Closed

crawl internet for pdf files

This project received 34 bids from talented freelancers with an average bid price of €979 EUR.

Get free quotes for a project like this
Employer working
Project Budget
€750 - €1500 EUR
Total Bids
34
Project Description

Specifications

Short description:
- Need a program that finds all hyperlinks for a certain URL (Domain) and checks whether one of the hyperlinks links to a pdf file. Scan all pages and subpages.

How is should work:
1. User enters a baseurl.
2. Baseurl is saved into a table
3. Webcrawler takes the baseurls given and start crawling the url to find all hyperlinks (main page and all subpages).
4. All found urls are saved to a hyperlink table where besides the base url all founds hyperlinks are stored and indentified.(relations between urls by parentid):
4a: When link is pdf file: mark in a column that the link is a pdf file (skip external domain check if the file is directly linked)
4b: When link is a link to another domain(for example from [url removed, login to view] to [url removed, login to view]: mark in a column that the links goes external. Skip crawling this url any deeper an continue to next url.

Extra specs:
1. User can enter how deep to crawl a website (parameter per baseurl) 0= only baseurl webpage, 1-998levels deep, 999 unlimitied
2. User can enter how fast too crawl a website (parameter per baseurl) 0= as fast a possible, 1-8000ms per check
3. User can define how to work through the baseurl queue: First in, First out or priority: High, Middle, Low.
4. User can define how long to crawl a baseurl. For example don't crawl the same baseurl(read server ip of the baseurl) for 0 to 600000 minutes. Then wait 0 to 600000 minutes before crawling again on this baseurl.
5. Application should work as a MS windows service that can be started and stopped
6. Data is stored in a mysql or mssql database
7. Application should be able to be started/stopped through commandline
8. New baseurls and parameters changes should be able to be given when application is running.
9. Application should be able to detect if crawler is being blocked. For example 25 times the same page response. Application should immediatly stop crawling and disable crawling on the baseurl/server ip until user want to start the crawling again.
10. Following should be logged in a logfile: start, stop, new cmd given, new baseurls given, parameter changes, crawler blocked, any basic errors, when pages as fully been crawled.
11. For future multithreading extension: a parameter where for each basurl it can be given which service on which server crawls the baseurl.
12. Prevent looping when crawling: For example by checking when a certain hyperlink already has been crawled.
13. Parameter to enable not follow bot
14. Parameters for db connection
15. Parameters for service/bot indentification
16. Parameters for logfile location
17. Prefer .NET solution but Python is also okay.
18. Code should be open and readable.

Looking to make some money?

  • Set your budget and the timeframe
  • Outline your proposal
  • Get paid for your work

Hire Freelancers who also bid on this project

    • Forbes
    • The New York Times
    • Time
    • Wall Street Journal
    • Times Online