@hasna/connectors 0.9.2 → 0.9.3

package/connectors/connect-epo/src/api/publications.ts

@@ -0,0 +1,383 @@
+import type { EPOClient } from './client';
+import type {
+  DocumentType,
+  DocumentFormat,
+  PublicationSearchRequest,
+  PublicationSearchResponse,
+  PublicationResult,
+  BiblioResponse,
+  BibliographicData,
+  AbstractResponse,
+  DescriptionResponse,
+  ClaimsResponse,
+  Claim,
+  ImagesResponse,
+  ImageInfo,
+} from '../types';
+
+// XML parsing helpers (simple regex-based for common patterns)
+function extractText(xml: string, tag: string): string | undefined {
+  const match = xml.match(new RegExp(`<${tag}[^>]*>([^<]*)</${tag}>`, 'i'));
+  return match ? match[1].trim() : undefined;
+}
+
+function extractAll(xml: string, tag: string): string[] {
+  const regex = new RegExp(`<${tag}[^>]*>([^<]*)</${tag}>`, 'gi');
+  const results: string[] = [];
+  let match;
+  while ((match = regex.exec(xml)) !== null) {
+    results.push(match[1].trim());
+  }
+  return results;
+}
+
+function extractAttribute(xml: string, tag: string, attr: string): string | undefined {
+  const match = xml.match(new RegExp(`<${tag}[^>]*\\s${attr}="([^"]*)"`, 'i'));
+  return match ? match[1] : undefined;
+}
+
+/**
+ * Publications API - Search and retrieve published patent data
+ */
+export class PublicationsApi {
+  constructor(private readonly client: EPOClient) {}
+
+  /**
+   * Search published patents using CQL query
+   * @param query - CQL query string (e.g., "ti=solar AND pa=tesla")
+   * @param options - Search options
+   */
+  async search(query: string, options?: { rangeBegin?: number; rangeEnd?: number }): Promise<PublicationSearchResponse> {
+    try {
+      const params: Record<string, string | number> = {
+        q: query,
+      };
+
+      if (options?.rangeBegin !== undefined && options?.rangeEnd !== undefined) {
+        params['Range'] = `${options.rangeBegin}-${options.rangeEnd}`;
+      }
+
+      const xml = await this.client.get<string>(
+        '/published-data/search',
+        params,
+        'application/xml'
+      );
+
+      return this.parseSearchResponse(xml);
+    } catch (error) {
+      return {
+        success: false,
+        error: String(error),
+      };
+    }
+  }
+
+  /**
+   * Get publication by document number
+   * @param type - Document type (publication, application, priority)
+   * @param format - Document format (docdb, epodoc, original)
+   * @param number - Document number
+   */
+  async getPublication(
+    type: DocumentType,
+    format: DocumentFormat,
+    number: string
+  ): Promise<BiblioResponse> {
+    try {
+      const path = `/published-data/${type}/${format}/${encodeURIComponent(number)}`;
+      const xml = await this.client.get<string>(path, undefined, 'application/xml');
+      return this.parseBiblioResponse(xml);
+    } catch (error) {
+      return {
+        success: false,
+        error: String(error),
+      };
+    }
+  }
+
+  /**
+   * Get bibliographic data for a publication
+   * @param type - Document type (publication, application, priority)
+   * @param format - Document format (docdb, epodoc, original)
+   * @param number - Document number
+   */
+  async getBiblio(
+    type: DocumentType,
+    format: DocumentFormat,
+    number: string
+  ): Promise<BiblioResponse> {
+    try {
+      const path = `/published-data/${type}/${format}/${encodeURIComponent(number)}/biblio`;
+      const xml = await this.client.get<string>(path, undefined, 'application/xml');
+      return this.parseBiblioResponse(xml);
+    } catch (error) {
+      return {
+        success: false,
+        error: String(error),
+      };
+    }
+  }
+
+  /**
+   * Get abstract for a publication
+   * @param type - Document type (publication, application, priority)
+   * @param format - Document format (docdb, epodoc, original)
+   * @param number - Document number
+   */
