add \"{name}\" to one of: {cats}." + f" This ensures it appears on the Code Architecture page." + )🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@toolchain/mfc/lint_docs.py` around lines 847 - 859, The loop in check_module_categories collects errors per file path, causing duplicate messages for the same module name across subdirs; fix by deduplicating on the module name and computing the category list string once. Before the for fpp in sorted(...) loop, compute cats = ", ".join(e["category"] for e in categories); then add a seen = set() and inside the loop compute name = fpp.stem; if name in seen: continue; add name to seen and only then check if name not in categorized and append a single error using the first seen relative path (rel = fpp.relative_to(repo_root)) so each missing module is reported once.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@docs/documentation/contributing.md`:
- Around line 539-541: check_page_refs in lint_docs.py fails because
"architecture.md" is gitignored and the function's page_ids set isn't pre-seeded
with "architecture"; update the check_page_refs function to add "architecture"
to the initial page_ids set (same one-line approach used to pre-seed "readme")
so the `@ref` architecture link is recognized during the precheck linter run and
both CI failures are resolved.
In `@docs/documentation/readme.md`:
- Line 27: The CI lint fails because the hardcoded page_ids set in
lint_docs.py::check_page_refs doesn't include the generated anchor for
architecture; update the page_ids set to add "architecture" (the {`#architecture`}
anchor) so check_page_refs will recognize `@ref` architecture in readme.md even
though architecture.md is gitignored at build time; modify the page_ids constant
in check_page_refs to include "architecture" and run the docs lint to verify the
CI gate passes.
---
Outside diff comments:
In `@toolchain/mfc/lint_docs.py`:
- Around line 373-375: Add the missing "architecture" entry to the hardcoded
page_ids set so generated/ignored architecture.md is treated like other
auto-generated pages; update the page_ids declaration (the set assigned to
variable page_ids in lint_docs.py) to include "architecture" alongside
"citelist", "parameters", "case_constraints", "physics_constraints", "examples",
and "cli-reference" so the subsequent doc_dir.glob("*.md") processing/logic
recognizes the generated architecture page.
---
Duplicate comments:
In `@CMakeLists.txt`:
- Around line 851-867: The documentation build can race because
documentation_doxygen depends on both gen_architecture and inject_page_dates
with no ordering; add a dependency so the inject_page_dates target depends on
gen_architecture (i.e., call add_dependencies(inject_page_dates
gen_architecture)) to ensure gen_architecture runs and produces architecture.md
before inject_page_dates runs; update the CMake file near the definitions of the
gen_architecture and inject_page_dates targets so inject_page_dates has an
explicit add_dependencies to gen_architecture while leaving
documentation_doxygen dependencies intact.
---
Nitpick comments:
In `@toolchain/mfc/lint_docs.py`:
- Around line 847-859: The loop in check_module_categories collects errors per
file path, causing duplicate messages for the same module name across subdirs;
fix by deduplicating on the module name and computing the category list string
once. Before the for fpp in sorted(...) loop, compute cats = ",
".join(e["category"] for e in categories); then add a seen = set() and inside
the loop compute name = fpp.stem; if name in seen: continue; add name to seen
and only then check if name not in categorized and append a single error using
the first seen relative path (rel = fpp.relative_to(repo_root)) so each missing
module is reported once.
- Add flow-over-dog entry (4 B200s on HiPerGator, 1h) to landing page and Doxygen gallery with thumbnail - Add shock-helium entry to landing page inline sims array (was only in simulations.json) - Fix nav button grid: grid-cols-7 → grid-cols-6 after API button removal YouTube: https://youtube.com/watch?v=Xhvr2kkQa30 Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Focused ultrasound pulse impinging a phantom kidney stone with stress visualization, by J.-S. Spratt. 10 V100s on Bridges-2, 2h. YouTube: https://youtube.com/watch?v=z8j3NH-Y6i0 Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
architecture.md is gitignored (generated at build time), so the linter glob never finds its anchor. Pre-seed it like the other generated pages to prevent CI failure. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Contributor
There was a problem hiding this comment.
Actionable comments posted: 1
🧹 Nitpick comments (1)
docs/index.html (1)
178-178:text-mdis not a valid Tailwind CSS utility — silently no-ops.
text-mddoes not exist in Tailwind's font-size scale, which ranges fromtext-xsthroughtext-9xlbut skipstext-md. The Play CDN silently ignores unknown utility classes, so elements usingtext-mdfall back to inherited or browser-default sizing. This class appears on lines 170, 178, 219 (asmd:text-md), and 252 (asmd:text-md). The correct equivalent istext-base.The responsive grid improvement on line 178 itself is correct.
