/**
* Pattern Extractor - Smart pattern-based extraction from NGSS PDFs
*/
import type {
ExtractedStandardCode,
ExtractedSection
} from '../types/ngss.js';
import { PATTERNS } from '../types/ngss.js';
import { PDFReader, parsePageContent } from './pdf-reader.js';
export type ExtractionPattern =
| 'standard_codes'
| 'sep_sections'
| 'dci_sections'
| 'ccc_sections';
export class PatternExtractor {
private pdfReader: PDFReader;
constructor() {
this.pdfReader = new PDFReader();
}
async extractStandardCodes(
pdfPath: string,
pages?: string
): Promise<ExtractedStandardCode[]> {
const content = pages
? await this.pdfReader.extractPages(pdfPath, pages)
: await this.pdfReader.extractAll(pdfPath);
const pageContents = parsePageContent(content, pages || 'all');
const results: ExtractedStandardCode[] = [];
for (const page of pageContents) {
const matches = page.content.matchAll(PATTERNS.STANDARD_CODE);
for (const match of matches) {
const code = match[0];
const contextStart = Math.max(0, match.index! - 50);
const contextEnd = Math.min(
page.content.length,
match.index! + code.length + 50
);
const context = page.content.slice(contextStart, contextEnd);
results.push({
code,
page: page.pageNumber,
context: context.trim()
});
}
}
const unique = results.filter(
(item, index, self) => index === self.findIndex((t) => t.code === item.code)
);
return unique;
}
async extractSEPSections(
pdfPath: string,
pages?: string
): Promise<ExtractedSection[]> {
return this.extractSectionsByPattern(pdfPath, 'SEP', pages);
}
async extractDCISections(
pdfPath: string,
pages?: string
): Promise<ExtractedSection[]> {
return this.extractSectionsByPattern(pdfPath, 'DCI', pages);
}
async extractCCCSections(
pdfPath: string,
pages?: string
): Promise<ExtractedSection[]> {
return this.extractSectionsByPattern(pdfPath, 'CCC', pages);
}
private async extractSectionsByPattern(
pdfPath: string,
sectionType: 'SEP' | 'DCI' | 'CCC',
pages?: string
): Promise<ExtractedSection[]> {
const content = pages
? await this.pdfReader.extractPages(pdfPath, pages)
: await this.pdfReader.extractAll(pdfPath);
const pageContents = parsePageContent(content, pages || 'all');
const results: ExtractedSection[] = [];
const pattern =
sectionType === 'SEP'
? PATTERNS.SEP_SECTION
: sectionType === 'DCI'
? PATTERNS.DCI_SECTION
: PATTERNS.CCC_SECTION;
for (const page of pageContents) {
const match = pattern.exec(page.content);
if (match) {
const startIndex = match.index! + match[0].length;
const nextSectionPattern =
/(?:Science and Engineering Practices|Disciplinary Core Ideas|Crosscutting Concepts|Connections to)/gi;
const restContent = page.content.slice(startIndex);
const nextMatch = nextSectionPattern.exec(restContent);
const endIndex = nextMatch ? nextMatch.index : restContent.length;
const content = restContent.slice(0, endIndex).trim();
results.push({
type: sectionType,
content,
page: page.pageNumber
});
}
}
return results;
}
async extractByPattern(
pdfPath: string,
pattern: ExtractionPattern,
pages?: string
): Promise<ExtractedStandardCode[] | ExtractedSection[]> {
switch (pattern) {
case 'standard_codes':
return this.extractStandardCodes(pdfPath, pages);
case 'sep_sections':
return this.extractSEPSections(pdfPath, pages);
case 'dci_sections':
return this.extractDCISections(pdfPath, pages);
case 'ccc_sections':
return this.extractCCCSections(pdfPath, pages);
}
}
}
