Commenting Code Properly

eccp wrote: post #5

Don't use regexes to parse HTML, use BeautifulSoup instead. Also, don't use "brogrammer" style comments, they are not funny enough.

Something like: https://gist.github.com/3850232

Re: Commenting Code Properly

#7

Post by sailfrog » Wed, Dec 12, 2012, 11:25 PM UTC

sneak wrote: post #2

The comments are stupid, the headline is wrong, and you're parsing HTML with regular expressions. It would be hard to craft a submission worse than this.

The submission was meant to be humorous, sorry to disappoint. No matter how you slice it parsing HTML from a 3rd party site is a major hack, though this does however work nicely (itch scratched). Lastly I have some other ideas for submissions so don't judge this as the worst just yet.

Re: Commenting Code Properly

#8

Post by talleyrand » Thu, Dec 13, 2012, 2:35 AM UTC

Love it! Code comments need more adjectives and Culture Club lyrics.

Re: Commenting Code Properly

#9

Post by sneak » Thu, Dec 13, 2012, 10:02 AM UTC

sailfrog wrote: post #7

sneak wrote: post #2

The comments are stupid, the headline is wrong, and you're parsing HTML with regular expressions. It would be hard to craft a submission worse than this.

The submission was meant to be humorous, sorry to disappoint. No matter how you slice it parsing HTML from a 3rd party site is a major hack, though this does however work nicely (itch scratched). Lastly I have some other ideas for submissions so don't judge this as the worst just yet.

> No matter how you slice it parsing HTML from a 3rd party site is a major hack

WebKit seems to do just fine. I think you're making excuses.

Re: Commenting Code Properly

#10

Post by sailfrog » Thu, Dec 13, 2012, 3:02 PM UTC

sneak wrote: post #9

sailfrog wrote: post #7

Earlier quoted context omitted.

The submission was meant to be humorous, sorry to disappoint. No matter how you slice it parsing HTML from a 3rd party site is a major hack, though this does however work nicely (itch scratched). Lastly I have some other ideas for submissions so don't judge this as the worst just yet.

> No matter how you slice it parsing HTML from a 3rd party site is a major hack WebKit seems to do just fine. I think you're making excuses.

Until the source site changes the URL, or the URL arguments, or the page structure, or the doctype, or the CSS selectors, or the element ids, or whatever it is you key on to ferret out the content you care about. Scraping data embedded in markup not bound to an API spec is fragile regardless of how "properly" you parse it because there is no guarantee of structural consistency.