#!/usr/bin/env bash
# Branch, PR body, and commit contract checker.
# v2: exits 1 on violation. Consumers that want warn-only set
# continue-on-error: true on the job (or stay on @v1).
set -euo pipefail

BRANCH="${HEAD_BRANCH:-}"
TITLE="${PR_TITLE:-}"
AUTHOR="${PR_AUTHOR:-}"
BASE="${BASE_SHA:-}"
HEAD="${HEAD_SHA:-}"
SERVER_URL="${GITEA_SERVER_URL:-}"
TOKEN="${GITEA_TOKEN:-}"
REPOSITORY="${GITEA_REPOSITORY:-}"
PR_NUMBER="${GITEA_PR_NUMBER:-}"

# Break-glass: dfritz is exempt from all naming checks.
if [ "${AUTHOR}" = "dfritz" ]; then
  echo "check-naming: dfritz break-glass — exempt"
  exit 0
fi

FAILED=0
BRANCH_KIND="invalid"
BRANCH_BUG=""

# Gitea's renderer is the visibility contract. Validate its output instead of
# maintaining a second Markdown parser in this action.
RENDERED_BODY=""
BODY=""
if [ -z "${SERVER_URL}" ] || [ -z "${TOKEN}" ] ||
  ! printf '%s' "${REPOSITORY}" | grep -qE '^[[:alnum:]_.-]+/[[:alnum:]_.-]+$' ||
  ! printf '%s' "${PR_NUMBER}" | grep -qE '^[1-9][0-9]*$'; then
  echo "FAIL[check-naming]: server-url, token, repository, and PR number are required to load the PR body"
  FAILED=1
elif ! BODY=$(curl --fail --silent --show-error \
  --connect-timeout 5 \
  --max-time 15 \
  --header "Authorization: token ${TOKEN}" \
  "${SERVER_URL%/}/api/v1/repos/${REPOSITORY}/pulls/${PR_NUMBER}" |
  jq -er '.body | if type == "string" then . else error("body is not a string") end'); then
  echo "FAIL[check-naming]: Gitea could not load the PR body"
  FAILED=1
elif ! RENDERED_BODY=$(printf '%s' "${BODY}" |
  jq -Rs '{Text: ., Mode: "gfm"}' |
  curl --fail --silent --show-error \
    --connect-timeout 5 \
    --max-time 15 \
    --header "Authorization: token ${TOKEN}" \
    --header "Content-Type: application/json" \
    --data-binary @- "${SERVER_URL%/}/api/v1/markdown"); then
  echo "FAIL[check-naming]: Gitea could not render the PR body"
  FAILED=1
fi

# A collapsed disclosure is valid supporting content, but provenance inside one
# is not visible by default. Remove only collapsed details subtrees; an open
# disclosure remains visible unless one of its ancestors is collapsed. Media
# fallback subtrees are likewise hidden when the browser supports the element.
VISIBLE_BODY=$(printf '%s\n' "${RENDERED_BODY}" |
  awk '
    {
      line=$0 "\n"
      for (i=1; i<=length(line); i++) {
        char=substr(line, i, 1)
        if (in_tag) {
          tag=tag char
          if (char == ">") {
            lower=tolower(tag)
            if (lower ~ /^<(video|audio)([[:space:]>])/) media_depth++
            else if (lower ~ /^<\/(video|audio)([[:space:]>])/) {
              if (media_depth > 0) media_depth--
            }
            else if (lower ~ /^<details([[:space:]>])/) {
              attributes=lower
              gsub(/"[^"]*"/, "", attributes)
              parent_hidden=(details_depth > 0 && hidden[details_depth])
              details_depth++
              hidden[details_depth]=(parent_hidden || attributes !~ /[[:space:]]open([[:space:]=>]|$)/)
            }
            else if (lower ~ /^<\/details([[:space:]>])/) {
              if (details_depth > 0) {
                delete hidden[details_depth]
                details_depth--
              }
            }
            else if (!hidden[details_depth] && media_depth == 0) printf "%s", tag
            in_tag=0
            tag=""
          }
        }
        else if (char == "<") {
          in_tag=1
          tag=char
        }
        else if (!hidden[details_depth] && media_depth == 0) printf "%s", char
      }
    }
  ')

hidden_content_guidance='move ## Tracking and ## Attribution outside collapsed <details> or <video>/<audio> fallback content'

# ---- branch form ----
# <role>/bug-<id>/<kebab>
if echo "${BRANCH}" | grep -qE "^(dev|ux|ops|security|perf|architect|support)/bug-[a-z0-9]+/[a-z0-9][a-z0-9-]*$"; then
  BRANCH_KIND="role-bug"
  BRANCH_BUG=$(echo "${BRANCH}" | sed -E 's|^[^/]+/(bug-[a-z0-9]+)/.*|\1|')
# chore/<kebab>
elif echo "${BRANCH}" | grep -qE "^chore/[a-z0-9][a-z0-9-]*$"; then
  BRANCH_KIND="chore"
else
  echo "FAIL[check-naming]: branch '${BRANCH}' does not match convention"
  echo "  expected: <role>/bug-<id>/<kebab>  (role: dev|ux|ops|security|perf|architect|support)"
  echo "         or: chore/<kebab>"
  FAILED=1
fi

