Started migrating workflow controller

This commit is contained in:
Felix Kaspar
2023-11-13 00:09:12 +01:00
parent 57142381ca
commit 5ae8cb77ac
29 changed files with 96 additions and 127 deletions

View File

@@ -0,0 +1,51 @@
import { PdfFile } from '../../wrappers/PdfFile';
import { PDFPageProxy } from "pdfjs-dist/types/src/display/api.js";
import { Image } from 'image-js';
import { getImagesOnPage } from "./getImagesOnPage.js";
export async function detectEmptyPages(file: PdfFile, whiteThreashold: number): Promise<number[]> {
const pdfDoc = await file.getAsPdfJs();
const emptyPages: number[] = [];
for (let i = 1; i <= pdfDoc.numPages; i++) {
const page = await pdfDoc.getPage(i);
console.log("Checking page " + i);
if(!await hasText(page)) {
console.log(`Found text on Page ${i}, page is not empty`);
continue;
}
if(!await areImagesBlank(page, whiteThreashold)) {
console.log(`Found non white image on Page ${i}, page is not empty`);
continue;
}
console.log(`Page ${i} is empty.`);
emptyPages.push(i - 1);
}
return emptyPages;
}
async function hasText(page: PDFPageProxy): Promise<boolean> {
const textContent = await page.getTextContent();
return textContent.items.length === 0;
}
async function areImagesBlank(page: PDFPageProxy, threshold: number): Promise<boolean> {
const images = await getImagesOnPage(page);
for (const image of images) {
if(!await isImageBlank(image, threshold))
return false;
}
return true;
}
async function isImageBlank(image: string | Uint8Array | ArrayBuffer, threshold: number): Promise<boolean> {
var img = await Image.load(image);
var grey = img.grey();
var mean = grey.getMean();
return mean[0] <= threshold;
}

View File

@@ -0,0 +1,15 @@
import { PDFPageProxy } from "pdfjs-dist/types/src/display/api.js";
import * as PDFJS from 'pdfjs-dist';
export async function getImagesOnPage(page: PDFPageProxy) {
const ops = await page.getOperatorList();
const images: any = [];
for (var j=0; j < ops.fnArray.length; j++) {
if (ops.fnArray[j] == PDFJS.OPS.paintImageXObject) {
const image = page.objs.get(ops.argsArray[j][0]);
images.push(image);
}
}
return images;
}

View File

@@ -0,0 +1,47 @@
import { PdfFile, convertAllToPdfLibFile } from '../../wrappers/PdfFile';
export async function sortPdfs(
files: PdfFile[],
sortType: "orderProvided"|"byFileName"|"byDateModified"|"byDateCreated"|"byPDFTitle" = "orderProvided"
): Promise<PdfFile[]> {
const pdfLibFiles = await convertAllToPdfLibFile(files);
switch(sortType) {
case "byFileName":
pdfLibFiles.sort((a, b) => {
if (!a || !b) return 0;
const ad = a.filename, bd = b.filename;
if (!ad || !bd) return 0;
return ad.localeCompare(bd);
});
break;
case "byDateModified":
pdfLibFiles.sort((a, b) => {
const ad = a.pdfLib?.getModificationDate()?.getTime();
const bd = b.pdfLib?.getModificationDate()?.getTime();
if (!ad || !bd) return 0;
return ad > bd ? 1 : -1
});
break;
case "byDateCreated":
pdfLibFiles.sort((a, b) => {
const ad = a.pdfLib?.getCreationDate()?.getTime();
const bd = b.pdfLib?.getCreationDate()?.getTime();
if (!ad || !bd) return 0;
return ad > bd ? 1 : -1
});
break;
case "byPDFTitle":
pdfLibFiles.sort((a, b) => {
const ad = a.pdfLib?.getTitle();
const bd = b.pdfLib?.getTitle();
if (!ad || !bd) return 0;
return ad.localeCompare(bd);
});
break;
}
return pdfLibFiles;
}

View File

@@ -0,0 +1,12 @@
export async function impose(snapshot: any, nup: number, format: string, pdfcpuWrapper: any) {
return await pdfcpuWrapper.oneToOne([
"pdfcpu.wasm",
"nup",
"-c",
"disable",
'f:' + format,
"/output.pdf",
String(nup),
"input.pdf",
], snapshot);
}

View File

@@ -0,0 +1,18 @@
import { PDFDocument } from 'pdf-lib';
import { PdfFile, convertAllToPdfLibFile, fromPdfLib } from '../wrappers/PdfFile';
export async function mergePDFs(files: PdfFile[]): Promise<PdfFile> {
const pdfLibFiles = await convertAllToPdfLibFile(files);
const mergedPdf = await PDFDocument.create();
for (let i = 0; i < pdfLibFiles.length; i++) {
const pdfToMerge = await pdfLibFiles[i].getAsPdfLib();
const copiedPages = await mergedPdf.copyPages(pdfToMerge, pdfToMerge.getPageIndices());
copiedPages.forEach((page) => mergedPdf.addPage(page));
}
return fromPdfLib(mergedPdf, files[0].filename);
};

View File

@@ -0,0 +1,26 @@
import { degrees } from 'pdf-lib';
import { PdfFile, fromPdfLib } from '../wrappers/PdfFile';
export async function rotatePages(file: PdfFile, rotation: number|number[]): Promise<PdfFile> {
const pdfDoc = await file.getAsPdfLib();
const pages = pdfDoc.getPages();
if (Array.isArray(rotation)) {
if (rotation.length != pages.length) {
throw new Error(`Number of given rotations '${rotation.length}' is not the same as the number of pages '${pages.length}'`)
}
for (let i=0; i<rotation.length; i++) {
const oldRotation = pages[i].getRotation().angle
pages[i].setRotation(degrees(oldRotation + rotation[i]))
}
} else {
pages.forEach(page => {
// Change page size
const oldRotation = page.getRotation().angle
page.setRotation(degrees(oldRotation + rotation))
});
}
return fromPdfLib(pdfDoc, file.filename);
};

View File

@@ -0,0 +1,36 @@
import { PDFPage } from 'pdf-lib';
import { PdfFile, fromPdfLib } from '../wrappers/PdfFile';
export async function scaleContent(file: PdfFile, scaleFactor: number|number[]): Promise<PdfFile> {
const pdfDoc = await file.getAsPdfLib();
const pages = pdfDoc.getPages();
if (Array.isArray(scaleFactor)) {
if (scaleFactor.length != pages.length) {
throw new Error(`Number of given scale factors '${scaleFactor.length}' is not the same as the number of pages '${pages.length}'`)
}
for (let i=0; i<scaleFactor.length; i++) {
scalePage(pages[i], scaleFactor[i]);
}
} else {
pages.forEach(page => scalePage(page, scaleFactor));
}
return fromPdfLib(pdfDoc, file.filename);
};
function scalePage(page: PDFPage, scaleFactor: number) {
const width = page.getWidth();
const height = page.getHeight();
// Scale content
page.scaleContent(scaleFactor, scaleFactor);
const scaled_diff = {
width: Math.round(width - scaleFactor * width),
height: Math.round(height - scaleFactor * height),
};
// Center content in new page format
page.translateContent(Math.round(scaled_diff.width / 2), Math.round(scaled_diff.height / 2));
}

