[SOLVED] Incorrect series name detection, even with correct content type - manual correction?

Support for Synology NAS, QNAP NAS and other Embedded Linux systems
Post Reply
xIntenso
Posts: 33
Joined: 14 Jan 2018, 23:39

[SOLVED] Incorrect series name detection, even with correct content type - manual correction?

Post by xIntenso »

Hi,

I've been using FileBot happily for some time now, and after years of sloppy naming, all my existing files are finally correctly named. Hurray!

However, I have stumbled upon some issues when trying to tackle my next challenge: making sure all new downloads are also correctly named.
I've been trying to use FileBot Node on my Synology DS218Play for this, but have ran into a couple of incorrect detections. Some examples:
Skipped [/volume1/video/Index/Kekkai Sensen & Beyond - 08 [1080p].mkv] because [/volume1/video/TV Shows/Blood Blockade Battlefront/Season 01/Blood Blockade Battlefront - S01E08 - Z's Longest Day - Part 1.mkv] already exists
In this case, FileBot incorrectly detects this as the 1st season of Blood Blockade Battlefrond, instead of the second season, which it actually is.
[TEST] From [/volume1/video/Index/Mahoutsukai no Yome - 11 [1080p].mkv] to [/volume1/video/TV Shows/A Certain Magical Index/Season 01/A Certain Magical Index - S01E11 - Sisters.mkv]
In this second case, the detected series is just plain wrong.

This result is achieved with content type set to TV Series, so they are parsed with TheTVDB (which is what I need, because Synology's DS Video app only supports TheTVDB indexing, and I prefer their naming scheme anyway).

Because most series are correctly detected, and I only add new series each quarter, I would not mind just manually correcting these anomalies, but I have so far not been able to find a way to add a "filter" to Node that fixes these detections (e.g. something that says "Kekkai Sensen & Beyond" => "Kekkai Sensen S2" or "Mahoutsukai no Yome" => "The Ancient Magus Bride").

Is there any way to achieve this?

Thanks in advance for your time.

Regards,
Niels
Last edited by xIntenso on 17 Jan 2018, 01:07, edited 1 time in total.
User avatar
rednoah
The Source
Posts: 22923
Joined: 16 Nov 2011, 08:59
Location: Taipei
Contact:

Re: Incorrect series name detection, even with correct content type - manual correction?

Post by rednoah »

1.
If the names/numbers in the file don't match the episode information online, then FileBot may not always be able to match things correctly.

This particular issue seems to have fixed itself already by now:

Code: Select all

$ filebot -rename 'Mahoutsukai no Yome - 11 [1080p].mkv' --action test --db thetvdb -non-strict
Rename episodes using [TheTVDB]
Auto-detected query: [mahoutsukai no yome]
Fetching episode data for [The Ancient Magus' Bride]
[TEST] from [Mahoutsukai no Yome - 11 [1080p].mkv] to [The Ancient Magus' Bride - 1x11 - Lovers ever run before the clock.mkv]
Processed 1 files

2.
Regardless, you can always use the --filter option to nudge FileBot in the right direction, by pre-filtering episodes that are considered for matching:
viewtopic.php?f=3&t=2127

e.g.

Code: Select all

$ filebot -rename 'Kekkai Sensen & Beyond - 08 [1080p].mkv' --action test --db thetvdb -non-strict --filter 's == 2' --log info
[TEST] from [Kekkai Sensen & Beyond - 08 [1080p].mkv] to [Blood Blockade Battlefront - 2x08 - Desperate Fight in the Macro Zone, Part 1.mkv]
:idea: Remove --log info if you want to see what's going on behind the scenes. ;)


3.
This might be helpful as well:
viewtopic.php?f=3&t=2769
:idea: Please read the FAQ and How to Request Help.
xIntenso
Posts: 33
Joined: 14 Jan 2018, 23:39

Re: Incorrect series name detection, even with correct content type - manual correction?

Post by xIntenso »

Thank you for the quick reply!

I'm much closer to getting things working, but I'm still not there yet.

The main problem is that I will be processing files from different series in a single folder, so I would like a way to apply multiple filters for certain cases.

