# ==================================================================== #
# TITLE:                                                               #
# AMR: An R Package for Working with Antimicrobial Resistance Data     #
#                                                                      #
# SOURCE CODE:                                                         #
# https://github.com/msberends/AMR                                     #
#                                                                      #
# PLEASE CITE THIS SOFTWARE AS:                                        #
# Berends MS, Luz CF, Friedrich AW, et al. (2022).                     #
# AMR: An R Package for Working with Antimicrobial Resistance Data.    #
# Journal of Statistical Software, 104(3), 1-31.                       #
# https://doi.org/10.18637/jss.v104.i03                                #
#                                                                      #
# Developed at the University of Groningen and the University Medical  #
# Center Groningen in The Netherlands, in collaboration with many      #
# colleagues from around the world, see our website.                   # 
#                                                                      #
# This R package is free software; you can freely use and distribute   #
# it for both personal and commercial purposes under the terms of the  #
# GNU General Public License version 2.0 (GNU GPL-2), as published by  #
# the Free Software Foundation.                                        #
# We created this package for both routine data analysis and academic  #
# research and it was publicly released in the hope that it will be    #
# useful, but it comes WITHOUT ANY WARRANTY OR LIABILITY.              #
#                                                                      #
# Visit our website for the full manual and a complete tutorial about  #
# how to conduct AMR data analysis: https://amr-for-r.org              #
# ==================================================================== #

on:
  push:
    branches: "main"

name: Update TODO Tracker

jobs:
  update-todo:
    runs-on: ubuntu-latest

    steps:
      - uses: actions/checkout@v4
        with:
          fetch-depth: 0  # full history required for git blame

      - name: Generate TODO report
        env:
          GH_TOKEN: ${{ secrets.GH_REPO_SCOPE }}
        run: |
          set -euo pipefail
          export TZ=Europe/Amsterdam

          REPO="msberends/AMR"
          REPO_URL="https://github.com/$REPO/blob/main"
          NOW=$(date +%s)
          LAST_UPDATED=$(date +"%e %B %Y %H:%M:%S %Z" | sed 's/^ *//')
          STALE_DAYS=180

          # ── helper: human-readable age ──────────────────────────────
          format_age() {
            local d=$1
            if [ "$d" -lt 0 ] 2>/dev/null; then echo "unknown"; return; fi
            local y=$((d / 365)) m=$(( (d % 365) / 30 ))
            if [ "$y" -gt 0 ] && [ "$m" -gt 0 ]; then echo "${y}y ${m}m"
            elif [ "$y" -gt 0 ]; then echo "${y}y"
            elif [ "$m" -gt 0 ]; then echo "${m}m"
            else echo "${d}d"
            fi
          }

          export -f format_age

          # ── step 1: find all markers ────────────────────────────────
          grep -rn \
            --include='*.R' --include='*.Rmd' --include='*.yaml' \
            --include='*.yml' --include='*.md' --include='*.css' \
            --include='*.js' \
            --exclude='todo-tracker.yml' --exclude='todo.md' \
            -E '\b(TODO|FIXME|HACK|XXX)\b' . > /tmp/raw.txt || true

          if [ ! -s /tmp/raw.txt ]; then
            echo -e "## \`TODO\` Report\n\n**Last Updated: ${LAST_UPDATED}**\n\nNo markers found." > todo.md
            exit 0
          fi

          # ── step 2: enrich with git blame & extract issue refs ──────
          > /tmp/enriched.tsv
          > /tmp/issues_seen.txt

          while IFS= read -r match; do
            clean=$(printf '%s\n' "$match" | sed 's|^\./||')
            file=$(printf '%s\n' "$clean" | cut -d: -f1)
            lineno=$(printf '%s\n' "$clean" | cut -d: -f2)
            text=$(printf '%s\n' "$clean" | cut -d: -f3-)

            # determine marker type (first match wins, TODO is default)
            marker="TODO"
            for m in FIXME HACK XXX; do
              if printf '%s\n' "$text" | grep -qw "$m"; then marker="$m"; break; fi
            done