View File

@@ -0,0 +1,56 @@
import { PDFPage } from 'pdf-lib';
import { PdfFile, fromPdfLib } from '../wrappers/PdfFile';
export async function scalePage(file: PdfFile, pageSize: {width?:number,height?:number}|{width?:number,height?:number}[]): Promise<PdfFile> {
const pdfDoc = await file.getAsPdfLib();
const pages = pdfDoc.getPages();
if (Array.isArray(pageSize)) {
if (pageSize.length != pages.length) {
throw new Error(`Number of given sizes '${pageSize.length}' is not the same as the number of pages '${pages.length}'`)
}
for (let i=0; i<pageSize.length; i++) {
resize(pages[i], pageSize[i]);
}
} else {
pages.forEach(page => resize(page, pageSize));
}
return fromPdfLib(pdfDoc, file.filename);
};
function resize(page: PDFPage, newSize: {width?:number,height?:number}) {
const calculatedSize = calculateSize(page, newSize);
page.setSize(calculatedSize.width, calculatedSize.height);
const xRatio = calculatedSize.width / page.getWidth();
const yRatio = calculatedSize.height / page.getHeight();
page.scaleContent(xRatio, yRatio);
}
function calculateSize(page: PDFPage, newSize: {width?:number,height?:number}): {width:number,height:number} {
if (!newSize.width && !newSize.height){
throw new Error(`Sizes '${newSize}' cannot have null width and null height`);
} else if (!newSize.width && newSize.height) {
const oldSize = page.getSize();
const ratio = oldSize.width / oldSize.height;
return { width: newSize.height * ratio, height: newSize.height };
} else if (newSize.width && !newSize.height) {
const oldSize = page.getSize();
const ratio = oldSize.height / oldSize.width;
return { width: newSize.width, height: newSize.width * ratio };
}
return { width: newSize.width!, height: newSize.height! };
}
export const PageSize = Object.freeze({
a4: {
width: 594.96,
height: 841.92
},
letter: {
width: 612,
height: 792
}
});

View File

@@ -0,0 +1,110 @@
import jsQR from "jsqr";
import { detectEmptyPages } from "./common/detectEmptyPages.js";
import { getImagesOnPage } from "./common/getImagesOnPage.js";
import { selectPages } from "./subDocumentFunctions";
import { PdfFile } from '../wrappers/PdfFile.js';
export async function splitOn(
file: PdfFile,
type: "BAR_CODE"|"QR_CODE"|"BLANK_PAGE",
whiteThreashold: number) {
let splitAtPages: number[] = [];
switch (type) {
case "BAR_CODE":
// TODO: Implement
throw new Error("This split-type has not been implemented yet");
case "QR_CODE":
splitAtPages = await getPagesWithQRCode(file);
break;
case "BLANK_PAGE":
splitAtPages = await detectEmptyPages(file, whiteThreashold);
break;
default:
throw new Error("An invalid split-type was provided.");
}
console.log("Split At Pages: ", splitAtPages);
// Remove detected Pages & Split
const pdfDoc = await file.getAsPdfLib();
const numberOfPages = pdfDoc.getPageCount();
let pagesArray: number[] = [];
let splitAfter = splitAtPages.shift();
const subDocuments: PdfFile[] = [];
for (let i = 0; i < numberOfPages; i++) {
console.log(i);
if(i == splitAfter) {
if(pagesArray.length > 0) {
subDocuments.push(await selectPages(file, pagesArray));
pagesArray = [];
}
splitAfter = splitAtPages.shift();
}
else { // Skip splitAtPage
console.log("PagesArray")
pagesArray.push(i);
}
}
if(pagesArray.length > 0) {
subDocuments.push(await selectPages(file, pagesArray));
}
pagesArray = [];
return subDocuments;
async function getPagesWithQRCode(file: PdfFile) {
const pdfDoc = await file.getAsPdfJs();
const pagesWithQR: number[] = [];
for (let i = 0; i < pdfDoc.numPages; i++) {
console.log("Page:", i, "/", pdfDoc.numPages);
const page = await pdfDoc.getPage(i + 1);
const images = await getImagesOnPage(page);
console.log("images:", images);
for (const image of images) {
const data = await checkForQROnImage(image);
if(data == "https://github.com/Frooodle/Stirling-PDF") {
pagesWithQR.push(i);
}
}
}
if(pagesWithQR.length == 0) {
console.warn("Could not find any QR Codes in the provided PDF.")
}
return pagesWithQR;
}
async function checkForQROnImage(image: any) {
// TODO: There is an issue with the jsQR package (The package expects rgba but sometimes we have rgb), and the package seems to be stale, we could create a fork and fix the issue. In the meanwhile we just force rgba:
// Check for rgb and convert to rgba
if(image.data.length == image.width * image.height * 3) {
const tmpArray = new Uint8ClampedArray(image.width * image.height * 4);
// Iterate through the original array and add an alpha channel
for (let i = 0, j = 0; i < image.data.length; i += 3, j += 4) {
tmpArray[j] = image.data[i]; // Red channel
tmpArray[j + 1] = image.data[i + 1]; // Green channel
tmpArray[j + 2] = image.data[i + 2]; // Blue channel
tmpArray[j + 3] = 255; // Alpha channel (fully opaque)
}
image.data = tmpArray;
}
const code = jsQR(image.data, image.width, image.height);
if(code)
return code.data;
else
return null;
}
};

View File

@@ -0,0 +1,29 @@
import { PDFDocument } from 'pdf-lib';
import { selectPages } from "./subDocumentFunctions";
import { PdfFile } from '../wrappers/PdfFile';
export async function splitPDF(file: PdfFile, splitAfterPageArray: number[]): Promise<PdfFile[]> {
const byteFile = await file.convertToPdfLibFile();
if (!byteFile?.pdfLib) return [];
const numberOfPages = byteFile.pdfLib.getPages().length;
let pagesArray: number[] = [];
let splitAfter = splitAfterPageArray.shift();
const subDocuments: PdfFile[] = [];
for (let i = 0; i < numberOfPages; i++) {
if(splitAfter && i > splitAfter && pagesArray.length > 0) {
subDocuments.push(await selectPages(byteFile, pagesArray));
splitAfter = splitAfterPageArray.shift();
pagesArray = [];
}
pagesArray.push(i);
}
subDocuments.push(await selectPages(byteFile, pagesArray));
pagesArray = [];
return subDocuments;
};