I have looked at all the links you have provided, and while they were very useful, I feel like I still lack the final bit of information I need to solve my problems.

I feel like the optimal solution for me would be to point FB Node to a .txt-file with multiple lines of filters, that it would then apply to every file. Something like

Code: Select all

age < 1
*magical code to force FileBot to interpret strings as something else* 
e.g. "Osomatsu-san" = "Mr. Osomatsu"
*magical code to force FileBot to only search a specific season for a certain series*
e.g. forceSeason("Kekkai Sensen",2)
I know these probably sound nonsensical to you, and the final syntax will look nothing like this, but this is the functionality that I am trying to achieve.

If none of this will work, would it be best to download each series to a separate subfolder, and make a task for each series separately?

I'll give you the full output of the folder I am currently trying to index:

Code: Select all

filebot -script 'fn:amc' /volume1/video/Index --output /volume1/video/ --action test -non-strict --conflict skip --filter 'age <= 1 ' --lang en --def 'ut_label=tv' 'skipExtract=y' 'minLengthMS=300000' 'minFileSize=0' 'seriesFormat={n} - {order.airdate.s00e00} - {t}' --log fine

------------------------------------------

Run script [fn:amc] at [Mon Jan 15 13:35:25 CET 2018]
Input: /volume1/video/Index/ Kekkai Sensen & Beyond - 08 [1080p].mkv
Input: /volume1/video/Index/ Kekkai Sensen & Beyond - 10 [1080p].mkv
Input: /volume1/video/Index/ Kekkai Sensen & Beyond - 11 [1080p].mkv
Input: /volume1/video/Index/ Kekkai Sensen & Beyond - 12 [1080p].mkv
Input: /volume1/video/Index/ Mahoutsukai no Yome - 11 [1080p].mkv
Input: /volume1/video/Index/ Mahoutsukai no Yome - 12 [1080p].mkv
Input: /volume1/video/Index/ Mahoutsukai no Yome - 13 [1080p].mkv
Input: /volume1/video/Index/ Osomatsu-san S2 - 13 [1080p].mkv
Input: /volume1/video/Index/ Osomatsu-san S2 - 14 [1080p].mkv
Input: /volume1/video/Index/ Overlord II - 01 [1080p].mkv
Input: /volume1/video/Index/ Violet Evergarden - 01 [1080p].mkv
Rename episodes using [TheTVDB]
Auto-detected query: [Kekkai Sensen, kekkai sensen beyond, sensen beyond]
Fetching episode data for [Blood Blockade Battlefront]
Fetching episode data for [Beyond]
Fetching episode data for [Batman Beyond]
Apply filter [age <= 1 ] on [98] items
No matching episode:  Kekkai Sensen & Beyond - 08 [1080p].mkv
No matching episode:  Kekkai Sensen & Beyond - 10 [1080p].mkv
No matching episode:  Kekkai Sensen & Beyond - 11 [1080p].mkv
No matching episode:  Kekkai Sensen & Beyond - 12 [1080p].mkv
CmdlineException: Failed to match files to episode data
Rename episodes using [TheTVDB]
Auto-detected query: [index]
Fetching episode data for [A Certain Magical Index]
Apply filter [age <= 1 ] on [54] items
No matching episode:  Mahoutsukai no Yome - 11 [1080p].mkv
No matching episode:  Mahoutsukai no Yome - 12 [1080p].mkv
No matching episode:  Mahoutsukai no Yome - 13 [1080p].mkv
No matching episode:  Osomatsu-san S2 - 13 [1080p].mkv
No matching episode:  Osomatsu-san S2 - 14 [1080p].mkv
No matching episode:  Violet Evergarden - 01 [1080p].mkv
CmdlineException: Failed to match files to episode data
Rename episodes using [TheTVDB]
Auto-detected query: [Overlord, index]
Fetching episode data for [Overlord]
Fetching episode data for [The Devil Is a Part-Timer!]
Fetching episode data for [A Certain Magical Index]
Apply filter [age <= 1 ] on [109] items
[TEST] From [/volume1/video/Index/ Overlord II - 01 [1080p].mkv] to [/volume1/video/Overlord - S02E02 - Departure.mkv]
Processed 1 files
------------------------------------------

[Process completed]
Note the folowing things:
1) Filebot still keeps insiting on "A certain magical index" being the correct series for "Mahoutsukai no Yome" along with a couple of other, unrelated series. If I manually remove this title from the job, it will simply not find any match at all.

