How much is the difference between html parsing an

问题:

I need to grab some data from websites in my django website. Now i am confused whether i should use python parsing libraries or web crawling libraries. Does search engine libraries also fall in same category

I want to know how much is the difference between the two and if i want to use those functions inside my website which should i use

回答1:

If you can get away with background web crawling use scrapy. If need to immediately grab something use html5lib (more robust) or lxml (faster). If you are going to be doing the later, use the awesome requests library. I would avoid using BeautifulSoup, mechanize, urllib2, httplib.

回答2:

HTML parse will parse the page and you can collect the links present in it. These links you can add to queue and visit these pages. Combine these steps in a loop and you made a basic crawler.

Crawling libraries are the ready to use solutions which do the crawling. They provide more features like detection of recursive links, cycles etc. A lot of features you would want to code would have already been done within these libraries.

However first option is preferred if you have some special requirements which libraries do not satisfy.

回答3:

I've done similar things previously. Web-crawlers were not usefull to me if I wanted the parsing to be done immediately in order to fetch something and be presented to the user. For batch-job stuff they're more appropriate. I found BeautifulSoup, lxml and mechanize to be quite usefull.