Replace BibBase embeds with locally generated FacultyPublications and add generation/verification scripts

docs/publications.mdx

@@ -2,7 +2,7 @@
 sidebar_label: 'Publications'
 custom_edit_url: 'https://github.com/vernamlab/vernamlab.github.io/tree/dev/docs/publications'
 ---
-import BibBaseEmbed from '@site/src/components/BibBaseEmbed';
+import FacultyPublications from '@site/src/components/FacultyPublications';
 
 {/* Start of JavaScript block to dynamically load publications */}
 {(() => {
@@ -68,16 +68,16 @@ This page provides a comprehensive list of publications from Vernam Lab.
 
 ### Publications by Fatemeh Ganji
 
-<BibBaseEmbed bibUrl="https://dblp.org/pid/137/6331.bib" />
+<FacultyPublications faculty="Fatemeh Ganji" />
 
 ### Publications by Patrick Schaumont
 
-<BibBaseEmbed bibUrl="https://dblp.org/pid/39/1269.bib" />
+<FacultyPublications faculty="Patrick Schaumont" />
 
 ### Publications by Berk Sunar
 
-<BibBaseEmbed bibUrl="https://dblp.org/pid/91/465.bib" />
+<FacultyPublications faculty="Berk Sunar" />
 
 ### Publications by Shahin Tajik
 
-<BibBaseEmbed bibUrl="https://dblp.org/pid/139/7378.bib" /> 
+<FacultyPublications faculty="Shahin Tajik" /> 

package.json

@@ -3,6 +3,9 @@
   "version": "0.0.0",
   "private": true,
   "scripts": {
+    "generate-publications": "python3 scripts/generate-faculty-publications.py",
+    "verify-publications": "python3 scripts/verify-faculty-publications.py",
+    "prebuild": "npm run generate-publications && npm run verify-publications",
     "docusaurus": "docusaurus",
     "start": "docusaurus start",
     "build": "docusaurus build",
@@ -102,4 +105,4 @@
       "js-yaml": "4.1.1"
     }
   }
-}
+}

scripts/generate-faculty-publications.py

@@ -0,0 +1,104 @@
+#!/usr/bin/env python3
+"""Generate the faculty publication data from the Scholar exports."""
+
+import json
+import re
+from pathlib import Path
+from xml.etree import ElementTree
+from zipfile import ZipFile
+
+
+ROOT = Path(__file__).resolve().parents[1]
+OUTPUT = ROOT / "src" / "data" / "facultyPublications.json"
+WORD_NS = {"w": "http://schemas.openxmlformats.org/wordprocessingml/2006/main"}
+
+SOURCES = {
+    "Fatemeh Ganji": ROOT / "scholar-data" / "ganji.docx",
+    "Patrick Schaumont": ROOT / "scholar-data" / "schaumont_cleaned.bib",
+    "Berk Sunar": ROOT / "scholar-data" / "sunar_cleaned.bib",
+    "Shahin Tajik": ROOT / "scholar-data" / "tajik_cleaned.bib",
+}
+
+
+def docx_bibliography(path):
+    with ZipFile(path) as archive:
+        document = ElementTree.fromstring(archive.read("word/document.xml"))
+    paragraphs = []
+    for paragraph in document.findall(".//w:body/w:p", WORD_NS):
+        paragraphs.append("".join(node.text or "" for node in paragraph.findall(".//w:t", WORD_NS)))
+    return "\n".join(paragraphs)
+
+
+def split_entries(bibliography):
+    starts = list(re.finditer(r"(?m)^@\w+\s*\{", bibliography))
+    return [bibliography[start.start() : starts[index + 1].start()].strip()
+            for index, start in enumerate(starts)
+            if index + 1 < len(starts)] + ([bibliography[starts[-1].start():].strip()] if starts else [])
+
+
+def field(entry, name):
+    match = re.search(rf"(?mi)^\s*{name}\s*=\s*", entry)
+    if not match:
+        return ""
+    position = match.end()
+    if entry[position] in "{\"":
+        opening = entry[position]
+        closing = "}" if opening == "{" else '"'
+        depth = 0
+        value_start = position + 1
+        for index in range(value_start, len(entry)):
+            character = entry[index]
+            if opening == "{" and character == "{" and (index == 0 or entry[index - 1] != "\\"):
+                depth += 1
+            elif character == closing and (index == 0 or entry[index - 1] != "\\"):
+                if depth == 0:
+                    return entry[value_start:index]
+                depth -= 1
+    return entry[position:].split(",", 1)[0].strip()
+
+
+def clean(value):
+    replacements = {
+        r"\&": "&", r"\L": "Ł", r'\"a': "ä", r'\"o': "ö", r'\"u': "ü",
+        r"\ss": "ß", "{": "", "}": "", "~": " ",
+    }
+    for old, new in replacements.items():
+        value = value.replace(old, new)
+    return re.sub(r"\s+", " ", value).strip()
+
+
+def publication(entry):
+    venue = field(entry, "journal") or field(entry, "booktitle") or field(entry, "howpublished")
+    if not venue:
+        venue = field(entry, "publisher") or field(entry, "note")
+    return {
+        "title": clean(field(entry, "title")),
+        "authors": clean(field(entry, "author")).replace(" and ", ", "),
+        "venue": clean(venue),
+        "year": int(field(entry, "year")) if field(entry, "year").isdigit() else None,
+    }
+
+
+def title_key(title):
+    return re.sub(r"[^a-z0-9]", "", title.casefold())
+
+
+def main():
+    output = {}
+    for faculty, path in SOURCES.items():
+        bibliography = docx_bibliography(path) if path.suffix == ".docx" else path.read_text(encoding="utf-8")
+        unique = {}
+        for entry in split_entries(bibliography):
+            item = publication(entry)
+            key = title_key(item["title"])
+            if key and key not in unique:
+                unique[key] = item
+        output[faculty] = sorted(unique.values(), key=lambda item: (-(item["year"] or 0), item["title"].casefold()))
+
+    OUTPUT.parent.mkdir(parents=True, exist_ok=True)
+    OUTPUT.write_text(json.dumps(output, ensure_ascii=False, indent=2) + "\n", encoding="utf-8")
+    print("Generated " + ", ".join(f"{name}: {len(items)}" for name, items in output.items()))
+
+
+if __name__ == "__main__":
+    main()