View File

@@ -0,0 +1,227 @@
import { PDFDocument } from 'pdf-lib';
import { PdfFile, fromPdfLib } from '../wrappers/PdfFile.js';
import { detectEmptyPages } from "./common/detectEmptyPages.js";
export async function sortPagesWithPreset(file: PdfFile, sortPreset: string, fancyPageSelector: string) {
if (sortPreset === "CUSTOM_PAGE_ORDER") {
return rearrangePages(file, fancyPageSelector);
}
const sortFunction = sorts[sortPreset];
if (!sortFunction) {
throw new Error("Operation not supported");
}
const byteFile = await file.convertToPdfLibFile();
if (!byteFile?.pdfLib) return byteFile;
const pageCount = byteFile.pdfLib.getPageCount();
const sortIndecies = sortFunction(pageCount);
return selectPages(byteFile, sortIndecies);
}
export async function rearrangePages(file: PdfFile, fancyPageSelector: string): Promise<PdfFile> {
const byteFile = await file.convertToPdfLibFile();
if (!byteFile?.pdfLib) return byteFile;
const pagesToExtractArray = parseFancyPageSelector(fancyPageSelector, byteFile.pdfLib.getPageCount());
const newDocument = selectPages(byteFile, pagesToExtractArray);
return newDocument;
};
export async function selectPages(file: PdfFile, pagesToExtractArray: number[]): Promise<PdfFile> {
const byteFile = await file.convertToPdfLibFile();
if (!byteFile?.pdfLib) return byteFile;
const subDocument = await PDFDocument.create();
// Check that array max number is not larger pdf pages number
if(Math.max(...pagesToExtractArray) >= byteFile.pdfLib.getPageCount()) {
throw new Error(`The PDF document only has ${byteFile.pdfLib.getPageCount()} pages and you tried to extract page ${Math.max(...pagesToExtractArray)}`);
}
const copiedPages = await subDocument.copyPages(byteFile.pdfLib, pagesToExtractArray);
for (let i = 0; i < copiedPages.length; i++) {
subDocument.addPage(copiedPages[i]);
}
return fromPdfLib(subDocument, file.filename);
}
export async function removePages(file: PdfFile, pagesToRemoveArray: number[]): Promise<PdfFile> {
const byteFile = await file.convertToPdfLibFile();
if (!byteFile?.pdfLib) return byteFile;
const pagesToExtractArray = invertSelection(pagesToRemoveArray, byteFile.pdfLib.getPageIndices())
return selectPages(byteFile, pagesToExtractArray);
}
export async function removeBlankPages(file: PdfFile, whiteThreashold: number) {
const emptyPages = await detectEmptyPages(file, whiteThreashold);
console.log("Empty Pages: ", emptyPages);
return removePages(file, emptyPages);
}
/**
* Parse the page selector string used in the 'PDF Page Organizer'
* @param pageOrderArr
* @param totalPages
* @returns
*/
function parseFancyPageSelector(pageNumbers: string, totalPages: number): number[] {
// Translated to JS from the original Java function
const pageOrderArr = pageNumbers.split(",")
const newPageOrder: number[] = [];
// loop through the page order array
pageOrderArr.forEach(element => {
if (element.toLocaleLowerCase() === "all") {
for (var i = 0; i < totalPages; i++) {
newPageOrder.push(i);
}
// As all pages are already added, no need to check further
return;
}
else if (element.match("\\d*n\\+?-?\\d*|\\d*\\+?n")) {
// Handle page order as a function
var coefficient = 0;
var constant = 0;
var coefficientExists = false;
var constantExists = false;
if (element.includes("n")) {
var parts = element.split("n");
if (!parts[0]) {
coefficient = parseInt(parts[0]);
coefficientExists = true;
}
if (parts.length > 1 && parts[1]) {
constant = parseInt(parts[1]);
constantExists = true;
}
} else if (element.includes("+")) {
constant = parseInt(element.replace("+", ""));
constantExists = true;
}
for (var i = 1; i <= totalPages; i++) {
var pageNum = coefficientExists ? coefficient * i : i;
pageNum += constantExists ? constant : 0;
if (pageNum <= totalPages && pageNum > 0) {
newPageOrder.push(pageNum - 1);
}
}
} else if (element.includes("-")) {
// split the range into start and end page
const range = element.split("-");
const start = parseInt(range[0]);
var end = parseInt(range[1]);
// check if the end page is greater than total pages
if (end > totalPages) {
end = totalPages;
}
// loop through the range of pages
for (var j = start; j <= end; j++) {
// print the current index
newPageOrder.push(j - 1);
}
} else {
// if the element is a single page
newPageOrder.push(parseInt(element) - 1);
}
});
return newPageOrder;
}
function invertSelection(selection: number[], pageIndecies: number[]): number[] {
const pageIndeciesCopy = [...pageIndecies];
return pageIndeciesCopy.filter(x => !selection.includes(x));
}
//////////////////
// Page Sorters //
//////////////////
function reverseSort(totalPages: number): number[] {
return [...Array(totalPages).keys()].reverse();
}
function duplexSort(totalPages: number): number[] {
// Translated to JS from the original Java function
const newPageOrder: number[] = [];
const half = Math.floor((totalPages + 1) / 2); // This ensures proper behavior with odd numbers of pages
for (let i = 1; i <= half; i++) {
newPageOrder.push(i - 1);
if (i <= totalPages - half) {
// Avoid going out of bounds
newPageOrder.push(totalPages - i);
}
}
return newPageOrder;
}
function bookletSort(totalPages: number): number[] {
const newPageOrder: number[] = [];
for (let i = 0; i < totalPages / 2; i++) {
newPageOrder.push(i);
newPageOrder.push(totalPages - i - 1);
}
return newPageOrder;
}
function sideStitchBooklet(totalPages: number): number[] {
const newPageOrder: number[] = [];
for (let i = 0; i < (totalPages + 3) / 4; i++) {
const begin = i * 4;
newPageOrder.push(Math.min(begin + 3, totalPages - 1));
newPageOrder.push(Math.min(begin, totalPages - 1));
newPageOrder.push(Math.min(begin + 1, totalPages - 1));
newPageOrder.push(Math.min(begin + 2, totalPages - 1));
}
return newPageOrder;
}
function oddEvenSplit(totalPages: number): number[] {
const newPageOrder: number[] = [];
for (let i = 1; i <= totalPages; i += 2) {
newPageOrder.push(i - 1);
}
for (let i = 2; i <= totalPages; i += 2) {
newPageOrder.push(i - 1);
}
return newPageOrder;
}
function removeFirst(totalPages: number): number[] {
return [...Array(totalPages-1).keys()].map(i => i+1);
}
function removeLast(totalPages: number): number[] {
return [...Array(totalPages-1).keys()];
}
function removeFirstAndLast(totalPages: number): number[] {
return [...Array(totalPages-2).keys()].map(i => i+1);
}
export type SortFunction = (totalPages: number) => number[];
type Sorts = {
[key: string]: SortFunction;
};
export const sorts: Sorts = Object.freeze({
"REVERSE_ORDER": reverseSort,
"DUPLEX_SORT": duplexSort,
"BOOKLET_SORT": bookletSort,
"SIDE_STITCH_BOOKLET_SORT": sideStitchBooklet,
"ODD_EVEN_SPLIT": oddEvenSplit,
"REMOVE_FIRST": removeFirst,
"REMOVE_LAST": removeLast,
"REMOVE_FIRST_AND_LAST": removeFirstAndLast,
});

View File

@@ -0,0 +1,56 @@
import { PDFDocument, ParseSpeeds } from 'pdf-lib';
import { PdfFile, fromPdfLib } from '../wrappers/PdfFile';
export type Metadata = {
deleteAll?: boolean, // Delete all metadata if set to true
author?: string, // The author of the document
creationDate?: Date, // The creation date of the document (format: yyyy/MM/dd HH:mm:ss)
creator?: string, // The creator of the document
keywords?: string, // The keywords for the document
modificationDate?: Date, // The modification date of the document (format: yyyy/MM/dd HH:mm:ss)
producer?: string, // The producer of the document
subject?: string, // The subject of the document
title?: string, // The title of the document
//trapped?: string, // The trapped status of the document
//allRequestParams?: {[key: string]: [key: string]}, // Map list of key and value of custom parameters. Note these must start with customKey and customValue if they are non-standard
}
export async function updateMetadata(file: PdfFile, metadata: Metadata|null): Promise<PdfFile> {
const pdfDoc = await file.getAsPdfLib();
if (!metadata || metadata.deleteAll) {
pdfDoc.setAuthor("");
pdfDoc.setCreationDate(new Date(0))
pdfDoc.setCreator("")
pdfDoc.setKeywords([])
pdfDoc.setModificationDate(new Date(0))
pdfDoc.setProducer("")
pdfDoc.setSubject("")
pdfDoc.setTitle("")
}
if (!metadata) {
return fromPdfLib(pdfDoc, file.filename);
}
if(metadata.author)
pdfDoc.setAuthor(metadata.author);
if(metadata.creationDate)
pdfDoc.setCreationDate(metadata.creationDate)
if(metadata.creator)
pdfDoc.setCreator(metadata.creator)
if(metadata.keywords)
pdfDoc.setKeywords(metadata.keywords.split(","))
if(metadata.modificationDate)
pdfDoc.setModificationDate(metadata.modificationDate)
if(metadata.producer)
pdfDoc.setProducer(metadata.producer)
if(metadata.subject)
pdfDoc.setSubject(metadata.subject)
if(metadata.title)
pdfDoc.setTitle(metadata.title)
// TODO add trapped and custom metadata. May need another library
return fromPdfLib(pdfDoc, file.filename);
};

