TipJar » ESP Visions

hypem.com

dz0ny — Sun, 23 May 2010 09:54:04 +0000

Cycle between songs

!set vision=a.play-ctrl.play

io9.com

Thomas Bondo Homburg — Sat, 08 May 2010 23:47:11 +0000

io9.com : a.dcolor

Sydney Morning Herald

martien.verbruggen — Sat, 08 May 2010 22:53:43 +0000

Selects the headlines on the front page of http://www.smh.com.au/, avoiding most of the stuff that I don’t want to read.

Unfortunately, this does not include the main articles on the category pages, because they use the same jQuery selector (div.wof>h3>a) as many of the things I want to avoid on the main page.

This works reasonably well for me:

div.cN-storyHeadlineLead:not(.storyRHS)>h3>a, ul.cT-wireNews>li>a, ul.featuredStoriesList>li>a, div.wof>p>a, h3.cN-storyHeadlineOnly>a

TypePad Blogs

Evan Sultanik — Mon, 26 Apr 2010 18:50:29 +0000

Cycles through posts on any TypePad blog:

typepad.com : h3.entry-header>a

The following version additionally highlights the links to the previous/next pages:

typepad.com : h3.entry-header>a,span.pager-left>a,span.pager-right>a

digg.com without ads

rpx2424f6f5a0c9d62d00b8c6c41b0921c6 — Mon, 26 Apr 2010 05:57:20 +0000

More effective procrastination

digg.com : div.news-summary:not(.sponsored) h3>a, div.news-full a.offsite

Github Dashboard and Commits

Ankit — Mon, 12 Apr 2010 18:06:40 +0000

Super tweaked selector for Github.

Selects issue, wiki or commit links on Dashboard, issues on issue list and commits in projects’ “Commit” view

github.com : .message>a:not(a:contains(Compare)),a:contains(issue),code>a,a:contains(News Feed):not(.selected):not(.feed),a:contains(Your Actions):not(.selected),div.message>pre>a,a.issue_title,span:contains(started) + a,span:contains(created) + a

torrentz search results

Dan Jones — Mon, 12 Apr 2010 17:17:18 +0000

torrentz.com/search : .results dl>dt>a

Go through torrentz.com search results.

Wikipedia Search Results

Dan Jones — Mon, 05 Apr 2010 18:44:43 +0000

wikipedia.org/w/index.php?title=Special : .mw-search-results a

This will cycle through search results on any English-language wikipedia search site. (Not just en.wikipedia.org, but also, for example, simple.wikipedia.org)

Change the part after title= to work on other languages, e.g. for german make it wikipedia.org/w/index.php?title=Spezial

You could also get rid of the part after title= to make it work for all languages, but that could have unexpected results on non-search pages. For example, if you do it on an edit page, you’ll get “Couldn’t find any elements on the page.”

Blogger blogs

toban — Wed, 24 Mar 2010 19:12:22 +0000

Cycles through posts on any Blogger blog.
blogspot.com : h3.post-title.entry-title>a

Facebook

Pyrohacker — Wed, 17 Mar 2010 22:43:00 +0000

Okay, this one’s for browsing Facebook. There are still a few problems, but it’s working well for the most part. The main roadblock was that the News Feed page works differently than viewing a single friend’s Wall.

News Feed:
Stories are selected with *.GenericStory
Story times (which are the only links to the actual story itself) are selected with *.GenericStory_Time

Friend’s Wall:
Stories are selected with *.UIStory (which can also work on the News Feed, strangely)
Story times are selected with *.UIIntentionalStory_Time

Both:
You can throw in the ‘Older Posts’ link at the bottom with *a:contains("Older Posts"), allowing you to move to – duh – older stories.

In Conclusion:

If you don’t care about the story times, then a good choice is an ESP Vision for just the stories:

div.GenericStory, div.UIStory, a:contains("Older Posts")

If you do want to select the story times, I would recommend 2 different scrapers:
1. ?story : *.GenericStory, .UIIntentionalStory, a:contains("Older")
2. ?time : *.GenericStory_Time, .UIIntentionalStory_Time, a:contains("Older")

Wow, that’s a lot of jQuery! Anyway, I hope somebody can make good use of these. Good luck!