scripts/verify-faculty-publications.py

@@ -0,0 +1,44 @@
+#!/usr/bin/env python3
+"""Verify generated publication data before Docusaurus builds the site."""
+
+import json
+import re
+from pathlib import Path
+
+
+DATA = Path(__file__).resolve().parents[1] / "src" / "data" / "facultyPublications.json"
+EXPECTED_FACULTY = (
+    "Fatemeh Ganji",
+    "Patrick Schaumont",
+    "Berk Sunar",
+    "Shahin Tajik",
+)
+
+
+def title_key(title):
+    return re.sub(r"[^a-z0-9]", "", title.casefold())
+
+
+def main():
+    publications = json.loads(DATA.read_text(encoding="utf-8"))
+    if tuple(publications) != EXPECTED_FACULTY:
+        raise SystemExit("Generated data does not contain exactly the four expected faculty lists")
+
+    for faculty, entries in publications.items():
+        if not entries:
+            raise SystemExit(f"Generated publication list is empty for {faculty}")
+        titles = [title_key(entry["title"]) for entry in entries]
+        if len(titles) != len(set(titles)):
+            raise SystemExit(f"Generated publication list contains duplicate titles for {faculty}")
+        expected_order = sorted(
+            entries,
+            key=lambda entry: (-(entry["year"] or 0), entry["title"].casefold()),
+        )
+        if entries != expected_order:
+            raise SystemExit(f"Generated publication list is not sorted for {faculty}")
+
+    print(", ".join(f"{faculty}: {len(entries)}" for faculty, entries in publications.items()))
+
+
+if __name__ == "__main__":
+    main()

src/components/FacultyPublications.js

@@ -0,0 +1,22 @@
+import React from 'react';
+import publications from '@site/src/data/facultyPublications.json';
+
+export default function FacultyPublications({faculty}) {
+  const facultyPublications = publications[faculty] || [];
+
+  return (
+    <div className="faculty-publications">
+      {facultyPublications.map((publication, index) => (
+        <div className="faculty-publication" key={`${publication.title}-${publication.year || 'unknown'}-${index}`}>
+          <div className="faculty-publication__title">{publication.title}</div>
+          {publication.authors && <div>{publication.authors}</div>}
+          {(publication.venue || publication.year) && (
+            <div className="faculty-publication__details">
+              {[publication.venue, publication.year].filter(Boolean).join(', ')}
+            </div>
+          )}
+        </div>
+      ))}
+    </div>
+  );
+}

src/css/custom.css

@@ -412,4 +412,17 @@
 
 .custom-dot-list-style {
     bottom: 10px;
-}
+}
+/* Locally rendered Google Scholar publication records. */
+.faculty-publication {
+  margin-bottom: 1rem;
+}
+
+.faculty-publication__title {
+  font-weight: 600;
+}
+
+.faculty-publication__details {
+  color: var(--ifm-color-emphasis-700);
+  font-style: italic;
+}