View File

@@ -0,0 +1,24 @@
// Import injected libraries here!
import { sortPagesWithPreset, rearrangePages, selectPages, removePages, removeBlankPages } from "./functions/subDocumentFunctions";
import { impose } from "./functions/impose";
import { mergePDFs } from './functions/mergePDFs';
import { rotatePages } from './functions/rotatePages';
import { scaleContent} from './functions/scaleContent';
import { scalePage } from './functions/scalePage';
import { splitOn } from './functions/splitOn';
import { splitPDF } from './functions/splitPDF';
import { updateMetadata } from "./functions/updateMetadata";
export default {
sortPagesWithPreset, rearrangePages, selectPages, removePages, removeBlankPages,
impose,
mergePDFs,
rotatePages,
scaleContent,
scalePage,
splitOn,
splitPDF,
updateMetadata,
}

File diff suppressed because one or more lines are too long

View File

@@ -0,0 +1,104 @@
// imports browserfs via index.html script-tag
let wasmLocation = "/wasm/pdfcpu/";
let fs;
let Buffer;
// TODO: This can later be defered to load asynchronously
configureFs();
loadWasm();
function configureFs() {
BrowserFS.configure(
{
fs: "InMemory",
},
function (e) {
if (e) {
// An error happened!
throw e;
}
fs = BrowserFS.BFSRequire("fs");
Buffer = BrowserFS.BFSRequire("buffer").Buffer;
window.fs = fs;
window.Buffer = Buffer;
}
);
}
function loadWasm() {
const script = document.createElement("script");
script.src = wasmLocation + "/wasm_exec.js";
script.async = true;
document.body.appendChild(script);
}
const runWasm = async (param) => {
if (window.cachedWasmResponse === undefined) {
const response = await fetch(wasmLocation + "/pdfcpu.wasm");
const buffer = await response.arrayBuffer();
window.cachedWasmResponse = buffer;
window.go = new Go();
}
const { instance } = await WebAssembly.instantiate(
window.cachedWasmResponse,
window.go.importObject
);
window.go.argv = param;
await window.go.run(instance);
return window.go.exitCode;
};
async function loadFileAsync(data) {
console.log(`Writing file to MemoryFS`);
await fs.writeFile(`/input.pdf`, data);
console.log(`Write done. Validating...`);
let exitcode = await runWasm([
"pdfcpu.wasm",
"validate",
"-c",
"disable",
`/input.pdf`,
]);
if (exitcode !== 0)
throw new Error("There was an error validating your PDFs");
console.log(`File is Valid`);
}
export async function impose(snapshot, nup, format) {
};
export async function oneToOne(wasmArray, snapshot) {
await loadFileAsync(Buffer.from(snapshot));
console.log("Nuping File");
let exitcode = await runWasm(wasmArray);
if (exitcode !== 0) {
console.error("There was an error nuping your PDFs");
return;
}
await fs.unlink("input.pdf");
const contents = fs.readFileSync("output.pdf");
fs.unlink("output.pdf");
console.log("Your File ist Ready!");
return new Uint8Array(contents);
}
export async function manyToOne() {
//TODO: Do this of neccesary for some operations
}
export async function oneToMany() {
//TODO: Do this of neccesary for some operations
}
export async function manyToMany() {
//TODO: Do this of neccesary for some operations
}

View File

@@ -0,0 +1,145 @@
import { WasmFs } from '@wasmer/wasmfs';
import path from "path";
let webWasmLocation = "/wasm/";
let nodeWasmLocation = "./public/wasm/";
let fs;
const wasmfs = new WasmFs();
// TODO: This can later be defered to load asynchronously
(async () => {
await loadWasm();
await configureFs();
})();
async function configureFs() {
// Can't use BrowserFS: https://github.com/jvilk/BrowserFS/issues/271
fs = wasmfs.fs;
global.fs = fs;
console.log("InMemoryFs configured");
}
async function loadWasm() {
global.crypto = (await import("crypto")).webcrypto; // wasm dependecy
await import("./wasm_exec.js");
}
const runWasm = async (param) => {
if (global.cachedWasmResponse === undefined) {
const buffer = (await import("fs")).readFileSync(nodeWasmLocation + "/pdfcpu.wasm");
global.cachedWasmResponse = buffer;
global.go = new Go();
}
const { instance } = await WebAssembly.instantiate(
global.cachedWasmResponse,
global.go.importObject
);
global.go.argv = param;
await global.go.run(instance);
return global.go.exitCode;
};
async function loadFileAsync(data) {
console.log(`Writing file to Disk`);
fs.writeFileSync(`input.pdf`, data);
console.log(`Write done. Validating...`);
let exitcode = await runWasm([
"pdfcpu.wasm",
"validate",
"-c",
"disable",
`input.pdf`,
]);
if (exitcode !== 0)
throw new Error("There was an error validating your PDFs");
// // Get logs of command
// wasmfs.getStdOut().then(response => {
// console.log(response);
// });
console.log(`File is Valid`);
}
export async function oneToOne(wasmArray, snapshot) {
await loadFileAsync(Buffer.from(snapshot));
console.log("Nuping File");
let exitcode = await runWasm(wasmArray);
if (exitcode !== 0) {
console.error("There was an error nuping your PDFs");
return;
}
console.log("Nuping Done");
await checkExistsWithTimeout("/output.pdf", 1000);
console.log("Write started...");
// TODO: Make this more elegant, this waits for the write to finish.
// Maybe replace wasmfs with https://github.com/streamich/memfs
let fileSize;
while (true) {
fileSize = fs.statSync("/output.pdf").size;
await new Promise((resolve, reject) => {
setTimeout(() => {
resolve();
}, 50);
});
if(fileSize > 0 && fileSize == fs.statSync("/output.pdf").size) // Wait for file Size not changing anymore.
break;
}
console.log("Could be done?");
fs.unlinkSync("input.pdf");
const data = fs.readFileSync("/output.pdf");
if(data.length == 0) {
throw Error("File Size 0 that should not happen. The write probably didn't finish in time.");
}
fs.unlinkSync("output.pdf");
console.log("Your File ist Ready!");
return new Uint8Array(data);
}
export async function manyToOne() {
//TODO: Do this if necessary for some pdfcpu operations
}
export async function oneToMany() {
//TODO: Do this if necessary for some pdfcpu operations
}
export async function manyToMany() {
//TODO: Do this if necessary for some pdfcpu operations
}
// THX: https://stackoverflow.com/questions/26165725/nodejs-check-file-exists-if-not-wait-till-it-exist
function checkExistsWithTimeout(filePath, timeout) {
return new Promise(function (resolve, reject) {
var timer = setTimeout(function () {
watcher.close();
reject(new Error('File did not exists and was not created during the timeout.'));
}, timeout);
fs.access(filePath, fs.constants.R_OK, function (err) {
if (!err) {
clearTimeout(timer);
watcher.close();
resolve();
}
});
var dir = path.dirname(filePath);
var watcher = fs.watch(dir, function (eventType, filename) {
clearTimeout(timer);
watcher.close();
resolve();
});
});
}

Binary file not shown.

