[SpyDIRmaN]

All questions of laziness aside, I would like to see this feature to use with my android stand-alone media center solution (minix neo x7 with local storage) running the android port of xbmc.

At present all the file sorting, auto-renaming options are not there on android (so far as I can tell)

My hope is to configure ttorrent with RSS feed support to download and dump all files on an attached USB drive. I will then require a way to ether auto-organize/rename those files into an Xbmc friendly format (which I don't believe presently exists on android), or (with this feature) xbmc could do the rest on its own (provided the shows have been downloaded with proper titles (mine are).

[morfraen]

I'd also like this feature. I don't see why this isn't the default behavior. Having none of your tv shows show up because they're not organized in some specific way is rather confusing to a new user.

It can't be that hard to extract the show name, season, episode for every file in a folder and then virtually organize them by show in the ui. At the very least they should get dumped in a generic or 'unsorted' category rather than the ui just ignoring them all.

[Ned Scott]

I'd also like this feature. I don't see why this isn't the default behavior. Having none of your tv shows show up because they're not organized in some specific way is rather confusing to a new user.

It can't be that hard to extract the show name, season, episode for every file in a folder and then virtually organize them by show in the ui. At the very least they should get dumped in a generic or 'unsorted' category rather than the ui just ignoring them all.

The UI doesn't ignore them. They're in Videos -> Files.

And yes, extracting the show name from the file name is not nearly standard enough for that to work so it is "that hard". Most new users are not confused by this and understand that XBMC cannot shit rainbows.

[morfraen]

The scraping scripts can already handle extracting the show and season from the filename, if they don't find that info, use the show->season that was captured while descending the folder structure, or the same logic flipped around. It's a trivial change for anyone who knows what they're doing.

You completely lack perspective if you think rearranging your entire media library is 'reasonable', no matter how many tools exist to help. The mere fact that those tools exist should make the existence of a fundamental problem blindingly obvious. The entire point of media center software is to take all of your media and present it in a convenient organized fashion.

But hey, way to be a dick about it. Typical behavior for an open source forum warrior.

[Ned Scott]

The scraping scripts can already handle extracting the show and season from the filename, if they don't find that info, use the show->season that was captured while descending the folder structure, or the same logic flipped around. It's a trivial change for anyone who knows what they're doing.

It's not a trivial change for XBMC's current scraper system. A new system is being worked on that will allow things like this, with much better logic handling, but it won't be available till at least XBMC V14.

You completely lack perspective if you think rearranging your entire media library is 'reasonable', no matter how many tools exist to help. The mere fact that those tools exist should make the existence of a fundamental problem blindingly obvious. The entire point of media center software is to take all of your media and present it in a convenient organized fashion.

I would argue that, given a lack of internal tag standards for videos, assuming a single folder for several shows with several different naming conventions would work out of the box, is more unreasonable. Even the scripts and tools out there typically will require human input when you come across a show with a similar name (remakes, unrelated shows with the same name, etc), or have a different naming convention (none of the anime groups use scene-naming, for example).

I think it's unreasonable to demand that a program work exactly as you use it without even attempting to listen to the reasons why it can't currently be done. It's not like it's a switch and we're just being assholes about turning the switch on. Like I said before, better scraping is coming to XBMC, but don't flip your lid because it's not done right away.

But hey, way to be a dick about it. Typical behavior for an open source forum warrior.

Fair point. My apologies.

[jjd-uk]

It's a trivial change for anyone who knows what they're doing.

If it's so trivial when can we expect the PR from you containing the code changes?

[texaco]

Less categorization leads to more guessing when identifying the episode, XBMC may use the folder name in the identifying process. It is as simple as that and leads to more misidentified files.

basicly you are asking XBMC to identify your episodes properly by giving it less information if all episodes are in one single folder.

You may write all the logic you want to compensate for this but in the end you end up with more complex code that leads to more complex results.

[ironic_monkey]

... and finally a sign of intelligence in this thread!

