#wikiteam 2019-04-25,Thu

↑back Search

Time Nickname Message
00:39 🔗 Zerote has quit IRC (Read error: Operation timed out)
04:57 🔗 kiska1 has quit IRC (Read error: Operation timed out)
05:13 🔗 kiska1 has joined #wikiteam
06:16 🔗 Zerote has joined #wikiteam
07:16 🔗 Zerote has quit IRC (Read error: Operation timed out)
07:20 🔗 Zerote has joined #wikiteam
14:20 🔗 Zerote has quit IRC (Ping timeout: 600 seconds)
15:22 🔗 Zerote has joined #wikiteam
20:04 🔗 phuzion has quit IRC (Remote host closed the connection)
20:47 🔗 phuzion has joined #wikiteam
21:37 🔗 Zerote I've encountered a weird issue with the dumpgenerator, when multiple images exist with the same name but different casing
21:38 🔗 Zerote In this case, the wiki has the two files "Alliance Handshake.jpg" and "Alliance handshake.jpg", it will save the first one, then not save the second, because it has the same name
21:38 🔗 Zerote However, if the scraper is paused or disconnects, and has to read the list of files, it will only find one of them
21:40 🔗 Zerote This means it stops once it can't find "Alliance handshake.jpg" and assumes only ~1500 images were downloaded, when it was more than 10k
22:20 🔗 Zerote Wait, I'm dumb, this might just be a problem with Windows
22:28 🔗 JAA That would make sense. NTFS is case-preserving, i.e. filenames with differing case are collisions but the case you use when creating a file is preserved.
22:57 🔗 Zerote Yeah, just got a couple Linux VMs running, I'll be redoing scrapes I did on Windows just to make sure stuff isn't missing because of that

irclogger-viewer