xinmi/.github/workflows/anydoc.yml

name: Anydoc

# Compiles the vendored Rust archive, audits it, and exercises the `anydoc`
# build tag. That is minutes of rustc + cargo-audit, so this workflow stays
# off PRs that only touch migrations, frontend, CLI, or unrelated Go packages.

on:
  push:
    branches: [main]
    paths:
      - "third_party/anydoc-go/**"
      - "scripts/build-anydoc-lib.sh"
      - "internal/infrastructure/docparser/**"
      - ".github/workflows/anydoc.yml"
  pull_request:
    paths:
      - "third_party/anydoc-go/**"
      - "scripts/build-anydoc-lib.sh"
      - "internal/infrastructure/docparser/**"
      - ".github/workflows/anydoc.yml"
  workflow_dispatch:

concurrency:
  group: ${{ github.workflow }}-${{ github.ref }}
  cancel-in-progress: true

permissions:
  contents: read

env:
  GO_VERSION: "1.26"
  GOPROXY: https://proxy.golang.org,direct

jobs:
  anydoc:
    name: anydoc parser engine
    runs-on: ubuntu-latest
    timeout-minutes: 30
    steps:
      - uses: actions/checkout@v6

      - uses: actions/setup-go@v6
        with:
          go-version: ${{ env.GO_VERSION }}
          cache: true

      - uses: dtolnay/rust-toolchain@stable

      - name: Cache cargo-audit
        id: cargo-audit-cache
        uses: actions/cache@v4
        with:
          path: ~/.cargo/bin/cargo-audit
          key: cargo-audit-${{ runner.os }}

      - name: Install cargo-audit
        if: steps.cargo-audit-cache.outputs.cache-hit != 'true'
        run: cargo install cargo-audit --locked

      # The archive is a build artifact, not a committed blob, so CI is what
      # keeps the vendored bindings in third_party/anydoc-go compiling.
      # This also materializes patched-anydoc/ (gitignored); cargo audit below
      # cannot load the crates-io patch until that tree exists.
      - name: Build the anydoc static archive
        run: ./scripts/build-anydoc-lib.sh

      # The vendored bindings pull anydoc and its PDF stack from crates.io, and
      # WeKnora parses untrusted uploads in-process. A Rust advisory there (as
      # in RUSTSEC-2026-0187, an uncatchable abort on a hostile PDF) has to
      # break the build, not wait for someone to notice. Vulnerabilities fail
      # the job; unmaintained-crate warnings only report, since a transitive
      # crate going unmaintained is not something this repo can act on.
      - name: Audit the vendored dependency tree
        working-directory: third_party/anydoc-go
        run: cargo audit

      - name: Vet, test, and build with the engine linked in
        run: |
          go vet -tags anydoc ./internal/infrastructure/docparser/...
          go test -tags anydoc -count=1 ./internal/infrastructure/docparser/...
          go build -tags anydoc ./cmd/server