Skip to main content

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: CUETools versions 1.9.5 through 2.1.6 (Read 1889481 times) previous topic - next topic
0 Members and 3 Guests are viewing this topic.

CUETools versions 1.9.5 through 2.1.5 (current)

Reply #775
OK, some requests and and questions/issues, all related to the AccurateRip verification functionality only.
Version 2.0.4 tried, on some ~ 70 folders in batch.



Issues/questions first:
-> Is the CRC32 "W/O NULL" done by setting offset for the entire disc so that it starts at first one?
IOW, if I want to scan my library for CDs that are identical modulo offset, then -- at least if there are sufficiently many zeroes in the beginning or at the end, and no data tracks -- it is necessary and sufficient that all these match?
Edit: And is the first one a CRC for the entire image, with the above holding true?

-> Some discs are allegedly not present in the database, even if they at time of ripping got confidence levels in the thirties. (dBpoweramp used, no single tracks needed re-ripping.)

-> Is it supposed to work well with dBpoweramp-esque HTOA track 0's?




Then requests:

-> CLI to accept folder name as input

-> A file report using ">" in the CLI.

-> In the batch GUI version, write the content of the report window as a file, with each line written as done (in case of crashes -- the process takes overnight here ...) or in the very least a "mark all".

-> The output giving, say "confidences 31 to 45" rather than referencing just the first file

-> In case of files already present: A "Yes to all" and "No to all" button. Or an "overwrite: [_] yes [_] no [_] ask" option before running.  And, in case no, report "skipped".

-> Drag'n'drop mode: A checkbox for "add" (vs "replace"), and the opportunity to mark and delete.

CUETools versions 1.9.5 through 2.1.5 (current)

Reply #776
Just found my old test notes, I had completely forgotten about them. Now I remember that I promised to report back some bugs I had found (+just found lots of new ones). Well, better late than never. Sorry about that.

Some of these points are already reported/requested by me (no answers), and there might be some things that others have reported/asked, sorry if there's duplicates.

Bugs/problems/usability/etc.: (hopefully this is readable, I split long lines)
Code: [Select]
1. Links in the about dialog doesn't work.

2. "Formats", "Decoders" and "Scripts" setting tabs doesn't have add/remove buttons like the "Encoders"
setting tab has.

3. Main window profiles dropdown menu: changing the profile doesn't change the "basket" icon to the
selected profile.. though the dropdown button shows the basket beside the selected profile name.

4. "Only if found" selected saving bug.

1) Clear CUETools appdata folder, i.e. start with "clean" settings before opening CUETools.
2) Start app and select e.g. "my music" folder so that you can access the "Action" settings in the main
window.
3) Select "Verify" and "only if found"
4) Open settings and click OK

= "only if found" reverted back to "default".

5. It's a bit confusing what is saved to a profile in the main window. The "Extra" settings "pregap" and
"data track" settings are global (gets changed to every profile) while "offset" is saved per profile. (Ok
ok, "pregap" and "data track" settings are in fact "session" only settings, no point of saving them, right?)

Same thing with the "audio output" settings; the 1st and 2nd dropdown selection boxes and "slider" settings
are global though the "lossless/lossy/hybrid/none" setting is per profile.

Also, "correct filenames" "mode" settings are global.

6. Why isn't the "format" (1st) dropdown selection box nor "compression slider" not disabled when "none" is
selected in the "audio output" section?

6.1. "Unhandled exception" bug when changing the "compression slider" setting while "none" selected in the
"audio output" section.

7. Is the small file browser "placeholder" seen in the far left designed to be that way? There's a very
small scroll button(s) seen in the bottom of it and one can select folder/files even though the browser is
"hidden".

8. "Data track" field disabled bug.

1) Clear CUETools appdata folder, i.e. start with "clean" settings before opening CUETools.
2) Start app and select e.g. "my music" folder so that you can access the "Action" settings in the main
window.
3) Change "audio output" to "none"
Side note: notice how "embedded" in the "mode" gets disabled. Still selected.. possible problem if
not noticed and changed before clicking "Go"?
4) Select "verify" in the "action" and back to "encode"
-> "Data track" gets disabled.
5) Select "lossless"

= "Data track" still disabled. Different types of main window "setting changes"/selections either does
or doesn't enable the "Data track" field, it seems that there's no logic to it at all.

9. "Empty" profile bug.