2) Filebot finds no matches for Kekkai Sensen with age <= 1, even though the airdates on TheTVDB are within that range. It does, however, work as intended for Overlord.

I know this is a lot to ask of you, so thanks once again for taking the time to deal with my questions!
User avatar
rednoah
The Source
Posts: 22923
Joined: 16 Nov 2011, 08:59
Location: Taipei
Contact:

Re: Incorrect series name detection, even with correct content type - manual correction?

Post by rednoah »

1.
It works for my simple test case. If there's other files, or xattr, or nfo files, then it might be picking up the name from there. You could try to move a few files into a separate folder and see how that changes behaviour.


2.
age <= 1 means less than 1 DAYS, so no episode is in range as of today:

Code: Select all

$ filebot -list --q 'Kekkai Sensen' --format '{age} {episode}'
...
22 Blood Blockade Battlefront - 2x12 - Spectral Eyes, Phantom Vision, Part 2

3.
The amc script may not be the best tool to deal with tricky files fully automatically. You could easily have your own bash script that calls filebot on certain files with certain parameters. You could have your bash script make various filebot calls for files that need special handling first, and then the amc script on whatever is left behind.

e.g.

Code: Select all

$ find . -iname 'Osomatsu*.mkv' -exec filebot -rename --action test --db thetvdb -non-strict --q 'Mr Osomatsu' {} +
Rename episodes using [TheTVDB]
Fetching episode data for [Mr. Osomatsu]
[TEST] from [Osomatsu-san S2 - 13 [1080p].mkv] to [Mr. Osomatsu - 2x13 - Year-end.mkv]
Processed 1 files
:idea: Please read the FAQ and How to Request Help.
xIntenso
Posts: 33
Joined: 14 Jan 2018, 23:39

Re: Incorrect series name detection, even with correct content type - manual correction?

Post by xIntenso »

1)
I (partially) found out what caused the behaviour: the folder containing my files was called "Index", thus probably causing the "A certain magical index" detection.

2)
Oh. :oops: Working fine now, thanks.

3)
That's a great idea! I've been trying to implement it like that for the last hour, and this is what I've cooked up so far:

Code: Select all

#!/bin/sh
find /volume1/video/Manual -iname '*Osomatsu*.mkv' -exec filebot -rename --action test --output /volume1/video/ --db thetvdb --format "TV Shows/{n}/{n} - {order.airdate.s00e00} - {t}" -non-strict --q 'Mr Osomatsu' {} + | tee "/volume1/video/result.txt"
find /volume1/video/Manual -iname '*Mahoutsukai*.mkv' -exec filebot -rename --action test --output /volume1/video/ --db thetvdb --format "TV Shows/{n}/{n} - {order.airdate.s00e00} - {t}" -non-strict --q 'Ancient Magus Bride' {} + | tee -a "/volume1/video/result.txt"
find /volume1/video/Manual -iname '*Violet*Evergarden*.mkv' -exec filebot -rename --action test --output /volume1/video/ --db thetvdb --format "TV Shows/{n}/{n} - {order.airdate.s00e00} - {t}" -non-strict --q 'Violet Evergarden' {} + | tee -a "/volume1/video/result.txt"
filebot -script 'fn:amc' /volume1/video/Manual --output /volume1/video/ --action test -non-strict --conflict skip --filter 'age <= 30' --lang en --def 'ut_label=tv' 'skipExtract=y' 'minLengthMS=300000' 'minFileSize=0' 'seriesFormat=TV Shows/{n}/{n} - {order.airdate.s00e00} - {t}' | tee -a "/volume1/video/result.txt"
However, I have to admit I am once again running into issues that I would like your help for:

A)
The three manual runs work perfectly, however AMC starts up fine from bash, but doesn't continue and eventually runs out of memory.
This happens during a test run via ssh, using the admin account (since that is the account the scheduler will be using):

