Reply to thread

Message

<blockquote data-quote="lwerndly" data-source="post: 340998" data-attributes="member: 64817">Ah time flies!Haven't been watching much anime lately, so haven't looked at this thread either.Getting back into it now and there are a few things that annoy me still, and have a few ideas on how to fix them.Tancor: no it is not possible to migrate unless you have exactly the same paths (absolute not relative)Although you could do something if you wanted to get your hands dirt in the SQL.The first thing I will be looking at is performance.The hashing does indeed take a long time. I am going to create an alternative matching system along the lines of file names.So if you download a file and keep it's original file name like this:[Chihiro]_Kyou_no_Go_no_Ni_09_[h264][C972C0C4].mkvIt will attempt to look up my alternative database (not AniDB), for an entry matching that name. If found, it will use that info instead of hashing the file.If it can't find it, you will need to revert to the usual hashing routine. Once this is done it will upload that data for future reference.Benefits1. If someone using this plugin has already hashed the file you won't need to (if file names are the same)2. You can move files around without need to worry that you will need to hash again.3. Information you upload will be associated with your user name and the search will first look for any information you upload. So you can rename files in your own format and get some benefit as well.After that I'll start looking at some of the bugs again.Oh, if anyone wants the source themselves let me know.</blockquote>

[QUOTE="lwerndly, post: 340998, member: 64817"] Ah time flies! Haven't been watching much anime lately, so haven't looked at this thread either. Getting back into it now and there are a few things that annoy me still, and have a few ideas on how to fix them. [B]Tancor[/B]: no it is not possible to migrate unless you have exactly the same paths (absolute not relative) Although you could do something if you wanted to get your hands dirt in the SQL. The first thing I will be looking at is performance. The hashing does indeed take a long time. I am going to create an alternative matching system along the lines of file names. So if you download a file and keep it's original file name like this: [B][SIZE="2"][Chihiro]_Kyou_no_Go_no_Ni_09_[h264][C972C0C4].mkv[/SIZE][/B] It will attempt to look up my alternative database (not AniDB), for an entry matching that name. If found, it will use that info instead of hashing the file. If it can't find it, you will need to revert to the usual hashing routine. Once this is done it will upload that data for future reference. Benefits 1. If someone using this plugin has already hashed the file you won't need to (if file names are the same) 2. You can move files around without need to worry that you will need to hash again. 3. Information you upload will be associated with your user name and the search will first look for any information you upload. So you can rename files in your own format and get some benefit as well. After that I'll start looking at some of the bugs again. Oh, if anyone wants the source themselves let me know. [/QUOTE]

Verification