zotero bibliography with tags

zotero-tagged-bib.sh

#!/bin/bash
# Zotero Better BibTeX URL → DOCX bibliography
# Usage: ./zotero-tagged-bib.sh "http://127.0.0.1:23119/better-bibtex/export?/group;id:6257749/journal_of_alignment.csljson" [output_name]
# Requires:
#
#  * Zotero with Better BibTeX extension
#  * curl
#  * pandoc with citeproc support
#  * jq

## You need the following postscript in user script to include tags in the CSL JSON export:
#
# if (Translator.BetterCSL) {
#   if (zotero.tags && zotero.tags.length) {
#     csl.keyword = zotero.tags.map(t => t.tag).join(', ');
#   }
# }


set -euo pipefail

EXPORT_URL="${1:-}"
OUTPUT_NAME="${2:-bibliography}"
TEMP_DIR="$(mktemp -d)"
BBT_HOST_DEFAULT="http://127.0.0.1:23119"

if [[ -z "${EXPORT_URL}" ]]; then
  echo "❌ Error: Missing Better BibTeX export URL"
  echo "Provide the exact URL from Zotero (right-click → Better BibTeX → Copy export URL)."
  echo "Example:"
  echo "  ./zotero-tagged-bib.sh \"${BBT_HOST_DEFAULT}/better-bibtex/export?/group;id:6257749/journal_of_alignment.json\" my-bib"
  exit 1
fi

# Basic sanity checks
if ! command -v curl >/dev/null 2>&1; then
  echo "❌ curl is required"; exit 1
fi
if ! command -v pandoc >/dev/null 2>&1; then
  echo "❌ pandoc is required (e.g., brew install pandoc or apt install pandoc)"; exit 1
fi

echo "📚 Zotero BBT URL → RTF"
echo "URL: ${EXPORT_URL}"
echo "Out: ${OUTPUT_NAME}.rtf"

# Create CSL style (simple author, title, year, URL, tags)
CSL_FILE="${TEMP_DIR}/simple-with-tags.csl"
cat > "${CSL_FILE}" << 'EOF'
<?xml version="1.0" encoding="utf-8"?>
<style xmlns="http://purl.org/net/xbiblio/csl" class="in-text" version="1.0">
  <info>
    <title>Simple Bibliography with Tags (Authors+Year, Bold Tags)</title>
    <id>http://www.zotero.org/styles/simple-with-tags-robust</id>
    <updated>2025-10-18T06:00:00+00:00</updated>
  </info>

  <!-- Author/editor fallback -->
  <macro name="names-or-title">
    <choose>
      <if variable="author editor" match="any">
        <names variable="author">
          <name and="text" delimiter=", " name-as-sort-order="first"/>
        </names>
        <choose>
          <if variable="author" match="none">
            <names variable="editor">
              <name and="text" delimiter=", " name-as-sort-order="first"/>
              <label form="short" prefix=" (" suffix=")" strip-periods="true"/>
            </names>
          </if>
        </choose>
      </if>
      <else>
        <text variable="title"/>
      </else>
    </choose>
  </macro>

  <!-- Year fallback -->
  <macro name="year">
    <choose>
      <if variable="issued">
        <date variable="issued">
          <date-part name="year"/>
        </date>
      </if>
      <else>
        <text value="n.d."/>
      </else>
    </choose>
  </macro>

  <!-- Title italicized -->
  <macro name="title-italic">
    <text variable="title" font-style="italic"/>
  </macro>

  <!-- URL, printed if present -->
  <macro name="url">
    <text variable="URL"/>
  </macro>

  <!-- Bold tags from keyword field -->
  <macro name="tags-bold">
    <choose>
      <if variable="keyword">
        <group delimiter=" ">
          <text value="Tags:" font-weight="bold"/>
          <text variable="keyword" font-weight="bold"/>
        </group>
      </if>
    </choose>
  </macro>

  <citation>
    <layout>
      <group delimiter=" ">
        <text macro="names-or-title"/>
        <text macro="year" prefix="(" suffix=")"/>
      </group>
    </layout>
  </citation>

  <bibliography hanging-indent="false" entry-spacing="1">
    <layout suffix=".">
      <group delimiter=". ">
        <text macro="names-or-title"/>
        <group delimiter=" ">
          <text macro="title-italic"/>
          <text macro="year" prefix="(" suffix=")"/>
        </group>
        <text macro="url"/>
        <text macro="tags-bold"/>
      </group>
    </layout>
  </bibliography>
</style>
EOF

# Minimal markdown that forces full bibliography rendering
MD_FILE="${TEMP_DIR}/bib.md"
cat > "${MD_FILE}" << 'EOF'
---
nocite: |
  @*
---

# Bibliography
EOF

# Fetch JSON
JSON_FILE="${TEMP_DIR}/library.json"
echo "🔄 Fetching export..."
HTTP_STATUS=$(curl -sS -w "%{http_code}" -H "User-Agent: zotero-script" "${EXPORT_URL}" -o "${JSON_FILE}" || true)

if [[ "${HTTP_STATUS}" != "200" ]]; then
  echo "❌ Export failed (HTTP ${HTTP_STATUS})."
  echo "Response snippet:"
  head -c 500 "${JSON_FILE}" || true
  echo
  echo "Tips:"
  echo " • Ensure Zotero is running with Better BibTeX enabled."
  echo " • Verify the URL in a browser; it should download JSON."
  echo " • Use .json in the URL to get CSL JSON output."
  exit 1
fi

JSON_FILE="${TEMP_DIR}/library.json"
CSL_FILE="${TEMP_DIR}/simple-with-tags.csl"
MD_FILE="${TEMP_DIR}/bib.md"

# Ensure absolute paths (Pandoc can be picky with cwd)
JSON_ABS="$(cd "$(dirname "$JSON_FILE")" && pwd)/$(basename "$JSON_FILE")"
CSL_ABS="$(cd "$(dirname "$CSL_FILE")" && pwd)/$(basename "$CSL_FILE")"

cat > "${MD_FILE}" << EOF
---
title: Bibliography
bibliography: ${JSON_ABS}
csl: ${CSL_ABS}
link-citations: true
nocite: |
  @*
bibliography-format: csljson
---

EOF


# Validate JSON looks like array/object
if ! grep -qE '^[[:space:]]*[\[\{]' "${JSON_FILE}"; then
  echo "❌ Export did not return JSON:"
  head -c 500 "${JSON_FILE}" || true
  exit 1
fi

# Optional: count items if jq exists
if command -v jq >/dev/null 2>&1; then
  COUNT=$(jq 'length' "${JSON_FILE}" 2>/dev/null || echo "unknown")
  echo "✅ Export OK (${COUNT} items)"
else
  echo "✅ Export OK"
fi

# Convert to DOCX
echo "🔧 Converting to DOCX (pandoc)..."
pandoc "${MD_FILE}" \
  --bibliography="${JSON_FILE}" \
  --csl="${CSL_FILE}" \
  --citeproc \
  --metadata bibliography-format=csljson \
  -o "${OUTPUT_NAME}.docx"

echo "✅ Done: ${OUTPUT_NAME}.docx"
echo "📋 Paste into Google Docs as needed."

# Cleanup
rm -rf "${TEMP_DIR}"