[Logo] Jaikoz and SongKong Forums
  [Search] Search   [Recent Topics] Recent Topics   [Members]  Member Listing   [Groups] Back to home page 
[Register] Register / 
[Login] Login 
Messages posted by: paultaylor  XML
Profile for paultaylor -> Messages posted by paultaylor [7487] Go to Page: Previous  1, 2, 3 ... 236, 237, 238 ... 248, 249, 250 Next 
Author Message
This is fixed in Jaikoz 1.12 which is released today.
The double up of columns is fixed in Jaikoz 1.12 which is released today.
This is fixed in Jaikoz 1.12 which is released today.
This release adds a load of small enhancements and bug fixes, many of them suggested on these forums, thanks !


boliver10 wrote:
1. During the auto-correct: please provide more feedback about progress:
a. It would be nice to see "tagging 405 out of 20,300 etc. for each step

Originally the Autocorrecter only contained the local correcter tasks such as 'Correct Artists' and these dont really work one record at time, and the corrections are quick so there wasnt a requirement for this. But now I have added in tasks such as Tag From MusicBrainz which are much slower and work one record at a time this would be useful.

boliver10 wrote:

b. It would also be nice to see what it is doing: once the auto tag starts you can't see anything on the screen or move the sliders etc.

To improve performance Jaikoz trys to minimize screen updates whilst running tasks, for this reason I can't allow you to scroll around whilst this was going because I would have to update the screen everytime there had been a change. I'll give you an example a customer was trying to delete the data from all the fields of about 4000 records, this equates to about 320,000 updates (4000 * 80 fields) if I updated each field immediately after deleting or just one update if i waited until the end !

boliver10 wrote:

c. an (accurate) estimation of completion would be nice (by task and by total)

This would be useful for the Remote Correcters. I can give you some estimates here.

Retrieve Acoustic Ids:
This is really cpu bound the processing of your file is cpu intensive, it then also has to contact MusicIP over the Internet. As a rough estimate assume 5 seconds per file (I may be able to optimize this a bit).

Tag from MusicBrainz:
The Terms of Service of MusicBrainz do not allow more than one request per second and this is strictly enforced. So every file you look up against MusicBrainz can take upto 3 seconds.
1 second for lookup by Acoustic Id
1 second for lookup by Metadata
If a match is found 1 second for additional album Info lookup

This is the worst case scenario but it is not normally this bad. The album lookups are cached so any particular album will only get looked up once.
The Acoustic Id is obviously only done if your file has an acoustic Id. If you reload files that already been matched you can ensure the match is not done again by selecting 'Action Settings/MusicBrainz Settings/Match/Do not Match Online if already has MusicBRainz Unique id' option. If you only want to match by Acoustic Id you can select the 'Action Settings/MusicBrainz Settings/AutoMatch/Do Not Match If do not have an Acoustic Ids' option.

boliver10 wrote:

2. In compilation albums, a particular song could conceivably be on multiple albums. But I don't want the program to auto-tag a song on a different album when clearly it is part of a compilation album as identified by the filename, directory or even the original tag! I find that jaikoz will disregard all of this info and put the song in another album because musicbrainz or another service identifies it as part of another album

Jaikoz doesnt disregard the tag at all, but it is just one factor. If a particular file has an entry in MusicBrainz as being in a compilation and the original album, and your album tag is for the compilation it would should pick the compilation album choice. However if MusicBrainz only has an entry for the file in the original album (and no match for the compilation album) Jaikoz will use that one IF it is has an acoustic id OR its metadata is sufficiently similar for there to be a match. I suggest you try out the Manual Match from MusicBrainz option that will give you a better idea of how jaikoz decides which is the best match. The Compilation Album issue is a complex one as almost everybody has different ideas about how they want compilation albums dealt with, another customer wants the match to always be done to the original album even if the track did actually come from a compilation.

boliver10 wrote:

3. It seems awfully slow - most of the time it is only using 2% of CPU: is this because it is spending all of its time accessing the internet?

See above.

boliver10 wrote:

4. If there is already a musicbrainz id for the file, Jaikoz seems to try to get it again, wasting time. Is this true or am I not understanding?

