New Case Study:See how Anthropic automated 95% of dependency reviews with Socket.Learn More
Socket
Sign inDemoInstall
Socket

website-scraper

Package Overview
Dependencies
Maintainers
1
Versions
60
Alerts
File Explorer

Advanced tools

Socket logo

Install Socket

Detect and block malicious and high-risk dependencies

Install

website-scraper - npm Package Compare versions

Comparing version 1.2.0 to 1.2.1

10

lib/config/resource-extensions-by-type.js

@@ -5,5 +5,11 @@ var types = require('./resource-types');

defaultExtensions[types.html] = '.html';
defaultExtensions[types.css] = '.css';
defaultExtensions[types.html] = {
defaultExtension: '.html',
possibleExtensions: [ '.html', '.htm' ]
};
defaultExtensions[types.css] = {
defaultExtension: '.css',
possibleExtensions: [ '.css' ]
};
module.exports = defaultExtensions;

11

lib/filename-generators/by-site-structure.js
var _ = require('lodash');
var path = require('path');
var utils = require('../utils');
var resourceTypes = require('../config/resource-types');
var resourceTypeExtensions = require('../config/resource-extensions-by-type');

@@ -11,4 +13,9 @@ module.exports = function generateFilename (resource, options) {

// If we have HTML from 'http://example.com/path' => set 'path/index.html' as filepath
if (resource.isHtml() && !extension) {
filePath = path.join(filePath, options.defaultFilename);
if (resource.isHtml()) {
var htmlExtensions = resourceTypeExtensions[resourceTypes.html].possibleExtensions;
var resourceHasHtmlExtension = _.includes(htmlExtensions, extension);
// add index.html only if filepath has ext != html '/path/test.com' => '/path/test.com/index.html'
if (!resourceHasHtmlExtension) {
filePath = path.join(filePath, options.defaultFilename);
}
}

@@ -15,0 +22,0 @@

var _ = require('lodash');
var path = require('path');
var utils = require('../utils.js');
var defaultExtensions = require('../config/resource-extensions-by-type');
var typeExtensions = require('../config/resource-extensions-by-type');

@@ -35,4 +35,4 @@ module.exports = function generateFilename (resource, options, occupiedFileNames) {

if (!extension && defaultExtensions[resourceType]) {
extension = defaultExtensions[resourceType];
if (!extension && typeExtensions[resourceType]) {
extension = typeExtensions[resourceType].defaultExtension;
filename += extension;

@@ -39,0 +39,0 @@ }

@@ -52,3 +52,4 @@ var url = require('url');

function getFilepathFromUrl (u) {
return url.parse(u).pathname.substring(1);
var normalizedUrl = normalizeUrl(u);
return url.parse(normalizedUrl).pathname.substring(1);
}

@@ -60,4 +61,10 @@

function getFilenameExtension (filename) {
return (typeof filename === 'string') ? path.extname(filename) : null;
/**
* Returns extension for given filepath
* Example: some/path/file.js => .js
* @param {string} filepath
* @returns {string|null} - extension
*/
function getFilenameExtension (filepath) {
return (typeof filepath === 'string') ? path.extname(filepath) : null;
}

@@ -64,0 +71,0 @@

{
"name": "website-scraper",
"version": "1.2.0",
"version": "1.2.1",
"description": "Download website to a local directory (including all css, images, js, etc.)",

@@ -5,0 +5,0 @@ "readmeFilename": "README.md",

SocketSocket SOC 2 Logo

Product

  • Package Alerts
  • Integrations
  • Docs
  • Pricing
  • FAQ
  • Roadmap
  • Changelog

Packages

npm

Stay in touch

Get open source security insights delivered straight into your inbox.


  • Terms
  • Privacy
  • Security

Made with ⚡️ by Socket Inc