import { initDb, addPaper } from './db.js';
import { readFileSync } from 'fs';
import { XMLParser } from 'fast-xml-parser';
import path from 'path';
import { fileURLToPath } from 'url';
const __dirname = path.dirname(fileURLToPath(import.meta.url));
interface PaperXML {
metadata: {
title: string;
author?: string;
year?: string;
periodical?: string;
pages?: string;
affiliation?: string;
subtitle?: string;
};
content: string;
path: string;
}
interface PapersDatabase {
PapersDatabase: {
Papers: {
Paper: PaperXML[];
};
};
}
async function importXML() {
const xmlPath = process.argv[2] || path.join(__dirname, '..', 'data', 'papers_database.xml');
console.log(`Reading XML from: ${xmlPath}`);
const xmlContent = readFileSync(xmlPath, 'utf-8');
console.log('Parsing XML...');
const parser = new XMLParser({
ignoreAttributes: false,
attributeNamePrefix: '@_',
textNodeName: '#text',
});
const parsed = parser.parse(xmlContent) as PapersDatabase;
const papers = parsed.PapersDatabase.Papers.Paper;
console.log(`Found ${papers.length} papers to import`);
await initDb();
let imported = 0;
let failed = 0;
for (const paper of papers) {
try {
const meta = paper.metadata;
const title = meta.title || 'Untitled';
const author = meta.author || 'Unknown';
const content = paper.content || '';
// Extract year from metadata or default
const year = meta.year || '1994';
const publicationDate = `${year}-01-01`;
// Build full title with subtitle if present
const fullTitle = meta.subtitle ? `${title}：${meta.subtitle}` : title;
// Use periodical name as publication
const publication = meta.periodical || '道風：漢語神學學刊';
// Generate topics based on content analysis
const topics = generateTopics(fullTitle, content);
// Generate keywords from title
const keywords = generateKeywords(fullTitle, content);
// Create abstract from first ~500 chars of content
const abstract = generateAbstract(content);
const id = addPaper({
title: fullTitle,
authors: author,
abstract,
content,
publication,
publication_date: publicationDate,
doi: null,
url: null,
topics,
keywords,
});
imported++;
if (imported % 50 === 0) {
console.log(`Imported ${imported} papers...`);
}
} catch (error) {
failed++;
console.error(`Failed to import paper: ${paper.metadata?.title || 'unknown'}`, error);
}
}
console.log(`\nImport complete!`);
console.log(`Successfully imported: ${imported}`);
console.log(`Failed: ${failed}`);
}
function generateTopics(title: string, content: string): string {
const topics: string[] = [];
const text = (title + ' ' + content).toLowerCase();
// Theology topics
if (text.includes('神學') || text.includes('theology')) topics.push('神學');
if (text.includes('基督') || text.includes('christ')) topics.push('基督教');
if (text.includes('聖經') || text.includes('bible') || text.includes('舊約') || text.includes('新約')) topics.push('聖經研究');
if (text.includes('哲學') || text.includes('philosophy')) topics.push('哲學');
if (text.includes('倫理') || text.includes('ethics')) topics.push('倫理學');
if (text.includes('歷史') || text.includes('history')) topics.push('歷史');
if (text.includes('文化') || text.includes('culture')) topics.push('文化研究');
if (text.includes('儒') || text.includes('confuci')) topics.push('儒學');
if (text.includes('佛') || text.includes('buddhis')) topics.push('佛學');
if (text.includes('道家') || text.includes('taois')) topics.push('道家');
if (text.includes('現代') || text.includes('modern')) topics.push('現代性');
if (text.includes('後現代') || text.includes('postmodern')) topics.push('後現代');
if (text.includes('宗教') || text.includes('religio')) topics.push('宗教學');
if (text.includes('教會') || text.includes('church')) topics.push('教會');
if (text.includes('社會') || text.includes('social')) topics.push('社會學');
if (text.includes('政治') || text.includes('politic')) topics.push('政治');
// Default topic if none matched
if (topics.length === 0) {
topics.push('漢語神學');
}
return topics.slice(0, 5).join(', ');
}
function generateKeywords(title: string, content: string): string {
// Extract key terms from title
const keywords: string[] = [];
// Add title words as keywords (filter short ones)
const titleWords = title.split(/[，、：\s]+/).filter(w => w.length >= 2);
keywords.push(...titleWords.slice(0, 5));
return keywords.join(', ');
}
function generateAbstract(content: string): string {
if (!content) return 'No abstract available.';
// Clean up content - remove page markers and extra whitespace
let cleaned = content
.replace(/Blank\s*Page\s*此頁為空白頁/g, '')
.replace(/\d+\s*《道風》漢語神學學刊/g, '')
.replace(/第.+期．\d+年．[春夏秋冬]/g, '')
.trim();
// Take first ~500 characters as abstract
if (cleaned.length > 500) {
cleaned = cleaned.substring(0, 500);
// Try to end at a sentence
const lastPeriod = Math.max(
cleaned.lastIndexOf('。'),
cleaned.lastIndexOf('！'),
cleaned.lastIndexOf('？')
);
if (lastPeriod > 300) {
cleaned = cleaned.substring(0, lastPeriod + 1);
} else {
cleaned += '...';
}
}
return cleaned || 'No abstract available.';
}
importXML().catch(console.error);
