3 ms·

I wrote a greasemonkey userscript for Firefox to reduce the number of non-technical articles. It works alright, I never meant it to be perfect, just get rid of

I wrote a greasemonkey userscript for Firefox to reduce the number of non-technical articles. It works alright, I never meant it to be perfect, just get rid of the stuff that never is fun for me to read or is about topics that come up too often. https://github.com/ErikHarmon/hncleanse/blob/master/hncleanse.user.js https://github.com/ErikHarmon/hncleanse/blob/master/hncleans...

PopeOfNope 11y ago
I did something very similar with dotjs before I realized it was introducing a 1-5 second delay in the loading of every webpage. How's the performance of this script?
- erik66132 11y ago
  It's imperceptible to me. The number of search terms is pretty low. My regex isn't as sophisticated as it could be because the script started as a simple terms list using indexof, so it could be improved a little bit as more terms are added. Mainly I filter by entire sites, which helps a lot. I won't miss many technical articles by filtering nytimes, and if I do, I can live with that. I'll take PRs or suggestions, this was a ten minute project to scratch an itch.
mcnamaratw 11y ago
Thanks for the link! I'll take a look!
cyphunk 11y ago
Your honesty made me giggle (and I do not mean that in a derogatory manner)... 'npr.org', // depressing 'nytimes.com', // will probably annoy me This patch also makes for an amusing HN experience and might answers the inverse "Ask HN: Is there a way to see only non-technical posts?" --- a/hncleanse.user.js +++ b/hncleanse.user.js @@ -148,7 +148,7 @@ while ((elem = result.snapshotItem(i)) != null) { if (shouldRemove || crap != null) { var reason = shouldRemove?('site '+sitebit):('terms '+crap); console.log(reason+': '+title); - + } else { var subtext = athing.nextSibling; subtext.parentNode.removeChild(subtext);

PopeOfNope 11y ago

I did something very similar with dotjs before I realized it was introducing a 1-5 second delay in the loading of every webpage. How's the performance of this script?

erik66132 11y ago
It's imperceptible to me. The number of search terms is pretty low. My regex isn't as sophisticated as it could be because the script started as a simple terms list using indexof, so it could be improved a little bit as more terms are added. Mainly I filter by entire sites, which helps a lot. I won't miss many technical articles by filtering nytimes, and if I do, I can live with that. I'll take PRs or suggestions, this was a ten minute project to scratch an itch.

mcnamaratw 11y ago

Thanks for the link! I'll take a look!

cyphunk 11y ago

Your honesty made me giggle (and I do not mean that in a derogatory manner)... 'npr.org', // depressing 'nytimes.com', // will probably annoy me This patch also makes for an amusing HN experience and might answers the inverse "Ask HN: Is there a way to see only non-technical posts?" --- a/hncleanse.user.js +++ b/hncleanse.user.js @@ -148,7 +148,7 @@ while ((elem = result.snapshotItem(i)) != null) { if (shouldRemove || crap != null) { var reason = shouldRemove?('site '+sitebit):('terms '+crap); console.log(reason+': '+title); - + } else { var subtext = athing.nextSibling; subtext.parentNode.removeChild(subtext);