View File

@@ -0,0 +1,872 @@
// Copyright 2018 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
(() => {
// Map multiple JavaScript environments to a single common API,
// preferring web standards over Node.js API.
//
// Environments considered:
// - Browsers
// - Node.js
// - Electron
// - Parcel
// - Webpack
console.log("imported")
if (typeof global !== "undefined") {
// global already exists
} else if (typeof window !== "undefined") {
window.global = window;
} else if (typeof self !== "undefined") {
self.global = self;
} else {
throw new Error("cannot export Go (neither global, window nor self is defined)");
}
let logFS = false
var handler = {
get: function (target, property) {
if (property in target && target[property] instanceof Function) {
return function () {
if (logFS) {
console.log(property, 'called', arguments);
}
// 将callback替换
if (arguments[arguments.length - 1] instanceof Function) {
var origCB = arguments[arguments.length - 1];
var newCB = function () {
if (logFS) {
console.log('callback for', property, 'get called with args:', arguments);
}
return Reflect.apply(origCB, arguments.callee, arguments);
}
arguments[arguments.length - 1] = newCB;
}
return Reflect.apply(target[property], target, arguments);
}
} else {
return target[property]
}
}
}
if (!global.require && typeof require !== "undefined") {
global.require = require;
}
if (!global.fs && global.require) {
//const fs = require("fs");
if (typeof fs === "object" && fs !== null && Object.keys(fs).length !== 0) {
global.fs = fs;
}
}
const enosys = () => {
const err = new Error("not implemented");
err.code = "ENOSYS";
return err;
};
if (!global.fs) {
let outputBuf = "";
global.fs = {
constants: {
O_WRONLY: -1,
O_RDWR: -1,
O_CREAT: -1,
O_TRUNC: -1,
O_APPEND: -1,
O_EXCL: -1
}, // unused
writeSync(fd, buf) {
outputBuf += decoder.decode(buf);
const nl = outputBuf.lastIndexOf("\n");
if (nl != -1) {
console.log(outputBuf.substr(0, nl));
outputBuf = outputBuf.substr(nl + 1);
}
return buf.length;
},
write(fd, buf, offset, length, position, callback) {
if (offset !== 0 || length !== buf.length || position !== null) {
callback(enosys());
return;
}
const n = this.writeSync(fd, buf);
callback(null, n);
},
chmod(path, mode, callback) {
callback(enosys());
},
chown(path, uid, gid, callback) {
callback(enosys());
},
close(fd, callback) {
callback(enosys());
},
fchmod(fd, mode, callback) {
callback(enosys());
},
fchown(fd, uid, gid, callback) {
callback(enosys());
},
fstat(fd, callback) {
callback(enosys());
},
fsync(fd, callback) {
callback(null);
},
ftruncate(fd, length, callback) {
callback(enosys());
},
lchown(path, uid, gid, callback) {
callback(enosys());
},
link(path, link, callback) {
callback(enosys());
},
lstat(path, callback) {
callback(enosys());
},
mkdir(path, perm, callback) {
callback(enosys());
},
open(path, flags, mode, callback) {
callback(enosys());
},
read(fd, buffer, offset, length, position, callback) {
callback(enosys());
},
readdir(path, callback) {
callback(enosys());
},
readlink(path, callback) {
callback(enosys());
},
rename(from, to, callback) {
callback(enosys());
},
rmdir(path, callback) {
callback(enosys());
},
stat(path, callback) {
callback(enosys());
},
symlink(path, link, callback) {
callback(enosys());
},
truncate(path, length, callback) {
callback(enosys());
},
unlink(path, callback) {
callback(enosys());
},
utimes(path, atime, mtime, callback) {
callback(enosys());
},
};
}
if (!global.process) {
global.process = {
getuid() {
return -1;
},
getgid() {
return -1;
},
geteuid() {
return -1;
},
getegid() {
return -1;
},
getgroups() {
throw enosys();
},
pid: -1,
ppid: -1,
umask() {
throw enosys();
},
cwd() {
throw enosys();
},
chdir() {
throw enosys();
},
}
}
if (!global.crypto && global.require) {
const nodeCrypto = require("crypto");
global.crypto = {
getRandomValues(b) {
nodeCrypto.randomFillSync(b);
},
};
}
if (!global.crypto) {
throw new Error("global.crypto is not available, polyfill required (getRandomValues only)");
}
if (!global.performance) {
global.performance = {
now() {
const [sec, nsec] = process.hrtime();
return sec * 1000 + nsec / 1000000;
},
};
}
if (!global.TextEncoder && global.require) {
global.TextEncoder = require("util").TextEncoder;
}
if (!global.TextEncoder) {
throw new Error("global.TextEncoder is not available, polyfill required");
}
if (!global.TextDecoder && global.require) {
global.TextDecoder = require("util").TextDecoder;
}
if (!global.TextDecoder) {
throw new Error("global.TextDecoder is not available, polyfill required");
}
const isNodeJS = global.process && global.process.title === "node";
if (!isNodeJS) {
// console.log("ini browser fs")
// var myfs = global.BrowserFS.BFSRequire('fs');
// global.Buffer = global.BrowserFS.BFSRequire('buffer').Buffer;
// global.fs = myfs;
global.fs.constants = {
O_RDONLY: 0,
O_WRONLY: 1,
O_RDWR: 2,
O_CREAT: 64,
O_CREATE: 64,
O_EXCL: 128,
O_NOCTTY: 256,
O_TRUNC: 512,
O_APPEND: 1024,
O_DIRECTORY: 65536,
O_NOATIME: 262144,
O_NOFOLLOW: 131072,
O_SYNC: 1052672,
O_DIRECT: 16384,
O_NONBLOCK: 2048,
};
let outputBuf = "";
global.fs.writeSyncOriginal = global.fs.writeSync
global.fs.writeSync = function (fd, buf) {
if (fd === 1 || fd === 2) {
outputBuf += decoder.decode(buf);
const nl = outputBuf.lastIndexOf("\n");
if (nl != -1) {
console.log(outputBuf.substr(0, nl));
outputBuf = outputBuf.substr(nl + 1);
}
return buf.length;
} else {
return global.fs.writeSyncOriginal(...arguments);
}
};
global.fs.writeOriginal = global.fs.write
global.fs.write = function (fd, buf, offset, length, position, callback) {
// (corresponding to STDOUT/STDERR)
if (fd === 1 || fd === 2) {
if (offset !== 0 || length !== buf.length || position !== null) {
throw new Error("not implemented");
}
const n = this.writeSync(fd, buf);
callback(null, n, buf);
} else {
// buf: read buf first
arguments[1] = global.Buffer.from(arguments[1]);
return global.fs.writeOriginal(...arguments);
}
};
global.fs.openOriginal = global.fs.open
global.fs.open = function (path, flags, mode, callback) {
var myflags = 'r';
var O = global.fs.constants;
// Convert numeric flags to string flags
// FIXME: maybe wrong...
console.log("open dir?", path, 'flag', flags, myflags)
if (flags & O.O_WRONLY) { // 'w'
myflags = 'w';
if (flags & O.O_EXCL) {
myflags = 'wx';
}
} else if (flags & O.O_RDWR) { // 'r+' or 'w+'
if (flags & O.O_CREAT && flags & O.O_TRUNC) { // w+
if (flags & O.O_EXCL) {
myflags = 'wx+';
} else {
myflags = 'w+';
}
} else { // r+
myflags = 'r+';
}
} else if (flags & O.O_APPEND) { // 'a'
console.log("append error")