[kbailie]

Hi there,

I think I have a similar issue. I use XBMC in one room but my media is accessed by a WDTV LIVE device in another and an iPad Mini as well. The WDTV just basically shows a file list so if I had folders for each show, it would make it difficult to know what shows were watched -- lots of hopping in and out of folders. The iPad Mini uses a program which access an SMB share and plays video that way -- so same basic issue -- having each show in its own subfolder is a bit painful.

I can't think of anything I could do which would make it easy to watch shows on all three devices.

thanks
Keith

And before someone tells me to install XBMC on my wife's ipad

1) I've jailbroken many a device and prefer them that way...but when it comes to my wife's ipad, I don't want to. And to tell the average Joe that "it works -- you just have to jailbreak your device and void the warranty...", well, that generally is okay with me, but that's not a good answer for the general public.

2) If you've got an iOS from the last year or so (6.13 and up), there's not a jailbreak even available, so oh well, too bad for those folks.

3) What do I do about my Xbox? Or PS4? Or Roku? Why do our other streaming devices have to be less useful in order to get XBMC to scrape TV shows?

Sorry to tell you dude but all your problems can be fixed on your end.
I have a setup with sickbeard, xbmc and Plex and literally dont have to do anything else than add a tvshow once in order to download, sort and make it available in my livingroom and on my ipad with scraped info and coverart.

Thats lazy!

[germanso]

I would like to add one more point of view. I'm sorry but I must line up with the unsatisfied. I don't mean to be ungrateful or disrespectful. I understand this is your program and you've spent many many many hours on it and still do, but I really think there have been many good points on this thread and I would like to add mine.

I've had a HTPC running for almost 7 years with Mediaportal until one week ago when it stopped working (I still can't understand what happens. I've tried many new installs from scratch). Anyway, not here to complain about Mediaportal...

At the HTPC there is also uTorrent with many series configured to download via RSS feeds (and I guess this can be a common installation for many HTPCs). So, when a new show is available it downloads automatically and goes to a common folder. As long as the show is shared by uTorrent it cannot be renamed or moved. This is important. It's not just a matter of lazyness, but of practical impossibility.

I was delighted when I just started my computer, went to the TVSeries plugin and there were all the new episodes ready to be watched. My wife and kids also were able to watch their favorite shows just minutes after starting the computer.

Later, some shows are organized by me on the way XBMC expects (Series/season/xxx.S01.E01.xxx.ext) to keep them and rewatch eventually or to share with friends later. But most of the series are just wiped out after being viewed.

Now, to use XBMC I'm supposed to go everyday to uTorrent, "release" the episodes, sort them and move each of them to an specific folder with great care to follow rather strict naming guidelines...

The main point of the computers is precisely they can take care of boring and repetitive tasks for us. I don't understand the logic of the user having to follow strict and arbitrary rule instead of "teaching" the computer a new trick valuable for most of the users.

As far as I understand, for the scrapper the full path is used to locate a series. Let's say for example "series\community\season 1\community.S01E01.Pilot.avi".

Then you use a regular expression like "<regexp>[Ss]([0-9]+)[][ ._-]*[Ee]([0-9]+)([^\\/]*)$</regexp>" to extract the exact number of the episode and season (I guess there are some other regular expressions to obtain the series name and the season from the file path). If the file is not identified, it checks on the next regular expression until it matches and, if none does, the file is just ignored.

I just don't see the difficulty on adding one final regular expression extracting the name from the filename before the SxxExx.

I only recall 4 naming patterns. For example, in naming episodes from "The Big Bang Theory":
"The Big Bang Theory S01E01" With spaces
"The.Big.Bang.Theory.S01E01" With dots
"The_Big_Bang_Theory_S01E01" With underscores
"TheBigBangTheoryS01E01" With capital letters (very rarely)

I guess the last one can be more difficult to process, but the 3 other seem equivalent for me and I don't see the real problem to feed them to the scrapper. I don't see the difference in having this information on the folder name or on the file name.

If a file is not recognized even then, no problem, unrecognized it stays, but at least we have a very good chance to find most of them (I would say close to 95% straight out of torrent...)

