add missing papers

Author: lacava

.githooks/pre-commit

@@ -0,0 +1,18 @@
+#!/bin/sh
+# Blocks commits that touch _data/publications.yaml unless every
+# La-Cava-authored publication has a matching PDF in assets/papers/.
+# Installed via: git config core.hooksPath .githooks
+
+if git diff --cached --name-only | grep -q '^_data/publications\.yaml$'; then
+  ruby "$(git rev-parse --show-toplevel)/scripts/check_publication_pdfs.rb"
+  status=$?
+  if [ $status -ne 0 ]; then
+    echo ""
+    echo "Commit blocked: copy the missing PDF(s) into assets/papers/ (filename must"
+    echo "match exactly, per Zotero's 'Rename Associated File' convention), then re-stage."
+    echo "To bypass once (not recommended): git commit --no-verify"
+    exit 1
+  fi
+fi
+
+exit 0

assets/papers/Liu et al. - 2025 - Equitable Survival Prediction A Fairness-Aware Su.pdf

@@ -0,0 +1,82 @@
+#!/usr/bin/env ruby
+# frozen_string_literal: true
+
+# Verifies that every publication in _data/publications.yaml (that lists
+# La Cava as an author, matching the filter in _pages/papers.md) has a
+# matching PDF checked into assets/papers/. Filenames are derived with the
+# exact same logic as _includes/pub-single.html so this stays in sync with
+# what the site actually links to (and with Zotero's own "Rename Associated
+# File" convention: "{Author(s)} - {Year} - {Title, 50 chars}.pdf").
+#
+# Usage:
+#   ruby scripts/check_publication_pdfs.rb            # error on missing PDFs
+#   ruby scripts/check_publication_pdfs.rb --list      # just list, exit 0
+
+require 'yaml'
+require 'pathname'
+
+ROOT = Pathname.new(__dir__).parent
+DATA_FILE = ROOT / '_data' / 'publications.yaml'
+PAPERS_DIR = ROOT / 'assets' / 'papers'
+
+def author_surname(author)
+  return author['literal'] if author['literal']
+
+  particle = author['non-dropping-particle']
+  family = author['family']
+  [particle, family].compact.join(' ').strip
+end
+
+def by_cava?(pub)
+  Array(pub['author']).any? do |author|
+    (author['family'] || '').include?('Cava') || (author['literal'] || '').include?('Cava')
+  end
+end
+
+def pdf_author(pub)
+  authors = Array(pub['author'])
+  first = author_surname(authors[0])
+  case authors.size
+  when 0 then ''
+  when 1 then first
+  when 2 then "#{first} and #{author_surname(authors[1])}"
+  else "#{first} et al."
+  end
+end
+
+# Mirrors Liquid's `truncate: 50, ''` followed by `strip`.
+def pdf_title(pub)
+  title = (pub['title'] || '').to_s
+  truncated = title.length > 50 ? title[0, 50] : title
+  truncated.strip
+end
+
+def expected_filename(pub)
+  year = pub.dig('issued', 0, 'year')
+  base = "#{pdf_author(pub)} - #{year} - #{pdf_title(pub)}.pdf"
+  base.delete(':').delete('?')
+end
+
+data = YAML.load_file(DATA_FILE)
+refs = data['references'] || []
+cava_refs = refs.select { |pub| by_cava?(pub) }
+
+missing = cava_refs.filter_map do |pub|
+  filename = expected_filename(pub)
+  path = PAPERS_DIR / filename
+  [pub['id'], filename] unless path.exist?
+end
+
+list_only = ARGV.include?('--list')
+
+if missing.empty?
+  puts "All #{cava_refs.size} publication PDFs are present in #{PAPERS_DIR.relative_path_from(ROOT)}."
+  exit 0
+end
+
+warn "#{missing.size} publication#{'s' if missing.size != 1} missing a PDF in #{PAPERS_DIR.relative_path_from(ROOT)}:"
+missing.each do |id, filename|
+  warn "  - [#{id}] #{filename}"
+end
+
+exit(list_only ? 0 : 1)