djoka/bentopdf
1
0
Fork 0
Code Issues Pull Requests Actions 4 Packages Projects Releases Wiki Activity

Reset branch to main

Browse Source
This commit is contained in:
Sebastian Espei
2026-03-09 17:02:30 +01:00
parent 68343a4c9a
commit e474d11730
606 changed files with 189976 additions and 43358 deletions
Download Patch File Download Diff File Expand all files Collapse all files

222
src/js/utils/helpers.ts
View File

@@ -2,8 +2,7 @@ import createModule from '@neslinesli93/qpdf-wasm';
import { showLoader, hideLoader, showAlert } from '../ui.js';
import { createIcons } from 'lucide';
import { state, resetState } from '../state.js';
import * as pdfjsLib from 'pdfjs-dist'
import * as pdfjsLib from 'pdfjs-dist';
const STANDARD_SIZES = {
A4: { width: 595.28, height: 841.89 },
@@ -50,14 +49,14 @@ export function convertPoints(points: any, unit: any) {
// Convert hex color to RGB
export function hexToRgb(hex: string): { r: number; g: number; b: number } {
const result = /^#?([a-f\d]{2})([a-f\d]{2})([a-f\d]{2})$/i.exec(hex)
const result = /^#?([a-f\d]{2})([a-f\d]{2})([a-f\d]{2})$/i.exec(hex);
return result
? {
r: parseInt(result[1], 16) / 255,
g: parseInt(result[2], 16) / 255,
b: parseInt(result[3], 16) / 255,
}
: { r: 0, g: 0, b: 0 }
r: parseInt(result[1], 16) / 255,
g: parseInt(result[2], 16) / 255,
b: parseInt(result[3], 16) / 255,
}
: { r: 0, g: 0, b: 0 };
}
export const formatBytes = (bytes: any, decimals = 1) => {
@@ -89,7 +88,10 @@ export const readFileAsArrayBuffer = (file: any) => {
});
};
export function parsePageRanges(rangeString: string, totalPages: number): number[] {
export function parsePageRanges(
rangeString: string,
totalPages: number
): number[] {
if (!rangeString || rangeString.trim() === '') {
return Array.from({ length: totalPages }, (_, i) => i);
}
@@ -128,11 +130,9 @@ export function parsePageRanges(rangeString: string, totalPages: number): number
}
}
return Array.from(indices).sort((a, b) => a - b);
}
/**
* Formats an ISO 8601 date string (e.g., "2008-02-21T17:15:56-08:00")
* into a localized, human-readable string.
@@ -198,7 +198,7 @@ export function formatStars(num: number) {
return (num / 1000).toFixed(1) + 'K';
}
return num.toLocaleString();
};
}
/**
* Truncates a filename to a maximum length, adding ellipsis if needed.
@@ -207,14 +207,18 @@ export function formatStars(num: number) {
* @param maxLength - Maximum length (default: 30)
* @returns Truncated filename with ellipsis if needed
*/
export function truncateFilename(filename: string, maxLength: number = 25): string {
export function truncateFilename(
filename: string,
maxLength: number = 25
): string {
if (filename.length <= maxLength) {
return filename;
}
const lastDotIndex = filename.lastIndexOf('.');
const extension = lastDotIndex !== -1 ? filename.substring(lastDotIndex) : '';
const nameWithoutExt = lastDotIndex !== -1 ? filename.substring(0, lastDotIndex) : filename;
const nameWithoutExt =
lastDotIndex !== -1 ? filename.substring(0, lastDotIndex) : filename;
const availableLength = maxLength - extension.length - 3; // 3 for '...'