1) Clear CUETools appdata folder, i.e. start with "clean" settings before opening CUETools.
2) Delete one by one all profiles (except the default, impossible to do that).
-> profile dropdown lists only default profile.
3) Close app and start it again.

= Profile dropdown lists default profile AND an empty profile with no name. Happens because the empty "Profiles="
setting in the "settings.txt" is parsed incorrectly?

4) Select the "empty" profile and Close app.
-> A "profile-.txt" file is created in the appdata folder for the empty profile.

10. ReadMe.txt old/incorrect filename and path formatting variables.

1) Clear CUETools appdata folder, i.e. start with "clean" settings before opening CUETools.
2) Set the "logname format" to "%F.accurip" (without quotes).
-> that setting is changed in the settings.txt after OK clicked.
3) Do verify operation.

= The file created is "%F.accurip" (without quotes).

4) Close app and restart.
5) Open settings dialog.
-> Notice how the "logname format" is now "%filename%.accurip" (without quotes).
6) Click OK.
-> Only after this, that setting is changed in the settings.txt.
7) Open settings dialog.
8) Set the "logname format" to "%Y.accurip" (without quotes).
-> that setting is changed in the settings.txt after OK clicked.
9) Close app and restart.
10) Open settings dialog.
-> Notice how the "logname format" is still "%Y.accurip" (without quotes).

So, it seems that only the old "%F" is detected as a "failsafe" (updating from previous versions), AND
only when opening the settings dialog.

11. "Verify" profile used when running "CUETools.exe /verify *.cue" + other oddness.

Prerequisite: a rip of an album that is not present in the AR database.

1) Clear CUETools appdata folder, i.e. start with "clean" settings before opening CUETools.
2) Run CUETools.exe /verify "filename.cue"

= "Error: Object reference not set to an instance of an object." Error when no appdata folder
settings.txt file present?

3) Open and close app.
-> "settings.txt" created, no settings file for the "Verify" profile created.
4) Run CUETools.exe /verify "filename.cue"

= Again, "Error: Object reference not set to an instance of an object.".

5) Open app and settings, enable the "In source folder" under "Write AccurateRip log" setting, click OK
and close app.
6) Run CUETools.exe /verify "filename.cue"

= No errors, CUETools makes a "only if found" AR verification. Notice that there's no "Verify" profile
file at all!! And the "settings.txt" (=default profile + some general settings (which are not in other
profile files)) file has both "DefaultVerifyScript" and "Script" set to "default", which means that
CUETools has some "internal" default settings, at least, for the "Verify" profile.

7) Open app, change to "verify" profile and back to "default", close app.

= Now, for some reason, "settings.txt" has "DefaultVerifyScript=only if found" (no change to "Script"
setting). The created new "profile-verify.txt" file has "DefaultVerifyScript=default" and "Script=only
if found". So, suddenly default profile "inherited" the "DefaultVerifyScript" setting from the "verify"
profile "Script" setting!?

8) Open app, check "action->verify: used script" from both "default" & "verify" profiles.

= Both profiles have "only if found" scripts set with the "verify" action! Look at the point 7 result
text, the 2 settings are different, why is both profiles using "only if found" scripts?!

9) Close app.

= Oddness continues: now the "settings.txt" profile "Script" setting is changed to "only if found"!?

10) Open app, select "verify" profile, delete it and close app.

= "profile-verify.txt" is not deleted though "verify" is removed from the "Profiles" setting in the
"settings.txt".

11) Open app, change the "verify" action script to "default", close app.

12) Run CUETools.exe /verify "filename.cue"

= Still using the "only if found" script when checking!

13) Delete "profile-verify.txt" file from the appdata folder.
14) Run CUETools.exe /verify "filename.cue"

= Still using the "only if found" script when checking! Again, "internal" default settings?

15) Open app and settings, disable the "In source folder" under "Write AccurateRip log" setting, click
OK and close app.
16) Run CUETools.exe /verify "filename.cue"

= Again, "Error: Object reference not set to an instance of an object.". So, settings in fact read from
default profile file and from "internal" default settings?


Bug/problem summary from this point only:

1. "CUETools.exe /verify" doesn't work without settings.txt file and when "In source folder"
setting is set to off.