♻️ Proposed fix — replace all occurrences of `text-md` with `text-base`
Line 170:
-<div class="w-full flex flex-col text-white text-md md:text-lg text-justify font-semibold gap-3"> +<div class="w-full flex flex-col text-white text-base md:text-lg text-justify font-semibold gap-3">Line 178:
-<div class="justify-center grid grid-cols-2 sm:grid-cols-3 lg:grid-cols-6 gap-x-4 md:gap-x-8 text-md md:text-xl text-center text-white font-medium"> +<div class="justify-center grid grid-cols-2 sm:grid-cols-3 lg:grid-cols-6 gap-x-4 md:gap-x-8 text-base md:text-xl text-center text-white font-medium">Line 219:
-<div class="text-sm md:text-md text-justify"> +<div class="text-sm md:text-base text-justify">Line 252:
-<div class="font-bold text-center text-sm md:text-md md:text-left"> +<div class="font-bold text-center text-sm md:text-base md:text-left">🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@docs/index.html` at line 178, Replace the invalid Tailwind utility "text-md" (and responsive variant "md:text-md") with the correct "text-base" (and "md:text-base") wherever used in the HTML; specifically update the class strings such as the div containing "justify-center grid ... text-md md:text-xl ..." and the other occurrences referenced (lines noted in the review) so all "text-md" become "text-base" and all "md:text-md" become "md:text-base".
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@docs/index.html`:
- Line 58: The HTML includes three simulation entries referencing missing image
assets (image: "res/simulations/t.png" for "Mach 1.5 shock-helium bubble",
image: "res/simulations/u.png" for "Flow over a dog (pressure)", and image:
"res/simulations/v.png" for "Focused ultrasound on kidney stone"), causing
broken images; fix by either adding the three PNG files into the repository at
the referenced paths (res/simulations/t.png, u.png, v.png) or remove/update
those simulation entries in docs/index.html to point to existing images or
placeholders so the simulation cards no longer reference nonexistent assets.
---
Nitpick comments:
In `@docs/index.html`:
- Line 178: Replace the invalid Tailwind utility "text-md" (and responsive
variant "md:text-md") with the correct "text-base" (and "md:text-base") wherever
used in the HTML; specifically update the class strings such as the div
containing "justify-center grid ... text-md md:text-xl ..." and the other
occurrences referenced (lines noted in the review) so all "text-md" become
"text-base" and all "md:text-md" become "md:text-base".
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## master #1169 +/- ##
=======================================
Coverage 44.05% 44.05%
=======================================
Files 70 70
Lines 20498 20498
Branches 1990 1990
=======================================
Hits 9030 9030
Misses 10329 10329
Partials 1139 1139 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
User description
Summary
docs/api/) that links to the Pre-Process, Simulation, and Post-Process API docs#nav-pathelement needed bynavtree.jsfor content height calculationTest plan
./mfc.sh build -j $(nproc) -t documentationdocs/mfc/api/index.htmlshows links to all three sub-projects🤖 Generated with Claude Code
CodeAnt-AI Description
Add a unified API landing page, simplify sidebar, and fix footer/tab titles
What Changed
Impact
✅ Unified API landing page✅ No blank space in footer✅ Consistent browser tab title💡 Usage Guide
Checking Your Pull Request
Every time you make a pull request, our system automatically looks through it. We check for security issues, mistakes in how you're setting up your infrastructure, and common code problems. We do this to make sure your changes are solid and won't cause any trouble later.
Talking to CodeAnt AI
Got a question or need a hand with something in your pull request? You can easily get in touch with CodeAnt AI right here. Just type the following in a comment on your pull request, and replace "Your question here" with whatever you want to ask:
This lets you have a chat with CodeAnt AI about your pull request, making it easier to understand and improve your code.
Example
Preserve Org Learnings with CodeAnt
You can record team preferences so CodeAnt AI applies them in future reviews. Reply directly to the specific CodeAnt AI suggestion (in the same thread) and replace "Your feedback here" with your input:
This helps CodeAnt AI learn and adapt to your team's coding style and standards.
Example
Retrigger review
Ask CodeAnt AI to review the PR again, by typing:
Check Your Repository Health
To analyze the health of your code repository, visit our dashboard at https://app.codeant.ai. This tool helps you identify potential issues and areas for improvement in your codebase, ensuring your repository maintains high standards of code health.
Summary by CodeRabbit
Documentation
Chores
Style