I started this to specifically dump my saved stories to an XML feed, which I hook up to http://ifttt.com and auto-import into Pinboard. Working great, so far.

I know there are a bunch of HN CLIs out there, but none handle saved stories (the only thing this supports, right now).

Mine only does saved stories. Though it's in Perl, not Python. It was whipped up in response to a complaint several months ago. It just never had a Show HN.

Http://github.com/bsandrow/hn-profile/

Shame you didn't do a Show HN because it looks interesting/promising so thanks for pointing me to it.

In return I can give you a clickable link :) http://github.com/bsandrow/hn-profile/

Item #N I'm learning on HN today: .netrc! I read that I need a "machine remote_hostname token pair". What should I use for remote_hostname in this case? Thanks!

Self-answer, for anyone interested. It's in the source code [1]:

  die "Error: No matches for news.ycombinator.com in ~/.netrc."

[1] https://github.com/bsandrow/hn-profile/blob/master/hn-profil...

Last week I had also written the same thing in Perl to login and handle saved stories, and then add them to a folder in Instapaper.

  in response to a complaint

What kind of complaint?

Someone was complaining in the comments that there was no easy way to get the full list of saved stories back out of HN. I whipped up this script to do that in response. It had been an idea I was batting around in the past, but the comment provided the motivation to create it.

edit: Looked up the thread, and turns out it was an inquiry, and not a complaint.

http://news.ycombinator.com/item?id=2479416

Thank you very much, I've been wanting to do that for a long time.

The next piece of this I'll be building out is a method for retrieving replies to your comments (I want an email when that happens).

You can also use http://hnnotify.com for this. Works great.

Ah, very cool - thanks!

  Retrieve all saved items [...]

This, by itself, is a wish come true for this programmer wannabe. Thank you wholeheartedly.

See also: https://github.com/scottjacksonx/hnsh

Saw that but it hasn't been updated in over a year. Also didn't look like there was a quick way to export saved stories as XML.

If anyone has any ideas as to how best to handle cookie caching, I'm all ears.

You might want to take a look into using something like twill (instead of requests) and BeautifulSoup instead of pyquery -- twill in particular will allow more control over cookies, etc.,

Thought about that, but using Requests + PyQuery was what made this project a joy to work with in the first place :)

I'm planning on just writing cookie data to a file and using that if available.

I prefer lxml to BeautifulSoup.

I can’t find the link now, but I remember something about BeautifulSoup being deprecated—it didn’t support HTML5 last time I checked. LXML is great. http://code.google.com/p/html5lib/ is also a nice parser for HTML5 documents.

Edit: pyquery wraps the aforementioned LXML. Seems like a good fit for jquery style selection…

I tried PyQuery ~1 year ago, and immediately found issues with it (IIRC it was having trouble selecting an element that had two classes, when the selector was only specifying one of those classes). I may have to revisit that if people are recommending it with positive reviews.

Give it a shot again. I've used it with several projects over the last 6 months and have had no issues. One of the projects also involved some fairly heinous malformed HTML, and PyQuery performed well.