Code: Select all

admin@CENSORED:~$ sh /volume1/video/filebot.sh
Rename episodes using [TheTVDB]
Fetching episode data for [Mr. Osomatsu]
Stripping invalid characters from new path: TV Shows/Mr. Osomatsu/Mr. Osomatsu - S02E14 - Weekly Drama Serial - Sanematsu-san: Episode 9 UMA Expedition Team The Choromatsu Incident
[TEST] From [/volume1/video/Manual/Mr. Osomatsu - 2x13 - Year-end - S02E13 - Year-end.mkv] to [/volume1/video/TV Shows/Mr. Osomatsu/Mr. Osomatsu - S02E13 - Year-end.mkv]
[TEST] From [/volume1/video/Manual/Mr. Osomatsu -S02E14 - Weekly Drama Serial - Sanematsu-san Episode 9 UMA Expedition Team The Choromatsu Incident.mkv] to [/volume1/video/TV Shows/Mr. Osomatsu/Mr. Osomatsu - S02E14 - Weekly Drama Serial - Sanematsu-san Episode 9 UMA Expedition Team The Choromatsu Incident.mkv]
Processed 2 files
Rename episodes using [TheTVDB]
Fetching episode data for [The Ancient Magus' Bride]
[TEST] From [/volume1/video/Manual/Mahoutsukai no Yome - 11 [1080p].mkv] to [/volume1/video/TV Shows/The Ancient Magus' Bride/The Ancient Magus' Bride - S01E11 - Lovers ever run before the clock..mkv]
[TEST] From [/volume1/video/Manual/Mahoutsukai no Yome - 12 [1080p].mkv] to [/volume1/video/TV Shows/The Ancient Magus' Bride/The Ancient Magus' Bride - S01E12 - Better to ask the way than go astray..mkv]
[TEST] From [/volume1/video/Manual/Mahoutsukai no Yome - 13 [1080p].mkv] to [/volume1/video/TV Shows/The Ancient Magus' Bride/The Ancient Magus' Bride - S01E13 - East, west, home's best..mkv]
Processed 3 files
Rename episodes using [TheTVDB]
Fetching episode data for [Violet Evergarden]
[TEST] From [/volume1/video/Manual/Violet Evergarden - 01 [1080p].mkv] to [/volume1/video/TV Shows/Violet Evergarden/Violet Evergarden - S01E01 - 'I Love You' and Auto Memoir Dolls.mkv]
Processed 1 files
Run script [fn:amc] at [Mon Jan 15 16:52:34 CET 2018]
Parameter: ut_label = tv
Parameter: skipExtract = y
Parameter: minLengthMS = 300000
Parameter: minFileSize = 0
Parameter: seriesFormat = TV Shows/{n}/{n} - {order.airdate.s00e00} - {t}
Argument[0]: /volume1/video/Manual
Ignore system path: /volume1/video/Manual/@eaDir
Input: /volume1/video/Manual/Mahoutsukai no Yome - 11 [1080p].mkv
Input: /volume1/video/Manual/Mahoutsukai no Yome - 12 [1080p].mkv
Input: /volume1/video/Manual/Mahoutsukai no Yome - 13 [1080p].mkv
Input: /volume1/video/Manual/Mr. Osomatsu - 2x13 - Year-end - S02E13 - Year-end.mkv
Input: /volume1/video/Manual/Mr. Osomatsu -S02E14 - Weekly Drama Serial - Sanematsu-san Episode 9 UMA Expedition Team The Choromatsu Incident.mkv
Input: /volume1/video/Manual/Violet Evergarden - 01 [1080p].mkv
Java heap space
java.lang.OutOfMemoryError: Java heap space
