Alrighty, I've added GLAMorous as a new source to Massviews. To use just enter the Commons category name. So for your category:
http://tools.wmflabs.org/massviews-test/?platform=all-access&agent=user&source=glamorous&target=Wikimedia_Israel_-_Channel_2_videos&range=latest-20&sort=views&direction=1&view=list&debug=true

Hopefully this is what you were looking for, I know some other GLAM folks requested this (T150507).

This is only available on the test version of Massviews because I would still very much consider this a quick demo. It seems to work fine for your example but it is by no means production-ready, and you may encounter random errors, especially with larger datasets. Some translations are also missing.

The source is called "GLAMorous" but that's actually a misnomer. I ended up implementing everything myself, so it does not use the GLAMorous tool at all. It also lacks features like selectively choosing projects, and it assumes you only want mainspace pages. I imagine this accounts for most use cases, though.

Let me know if you run into any problems or have any feedback. I'm sure you'd like to see pageviews totals grouped by the source file, which I will try to work on soon, among other features, before officially releasing this.


Best,

~MA

On Sun, Feb 26, 2017 at 12:32 PM, Federico Leva (Nemo) <nemowiki@gmail.com> wrote:
Itzik - Wikimedia Israel, 26/02/2017 17:56:
ammm.. maybe a easier way for someone who don't want to play with code
and download dumps? :)

Does a standard command like grep qualify as easier? :-) On a computer with some bandwidth I did something like:

wget -r -np -nH -nd -A bz2 http://ftp.acc.umu.se/mirror/wikimedia.org/other/mediacounts/daily/2016/ ; find -name "mediacounts*bz2" -print0 | xargs -0 -P8 -I§ -n1 bzgrep webm § | grep -E '/Channel_?2.+webm' > 2016-12-channel2.csv

Which gives me about 650k accesses during December 2016, of which 10500 downloads as complete file and 8k streamed plays.

Nemo

_______________________________________________
Analytics mailing list
Analytics@lists.wikimedia.org
https://lists.wikimedia.org/mailman/listinfo/analytics