smartpdf/ts/smartpdf.classes.smartpdf.ts
2022-03-24 14:32:49 +01:00

203 lines
6.1 KiB
TypeScript

import * as plugins from './smartpdf.plugins.js';
import * as paths from './smartpdf.paths.js';
import { Server } from 'http';
import { PdfCandidate } from './smartpdf.classes.pdfcandidate.js';
declare const document: any;
import * as interfaces from './interfaces/index.js';
export class SmartPdf {
htmlServerInstance: Server;
serverPort: number;
headlessBrowser: plugins.smartpuppeteer.puppeteer.Browser;
externalBrowserBool: boolean = false;
private _readyDeferred: plugins.smartpromise.Deferred<void>;
private _candidates: { [key: string]: PdfCandidate } = {};
constructor() {
this._readyDeferred = new plugins.smartpromise.Deferred();
}
async start(headlessBrowserArg?: plugins.smartpuppeteer.puppeteer.Browser) {
// lets set the external browser in case one is provided
this.headlessBrowser = headlessBrowserArg;
// setup puppeteer
if (this.headlessBrowser) {
this.externalBrowserBool = true;
} else {
this.headlessBrowser = await plugins.smartpuppeteer.getEnvAwareBrowserInstance({
forceNoSandbox: true,
});
}
// setup server
const app = plugins.express();
app.get('/:pdfId', (req, res) => {
res.setHeader('PDF-ID', this._candidates[req.params.pdfId].pdfId);
res.send(this._candidates[req.params.pdfId].htmlString);
});
this.htmlServerInstance = plugins.http.createServer(app);
const smartnetworkInstance = new plugins.smartnetwork.SmartNetwork();
const portAvailable = smartnetworkInstance.isLocalPortUnused(3210);
this.htmlServerInstance.listen(3210, 'localhost');
this.htmlServerInstance.on('listening', () => {
this._readyDeferred.resolve();
});
}
// stop
async stop() {
const done = plugins.smartpromise.defer<void>();
this.htmlServerInstance.close(() => {
done.resolve();
});
if (!this.externalBrowserBool) {
await this.headlessBrowser.close();
}
await done.promise;
}
/**
* returns a pdf for a given html string;
*/
async getA4PdfResultForHtmlString(htmlStringArg: string): Promise<interfaces.IPdfResult> {
await this._readyDeferred.promise;
const pdfCandidate = new PdfCandidate(htmlStringArg);
this._candidates[pdfCandidate.pdfId] = pdfCandidate;
const page = await this.headlessBrowser.newPage();
await page.setViewport({
width: 794,
height: 1122,
});
const response = await page.goto(`http://localhost:3210/${pdfCandidate.pdfId}`, {
waitUntil: 'networkidle2',
});
const headers = response.headers();
if (headers['pdf-id'] !== pdfCandidate.pdfId) {
console.log('Error! Headers do not match. For security reasons no pdf is being emitted!');
return;
} else {
console.log(`id security check passed for ${pdfCandidate.pdfId}`);
}
const pdfBuffer = await page.pdf({
width: 794,
height: 1122,
printBackground: true,
displayHeaderFooter: false,
});
await page.close();
delete this._candidates[pdfCandidate.pdfId];
pdfCandidate.doneDeferred.resolve();
await pdfCandidate.doneDeferred.promise;
return {
id: pdfCandidate.pdfId,
name: `${pdfCandidate.pdfId}.js`,
metadata: {
textExtraction: await this.extractTextFromPdfBuffer(pdfBuffer),
},
buffer: pdfBuffer,
};
}
async getPdfResultForWebsite(websiteUrl: string): Promise<interfaces.IPdfResult> {
const page = await this.headlessBrowser.newPage();
await page.setViewport({
width: 1980,
height: 1200,
});
await page.emulateMediaType('screen');
const response = await page.goto(websiteUrl, { waitUntil: 'networkidle2' });
const pdfId = plugins.smartunique.shortId();
const { documentHeight, documentWidth } = await page.evaluate(() => {
return {
documentHeight: document.height,
documentWidth: document.width,
};
});
const pdfBuffer = await page.pdf({
height: documentHeight,
width: documentWidth,
printBackground: true,
displayHeaderFooter: false,
});
await page.close();
return {
id: pdfId,
name: `${pdfId}.js`,
metadata: {
textExtraction: await this.extractTextFromPdfBuffer(pdfBuffer),
},
buffer: pdfBuffer,
};
}
async getFullWebsiteAsSinglePdf(websiteUrl: string): Promise<interfaces.IPdfResult> {
const page = await this.headlessBrowser.newPage();
await page.setViewport({
width: 1920,
height: 1200,
});
page.emulateMediaType('screen');
const response = await page.goto(websiteUrl, { waitUntil: 'networkidle2' });
const pdfId = plugins.smartunique.shortId();
const { documentHeight, documentWidth } = await page.evaluate(() => {
return {
documentHeight: document.body.scrollHeight,
documentWidth: document.body.clientWidth,
};
});
await page.setViewport({
width: 1920,
height: documentHeight,
});
const pdfBuffer = await page.pdf({
height: documentHeight,
width: 1920,
printBackground: true,
displayHeaderFooter: false,
scale: 1,
pageRanges: '1'
});
await page.close();
return {
id: pdfId,
name: `${pdfId}.js`,
metadata: {
textExtraction: await this.extractTextFromPdfBuffer(pdfBuffer),
},
buffer: pdfBuffer,
};
}
public async mergePdfBuffers(pdfBuffers: Buffer[]): Promise<Buffer> {
const merger = new plugins.pdfMerger();
for (const buffer of pdfBuffers) {
merger.add(buffer);
}
return merger.saveAsBuffer();
}
public async extractTextFromPdfBuffer(pdfBufferArg: Buffer): Promise<string> {
const deferred = plugins.smartpromise.defer<string>();
const pdfParser: any = new plugins.pdf2json();
pdfParser.on('pdfParser_dataReady', (pdfData: any) => {
let finalText = ''
for (const page of pdfData.Pages) {
for(const text of page.Texts) {
for (const letter of text.R) {
finalText = finalText + letter.T;
}
};
}
deferred.resolve(finalText);
});
pdfParser.parseBuffer(pdfBufferArg);
return deferred.promise;
}
}