        at com.ibm.icu.impl.ICUBinary.getByteBufferFromInputStreamAndCloseStream(ICUBinary.java:711)
        at com.ibm.icu.impl.ICUBinary.getData(ICUBinary.java:499)
        at com.ibm.icu.impl.ICUBinary.getData(ICUBinary.java:437)
        at com.ibm.icu.impl.ICUResourceBundleReader$ReaderCache.createInstance(ICUResourceBundleReader.java:188)
        at com.ibm.icu.impl.ICUResourceBundleReader$ReaderCache.createInstance(ICUResourceBundleReader.java:177)
        at com.ibm.icu.impl.SoftCache.getInstance(SoftCache.java:65)
        at com.ibm.icu.impl.ICUResourceBundleReader.getReader(ICUResourceBundleReader.java:232)
        at com.ibm.icu.impl.ICUResourceBundle.createBundle(ICUResourceBundle.java:1280)
        at com.ibm.icu.impl.ICUResourceBundle$4.load(ICUResourceBundle.java:1153)
        at com.ibm.icu.impl.ICUResourceBundle$1.createInstance(ICUResourceBundle.java:92)
        at com.ibm.icu.impl.ICUResourceBundle$1.createInstance(ICUResourceBundle.java:89)
        at com.ibm.icu.impl.SoftCache.getInstance(SoftCache.java:65)
        at com.ibm.icu.impl.ICUResourceBundle.instantiateBundle(ICUResourceBundle.java:1140)
        at com.ibm.icu.impl.ICUResourceBundle.access$600(ICUResourceBundle.java:35)
        at com.ibm.icu.impl.ICUResourceBundle$4.load(ICUResourceBundle.java:1201)
        at com.ibm.icu.impl.ICUResourceBundle$1.createInstance(ICUResourceBundle.java:92)
        at com.ibm.icu.impl.ICUResourceBundle$1.createInstance(ICUResourceBundle.java:89)
        at com.ibm.icu.impl.SoftCache.getInstance(SoftCache.java:65)
        at com.ibm.icu.impl.ICUResourceBundle.instantiateBundle(ICUResourceBundle.java:1140)
        at com.ibm.icu.impl.ICUResourceBundle.access$600(ICUResourceBundle.java:35)
        at com.ibm.icu.impl.ICUResourceBundle$4.load(ICUResourceBundle.java:1177)
        at com.ibm.icu.impl.ICUResourceBundle$1.createInstance(ICUResourceBundle.java:92)
        at com.ibm.icu.impl.ICUResourceBundle$1.createInstance(ICUResourceBundle.java:89)
        at com.ibm.icu.impl.SoftCache.getInstance(SoftCache.java:65)
        at com.ibm.icu.impl.ICUResourceBundle.instantiateBundle(ICUResourceBundle.java:1140)
        at com.ibm.icu.impl.ICUResourceBundle.access$600(ICUResourceBundle.java:35)
        at com.ibm.icu.impl.ICUResourceBundle$4.load(ICUResourceBundle.java:1183)
        at com.ibm.icu.impl.ICUResourceBundle$1.createInstance(ICUResourceBundle.java:92)
        at com.ibm.icu.impl.ICUResourceBundle$1.createInstance(ICUResourceBundle.java:89)
        at com.ibm.icu.impl.SoftCache.getInstance(SoftCache.java:65)
        at com.ibm.icu.impl.ICUResourceBundle.instantiateBundle(ICUResourceBundle.java:1140)
        at com.ibm.icu.impl.ICUResourceBundle.getBundleInstance(ICUResourceBundle.java:1113)