If you reload files that already been matched you can ensure the match is not done again by selecting 'Action Settings/MusicBrainz Settings/Match/Do not Match Online if already has MusicBrainz Unique id' option. Also note you can limit what files are processed by using the popmenu on the selected files, or by filtering your records using the tagbrowser.

boliver10 wrote:

5. Many music collectors are using Monkeys Audio .ape and Flac .flac files. Support for them should be easy since their converters are free to be used in commercial products. I would pay extra for this feature.

Work still has to be done to read and write to these files, there is no reliable Java library for this, so I have to do it myself. Ive already started looking at Flac and I expect to do .ape at some point. When these are available they will be available to all existing users at no extra cost. But if you want a particular feature there is a customization option that lets you pay a oneoff payment to get a job worked on more quickly.

boliver10 wrote:

The above said, I can't think of ANY program that does everything this program does. WONDERFUL! 

Personally I completely agree, although I don't think this is the general view of the MusicBrainz Community. I will be looking at this and minimizing the number of albums songs are spread over in a later release.
Hi, well this error is not meant to occur if someone could actually send me their log files I might be able to resolve the problem
The manual tag from musicbrainz doesnt expect all the criteria to match, it is weighted so Im suprised your not getting any matches. Do you think you could give me an example please next time you do this, I would need the information held in the tag, and the match on Musicbrainz
Im afraid not at the moment, I realize this is an inconvienience, this is on the list to be done.
There was a problem with the MBServer, fixed now please try again.
Hi actually I think there is a temporary problem with the MusicBrainz server getting overloaded at the moment, this may be the problem

Tank wrote:
Hi. I've got a problem that when ever i do a manual musicbrainz lookup that the screen shows up normally but without any details fill in the table.

Also how do i get the musicbrainz unique id to get updated automatically or do i have to go on the musicbrainz website and copy and past the musicbrainz unique id code for each file. The files affected are usually one or two files out of an album which don't have the musicbrainz unique id which is weird as i'd thought it would update from the id3 tags which are already filled in and correct.

Hi, have you changed any of the MusicBrainz settings, could you send me a screenshot of the problem.

if you do AutoCorrectTag From MusicBrainz and it finds a match it populates the Musicbrainz Unique Id, as it does if you run Manual Correct Tag from MusicBrainz and you select a possible match. Are you getting confused between AutoCorrectTag From MusicBrainz and Retrieve Acoustic Ids ?
Yes, what you could do is:

Load your files
Change the Version field of the first record to ID3v1
Copy and paste this value to all other rows
Then do 'File/Force Save' menu item

this will save all your files to ID3v1 format, ID3v1 can only hold album, artist, title, track no, genre, comment so everything else will get lost.

If you reopen your files the version will be set to ID3v23 by default (See 'Settings/Save Settings/ID3 Tag V2 Tab' menu item) so if you now start adding extra info this will be saved on a subsequent save.

Hi, Ive taken a look at the logs. I dont know how technical you are but the problem is that every field you delete is processed seperately, and if the field you delete is in the Artist,Album or Genre column then the data has to be reindexed in a separate thread to make sure the Browser is synced with the data, it is running out of memory generating threads. I need to change the way of doing this so it works better when you modify many fields, however I was not expecting so many fields to be deleted in one go.

In your log file you load 4,643 files but then delete 264,651 fields in one go, I assume you are basically blanking out all the fields ?

greengeek wrote:
I then hide all the fields I do not want to modify and select all the visible fields to delete which I do by right clicking and choosing Delete. Half the time it will take about 10 seconds then complete and I can then save the files, the other half of the time it will up with an error stating "Out of memory, Jaikoz must be restarted, to increase the memory allocated to jaikoz modify its memory settings in the start up file and then restart."

This doesn't sound like it really is a memory error, there may be a different underlying cause, please send me your jaikozuser0-0.log and jaikozdebug0-0.log so I can investigate further.

greengeek wrote:

Is there anyway I can turn off this check, or not limit the amount of ram it thinks i can only use? Also I noticed the batch file and I modified those numbers in there of both the xms and xmx and that didnt see to make much difference.