2. Using of "internal" profile default settings.. + using/mixing those settings with settings that
ARE saved to a file (e.g. "ArLogToSourceFolder" read from default profile, "only if found" script
setting read from "internal" default setting AND/OR from saved "verify" profile (ignoring the
"ArLogToSourceFolder" from that same profile)). Messy, eh?

3. "DefaultVerifyScript" and "Script" settings jumping across profiles & across those settings.

4. Profile files not deleted when deleting profiles from the GUI.
Maybe the profile/settings system needs a rethink/redesign? I'm really sorry to say this but, nicely put, currently it's a big mess (judging from black-box testing). There might be more bugs lurking because of the profile/settings design. Points 4, 5, 9, 10 & 11 attribute to this concern.


Feature reqs/misc.:

12. GUI beautification(?): CUETools started with /verify and filename -> window has no "proper" title bar (not really needed I guess) and shows "default" VS icon in the taskbar.

13. GUI beautification(?): "Tagging" settings tab: "Album art", "resize if resolution >" text field goes slightly over the above input field.

14. "Set read-only attribute to AR log file" feature in the options? Or how to amend the "only if found" script to do it?

14.1. Is there a manual of using the scripting feature? Kind of pointless to offer a script editing/adding feature if there's no language syntax nor method names to call for documented?

15. Include the version of CUETools in the log. IMO, it's always a good practice to log the app name+version.


Questions/notes:

16. Just noticed that there's an update to CUETools, the 2.0.4a version. I've been using 2.0.4 for some time.. with the annoying "logname setting not saved" bug (I wrote this part first.. all bugs/etc. listed in this post have been tested with the latest 2.0.4a version). Completely missed that ("silent") update. And "check for updates" in the app failed to notify me.  Version value problems? (from the "2.0.4a" version package: CUETools.exe: 2.0.3.1, settings.txt: Version=203 and motd.txt: CUETools 2.0.4) Also there's no mention in the changelog that you made the "only if found" script to work with the "CUETools.exe /verify" (=changes to profiles system). Found it by accident just now when testing the new bugs I found. Maybe other people don't like to read changelogs but I do. I don't want to guess what's changed, or being surprised by them (especially if that happens later after they have already affected my "process").

17. What's the situation with CUETools? Are you still developing it? IIRC you wrote that you have moved on to other projects and you see CUETools ready? IMHO it's weird that the 1st post mentions that "Current stable version: 1.9.5a", and the last version changelog says "Experimental profiles system". Is the 2.0.4a still an unstable/experimental version? Also, maybe an installer to finish off the app? Manual (online)? E.g. the "%filename%['('%unique%')'].accurip" should be documented (picked that out from this topic). IMHO, lots of polish needed to call this app ready.

18. Does CUETools use the local AR "cache" (it's sometimes a stale cache) in "*\Application Data\AccurateRip\AccurateRipCache"?

19. It seems that "ArCueDotNet.exe" does not/can't use any profiles, is this true?



P.S. Congrats on the very high number of downloads, CUETools is a success.

P.P.S. Sorry about this "bomb".. I wouldn't have done this much free work for a program I don't like. (Here's hoping that it doesn't go to waste.)

CUETools versions 1.9.5 through 2.1.5 (current)

Reply #777
-> Is the CRC32 "W/O NULL" done by setting offset for the entire disc so that it starts at first one?
IOW, if I want to scan my library for CDs that are identical modulo offset, then -- at least if there are sufficiently many zeroes in the beginning or at the end, and no data tracks -- it is necessary and sufficient that all these match?
Edit: And is the first one a CRC for the entire image, with the above holding true?


Hmmm ... might it be that the range of offsets should have been bigger?

What is this variable called, and what do I need to compile the source code? (As mentioned before, it crashes on my computer, but maybe a locally compiled .exe will run?)

CUETools versions 1.9.5 through 2.1.5 (current)

Reply #778
-> Some discs are allegedly not present in the database, even if they at time of ripping got confidence levels in the thirties. (dBpoweramp used, no single tracks needed re-ripping.)

Sounds a bit familiar, have you seen this topic? Related?

CUETools versions 1.9.5 through 2.1.5 (current)

Reply #779
Sounds a bit familiar, have you seen this topic? Related?

Interesting. I didn't realize the offset field in CUETools worked that way, so having tried Gregory's advise, I see that too small offset is not the issue.

PS: Sweet avatar

CUETools versions 1.9.5 through 2.1.5 (current)

Reply #780
Can someone help me with some scripting that would get CUETools to move complete folders based on whether they were Accurately Ripped, not Accurately Ripped, or not in database?

