Yellow-Pages-Scraper-withEmail
Try for free
No credit card required
Go to Store
This Actor may be unreliable while under maintenance. Would you like to try a similar Actor instead?
See alternative ActorsYellow-Pages-Scraper-withEmail
krish_patel/yellow-pages-scraper-withemail
Try for free
No credit card required
Allows you to gather leads from yellow pages along with their corresponding email if any.
.editorconfig
1root = true
2
3[*]
4indent_style = space
5indent_size = 4
6charset = utf-8
7trim_trailing_whitespace = true
8insert_final_newline = true
9end_of_line = lf
.eslintrc
1{
2 "extends": "@apify"
3}
.gitignore
1# This file tells Git which files shouldn't be added to source control
2
3.idea
4node_modules
Dockerfile
1# First, specify the base Docker image. You can read more about
2# the available images at https://sdk.apify.com/docs/guides/docker-images
3# You can also use any other image from Docker Hub.
4FROM apify/actor-node:16
5
6# Second, copy just package.json and package-lock.json since it should be
7# the only file that affects "npm install" in the next step, to speed up the build
8COPY package*.json ./
9
10# Install NPM packages, skip optional and development dependencies to
11# keep the image small. Avoid logging too much and print the dependency
12# tree for debugging
13RUN npm --quiet set progress=false \
14 && npm install --only=prod --no-optional \
15 && echo "Installed NPM packages:" \
16 && (npm list --only=prod --no-optional --all || true) \
17 && echo "Node.js version:" \
18 && node --version \
19 && echo "NPM version:" \
20 && npm --version
21
22# Next, copy the remaining files and directories with the source code.
23# Since we do this after NPM install, quick build will be really fast
24# for most source file changes.
25COPY . ./
26
27# Optionally, specify how to launch the source code of your actor.
28# By default, Apify's base Docker images define the CMD instruction
29# that runs the Node.js source code using the command specified
30# in the "scripts.start" section of the package.json file.
31# In short, the instruction looks something like this:
32#
33# CMD npm start
INPUT_SCHEMA.json
1{
2 "title": "Yellowpages scraper input schema",
3 "type": "object",
4 "schemaVersion": 1,
5 "properties": {
6 "search": {
7 "title": "Search",
8 "type": "string",
9 "editor": "textfield",
10 "prefill": "Dentist",
11 "description": "Query to search on Yellow Pages"
12 },
13 "location": {
14 "title": "Location",
15 "type": "string",
16 "editor": "textfield",
17 "prefill": "Los Angeles",
18 "description": "Location to search entries in"
19 },
20 "startUrls": {
21 "title": "Start URLs",
22 "type": "array",
23 "editor": "requestListSources",
24 "description": "List of URLs that will be crawled."
25 },
26 "maxItems": {
27 "title": "Max items",
28 "type": "integer",
29 "minimum": 0,
30 "prefill": 200,
31 "description": "Maximum number of outputted results"
32 },
33 "extendOutputFunction": {
34 "title": "Extend output function",
35 "type": "string",
36 "editor": "javascript",
37 "description": "A function which result will get merged with the default result",
38 "prefill": "($, record) => {\n return {};\n}",
39 "sectionCaption": "Advanced options"
40 },
41 "proxyConfiguration": {
42 "title": "Proxy configuration",
43 "type": "object",
44 "editor": "proxy",
45 "description": "In this case the best option is usually no proxy. Otherwise US or UK proxies work best.",
46 "prefill": { "useApifyProxy": false }
47 }
48 }
49}
apify.json
1{
2 "env": { "npm_config_loglevel": "silent" }
3}
main.js
1// This is the main Node.js source code file of your actor.
2
3// Import Apify SDK. For more information, see https://sdk.apify.com/
4const Apify = require('apify');
5
6Apify.main(async () => {
7 const input= await Apify.getInput();
8 const run=await Apify.call('petr_cermak/yellow-pages-scraper', input);
9 const axios=require('axios');
10 const cheerio=require('cheerio');
11 var dataset=await Apify.openDataset(run.defaultDatasetId);
12 var ar=await dataset.getData();
13 var data=ar.items;
14 for(var i=0;i<data.length;i++){
15 var obj=data[i];
16 var res=await axios.get(obj.url);
17 var $=cheerio.load(res.data);
18 var emails=$(".email-business").attr("href");
19 obj.email="";
20 if(emails==undefined){
21 obj.email="none found";
22 }else{
23 var array=emails.split(":");
24 obj.email=array[1];
25 }
26 await Apify.pushData(obj);
27 }
28});
package.json
1{
2 "name": "project-empty",
3 "version": "0.0.1",
4 "description": "This is a boilerplate of an Apify actor.",
5 "dependencies": {
6 "apify": "^2.3.2"
7 },
8 "devDependencies": {
9 "@apify/eslint-config": "^0.1.3",
10 "eslint": "^7.0.0"
11 },
12 "scripts": {
13 "start": "node main.js",
14 "lint": "./node_modules/.bin/eslint ./src --ext .js,.jsx",
15 "lint:fix": "./node_modules/.bin/eslint ./src --ext .js,.jsx --fix",
16 "test": "echo \"Error: oops, the actor has no tests yet, sad!\" && exit 1"
17 },
18 "author": "It's not you it's me",
19 "license": "ISC"
20}
Developer
Maintained by Community
Actor Metrics
20 monthly users
-
2 stars
52% runs succeeded
Created in Aug 2022
Modified 2 years ago
Categories