Closed georgedorn closed 6 years ago
+1
Thanks, I've reproduced this and will fix the pagination issue. Probably I won't include screen names, for two reasons:
I was looking at the code but couldn't find the CSV generation at first glance. I'm building something that would benefit enormously of the lists as a training data mass, so, if you point me the right direction, I can prepare a PR.
Since I'll be working either on Ubuntu or a Mac with MacPorts, I can also update the readme (but that should be a separate PR, of course).
Also +1 here. My block list got too large before I realized that it was unshareable, and I can't figure out how to deal with it except to throw the whole thing out and start over. If I could download, I could arbitrarily pick the oldest 50,000 to unblock, and restart from there.
There is a workaround: If you get CSV from URLs like https://blocktogether.org/show-blocks/TJl_eDUojyl7wgjZOaRDTBqQTgamC4_Z3PJo1ONW.csv?page=n
where n is the page you are looking at, it'll retrieve a unique set of IDs. This way, the whole can be exported.
for i in $(seq 15)
do
curl https://blocktogether.org/show-blocks/TJl_eDUojyl7wgjZOaRDTBqQTgamC4_Z3PJo1ONW.csv?page=$i >> naziscore.csv
done
does the trick for this list, at least. It has 15 pages and the resulting file has 7271 lines (now)
Thank you, I can confirm that it does download a list of numbers and that they are all different, so this is probably working.
Thanks! This is now fixed. For block lists up to 250,000 you can download the full list.
I went here: https://blocktogether.org/show-blocks/TJl_eDUojyl7wgjZOaRDTBqQTgamC4_Z3PJo1ONW
I clicked 'Download CSV'.
I got a file that looks like this:
I expected there to be >7000 items, but there's only the 500 from the pagination. Also, user IDs alone aren't that useful; I'd prefer to have the same contents as the table (or at least current usernames), but with every user in the blocklist.