@@ -225,7 +229,10 @@ export function truncateFilename(filename: string, maxLength: number = 25): stri
return nameWithoutExt.substring(0, availableLength) + '...' + extension;
}
export function formatShortcutDisplay(shortcut: string, isMac: boolean): string {
export function formatShortcutDisplay(
shortcut: string,
isMac: boolean
): string {
if (!shortcut) return '';
return shortcut
.replace('mod', isMac ? '⌘' : 'Ctrl')
@@ -233,7 +240,7 @@ export function formatShortcutDisplay(shortcut: string, isMac: boolean): string
.replace('alt', isMac ? '⌥' : 'Alt')
.replace('shift', 'Shift')
.split('+')
.map(k => k.charAt(0).toUpperCase() + k.slice(1))
.map((k) => k.charAt(0).toUpperCase() + k.slice(1))
.join(isMac ? '' : '+');
}
@@ -263,7 +270,7 @@ export function resetAndReloadTool(preResetCallback?: () => void) {
export function getPDFDocument(src: any) {
let params = src;
// Handle different input types similar to how getDocument handles them,
// Handle different input types similar to how getDocument handles them,
// but we ensure we have an object to attach wasmUrl to.
if (typeof src === 'string') {
params = { url: src };
@@ -285,20 +292,171 @@ export function getPDFDocument(src: any) {
}
/**
* Returns a sanitized PDF filename.
*
* The provided filename is processed as follows:
* - Removes a trailing `.pdf` file extension (case-insensitive)
* - Trims leading and trailing whitespace
* - Truncates the name to a maximum of 80 characters
*
* @param filename The original filename (including extension)
* @returns The sanitized filename without the `.pdf` extension, limited to 80 characters
* Escape HTML special characters to prevent XSS
* @param text - The text to escape
* @returns The escaped text
*/
export function getCleanPdfFilename(filename: string): string {
let clean = filename.replace(/\.pdf$/i, '').trim();
if (clean.length > 80) {
clean = clean.slice(0, 80);
}
return clean;
export function escapeHtml(text: string): string {
const map: Record<string, string> = {
'&': '&amp;',
'<': '&lt;',
'>': '&gt;',
'"': '&quot;',
"'": '&#039;',
};
return text.replace(/[&<>"']/g, (m) => map[m]);
}
export function uint8ArrayToBase64(bytes: Uint8Array): string {
const CHUNK_SIZE = 0x8000;
const chunks: string[] = [];
for (let i = 0; i < bytes.length; i += CHUNK_SIZE) {
const chunk = bytes.subarray(i, Math.min(i + CHUNK_SIZE, bytes.length));
chunks.push(String.fromCharCode(...chunk));
}
return btoa(chunks.join(''));
}
export function sanitizeEmailHtml(html: string): string {
if (!html) return html;
let sanitized = html;
sanitized = sanitized.replace(/<head[^>]*>[\s\S]*?<\/head>/gi, '');
sanitized = sanitized.replace(/<style[^>]*>[\s\S]*?<\/style>/gi, '');
sanitized = sanitized.replace(/<script[^>]*>[\s\S]*?<\/script>/gi, '');
sanitized = sanitized.replace(/<link[^>]*>/gi, '');
sanitized = sanitized.replace(/\s+style=["'][^"']*["']/gi, '');
sanitized = sanitized.replace(/\s+class=["'][^"']*["']/gi, '');
sanitized = sanitized.replace(/\s+data-[a-z-]+=["'][^"']*["']/gi, '');
sanitized = sanitized.replace(
/<img[^>]*(?:width=["']1["'][^>]*height=["']1["']|height=["']1["'][^>]*width=["']1["'])[^>]*\/?>/gi,
''
);
sanitized = sanitized.replace(
/href=["']https?:\/\/[^"']*safelinks\.protection\.outlook\.com[^"']*url=([^&"']+)[^"']*["']/gi,
(match, encodedUrl) => {
