[10:39:53] *** Joins: majora (~majora@gateway/vpn/privateinternetaccess/majora) [10:40:06] I removed motherfucking android studio and the sdk and got back 30gb of space on my laptop [11:17:01] *** Quits: majora (~majora@gateway/vpn/privateinternetaccess/majora) (Quit: Leaving) [13:42:53] most of it is the emulator images [13:43:25] usually if only have one SDK version installed and one image from that SDK, it's not too bad [14:19:05] sivoais: if you have some time could you take a look at this code: https://github.com/zneak/x86doc/blob/master/x86manual.py [14:19:05] Title: x86doc/x86manual.py at master · zneak/x86doc · GitHub [14:19:28] it's somehow able to do what i haven't been able to figure out, which is group elements in this manual in a sane way [14:19:50] the default ordering given by pdfminer was all over the place, so i couldn't group sections together [14:20:17] so i tried sorting by y and then x, but millimeter differences in layout fuck things out, not to mention lines that wrap in a single line of a table [14:20:25] but that guys code seems to be doing everything right [14:20:29] i just don't get how he's doing it [14:21:00] was wondering if it was related to one of the PDF algorithms you mentioned before, although i suspect most of that is taken care of by pdfminer [15:36:46] i mean, he did write 800 lines of python to just scrape some text from a pdf [15:57:18] lol, a good chunk of my / is taken up by xilinx vivado [16:26:57] clinth: he did a really good job though [16:27:06] trying to figure out exactly how he's doing it [17:13:04] from a tiny initial look, it seems a lot of workarounds for specific tables and stuff [17:13:17] :( living out of a hotel room is totally not as fun as it sounds [21:53:33] lol no it is not too fun [22:03:18] and leaving it is kinda torture today too [22:03:21] because nyc is hot as hell [22:17:00] philly too