throw new Error("Not implmented");
} else {
// 打开文件
myflags = 'r+';
console.log("open dir?", path, 'flag', flags, myflags)
}
return global.fs.openOriginal(path, myflags, mode, callback);
};
global.fs.fstatOriginal = global.fs.fstat;
global.fs.fstat = function (fd, callback) {
return global.fs.fstatOriginal(fd, function () {
var retStat = arguments[1];
delete retStat['fileData'];
retStat.atimeMs = retStat.atime.getTime();
retStat.mtimeMs = retStat.mtime.getTime();
retStat.ctimeMs = retStat.ctime.getTime();
retStat.birthtimeMs = retStat.birthtime.getTime();
return callback(arguments[0], retStat);
});
};
global.fs.closeOriginal = global.fs.close;
global.fs.close = function (fd, callback) {
return global.fs.closeOriginal(fd, function () {
if (typeof arguments[0] === 'undefined') arguments[0] = null;
return callback(...arguments);
});
}
// global.fs.renameOriginal = global.fs.rename
// global.fs.rename = function (from, to, callback) {
// console.log("rename a0", arguments[0])
// global.fs.renameOriginal(from, to);
// callback(arguments[0])
// }
// global.fs.renameSyncOriginal = global.fs.renameSync
// global.fs.renameSync = function(fd, options) {
// console.log("Sync")
// }
global.fs = new Proxy(global.fs, handler);
}
// End of polyfills for common API.
const encoder = new TextEncoder("utf-8");
const decoder = new TextDecoder("utf-8");
global.Go = class {
constructor() {
this.argv = ["js"];
this.env = {};
this.exit = (code) => {
this.exitCode = code;
if (code !== 0) {
console.warn("exit code:", code);
}
};
this._exitPromise = new Promise((resolve) => {
this._resolveExitPromise = resolve;
});
this._pendingEvent = null;
this._scheduledTimeouts = new Map();
this._nextCallbackTimeoutID = 1;
const setInt64 = (addr, v) => {
this.mem.setUint32(addr + 0, v, true);
this.mem.setUint32(addr + 4, Math.floor(v / 4294967296), true);
}
const getInt64 = (addr) => {
const low = this.mem.getUint32(addr + 0, true);
const high = this.mem.getInt32(addr + 4, true);
return low + high * 4294967296;
}
const loadValue = (addr) => {
const f = this.mem.getFloat64(addr, true);
if (f === 0) {
return undefined;
}
if (!isNaN(f)) {
return f;
}
const id = this.mem.getUint32(addr, true);
return this._values[id];
}
const storeValue = (addr, v) => {
const nanHead = 0x7FF80000;
if (typeof v === "number" && v !== 0) {
if (isNaN(v)) {
this.mem.setUint32(addr + 4, nanHead, true);
this.mem.setUint32(addr, 0, true);
return;
}
this.mem.setFloat64(addr, v, true);
return;
}
if (v === undefined) {
this.mem.setFloat64(addr, 0, true);
return;
}
let id = this._ids.get(v);
if (id === undefined) {
id = this._idPool.pop();
if (id === undefined) {
id = this._values.length;
}
this._values[id] = v;
this._goRefCounts[id] = 0;
this._ids.set(v, id);
}
this._goRefCounts[id]++;
let typeFlag = 0;
switch (typeof v) {
case "object":
if (v !== null) {
typeFlag = 1;
}
break;
case "string":
typeFlag = 2;
break;
case "symbol":
typeFlag = 3;
break;
case "function":
typeFlag = 4;
break;
}
this.mem.setUint32(addr + 4, nanHead | typeFlag, true);
this.mem.setUint32(addr, id, true);
}
const loadSlice = (addr) => {
const array = getInt64(addr + 0);
const len = getInt64(addr + 8);
return new Uint8Array(this._inst.exports.mem.buffer, array, len);
}
const loadSliceOfValues = (addr) => {
const array = getInt64(addr + 0);
const len = getInt64(addr + 8);
const a = new Array(len);
for (let i = 0; i < len; i++) {
a[i] = loadValue(array + i * 8);
}
return a;
}
const loadString = (addr) => {
const saddr = getInt64(addr + 0);
const len = getInt64(addr + 8);
return decoder.decode(new DataView(this._inst.exports.mem.buffer, saddr, len));
}
const timeOrigin = Date.now() - performance.now();
this.importObject = {
go: {
// Go's SP does not change as long as no Go code is running. Some operations (e.g. calls, getters and setters)
// may synchronously trigger a Go event handler. This makes Go code get executed in the middle of the imported
// function. A goroutine can switch to a new stack if the current stack is too small (see morestack function).
// This changes the SP, thus we have to update the SP used by the imported function.
// func wasmExit(code int32)
"runtime.wasmExit": (sp) => {
sp >>>= 0;
const code = this.mem.getInt32(sp + 8, true);
this.exited = true;
delete this._inst;
delete this._values;
delete this._goRefCounts;
delete this._ids;
delete this._idPool;
this.exit(code);
},
// func wasmWrite(fd uintptr, p unsafe.Pointer, n int32)
"runtime.wasmWrite": (sp) => {
sp >>>= 0;
const fd = getInt64(sp + 8);
const p = getInt64(sp + 16);
const n = this.mem.getInt32(sp + 24, true);
fs.writeSync(fd, new Uint8Array(this._inst.exports.mem.buffer, p, n));
},
// func resetMemoryDataView()
"runtime.resetMemoryDataView": (sp) => {
sp >>>= 0;
this.mem = new DataView(this._inst.exports.mem.buffer);
},
// func nanotime1() int64
"runtime.nanotime1": (sp) => {
sp >>>= 0;
setInt64(sp + 8, (timeOrigin + performance.now()) * 1000000);
},
// func walltime1() (sec int64, nsec int32)
"runtime.walltime1": (sp) => {
sp >>>= 0;
const msec = (new Date).getTime();
setInt64(sp + 8, msec / 1000);
this.mem.setInt32(sp + 16, (msec % 1000) * 1000000, true);
},
// func scheduleTimeoutEvent(delay int64) int32
"runtime.scheduleTimeoutEvent": (sp) => {
sp >>>= 0;
const id = this._nextCallbackTimeoutID;
this._nextCallbackTimeoutID++;
this._scheduledTimeouts.set(id, setTimeout(
() => {
this._resume();
while (this._scheduledTimeouts.has(id)) {
// for some reason Go failed to register the timeout event, log and try again
// (temporary workaround for https://github.com/golang/go/issues/28975)
console.warn("scheduleTimeoutEvent: missed timeout event");
this._resume();
}
},
getInt64(sp + 8) + 1, // setTimeout has been seen to fire up to 1 millisecond early
));
this.mem.setInt32(sp + 16, id, true);
},
// func clearTimeoutEvent(id int32)
"runtime.clearTimeoutEvent": (sp) => {
sp >>>= 0;
const id = this.mem.getInt32(sp + 8, true);
clearTimeout(this._scheduledTimeouts.get(id));
this._scheduledTimeouts.delete(id);
},
// func getRandomData(r []byte)
"runtime.getRandomData": (sp) => {
sp >>>= 0;
crypto.getRandomValues(loadSlice(sp + 8));
},
// func finalizeRef(v ref)
"syscall/js.finalizeRef": (sp) => {
sp >>>= 0;
const id = this.mem.getUint32(sp + 8, true);
this._goRefCounts[id]--;
if (this._goRefCounts[id] === 0) {
const v = this._values[id];
this._values[id] = null;
this._ids.delete(v);
this._idPool.push(id);
}
},
// func stringVal(value string) ref
"syscall/js.stringVal": (sp) => {
sp >>>= 0;
storeValue(sp + 24, loadString(sp + 8));
},
// func valueGet(v ref, p string) ref