# ---- title form ----
TITLE_BUG=""
if echo "${TITLE}" | grep -qE "^\[bug-[a-z0-9]+\] ."; then
  TITLE_BUG=$(echo "${TITLE}" | sed -E 's|^\[(bug-[a-z0-9]+)\].*|\1|')
fi

if [ "${BRANCH_KIND}" = "role-bug" ]; then
  if [ -z "${TITLE_BUG}" ]; then
    echo "FAIL[check-naming]: title must lead with [${BRANCH_BUG}]"
    echo "  have: ${TITLE}"
    echo "  want: [${BRANCH_BUG}] ${TITLE}"
    FAILED=1
  elif [ "${TITLE_BUG}" != "${BRANCH_BUG}" ]; then
    echo "FAIL[check-naming]: bug-id mismatch — branch carries '${BRANCH_BUG}' but title carries '${TITLE_BUG}'"
    FAILED=1
  elif ! printf '%s\n' "${TITLE}" | grep -qE "^\\[${BRANCH_BUG}\\] [a-z][a-z0-9-]*\\([a-z0-9][a-z0-9-]*\\)!?: .+"; then
    echo "FAIL[check-naming]: Bug-backed title must use '[${BRANCH_BUG}] type(scope): description'"
    FAILED=1
  fi

  tracking=$(printf '%s\n' "${VISIBLE_BODY}" | awk '
    /<h2[^>]*>Tracking<\/h2>/ { in_section=1; next }
    /<h2[^>]*>/ && in_section { exit }
    in_section { print }
  ')
  if [ -z "${tracking}" ]; then
    echo "FAIL[check-naming]: PR body must contain a non-empty ## Tracking section visible by default; ${hidden_content_guidance}"
    FAILED=1
  else
    if ! printf '%s\n' "${tracking}" | grep -qE "(^|[[:space:]>])Fixes[[:space:]]+${BRANCH_BUG}([^a-z0-9]|$)"; then
      echo "FAIL[check-naming]: ## Tracking must contain the literal token 'Fixes ${BRANCH_BUG}'"
      FAILED=1
    fi
    if ! printf '%s\n' "${tracking}" | grep -Fq "https://agenthub.fritzlab.net/${BRANCH_BUG}"; then
      echo "FAIL[check-naming]: ## Tracking must link https://agenthub.fritzlab.net/${BRANCH_BUG}"
      FAILED=1
    fi
  fi

elif [ "${BRANCH_KIND}" = "chore" ] && [ -n "${TITLE_BUG}" ]; then
  echo "FAIL[check-naming]: chore branch should not carry a [bug-id] title prefix"
  FAILED=1
fi

if [ "${BRANCH_KIND}" != "invalid" ]; then
  attribution=$(printf '%s\n' "${VISIBLE_BODY}" | awk '
    /<h2[^>]*>Attribution<\/h2>/ { in_section=1; next }
    /<h2[^>]*>/ && in_section { exit }
    in_section { print }
  ')
  watermark_re='Authored-By: .+ \(.+\) <a href="mailto:noreply@[[:alnum:].-]+"[^>]*>noreply@[[:alnum:].-]+</a>'
  if ! printf '%s\n' "${attribution}" | grep -qE "${watermark_re}"; then
    echo "FAIL[check-naming]: visible ## Attribution must contain an Authored-By product/model watermark; ${hidden_content_guidance}"
    FAILED=1
  fi

  if ! printf '%s\n%s\n' "${BASE}" "${HEAD}" | grep -qEv '^[0-9a-f]{40,64}$'; then
    for revision in "${BASE}" "${HEAD}"; do
      if ! git cat-file -e "${revision}^{commit}" 2>/dev/null; then
        echo "FAIL[check-naming]: base-sha and head-sha must name available commits"
        FAILED=1
        break
      fi
    done
    if git cat-file -e "${BASE}^{commit}" 2>/dev/null && git cat-file -e "${HEAD}^{commit}" 2>/dev/null; then
      commit_count=0
      while IFS= read -r commit; do
        [ -n "${commit}" ] || continue
        commit_count=$((commit_count + 1))
        message=$(git log -1 --format=%B "${commit}")
        if ! printf '%s\n' "${message}" | awk '
          { line[NR]=$0 }
          END {
            n=NR
            while (n > 0 && line[n] == "") n--
            if (n < 3 || line[n-1] != "" || line[n] !~ /^Authored-By: .+ \(.+\) <noreply@[[:alnum:].-]+>$/) exit 1
          }
        '; then
          echo "FAIL[check-naming]: commit ${commit} must end with an Authored-By product/model trailer"
          FAILED=1
        fi
      done < <(git rev-list --reverse "${BASE}..${HEAD}" 2>/dev/null)
      if [ "${commit_count}" -eq 0 ]; then
        echo "FAIL[check-naming]: base-sha..head-sha contains no PR commits"
        FAILED=1
      fi
    fi
  else
    echo "FAIL[check-naming]: base-sha and head-sha must be lowercase hexadecimal commit SHAs"
    FAILED=1
  fi
fi

if [ "${FAILED}" -eq 0 ]; then
  echo "check-naming: ok"
fi

exit "${FAILED}"