try {
const decodedUrl = decodeURIComponent(encodedUrl);
return `href="${decodedUrl}"`;
} catch {
return match;
}
}
);
sanitized = sanitized.replace(/\s+originalsrc=["'][^"']*["']/gi, '');
sanitized = sanitized.replace(
/href=["']([^"']{500,})["']/gi,
(match, url) => {
const baseUrl = url.split('?')[0];
if (baseUrl && baseUrl.length < 200) {
return `href="${baseUrl}"`;
}
return `href="${url.substring(0, 200)}"`;
}
);
sanitized = sanitized.replace(
/\s+(cellpadding|cellspacing|bgcolor|border|valign|align|width|height|role|dir|id)=["'][^"']*["']/gi,
''
);
sanitized = sanitized.replace(/<\/?table[^>]*>/gi, '<div>');
sanitized = sanitized.replace(/<\/?tbody[^>]*>/gi, '');
sanitized = sanitized.replace(/<\/?thead[^>]*>/gi, '');
sanitized = sanitized.replace(/<\/?tfoot[^>]*>/gi, '');
sanitized = sanitized.replace(/<tr[^>]*>/gi, '<div>');
sanitized = sanitized.replace(/<\/tr>/gi, '</div>');
sanitized = sanitized.replace(/<td[^>]*>/gi, '<span> ');
sanitized = sanitized.replace(/<\/td>/gi, ' </span>');
sanitized = sanitized.replace(/<th[^>]*>/gi, '<strong> ');
sanitized = sanitized.replace(/<\/th>/gi, ' </strong>');
sanitized = sanitized.replace(/<div>\s*<\/div>/gi, '');
sanitized = sanitized.replace(/<span>\s*<\/span>/gi, '');
sanitized = sanitized.replace(/(<div>)+/gi, '<div>');
sanitized = sanitized.replace(/(<\/div>)+/gi, '</div>');
sanitized = sanitized.replace(
/<a[^>]*href=["']\s*["'][^>]*>([^<]*)<\/a>/gi,
'$1'
);
const MAX_HTML_SIZE = 100000;
if (sanitized.length > MAX_HTML_SIZE) {
const truncateAt = sanitized.lastIndexOf('</div>', MAX_HTML_SIZE);
if (truncateAt > MAX_HTML_SIZE / 2) {
sanitized = sanitized.substring(0, truncateAt) + '</div></body></html>';
} else {
sanitized = sanitized.substring(0, MAX_HTML_SIZE) + '...</body></html>';
}
}
return sanitized;
}
/**
* Formats a raw RFC 2822 date string into a nicer human-readable format,
* while preserving the original timezone and time.
* Example input: "Sun, 8 Jan 2017 20:37:44 +0200"
* Example output: "Sunday, January 8, 2017 at 8:37 PM (+0200)"
*/
export function formatRawDate(raw: string): string {
try {
const match = raw.match(
/([A-Za-z]{3}),\s+(\d{1,2})\s+([A-Za-z]{3})\s+(\d{4})\s+(\d{2}):(\d{2})(?::(\d{2}))?\s+([+-]\d{4})/
);
if (match) {
const [
,
dayAbbr,
dom,
monthAbbr,
year,
hoursStr,
minsStr,
secsStr,
timezone,
] = match;
const days: Record<string, string> = {
Sun: 'Sunday',
Mon: 'Monday',
Tue: 'Tuesday',
Wed: 'Wednesday',
Thu: 'Thursday',
Fri: 'Friday',
Sat: 'Saturday',
};
const months: Record<string, string> = {
Jan: 'January',
Feb: 'February',
Mar: 'March',
Apr: 'April',
May: 'May',
Jun: 'June',
Jul: 'July',
Aug: 'August',
Sep: 'September',
Oct: 'October',
Nov: 'November',
Dec: 'December',
};
const fullDay = days[dayAbbr] || dayAbbr;
const fullMonth = months[monthAbbr] || monthAbbr;
let hours = parseInt(hoursStr, 10);
const ampm = hours >= 12 ? 'PM' : 'AM';
hours = hours % 12;
hours = hours ? hours : 12;
const tzSign = timezone.substring(0, 1);
const tzHours = timezone.substring(1, 3);
const tzMins = timezone.substring(3, 5);
const formattedTz = `UTC${tzSign}${tzHours}:${tzMins}`;
return `${fullDay}, ${fullMonth} ${dom}, ${year} at ${hours}:${minsStr} ${ampm} (${formattedTz})`;
}
} catch (e) {
// Fallback to raw string if parsing fails
}
return raw;
}