"syscall/js.valueGet": (sp) => {
sp >>>= 0;
const result = Reflect.get(loadValue(sp + 8), loadString(sp + 16));
sp = this._inst.exports.getsp() >>> 0; // see comment above
storeValue(sp + 32, result);
},
// func valueSet(v ref, p string, x ref)
"syscall/js.valueSet": (sp) => {
sp >>>= 0;
Reflect.set(loadValue(sp + 8), loadString(sp + 16), loadValue(sp + 32));
},
// func valueDelete(v ref, p string)
"syscall/js.valueDelete": (sp) => {
sp >>>= 0;
Reflect.deleteProperty(loadValue(sp + 8), loadString(sp + 16));
},
// func valueIndex(v ref, i int) ref
"syscall/js.valueIndex": (sp) => {
sp >>>= 0;
storeValue(sp + 24, Reflect.get(loadValue(sp + 8), getInt64(sp + 16)));
},
// valueSetIndex(v ref, i int, x ref)
"syscall/js.valueSetIndex": (sp) => {
sp >>>= 0;
Reflect.set(loadValue(sp + 8), getInt64(sp + 16), loadValue(sp + 24));
},
// func valueCall(v ref, m string, args []ref) (ref, bool)
"syscall/js.valueCall": (sp) => {
sp >>>= 0;
try {
const v = loadValue(sp + 8);
const m = Reflect.get(v, loadString(sp + 16));
const args = loadSliceOfValues(sp + 32);
const result = Reflect.apply(m, v, args);
sp = this._inst.exports.getsp() >>> 0; // see comment above
storeValue(sp + 56, result);
this.mem.setUint8(sp + 64, 1);
} catch (err) {
storeValue(sp + 56, err);
this.mem.setUint8(sp + 64, 0);
}
},
// func valueInvoke(v ref, args []ref) (ref, bool)
"syscall/js.valueInvoke": (sp) => {
sp >>>= 0;
try {
const v = loadValue(sp + 8);
const args = loadSliceOfValues(sp + 16);
const result = Reflect.apply(v, undefined, args);
sp = this._inst.exports.getsp() >>> 0; // see comment above
storeValue(sp + 40, result);
this.mem.setUint8(sp + 48, 1);
} catch (err) {
storeValue(sp + 40, err);
this.mem.setUint8(sp + 48, 0);
}
},
// func valueNew(v ref, args []ref) (ref, bool)
"syscall/js.valueNew": (sp) => {
sp >>>= 0;
try {
const v = loadValue(sp + 8);
const args = loadSliceOfValues(sp + 16);
const result = Reflect.construct(v, args);
sp = this._inst.exports.getsp() >>> 0; // see comment above
storeValue(sp + 40, result);
this.mem.setUint8(sp + 48, 1);
} catch (err) {
storeValue(sp + 40, err);
this.mem.setUint8(sp + 48, 0);
}
},
// func valueLength(v ref) int
"syscall/js.valueLength": (sp) => {
sp >>>= 0;
setInt64(sp + 16, parseInt(loadValue(sp + 8).length));
},
// valuePrepareString(v ref) (ref, int)
"syscall/js.valuePrepareString": (sp) => {
sp >>>= 0;
const str = encoder.encode(String(loadValue(sp + 8)));
storeValue(sp + 16, str);
setInt64(sp + 24, str.length);
},
// valueLoadString(v ref, b []byte)
"syscall/js.valueLoadString": (sp) => {
sp >>>= 0;
const str = loadValue(sp + 8);
loadSlice(sp + 16).set(str);
},
// func valueInstanceOf(v ref, t ref) bool
"syscall/js.valueInstanceOf": (sp) => {
sp >>>= 0;
this.mem.setUint8(sp + 24, (loadValue(sp + 8) instanceof loadValue(sp + 16)) ? 1 : 0);
},
// func copyBytesToGo(dst []byte, src ref) (int, bool)
"syscall/js.copyBytesToGo": (sp) => {
sp >>>= 0;
const dst = loadSlice(sp + 8);
const src = loadValue(sp + 32);
if (!(src instanceof Uint8Array || src instanceof Uint8ClampedArray)) {
this.mem.setUint8(sp + 48, 0);
return;
}
const toCopy = src.subarray(0, dst.length);
dst.set(toCopy);
setInt64(sp + 40, toCopy.length);
this.mem.setUint8(sp + 48, 1);
},
// func copyBytesToJS(dst ref, src []byte) (int, bool)
"syscall/js.copyBytesToJS": (sp) => {
sp >>>= 0;
const dst = loadValue(sp + 8);
const src = loadSlice(sp + 16);
if (!(dst instanceof Uint8Array || dst instanceof Uint8ClampedArray)) {
this.mem.setUint8(sp + 48, 0);
return;
}
const toCopy = src.subarray(0, dst.length);
dst.set(toCopy);
setInt64(sp + 40, toCopy.length);
this.mem.setUint8(sp + 48, 1);
},
"debug": (value) => {
console.log(value);
},
}
};
}
async run(instance) {
if (!(instance instanceof WebAssembly.Instance)) {
throw new Error("Go.run: WebAssembly.Instance expected");
}
this._inst = instance;
this.mem = new DataView(this._inst.exports.mem.buffer);
this._values = [ // JS values that Go currently has references to, indexed by reference id
NaN,
0,
null,
true,
false,
global,
this,
];
this._goRefCounts = new Array(this._values.length).fill(Infinity); // number of references that Go has to a JS value, indexed by reference id
this._ids = new Map([ // mapping from JS values to reference ids
[0, 1],
[null, 2],
[true, 3],
[false, 4],
[global, 5],
[this, 6],
]);
this._idPool = []; // unused ids that have been garbage collected
this.exited = false; // whether the Go program has exited
// Pass command line arguments and environment variables to WebAssembly by writing them to the linear memory.
let offset = 4096;
const strPtr = (str) => {
const ptr = offset;
const bytes = encoder.encode(str + "\0");
new Uint8Array(this.mem.buffer, offset, bytes.length).set(bytes);
offset += bytes.length;
if (offset % 8 !== 0) {
offset += 8 - (offset % 8);
}
return ptr;
};
const argc = this.argv.length;
const argvPtrs = [];
this.argv.forEach((arg) => {
argvPtrs.push(strPtr(arg));
});
argvPtrs.push(0);
const keys = Object.keys(this.env).sort();
keys.forEach((key) => {
argvPtrs.push(strPtr(`${key}=${this.env[key]}`));
});
argvPtrs.push(0);
const argv = offset;
argvPtrs.forEach((ptr) => {
this.mem.setUint32(offset, ptr, true);
this.mem.setUint32(offset + 4, 0, true);
offset += 8;
});
this._inst.exports.run(argc, argv);
if (this.exited) {
this._resolveExitPromise();
}
await this._exitPromise;
}
_resume() {
if (this.exited) {
throw new Error("Go program has already exited");
}
this._inst.exports.resume();
if (this.exited) {
this._resolveExitPromise();
}
}
_makeFuncWrapper(id) {
const go = this;
return function () {
const event = {
id: id,
this: this,
args: arguments
};
go._pendingEvent = event;
go._resume();
return event.result;
};
}
}
if (
typeof module !== "undefined" &&
global.require &&
global.require.main === module &&
global.process &&
global.process.versions &&
!global.process.versions.electron
) {
if (process.argv.length < 3) {
console.error("usage: go_js_wasm_exec [wasm binary] [arguments]");
process.exit(1);
}
const go = new Go();
go.argv = process.argv.slice(2);
go.env = Object.assign({
TMPDIR: require("os").tmpdir()
}, process.env);
go.exit = process.exit;
WebAssembly.instantiate(fs.readFileSync(process.argv[2]), go.importObject).then((result) => {
process.on("exit", (code) => { // Node.js exits if no event handler is pending
if (code === 0 && !go.exited) {
// deadlock, make Go print error and stack traces
go._pendingEvent = {
id: 0
};
go._resume();
}
});
return go.run(result.instance);
}).catch((err) => {
console.error(err);
process.exit(1);
});
}
})();