+  async getAbstract(
+    type: DocumentType,
+    format: DocumentFormat,
+    number: string
+  ): Promise<AbstractResponse> {
+    try {
+      const path = `/published-data/${type}/${format}/${encodeURIComponent(number)}/abstract`;
+      const xml = await this.client.get<string>(path, undefined, 'application/xml');
+      return this.parseAbstractResponse(xml, number);
+    } catch (error) {
+      return {
+        success: false,
+        error: String(error),
+      };
+    }
+  }
+
+  /**
+   * Get description for a publication
+   * @param type - Document type (publication, application, priority)
+   * @param format - Document format (docdb, epodoc, original)
+   * @param number - Document number
+   */
+  async getDescription(
+    type: DocumentType,
+    format: DocumentFormat,
+    number: string
+  ): Promise<DescriptionResponse> {
+    try {
+      const path = `/published-data/${type}/${format}/${encodeURIComponent(number)}/description`;
+      const xml = await this.client.get<string>(path, undefined, 'application/xml');
+      return this.parseDescriptionResponse(xml, number);
+    } catch (error) {
+      return {
+        success: false,
+        error: String(error),
+      };
+    }
+  }
+
+  /**
+   * Get claims for a publication
+   * @param type - Document type (publication, application, priority)
+   * @param format - Document format (docdb, epodoc, original)
+   * @param number - Document number
+   */
+  async getClaims(
+    type: DocumentType,
+    format: DocumentFormat,
+    number: string
+  ): Promise<ClaimsResponse> {
+    try {
+      const path = `/published-data/${type}/${format}/${encodeURIComponent(number)}/claims`;
+      const xml = await this.client.get<string>(path, undefined, 'application/xml');
+      return this.parseClaimsResponse(xml, number);
+    } catch (error) {
+      return {
+        success: false,
+        error: String(error),
+      };
+    }
+  }
+
+  /**
+   * Get images metadata for a publication
+   * @param type - Document type (publication, application, priority)
+   * @param format - Document format (docdb, epodoc, original)
+   * @param number - Document number
+   */
+  async getImages(
+    type: DocumentType,
+    format: DocumentFormat,
+    number: string
+  ): Promise<ImagesResponse> {
+    try {
+      const path = `/published-data/${type}/${format}/${encodeURIComponent(number)}/images`;
+      const xml = await this.client.get<string>(path, undefined, 'application/xml');
+      return this.parseImagesResponse(xml, number);
+    } catch (error) {
+      return {
+        success: false,
+        error: String(error),
+      };
+    }
+  }
+
+  // ============================================
+  // Response Parsing
+  // ============================================
+
+  private parseSearchResponse(xml: string): PublicationSearchResponse {
+    const totalMatch = xml.match(/total-result-count="(\d+)"/);
+    const totalResults = totalMatch ? parseInt(totalMatch[1]) : 0;
+
+    const results: PublicationResult[] = [];
+
+    // Extract exchange documents
+    const docRegex = /<exchange-document[^>]*>([\s\S]*?)<\/exchange-document>/gi;
+    let match;
+    while ((match = docRegex.exec(xml)) !== null) {
+      const docXml = match[0];
+
+      const country = extractAttribute(docXml, 'exchange-document', 'country');
+      const docNumber = extractAttribute(docXml, 'exchange-document', 'doc-number');
+      const kind = extractAttribute(docXml, 'exchange-document', 'kind');
+
+      const title = extractText(docXml, 'invention-title');
+
+      results.push({
+        documentId: `${country || ''}${docNumber || ''}${kind || ''}`,
+        country,
+        docNumber,
+        kind,
+        title,
+      });
+    }
+
+    return {
+      success: true,
+      totalResults,
+      results,
+    };
+  }
+
+  private parseBiblioResponse(xml: string): BiblioResponse {