After all we don't really expect XBMC to "shit rainbows" and to identify correctly every one of the files we may have, but this functionality would increase immensely the usefulness of your software and there are other programs (like boxee and Mediaportal) who can manage this since many years ago.

Again, I'm nothing but grateful from your excellent program, and please excuse my ignorance.

[scudlee]

At the HTPC there is also uTorrent with many series configured to download via RSS feeds (and I guess this can be a common installation for many HTPCs). So, when a new show is available it downloads automatically and goes to a common folder. As long as the show is shared by uTorrent it cannot be renamed or moved. This is important. It's not just a matter of lazyness, but of practical impossibility.

If you're downloading using the RSS downloader in utorrent, then it's perfectly possible to set a specific folder to download to for each show in the filter settings.

[Ned Scott]

You can retarget the files in most BitTorrent clients if they get moved or even renamed. Torrent clients have been able to do this for years. I do it every day. So much for impossible.

See, I use torrents that are fees with RSS feeds as well. I use them with anime files that even get renamed for their episode numbers. They also get moved into show folders. I don't have to manually touch anything. My computer already takes care of the boring and repetitive tasks for me.

Keep in mind, I'm not a dev, and I don't oppose anyone who would like to add this if they choose to spend their time on it. Expanding XBMC's abilities for this isn't a bad idea. I just don't see the argument for the "need" or that people don't have an alternative. Those alternatives work much better than keeping all files in a huge orgy pile of chaos.

But hey, if you want to fight for your right to make your life harder, all the more power to you.

[framerate]

Hey hey!

I'm looking for a simple answer. Is there a scraper or add-on that can scrape a single folder containing multiple files and if so where is it?

If there's not, can someone explain to me the reasoning?

I have an automated system that downloads all my files remotely via rtorrent and rsync's them to a "TV SHOWS" folder locally. XBMC points at that. All the files are named "scene" i.e:

Game.of.Thrones.s01.e01-ReLeAsEgRoUp.rXX

I'm curious the technical reasoning (or product decision) that we can't just pattern recognize the season and episode and title. Some scene releases I think do Title.Year.Season.Episode, but it should be trivial to write functions for IsDate(), IsSeason(), isEpisode() after splitting the string by periods.

Does anyone know how to get this to work? Manually moving files into "TV SHOW\Season\" folder structure defeats the whole purpose of having an auotmated system and I'm hard pressed to believe I need to write my own python script to parse this string.

I can parse the string easily to get title/year/season/episode from scene releases, can XBMC?

Thanks so much for the help!

[jmarshall]

There's various scripts and renamers out there that can do this for you.

If you wish to write a patch for XBMC, please feel free to do so. Obviously you have to support the case where the tvshow is not to be found at all in the filename, given that that's not very uncommon.

[framerate]

This answer doesn't help, as I know there are programs that do this. I asked for specifics that would automate it, as my current low power automated solution does everything except this last step.

Imagine if every time you turned on your cable box DVR you also had to go in your room, boot up your PC and run an organizing program then go BACK to your living room to have your DVR show your shows.

Newsgroups do well with sickbeard + sabnzbd but I'm trying to replicate this full automation with my bittorrent setup.

I believe I understand why XBMC doesn't do this (can't drop meta data for multiple shows into the root folder) so shortly after I wrote my reply here I wrote a command line application that will take a folder containing all your downloaded files and reorganize it into Show/Season folders for you automatically based on the file name. Once I clean up the code a bit I'll post it in case anyone wants to use it.

---

I realize the tone there sounded a bit harsh, I didn't mean it. I'm sure you're sick of answering scraper posts just as much as I'm sick of people telling me to run a program on my PC to do it

Also, my app is written in node.js. To solve the problem you mentioned I'm forcing case-sensitive show names in the config file so someone running my script will have to plug them in. It's meant to be run on a directory of "scene" releases, so they're all "TvShow.Year.s00.e00-CrAzYsCeNeName" or "TvShow.s00.e00-CrAzYsCeNeName" which is fairly easy to parse.

Also, was I right about the meta data nfo being the reason you can't scrape a directory containing multiple shows?