View File

@@ -0,0 +1,44 @@
import { Operation } from "../../declarations/Operation";
export function organizeWaitOperations(operations: Operation[]) {
// Initialize an object to store the counts and associated "done" operations
const waitCounts = {};
const doneOperations = {};
// Function to count "type: wait" operations and associate "done" operations per id
function countWaitOperationsAndDone(operations: Operation[]) {
for (const operation of operations) {
if (operation.type === "wait") {
const id = operation.values.id;
if (id in waitCounts) {
waitCounts[id]++;
} else {
waitCounts[id] = 1;
}
}
if (operation.type === "done") {
const id = operation.values.id;
doneOperations[id] = operation;
}
if (operation.operations) {
countWaitOperationsAndDone(operation.operations);
}
}
}
// Start counting and associating from the root operations
countWaitOperationsAndDone(operations);
// Combine counts and associated "done" operations
const result = {};
for (const id in waitCounts) {
result[id] = {
waitCount: waitCounts[id],
doneOperation: doneOperations[id],
input: []
};
}
return result;
}

View File

@@ -0,0 +1,183 @@
import { organizeWaitOperations } from "./organizeWaitOperations.js";
import { Operation } from "../../declarations/Operation.js";
import { PDF } from "../../declarations/PDF.js";
export async function * traverseOperations(operations: Operation[], input: PDF[] | PDF, Operations: AllOperations) {
const waitOperations = organizeWaitOperations(operations);
let results: PDF[] = [];
yield* nextOperation(operations, input);
return results;
async function * nextOperation(operations: Operation[], input: PDF[] | PDF) {
if(Array.isArray(operations) && operations.length == 0) { // isEmpty
if(Array.isArray(input)) {
console.log("operation done: " + input[0].fileName + (input.length > 1 ? "+" : ""));
results = results.concat(input);
return;
}
else {
console.log("operation done: " + input.fileName);
results.push(input);
return;
}
}
for (let i = 0; i < operations.length; i++) {
yield* computeOperation(operations[i], structuredClone(input));
}
}
async function * computeOperation(operation: Operation, input: PDF|PDF[]) {
yield "Starting: " + operation.type;
switch (operation.type) {
case "done": // Skip this, because it is a valid node.
break;
case "wait":
const waitOperation = waitOperations[operation.values.id];
if(Array.isArray(input)) {
waitOperation.input.concat(input); // TODO: May have unexpected concequences. Needs further testing!
}
else {
waitOperation.input.push(input);
}
waitOperation.waitCount--;
if(waitOperation.waitCount == 0) {
yield* nextOperation(waitOperation.doneOperation.operations, waitOperation.input);
}
break;
case "extract":
yield* nToN(input, operation, async (input) => {
input.fileName += "_extractedPages";
input.buffer = await Operations.extractPages(input.buffer, operation.values["pagesToExtractArray"]);
});
break;
case "impose":
yield* nToN(input, operation, async (input) => {
input.fileName += "_imposed";
input.buffer = await Operations.impose(input.buffer, operation.values["nup"], operation.values["format"]);
});
break;
case "merge":
yield* nToOne(input, operation, async (inputs) => {
return {
originalFileName: inputs.map(input => input.originalFileName).join("_and_"),
fileName: inputs.map(input => input.fileName).join("_and_") + "_merged",
buffer: await Operations.mergePDFs(inputs.map(input => input.buffer))
}
});
break;
case "rotate":
yield* nToN(input, operation, async (input) => {
input.fileName += "_turned";
input.buffer = await Operations.rotatePages(input.buffer, operation.values["rotation"]);
});
break;
case "split":
// TODO: A split might break the done condition, it may count multiple times. Needs further testing!
yield* oneToN(input, operation, async (input) => {
const splitResult = await Operations.splitPDF(input.buffer, operation.values["pagesToSplitAfterArray"]);
const splits = [];
for (let j = 0; j < splitResult.length; j++) {
splits.push({
originalFileName: input.originalFileName,
fileName: input.fileName + "_split" + j,
buffer: splitResult[j]
})
}
return splits;
});
break;
case "updateMetadata":
yield* nToN(input, operation, async (input) => {
input.fileName += "_metadataEdited";
input.buffer = await Operations.updateMetadata(input.buffer, operation.values["metadata"]);
});
break;
case "organizePages":
yield* nToN(input, operation, async (input) => {
input.fileName += "_pagesOrganized";
input.buffer = await Operations.organizePages(input.buffer, operation.values["operation"], operation.values["customOrderString"]);
});
break;
case "removeBlankPages":
yield* nToN(input, operation, async (input) => {
input.fileName += "_removedBlanks";
input.buffer = await Operations.removeBlankPages(input.buffer, operation.values["whiteThreashold"]);
});
break;
case "splitOn":
yield* oneToN(input, operation, async (input) => {
const splitResult = await Operations.splitOn(input.buffer, operation.values["type"], operation.values["whiteThreashold"]);
const splits = [];
for (let j = 0; j < splitResult.length; j++) {
splits.push({
originalFileName: input.originalFileName,
fileName: input.fileName + "_split" + j,
buffer: splitResult[j]
})
}
return splits;
});
break;
default:
throw new Error(`${operation.type} not implemented yet.`);
break;
}
}
/**
*
* @param {PDF|PDF[]} input
* @param {JSON} operation
* @returns {undefined}
*/
async function * nToOne(inputs, operation, callback) {
inputs = Array.from(inputs); // Convert single values to array, keep arrays as is.
inputs = await callback(inputs);
yield* nextOperation(operation.operations, inputs);
}
/**
*
* @param {PDF|PDF[]} input
* @param {JSON} operation
* @returns {undefined}
*/
async function * oneToN(input, operation, callback) {
if(Array.isArray(input)) {
let output = [];
for (let i = 0; i < input.length; i++) {
output = output.concat(await callback(input[i]));
}
yield* nextOperation(operation.operations, output);
}
else {
input = await callback(input);
yield* nextOperation(operation.operations, input);
}
}
/**
*
* @param {PDF|PDF[]} input
* @param {JSON} operation
* @returns {undefined}
*/
async function * nToN(input, operation, callback) {
if(Array.isArray(input)) {
for (let i = 0; i < input.length; i++) {
await callback(input[i]);
}
yield* nextOperation(operation.operations, input);
}
else {
await callback(input);
yield* nextOperation(operation.operations, input);
}
}
}

View File

@@ -0,0 +1,96 @@
import { PDFDocument } from 'pdf-lib';
import * as PDFJS from 'pdfjs-dist';
import { PDFDocumentProxy } from 'pdfjs-dist/types/src/display/api';
export class PdfFile {
byteArray: Uint8Array | null;
pdfLib: PDFDocument | null;
pdfJs: PDFDocumentProxy | null;
filename: string;
constructor() {
this.byteArray = null;
this.pdfLib = null;
this.pdfJs = null;
this.filename = "";
}
async convertToByteArrayFile(): Promise<PdfFile> {
if (this.byteArray) return this;
var byteArray: Uint8Array|null = null;
if (this.pdfLib) {
byteArray = await this.pdfLib.save();
} else if (this.pdfJs) {
byteArray = await this.pdfJs.getData();
}
return fromUint8Array(byteArray!, this.filename);
}
async convertToPdfLibFile(): Promise<PdfFile> {
if (this.pdfLib) return this;
const byteFile = await this.convertToByteArrayFile();
const pdfLib = await PDFDocument.load(byteFile.byteArray!, {
updateMetadata: false,
});
return fromPdfLib(pdfLib, this.filename);
}
async convertToPdfJsFile(): Promise<PdfFile> {
if (this.pdfJs) return this;
const byteFile = await this.convertToByteArrayFile();
const pdfJs = await PDFJS.getDocument(byteFile.byteArray!).promise;
return fromPdfJs(pdfJs, this.filename);
}
async getAsByteArray(): Promise<Uint8Array> {
const file = await this.convertToByteArrayFile();
return file.byteArray!;
}
async getAsPdfLib(): Promise<PDFDocument> {
const file = await this.convertToPdfLibFile();
return file.pdfLib!;
}
async getAsPdfJs(): Promise<PDFDocumentProxy> {
const file = await this.convertToPdfJsFile();
return file.pdfJs!;
}
}
export function fromMulterFile(value: Express.Multer.File): PdfFile {
return fromUint8Array(value.buffer, value.originalname)
}
export function fromUint8Array(value: Uint8Array, filename: string): PdfFile {
const out = new PdfFile();
out.byteArray = value;
out.filename = filename;
return out;
}
export function fromPdfLib(value: PDFDocument, filename: string): PdfFile {
const out = new PdfFile();
out.pdfLib = value;
out.filename = filename;
return out;
}
export function fromPdfJs(value: PDFDocumentProxy, filename: string): PdfFile {
const out = new PdfFile();
out.pdfJs = value;
out.filename = filename;
return out;
}
export async function convertAllToByteArrayFile(files: PdfFile[]): Promise<(PdfFile)[]> {
const pdfPromises = files.map(s => s.convertToByteArrayFile());
return await Promise.all(pdfPromises);
}
export async function convertAllToPdfLibFile(files: PdfFile[]): Promise<(PdfFile)[]> {
const pdfPromises = files.map(s => s.convertToPdfLibFile());
return await Promise.all(pdfPromises);
}
export async function convertAllToPdfJsFile(files: PdfFile[]): Promise<(PdfFile)[]> {
const pdfPromises = files.map(s => s.convertToPdfJsFile());
return await Promise.all(pdfPromises);
}