2013-02-24, 20:58
Hi. Since version 3 of this addon there are two annoying problems with the provider Subdivx (one of the best Spanish subtitles sites):
1) The first is related with the way they represent the number of downloads. They use a colon to separate thousands, which makes the line
downloads = int (matches.group(2)) / 1000 fail, as int will receive a number like 1,200. This wasn't a problem until the number of downloads exceeded 999, which happens now in every subtitle. The solution (found elsewhere in the net) is to change by:
downloads = int(matches.group(2).replace(",","")) / 1000
This works and I implemented it in every update, but it would be desirable to have it corrected in the addon, since it is rather cumbersome to fix it every time.
2) The second problem is far worse. There are some subtitles that contain (in addition to the corresponding subtitle file) a text file with a name using accented characters (i.e. á, é, í, ó, ú) or the Spanish character ñ. This makes the function which iterates through the files fail, so the .rar or .zip file is decompressed but the subtitle is never processed. The relevant part of the debug log is:
log removed
The file that triggered this problem is named "Importante para la utilización de este archivo.txt" (of course it is a bad idea to use such a file name, but there is nothing that I can do about it). There are also some subtitles which contains such a character(s) in the filename. Even more annoying is that once one of these files is in the sub_tmp folder the it makes impossible to use it until the file is removed from the operating system (which, BTW, is an Ubuntu 12.04). I didn't find a solution for this problem yet (I didn't spend a lot of time either...).
1) The first is related with the way they represent the number of downloads. They use a colon to separate thousands, which makes the line
downloads = int (matches.group(2)) / 1000 fail, as int will receive a number like 1,200. This wasn't a problem until the number of downloads exceeded 999, which happens now in every subtitle. The solution (found elsewhere in the net) is to change by:
downloads = int(matches.group(2).replace(",","")) / 1000
This works and I implemented it in every update, but it would be desirable to have it corrected in the addon, since it is rather cumbersome to fix it every time.
2) The second problem is far worse. There are some subtitles that contain (in addition to the corresponding subtitle file) a text file with a name using accented characters (i.e. á, é, í, ó, ú) or the Spanish character ñ. This makes the function which iterates through the files fail, so the .rar or .zip file is decompressed but the subtitle is never processed. The relevant part of the debug log is:
log removed
The file that triggered this problem is named "Importante para la utilización de este archivo.txt" (of course it is a bad idea to use such a file name, but there is nothing that I can do about it). There are also some subtitles which contains such a character(s) in the filename. Even more annoying is that once one of these files is in the sub_tmp folder the it makes impossible to use it until the file is removed from the operating system (which, BTW, is an Ubuntu 12.04). I didn't find a solution for this problem yet (I didn't spend a lot of time either...).