Failure (°_°)
Edit: Please ignore question B
B)
Another thing is that, apparently, when running bash from Synology's scheduler, I will need to provide full paths to all files/programs.
I tried using "/volume1/@appstore/filebot/FileBot.jar" instead of "filebot" in order to get the scheduler to actually run filebot, but then I get a permission denied error when testing the script using SSH, even when using the admin account.
Do you know how I need to call filebot from the scheduler, so that it is acutally ran?

Edit: has been fixed by renaming "find" to "/usr/bin/find" in the Bash script

C)
Finally a strange error that is technically not even FileBot territory, but I would like to ask you regardless since you seem knowledegable, and Googling this phenomenon yields no results:
As you might have noticed from my script code, I'm trying to write the console output of the bash file to a file "result.txt", but it behaves strangely.
The first three manual runs create a result.txt that cannot be opened in NotePad++ (or any other text editor) from Windows using SMB (Windows gives a Y:\result.txt - path does not exist error [Y is my SMB mount point]), and when downloaded from the NAS using File Station, this file has name "result.txt%0D".
The AMC run creates a perfectly working result.txt file next to the other file, so I now have two files named result.txt, of which one is corrupt.

Do you have any idea what might be causing this?

I would once again like to thank you for taking the time to provide your (excellent indeed) support.

Edit: Solved question B on my own
User avatar
rednoah
The Source
Posts: 22923
Joined: 16 Nov 2011, 08:59
Location: Taipei
Contact:

Re: Incorrect series name detection, even with correct content type - manual correction?

Post by rednoah »

1.
Please run filebot -script fn:sysinfo and post the output. If you run out of memory, then maybe the device just doesn't have enough RAM, or you're not giving the Java process enough RAM via the -Xmx option (by default, the filebot startup script should set -Xmx to 70% of your RAM).

:idea: 512 MB is definitely enough because I've got a DS213J myself. Less might work, but may not always work in all cases, depending on the situation.


2.
The profile is different so the environment is different, so things that you expect to be in the $PATH may not be. It's just how cron works. You can compare printenv output between your SSH session and a script called by cron and then you'll see why things behave differently.


3.
"result.txt%0D" probably means there's special characters at the end of the filename that Windows can't deal with (but Linux can).

@see https://en.wikipedia.org/wiki/Newline

i.e. you're doing this:

Code: Select all

touch "file.txt
"
which is distinctively not the same as this:

Code: Select all

touch "file.txt"
because you can have a newline in the filename, or control key codes, or emoji, etc. That's perfectly cool on Linux, but it'll cause compatibility issues when accessing files via SMB from Windows. ;)
:idea: Please read the FAQ and How to Request Help.
User avatar
rednoah
The Source
Posts: 22923
Joined: 16 Nov 2011, 08:59
Location: Taipei
Contact:

Re: Incorrect series name detection, even with correct content type - manual correction?

Post by rednoah »

:idea: I know why you get CR (0x0D) as last character.

You're probably using Windows to edit your script files. This means you're using a Windows editor that's using \r\n (Windows style) instead of \n (Unix style) for newline.

bash in particular does not support \r\n and will only interpret \n as newline. Therefore, if you type a line on Windows, you'll be writing this:

Code: Select all

test.txt\r\n
But bash only interprets \n as newline, so everything beforehand, test.txt\r, is interpreted as argument value, in this case a file path, and so you're creating a filename where the last character of the filename is carriage return, and the key code of carriage return is 0x0D.

QED.
:idea: Please read the FAQ and How to Request Help.
xIntenso
Posts: 33
Joined: 14 Jan 2018, 23:39

Re: Incorrect series name detection, even with correct content type - manual correction?

Post by xIntenso »

You were right on the money with the .txt-file anomalies. After converting EOL to Unix in NotePad++, it works as intended! :D
Many thanks for taking the time to think about this, because the solution would have been about the last thing that would have come to my mind.

The only thing left now is AMC going "out of memory" when called through commandline/script.
My NAS (DS218Play) has 1GB of ram, and execution using the GUI instead of the script works just fine (in fact, that is where I got the syntax in the Bash script from).
Using the GUI gives the following result:

Code: Select all

filebot -script 'fn:amc' /volume1/video/Manual --output /volume1/video/ --action test -non-strict --conflict skip --filter 'age <= 30' --lang en --def 'ut_label=tv' 'skipExtract=y' 'minLengthMS=300000' 'minFileSize=0' 'seriesFormat=TV Shows/{n}/{fn} - {order.airdate.s00e00} - {t}' --log fine