Unfortunately with Java you have to set a maximum memory heap size, you cant just set it to unlimited. If it really is a memory problem increasing these numbers should make a difference, however editing the batch file will only have an effect if you run Jaikoz.bat. If you run Jaikoz.exe you need to edit the


fields in jaikoz.lax instead, by default it is set to mimimum heap size of 64Mb and maximum of 256Mb, so you could increase the second figure to 1000000000

allowing heap size of upto 1GB

greengeek wrote:

I also noticed there was a bunch of other parimeters that the batch file called. Is there a list of what modifiers we could use on the exe file?

Once again, if using the exe file you need to modify a value in Jaikoz.lax
rather than the batch file.

lax.command.line.args=-l2 -m2 -f

In either case these other parameters refer to logging

-l outputs information in Jaikoz, so this is concerned with the various tasks that can be performed in Jaikoz.

-m outputs information in the the jaudiotagger library, so this is concerned with the reading and writing of files.

Both accept values of 1-7, the higher the value the more output will be generated, if you set to higher than 4 jaikoz will slow down noticeably in some circumstances.

-f ensures the debugging in written to the jaikozdebug.log file

greengeek wrote:

The other thing that I am wondering is there is a save settings option, but iI can not tell what settings it is saving. Is there anyway once i get my edit tab looking the way I want it with some fields showing and some not showing to save that view so that it shows that way every time i bring up the app?

Yes, just moving columns around in the diaply does not automtically restore them on next invocation of Jaikoz. You need to go to 'Settings/Table Settings' menu item to define the columsn that should be displayed by default on startup, you can use 'From Display' button to use the existing setup as the default.

The 'File/Save Settings' menu item will then tave ANY settings/preferences you have made during this session of Jaikoz.

itools wrote:
I've got a bunch of songs that either do not have a title, or the title is prefixed with 01 or 02.

My best information for those is in the file name. And while some of the file names also are prefexed with track number or other garbage, I can use Renamer4mac to apply regular expressions to clean up.

So I want to select those songs, and select an option to move the file name data into the title field. Minus the trailing file name suffix of mp3 or m4a, etc.

You can use 'Action/File and Folder COrrect/Tag From Filename' to extract file info into tags, and in your case I would think this would move the info directly into the title field.

itools wrote:

Also on the Find and Replace function, I want to replace with nothing.  

This has already been reported and wil be fixed.

ErwinB wrote:
Personally I would prefer the "one file at a time" approach I think Focher suggested in another thread I can't find at the moment.

Maybe I need two Manual Corrects, a simple view and an advanced view with the advanced view only showing one file at a time because of the amount of space required to fit everything in. Im hesitant to do this by default as I think it will be too complicated for some people to understand, and I think one file at a time is slower.

ErwinB wrote:

- Make it possible to undo a selection. If you make an error, now the only way to correct it is to cancel the lookup and redo it. In other words click a line once selects it, a second click deselects it

Yep, that is needed

ErwinB wrote:

- Color coding differences so you can for example detect spelling errors even without selecting a suggested match. Which would be really great in combination with:

Thats an interesting idea that could work

ErwinB wrote:

- Placing a row of checkboxes beneath every column of every block of suggestions. The default settings for the checkboxes would be the selection made in the musicbrainz preferences. I hope you get what I mean. This would make it possible to customize the fields taken over from musicbrainz per song looked up.

Ok, I understand I can see the advantage of that , though the screen is getting cluttered now.

ErwinB wrote:

On a sidenote, the automatic sizing of the window is not always right currently. If you lookup just 2 items sometimes the first is shown completely, the second only partly while there is ample room on the screen.  

Added to enhancement list
As I said earlier there used to be one, but it caused confusion. Please explain how you want it to work and I will look into reintroducing a save folder. For example does it just provide a default starting folder for when you use the 'Save and move' option , or does it effect the values of the Base Folder column in the Edit tab when you first load the files ?
Auto Sorting:I don't think they are being sorted, but the filtering is getting lost, this shoudnt be happening.

Loss of Selection:if you change the value of the filtered value, then the value will disappear from the list wont it ?