Right now I can sort of achieve this for the CDs that were Accurately Ripped by re-encoding to a different folder, but the accompanying files (cover scans, etc.) don't move along.

you might use foobar2000 for this
CUETools check the rips and write corresponding tags to the files
Then you open these files in fb2k, sort them according to these tags and finally move each category using a different method for each
I use the tag and move features of fb2k a lot, it is very powerful and works perfectly even on Linux through Wine


Brilliant! I can't believe I never thought of this. I use foobar2000 religiously, but very early on I turned off all the tags that CueTools adds and never remembered that they are there.

Thanks!

CUETools versions 1.9.5 through 2.1.5 (current)

Reply #781
There will be at least one more release in the nearest future, incorporating Flacuda/Flake/ALACEnc.
Hi Gregory, any chance of a build with ALACEnc ? I have been using the latest version from sourceforge and Flake and ALAC work very well, but as I could not get the C part to compile I cannot process APE files (must decompress them to WAV first). Cheers.

CUETools versions 1.9.5 through 2.1.5 (current)

Reply #782
As soon as i can. A bit overwhelmed at work right now.
CUETools 2.1.6

CUETools versions 1.9.5 through 2.1.5 (current)

Reply #783
As soon as i can. A bit overwhelmed at work right now.
The APE file is not a blocking issue, thanks for all your work so far. Cheers.

CUETools versions 1.9.5 through 2.1.5 (current)

Reply #784
Is there a way with fb2k & the accuraterip tags cuetools makes, to tell if a rip is not accurate? You know similar to how it says it in the batch log "rip not accurate"?

CUETools versions 1.9.5 through 2.1.5 (current)

Reply #785
Seems that CUETools refuses to generate cuesheets for single-track albums (or query AccurateRip).
Is there any setting I might have screwed up, or is it a bug?

CUETools versions 1.9.5 through 2.1.5 (current)

Reply #786
Seems that CUETools refuses to generate cuesheets for single-track albums...

I reported that problem when the v. 1.95 was new: http://www.hydrogenaudio.org/forums/index....st&p=618837

Gregory fixed it in 1.95a: http://www.hydrogenaudio.org/forums/index....st&p=620986

I have not tried single-track albums since then (I have only two such CDs), but could the problem have sneaked back in?


Or are you tying to create a cue sheet for a single audio file that doesn't already have a cue sheet? That is not possible because the folder must contain at least two files in the same format before the "album" appears in CUETools' internal browser.

CUETools versions 1.9.5 through 2.1.5 (current)

Reply #787
are you tying to create a cue sheet for a single audio file that doesn't already have a cue sheet? That is not possible because the folder must contain at least two files in the same format before the "album" appears in CUETools' internal browser.


Yes. I drag and drop the folder. No matter whether I try to create cuesheets or verify, nothing happens.

The use of a cuesheet for a single track album, is to have only .cue files in my media player library, slashing a zero from a file count of some 70 K.
(My albums are organized as one folder per physical CD.)



Using 2.04a.

CUETools versions 1.9.5 through 2.1.5 (current)

Reply #788
Heello
After i have installed .Net Framework 2.0 & that C++ shit i received another bug message

See the picture below :



Is this program really working ???


CUETools versions 1.9.5 through 2.1.5 (current)

Reply #790
I tried first the latest versiion but it gave me the same mistake.
& i tried again with 1.9.5  , because it writes its the latest stable versiion
but the problem remains...
some opinions whats the reason ?

CUETools versions 1.9.5 through 2.1.5 (current)

Reply #791
A quote from the above link:
Quote
IMPORTANT: .NET Framework 2.0 (SP2) required.

Did you install the SP2 update?
What is your Windows version?

CUETools versions 1.9.5 through 2.1.5 (current)

Reply #792
I don't think i've seen such error before. Please, give more details. When exactly does it appear, etc.
CUETools 2.1.6

CUETools versions 1.9.5 through 2.1.5 (current)

Reply #793
Yes. I drag and drop the folder. No matter whether I try to create cuesheets or verify, nothing happens.

