A scrapy
-based Hacker News crawler.
Introduction
HNCrawl is a tiny, simple scrapy
-based crawler which grabs the html
content of pages linked to the front page of hacker news.
$ pip install scrapy
$ git clone [email protected]:mvanveen/hncrawl.git
Note: Please be sure to keep in mind that the Crawl-Delay
value set in the HN robots.txt
file is set to 30 seconds. Please be sure to avoid using the scraper more than once per 30 seconds!
Scrape the links from the front page of HN
$ scrapy crawl hnspider
Scrape items and return json summary of items scraped into items.json
$ scrapy crawl alias_scrape -o items.json -t json
Here is an example file hierarchy. Folders are a hex digest of the SHA1 hash of the hacker news item url.
├── out
│ ├── 000f86c7547b47a700dee0879a0fe08b4597360f
│ │ └── index.html
│ ├── 0190cbad182ab3bc9a92482d169f38e363ca3c57
│ │ └── index.html
│ ├── 02bae9642c8dd4b75a593c1c42beff62824ee8fc
│ │ └── index.html
│ ├── 05c1460571f0ac45f77bf2ecbd3cba8b85c20621
│ │ └── index.html
│ ├── 0b1587a3dbe9996d10a0fd3250f75462ebd59a0b
│ │ └── index.html
│ ├── 0c5c67585004e03341e6a87d2db5257b93337b86
│ │ └──
The JSON summary of news items look like this:
{'title': u'EFF Wins Protection for Time Zone Database',
'url': u'https://www.eff.org/press/releases/eff-wins-protection-time-zone-database'}
HNCrawl is MIT licensed.