Im not sure can you send me a problem file and your log files please

Focher wrote:
Minor bug in that when you do a Find and Replace, the drop down list in the dialog window that lets you limit your search to specific column shows each column name twice.

oops, Ill sort that out for next release.

Focher wrote:

I also had my Jaikoz freeze when I tried to execute the find and replace, but I have not re-tested to see if that was a recurring issue.

It appears to be working ok
HI, Im unable to reproduce this problem , please send me a screenshot and double check you can replicate this consistently
Control clicking shouldnt work at all actually , its not something Ive added and I dont know if things work properly if you select non-contiguous selection.

ErwinB wrote:

If you select a file by highlighting a field and then rightclick it and select "run external app." nothing happens.

Thanks yes youv'e found a bug, when you right click on a field in the edit tab it actually selects the file that is selected in the index header(if any) instead of the one selected. This is now fixed for next release.
Hi, nothing in the logs unfortunately - I am at a loss to know the problem but there are a couple of other things we could do.

1. Run Jaikoz directly from the Disk Image, rather than copying it to Applications and see if that works.

2. Increase debugging level

Edit Applications/Jaikoz.app/Contents/Info.plist, changing the line
<string>-l2 -m2 -f<string>
<string>-l7 -m2 -f<string>

then retrying , and send me the logs again.

(The debugging will slow Jaikoz down, so only do this temporarily for diagnostic purposes)

3. is there anyway I could connect to your Mac. (I did this for other customer using DesktopTransporter).

radellaf wrote:
Main bug:
Filenames from tags works the first time but if the setting is changed, invoking it again makes no changes to the filenames. Revert to saved and invoking it does use the new pattern. Maybe bug?

Yes this is a known regression bug that has already been fixed for the next release.

radellaf wrote:

The edit table reverts to default column arrangement on restart. Should remember the way columns were last set up, no?

Possibly, but if you go to Settings/Table Settings Menu and make adjustments there (you can use 'From Display' to match current screen) and save, the display will be saved on next invocation

radellaf wrote:

"%A-%D" and "%A - %D" both generate filenames without the spaces around the dash. Is this a bug, or is there a code, not mentioned in the filenames-from-tags portion of the manual, for putting a space in the pattern?

It is not doing this for me, I think you having this problem because of the first bug you mentioned (changes in mask dont take effect until restart Jaikoz.

radellaf wrote:

Possibly a feature request, though kind of a "deal breaker" level of essential:
Commands seem to act on ALL loaded files, even if only some are selected. Am I doing something wrong or is there no facility to, say, run "Correct tags from musicbrainz" on only the selected rows?

The main menu commands work on all files, whereas the popmenus and popup shortcuts) work on the selected files. Does anybody else have a problem with the way it works ?

radellaf wrote:

Evaluation complaint:
The 30 day limit, OK, but I can't really evaluate how well this works with a 50 file limit. My test case for a tagger here is a miscellaneous music folder with 475 files. Any possibility of more generous terms? Or, perhaps, buy a license, but with 30 day "return privilege"? 

The 50 file limit only applies to saving files, there is nothing to stop you loading your 475 folder and using all the tasks in Jaikoz (except save), as all changes are made in memory until you save back to disk. So I don't really this as a major restriction. If I was to allow an unlimited number of files saves I would have many potential customers just fixing their whole collection and not making a purchase. I cant give a buy and return if you dont like it because once you have a license there is little I can do to prevent using it even if I issue a refund.

Sorry you need to run ./jaikoz.sh NOT just jaikoz.sh

sbrixey wrote:

If you still think sending along a console log will help you out I'm still happy to send it along.

Can you empty the console log, then try Jaikoz again, and then immediately after that send the console log.

You could also in a terminal window
cd /Applications/Jaikoz.app/Contents/Resources/java

and see if that works

Is this since upgrading Jaikoz ?

There should be at least a jaikozuser0-0.log in Home/Library/Logs/Jaikoz and you will definently have a console log called console.log which is easiest to find by running the Applications/Utilities/Console application.
Profile for paultaylor -> Messages posted by paultaylor [7487] Go to Page: Previous  1, 2, 3 ... 236, 237, 238 ... 248, 249, 250 Next 
Go to:   
Powered by JForum 2.1.6 © JForum Team