CUETools should be able to process cue sheets for one-track albums, at least i test it on Mike Oldfield's Amarok
It cannot automatically create cue sheets for a single file, but you can create it manually, it's very easy: all one-track cue sheets (minus pregap) are the same:
Code: [Select]
FILE "Image.wav" WAVE
  TRACK 01 AUDIO
    INDEX 01 00:00:00
CUETools 2.1.6

CUETools versions 1.9.5 through 2.1.5 (current)

Reply #794
Windows Service Pack 2 needed   
I have neither SP1 nor SP2
.Net Framework 2.0 needed
some C++ needed
Sp1 needed
sp2 needed
Imho these are to many limitations , that require to be accomplished  in order to work correctly..
no thanks , for cue-ripping i will switch to Foobar 2000...

CUETools versions 1.9.5 through 2.1.5 (current)

Reply #795
Gosh. Well, i'm sorry, but i just can't support 10 years old operating systems that never have been updated. Updating to a latest service pack is always a good idea, even if you don't need CUETools
CUETools 2.1.6

CUETools versions 1.9.5 through 2.1.5 (current)

Reply #796
OK, seems like some moderator deletes wishlist entries (s)he disagrees with 


Anyway, I wish for better features to (I) retro-check rips, and (II) pairwise-compare rips. Looks like if there are non-null samples at the very end of a live recording, then the CRCs of two different pressings not found in AccurateRip, will all be different.


So for topic (II), what about:
- When two folders are dragged and dropped (could be restricted to the case where there are exactly two there!), what about checking how far they agree, starting from first nonzero sample? That would reveal duplicates not found in AccurateRip, as well as the case where one release has one or more bonus tracks (numbers > N), and the other CD consists of the first N tracks, bit-identical modulo offset.
- When reporting the results of this, report whether they are identical up to and including track N with or without the end-of-CD truncation that is done by AccurateRip-CRCs.


For topic (I):
- There are cases where the appropriate disc ID is known, for example does dBpoweramp write it to tag (I don't know if EAC does, but it might be?). So, the option to read disc ID from tag, or maybe even from discid.txt in the folder to be checked, would certainly help. (In the presence of data tracks might be some issues concerning whether the audio starts at track 1 or 2 though -- that could be checked by just comparing to the AR-CRC for track n+1 provided the first track has number 1.)


And finally, relevant to both (I) and (II):
- If the number of folders is two, scan either with its CUETools-calculated TOC [or ID read from files], and whenever one is failed to verify as accurate, then scan with the other's TOC [or ID read from files].
This covers cases where two CDs differ only in track separation. There seem to be a few such cases where there are no sign of remastring from analogue tapes, but where the knife between two tracks is set slightly different.


Edit: And thanks for the info on single track discs. I can of course create it manually yes, and maybe I am the only one who doesn't even know how many single-track discs I actually have in my collection?

CUETools versions 1.9.5 through 2.1.5 (current)

Reply #797
OK, seems like some moderator deletes wishlist entries (s)he disagrees with 

No, just that "some moderator" split it off because it was going too far off-topic but failed to provide a link.

CUETools versions 1.9.5 through 2.1.5 (current)

Reply #798
OK, seems like some moderator deletes wishlist entries (s)he disagrees with 

No, just that "some moderator" split it off because it was going too far off-topic but failed to provide a link.

OK; in that case, dare I reiterate the wishlist item at http://www.hydrogenaudio.org/forums/index....st&p=683133 ? Need not be in log file (if log file is set to mimic EAC actions), might as well be a pop-up if precisely two folders are dragged-and-dropped.

This is not about what EAC does and doesn't, it is about this application (edit: CUETools, that is) as of now being my most efficient way of finding two folders whose audio matches modulo offset. Scan a library, scan accurip files pipe grep the "--" string, extract the two image CRCs; if both match (well they probably don't, as I have removed a few of them before starting to use CUETools), then definitely equal, if the latter matches, then well, probably still equal (but have a look).

CUETools versions 1.9.5 through 2.1.5 (current)

Reply #799
Yes. I drag and drop the folder. No matter whether I try to create cuesheets or verify, nothing happens.

CUETools should be able to process cue sheets for one-track albums, at least i test it on Mike Oldfield's Amarok
It cannot automatically create cue sheets for a single file, but you can create it manually, it's very easy: all one-track cue sheets (minus pregap) are the same:
Code: [Select]
FILE "Image.wav" WAVE
  TRACK 01 AUDIO
    INDEX 01 00:00:00



If it is just for creating cue files based on directory contents the tool cuecreator is very handy as well.

-
aleg