------------------------------------------

Run script [fn:amc] at [Tue Jan 16 02:43:01 CET 2018]
Input: /volume1/video/Manual/Mahoutsukai no Yome - 11 [1080p].mkv
Input: /volume1/video/Manual/Mahoutsukai no Yome - 12 [1080p].mkv
Input: /volume1/video/Manual/Mahoutsukai no Yome - 13 [1080p].mkv
Input: /volume1/video/Manual/Mr. Osomatsu - 2x13 - Year-end - S02E13 - Year-end.mkv
Input: /volume1/video/Manual/Mr. Osomatsu -S02E14 - Weekly Drama Serial - Sanematsu-san Episode 9 UMA Expedition Team The Choromatsu Incident.mkv
Input: /volume1/video/Manual/Violet Evergarden - 01 [1080p].mkv
Rename episodes using [TheTVDB]
Auto-detected query: [manual]
Fetching episode data for [Owner's Manual]
Fetching episode data for [The Herbalist's Manual]
Fetching episode data for [Earth: The Operators Manual]
Fetching episode data for [You: The Owner's Manual with Dr. Mehmet Oz]
Fetching episode data for [Manimal]
Apply filter [age <= 30] on [48] items
No matching episode: Mahoutsukai no Yome - 11 [1080p].mkv
No matching episode: Mahoutsukai no Yome - 12 [1080p].mkv
No matching episode: Mahoutsukai no Yome - 13 [1080p].mkv
No matching episode: Violet Evergarden - 01 [1080p].mkv
CmdlineException: Failed to match files to episode data
Rename episodes using [TheTVDB]
Auto-detected query: [Mr. Osomatsu]
Fetching episode data for [Mr. Osomatsu]
Fetching episode data for [Little Osomatsu]
Apply filter [age <= 30] on [115] items
Stripping invalid characters from new path: TV Shows/Mr. Osomatsu/Mr. Osomatsu -S02E14 - Weekly Drama Serial - Sanematsu-san Episode 9 UMA Expedition Team The Choromatsu Incident - S02E14 - Weekly Drama Serial - Sanematsu-san: Episode 9 UMA Expedition Team The Choromatsu Incident
[TEST] From [/volume1/video/Manual/Mr. Osomatsu - 2x13 - Year-end - S02E13 - Year-end.mkv] to [/volume1/video/TV Shows/Mr. Osomatsu/Mr. Osomatsu - 2x13 - Year-end - S02E13 - Year-end - S02E13 - Year-end.mkv]
[TEST] From [/volume1/video/Manual/Mr. Osomatsu -S02E14 - Weekly Drama Serial - Sanematsu-san Episode 9 UMA Expedition Team The Choromatsu Incident.mkv] to [/volume1/video/TV Shows/Mr. Osomatsu/Mr. Osomatsu -S02E14 - Weekly Drama Serial - Sanematsu-san Episode 9 UMA Expedition Team The Choromatsu Incident - S02E14 - Weekly Drama Serial - Sanematsu-san Episode 9 UMA Expedition Team The Choromatsu Incident.mkv]
Processed 2 files
------------------------------------------

[Process completed]
While it is stuck, the NAS' Resource Monitor shows Java at ~25% CPU (it is a 4-core system, so probably full load on 1 core, since my guess is that FileBot is single-threaded) and 200MB private memory (~5MB shared). Total system RAM used is about ~55% (550MB).
Running the command directly in SSH also yields the same crash result, so it's most likely not the fact that FileBot is being called from a script.

By the way, is there a way to get FileBot to ignore the folder name when indexing?

Thanks once again!
xIntenso
Posts: 33
Joined: 14 Jan 2018, 23:39

Re: Incorrect series name detection, even with correct content type - manual correction?

Post by xIntenso »

Oh, I almost forgot, here is the output of the command you requested:

Code: Select all

filebot -script fn:sysinfo
FileBot 4.7.9 (r4984)
JNA Native: java.lang.UnsatisfiedLinkError: Native library (com/sun/jna/linux-aarch64/libjnidispatch.so) not found in resource path ([file:/volume1/@appstore/filebot/FileBot.jar])
MediaInfo: net.filebot.mediainfo.MediaInfoException: Unable to load aarch64 (64-bit) native library libmediainfo.so: Could not initialize class com.sun.jna.Native
Apache Commons VFS: [zip, rar]
Chromaprint: java.io.IOException: Cannot run program "fpcalc": error=2, No such file or directory
Extended Attributes: OK
Unicode Filesystem: OK
Script Bundle: 2017-11-21 (r512)
Groovy: 2.4.10
JRE: Java(TM) SE Runtime Environment 1.8.0_131
JVM: 64-bit Java HotSpot(TM) 64-Bit Server VM
CPU/MEM: 4 Core / 156 MB Max Memory / 10 MB Used Memory
OS: Linux (aarch64)
Package: SPK
uname: Linux [CENSORED] 4.4.15+ #15217 SMP Wed Dec 20 17:38:18 CST 2017 aarch64 GNU/Linux synology_rtd1296_ds218play
Done ヾ(@⌒ー⌒@)ノ
xIntenso
Posts: 33
Joined: 14 Jan 2018, 23:39

Re: Incorrect series name detection, even with correct content type - manual correction?