+    const country = extractAttribute(xml, 'exchange-document', 'country');
+    const docNumber = extractAttribute(xml, 'exchange-document', 'doc-number');
+    const kind = extractAttribute(xml, 'exchange-document', 'kind');
+
+    const titles = extractAll(xml, 'invention-title');
+
+    // Extract applicants
+    const applicantNames = extractAll(xml, 'name');
+    const applicants = applicantNames.slice(0, Math.ceil(applicantNames.length / 2)).map((name, i) => ({
+      name,
+      sequence: i + 1,
+    }));
+
+    // Extract classifications
+    const classificationSymbols = extractAll(xml, 'classification-symbol');
+    const classifications = classificationSymbols.map(symbol => ({
+      scheme: symbol.startsWith('A') || symbol.startsWith('B') ? 'IPC' : 'CPC',
+      symbol,
+    }));
+
+    const data: BibliographicData = {
+      documentId: `${country || ''}${docNumber || ''}${kind || ''}`,
+      country,
+      docNumber,
+      kind,
+      title: titles[0],
+      titles: titles.map((text, i) => ({ lang: i === 0 ? 'en' : 'other', text })),
+      applicants,
+      classifications,
+    };
+
+    return {
+      success: true,
+      data,
+    };
+  }
+
+  private parseAbstractResponse(xml: string, documentId: string): AbstractResponse {
+    const abstracts: { lang: string; text: string }[] = [];
+
+    const abstractRegex = /<abstract[^>]*lang="([^"]*)"[^>]*>([\s\S]*?)<\/abstract>/gi;
+    let match;
+    while ((match = abstractRegex.exec(xml)) !== null) {
+      const lang = match[1];
+      const content = match[2].replace(/<[^>]+>/g, ' ').replace(/\s+/g, ' ').trim();
+      abstracts.push({ lang, text: content });
+    }
+
+    // Fallback: try to get abstract without lang attribute
+    if (abstracts.length === 0) {
+      const simpleMatch = xml.match(/<abstract[^>]*>([\s\S]*?)<\/abstract>/i);
+      if (simpleMatch) {
+        const content = simpleMatch[1].replace(/<[^>]+>/g, ' ').replace(/\s+/g, ' ').trim();
+        abstracts.push({ lang: 'en', text: content });
+      }
+    }
+
+    return {
+      success: true,
+      data: {
+        documentId,
+        abstracts,
+      },
+    };
+  }
+
+  private parseDescriptionResponse(xml: string, documentId: string): DescriptionResponse {
+    const descMatch = xml.match(/<description[^>]*>([\s\S]*?)<\/description>/i);
+    const description = descMatch
+      ? descMatch[1].replace(/<[^>]+>/g, ' ').replace(/\s+/g, ' ').trim()
+      : '';
+
+    const langMatch = xml.match(/<description[^>]*lang="([^"]*)"/i);
+    const lang = langMatch ? langMatch[1] : undefined;
+
+    return {
+      success: true,
+      data: {
+        documentId,
+        description,
+        lang,
+      },
+    };
+  }
+
+  private parseClaimsResponse(xml: string, documentId: string): ClaimsResponse {
+    const claims: Claim[] = [];
+
+    const claimRegex = /<claim[^>]*num="(\d+)"[^>]*>([\s\S]*?)<\/claim>/gi;
+    let match;
+    while ((match = claimRegex.exec(xml)) !== null) {
+      const num = parseInt(match[1]);
+      const content = match[2].replace(/<[^>]+>/g, ' ').replace(/\s+/g, ' ').trim();
+
+      claims.push({
+        number: num,
+        text: content,
+        type: num === 1 ? 'independent' : 'dependent',
+      });
+    }
+
+    return {
+      success: true,
+      data: {
+        documentId,
+        claims,
+      },
+    };
+  }
+
+  private parseImagesResponse(xml: string, documentId: string): ImagesResponse {
+    const images: ImageInfo[] = [];
+
+    const imageRegex = /<document-instance[^>]*desc="([^"]*)"[^>]*number-of-pages="(\d+)"[^>]*>/gi;
+    let match;
+    while ((match = imageRegex.exec(xml)) !== null) {
+      images.push({
+        type: match[1],
+        format: 'PDF',
+        pages: parseInt(match[2]),
+      });
+    }
+
+    return {
+      success: true,
+      data: {
+        documentId,
+        images,
+      },
+    };
+  }
+}