            # git blame timestamp
            blame_ts=$(git blame -L "${lineno},${lineno}" --porcelain -- "$file" 2>/dev/null \
              | awk '/^author-time/{print $2}' || echo "0")
            blame_ts=${blame_ts:-0}

            if [ "$blame_ts" -gt 0 ] 2>/dev/null; then
              age_days=$(( (NOW - blame_ts) / 86400 ))
            else
              age_days=-1
            fi

            # extract issue references (#NNN)
            issues=$(printf '%s\n' "$text" | grep -oE '#[0-9]+' | sed 's/#//' | tr '\n' ',' | sed 's/,$//' || true)
            if [ -n "$issues" ]; then
              for inum in $(echo "$issues" | tr ',' ' '); do
                echo "$inum" >> /tmp/issues_seen.txt
              done
            fi

            printf '%s\t%s\t%s\t%s\t%s\t%s\n' \
              "$file" "$lineno" "$marker" "$age_days" "$issues" "$text" >> /tmp/enriched.tsv
          done < /tmp/raw.txt

          # ── step 3: query GitHub API for referenced issues ──────────
          > /tmp/issue_info.tsv
          if [ -s /tmp/issues_seen.txt ]; then
            sort -un /tmp/issues_seen.txt | while read -r inum; do
              info=$(gh api "/repos/$REPO/issues/$inum" \
                --jq '"\(.state)\t\(.title)"' 2>/dev/null \
                || echo "unknown	(could not fetch)")
              printf '%s\t%s\n' "$inum" "$info" >> /tmp/issue_info.tsv
            done
          fi

          # ── step 4: build the report ────────────────────────────────
          {
            # ── header ──
            echo "## \`TODO\` Report"
            echo ""
            echo "**Last Updated: ${LAST_UPDATED}**"
            echo ""
            echo "_This overview is automatically updated on each push to \`main\`. It scans for \`TODO\`, \`FIXME\`, \`HACK\`, and \`XXX\` markers across the codebase._"
            echo ""

            # ── summary table ──
            total=$(wc -l < /tmp/enriched.tsv | tr -d ' ')
            files_affected=$(awk -F'\t' '{print $1}' /tmp/enriched.tsv | sort -u | wc -l | tr -d ' ')
            todo_n=$(awk -F'\t' '$3=="TODO"' /tmp/enriched.tsv | wc -l | tr -d ' ')
            fixme_n=$(awk -F'\t' '$3=="FIXME"' /tmp/enriched.tsv | wc -l | tr -d ' ')
            hack_n=$(awk -F'\t' '$3=="HACK"' /tmp/enriched.tsv | wc -l | tr -d ' ')
            xxx_n=$(awk -F'\t' '$3=="XXX"' /tmp/enriched.tsv | wc -l | tr -d ' ')
            stale_n=$(awk -F'\t' -v s="$STALE_DAYS" '$4 > s' /tmp/enriched.tsv | wc -l | tr -d ' ')
            linked_n=$(awk -F'\t' '$5 != ""' /tmp/enriched.tsv | wc -l | tr -d ' ')
            unlinked_n=$(awk -F'\t' '$5 == ""' /tmp/enriched.tsv | wc -l | tr -d ' ')

            # oldest marker
            oldest_line=$(awk -F'\t' '$4 >= 0' /tmp/enriched.tsv | sort -t$'\t' -k4 -rn | head -1)
            oldest_days=$(echo "$oldest_line" | cut -f4)
            oldest_file=$(echo "$oldest_line" | cut -f1)
            oldest_lineno=$(echo "$oldest_line" | cut -f2)
            oldest_age=$(format_age "$oldest_days")

            echo "### Summary"
            echo ""
            echo "| Metric | Value |"
            echo "|:---|---:|"
            echo "| Total markers | **${total}** |"
            [ "$todo_n" -gt 0 ] && echo "| \`TODO\` | ${todo_n} |"
            [ "$fixme_n" -gt 0 ] && echo "| \`FIXME\` | ${fixme_n} |"
            [ "$hack_n" -gt 0 ] && echo "| \`HACK\` | ${hack_n} |"
            [ "$xxx_n" -gt 0 ] && echo "| \`XXX\` | ${xxx_n} |"
            echo "| Files affected | ${files_affected} |"
            echo "| Stale (> 6 months) | ${stale_n} |"
            echo "| Oldest marker | ${oldest_age}, \`${oldest_file}\` L${oldest_lineno} |"
            echo "| Linked to issues | ${linked_n} |"
            echo "| Unlinked (no issue ref) | ${unlinked_n} |"
            echo ""