Post by xIntenso »

From the log I can see that only 156MB of memory is allocated by default. Any idea why it might do that? It doesn't seem to happen when calling FileBot from GUI.

Adding the following line of code that increases the amount of allowed memory to my script before calling FileBot made it work:

Code: Select all

export JAVA_OPTS="-Xmx512m"
For any future people looking at this thread with the same issues, this is my final script:

Code: Select all

#!/bin/sh
export JAVA_OPTS="-Xmx512m"
/usr/bin/find /volume1/video/Manual -iname '*Osomatsu*.mkv' -exec filebot -rename --action test --output /volume1/video/ --db thetvdb --format "TV Shows/{n}/{n} - {order.airdate.s00e00} - {t}" -non-strict --q 'Mr Osomatsu' {} + | tee "/volume1/video/result.txt"
/usr/bin/find /volume1/video/Manual -iname '*Mahoutsukai*.mkv' -exec filebot -rename --action test --output /volume1/video/ --db thetvdb --format "TV Shows/{n}/{n} - {order.airdate.s00e00} - {t}" -non-strict --q 'Ancient Magus Bride' {} + | tee -a "/volume1/video/result.txt"
/usr/bin/find /volume1/video/Manual -iname '*Violet*Evergarden*.mkv' -exec filebot -rename --action test --output /volume1/video/ --db thetvdb --format "TV Shows/{n}/{n} - {order.airdate.s00e00} - {t}" -non-strict --q 'Violet Evergarden' {} + | tee -a "/volume1/video/result.txt"
filebot -script 'fn:amc' /volume1/video/Manual --output /volume1/video/ --action test -non-strict --conflict skip --filter 'age <= 30' --lang en --def 'ut_label=tv' 'skipExtract=y' 'minLengthMS=300000' 'minFileSize=0' 'seriesFormat=TV Shows/{n}/{n} - {order.airdate.s00e00} - {t}' --log fine | tee -a "/volume1/video/result.txt"
So I guess my final question now is: Is there any way to get AMC to ignore the name of the folder I am trying to index?

Thanks once again for pointing me in the right direction!
User avatar
rednoah
The Source
Posts: 22923
Joined: 16 Nov 2011, 08:59
Location: Taipei
Contact:

Re: Incorrect series name detection, even with correct content type - manual correction?

Post by rednoah »

No, but if you rename Index to Complete or Staging then it'll work.
:idea: Please read the FAQ and How to Request Help.
xIntenso
Posts: 33
Joined: 14 Jan 2018, 23:39

Re: Incorrect series name detection, even with correct content type - manual correction?

Post by xIntenso »

All right, then all of my questions have been answered!
Once again, thank you for the great, quick support, and have a nice day!
felho001
Posts: 1
Joined: 27 Feb 2018, 14:23

Re: [SOLVED] Incorrect series name detection, even with correct content type - manual correction?

Post by felho001 »

Hi my synology nas works fine filebot node but few months ago gone settings line when i started filebot on safari 11.0.3. Some one could you help me? https://www.dropbox.com/s/2sjf0aqwdqxnh ... 4.png?dl=0
If I use chrome I see the settings row. Why?
User avatar
rednoah
The Source
Posts: 22923
Joined: 16 Nov 2011, 08:59
Location: Taipei
Contact:

Re: [SOLVED] Incorrect series name detection, even with correct content type - manual correction?

Post by rednoah »

The command configuration is stored in the browser, so if you change browser, or clear local storage, then those settings will be lost.
:idea: Please read the FAQ and How to Request Help.
Post Reply