            # ── by referenced issue ──
            if [ -s /tmp/issue_info.tsv ]; then
              echo "### By Referenced Issue"
              echo ""

              has_closed=false

              while IFS=$'\t' read -r inum state title; do
                count=$(awk -F'\t' -v n="$inum" '$5 ~ "(^|,)"n"(,|$)"' /tmp/enriched.tsv | wc -l | tr -d ' ')
                [ "$state" = "closed" ] && has_closed=true

                state_icon=""
                [ "$state" = "closed" ] && state_icon=" :warning:"

                echo "<details><summary><b>#${inum}</b> (${state}): <i>${title}</i> &mdash; ${count} marker(s)${state_icon}</summary>"
                echo ""

                awk -F'\t' -v n="$inum" '$5 ~ "(^|,)"n"(,|$)"' /tmp/enriched.tsv \
                | while IFS=$'\t' read -r f l m d refs txt; do
                  age_str=$(format_age "$d")
                  flag=""
                  [ "$d" -gt "$STALE_DAYS" ] 2>/dev/null && flag=" :warning:"
                  # re-read the actual source line and trim leading/trailing whitespace
                  src_text=$(sed -n "${l}p" "$f" 2>/dev/null | sed 's/^[[:space:]]*//;s/[[:space:]]*$//' || true)
                  echo "- [\`${f}\` L${l}](${REPO_URL}/${f}#L${l}) (${age_str} ago)${flag}"
                  [ -n "$src_text" ] && echo "  \`${src_text}\`"
                done

                echo ""
                echo "</details>"
                echo ""
              done < /tmp/issue_info.tsv

              if [ "$has_closed" = true ]; then
                echo "> **Warning:** some markers reference closed issues and may be stale."
                echo ""
              fi
            fi

            # ── by file ──
            echo "### By File"
            echo ""

            prev_file=""
            prev_lineno=-99

            while IFS=$'\t' read -r file lineno marker age_days issues text; do
              if [ "$file" != "$prev_file" ]; then
                # close previous code block
                if [ -n "$prev_file" ]; then
                  echo '```'
                  echo ""
                fi

                file_count=$(awk -F'\t' -v f="$file" '$1==f' /tmp/enriched.tsv | wc -l | tr -d ' ')
                echo "#### [\`${file}\`](${REPO_URL}/${file}) &mdash; ${file_count} marker(s)"
                echo '```r'

                prev_lineno=-99
              fi

              # blank line between non-sequential lines (visual grouping)
              if [ "$file" = "$prev_file" ] && [ $((lineno - prev_lineno)) -gt 1 ]; then
                echo ""
              fi

              age_str=$(format_age "$age_days")
              flag=""
              [ "$age_days" -gt "$STALE_DAYS" ] 2>/dev/null && flag=" !!"

              # re-read the actual source line to avoid TSV round-trip corruption
              src_line=$(sed -n "${lineno}p" "$file" 2>/dev/null | sed 's/[[:space:]]*$//' || true)
              printf 'L%s: %s  ◁ %s ago%s\n' "$lineno" "$src_line" "$age_str" "$flag"

              prev_file="$file"
              prev_lineno="$lineno"
            done < <(sort -t$'\t' -k1,1 -k2,2n /tmp/enriched.tsv)

            # close final code block
            if [ -n "$prev_file" ]; then
              echo '```'
            fi

          } > todo.md

      - name: Update GitHub issue
        uses: peter-evans/create-or-update-comment@v4
        with:
          token: ${{ secrets.GH_REPO_SCOPE }}
          issue-number: 231
          comment-id: 3253439219
          body-file: todo.md
          edit-mode: replace