Beta [webif] Web Interface 1.4.x

[af123]

I was unsure whether to report this as I cannot replicate but...

It's a useful report, thanks.
I don't know why the Enc flag wasn't automatically removed at the end of the recording (it can sometimes happen with chase play but that doesn't seem to be the case here). The next reboot fixed the problem as auto-unprotect does a full disk scan at boot.

With the new Auto structure it would probably worth retrying unprotect from within Decrypt processing if it hasn't happened by the first retry

Good idea. In the case where it's going to use a DLNA helper to do the decryption it doesn't even need to care about the Enc flag, otherwise it needs to unset and fix-up the DLNA database, both of which will be easy enough to do automatically..

Edit: I have a working solution in place for the release version, thanks again.

 

[MymsMan]

Tks MymsMan - nothing in the recmon.log since 17/01/17 at 17:05 as below:

4 at file "/mod/etc/recmon.d/rslog", line 16
3 /mod/etc/recmon.d/rslog:16: Error: invalid command name "0"
2 at file "/mod/etc/recmon.d/webiflog", line 16
1 /mod/etc/recmon.d/webiflog:16: Error: invalid command name "0"

I suspect you wont find a 'Recorded' message for that program in activity.log - 17:05 is close to but a bit later than I would expect for the end of A place in the sun.

For some reason the recording files were not properly accessible so the webif couldn't create a ts object for the recmon processes to use.
Some like auto-unprotect fail quietly because they contain a check

if {$ts eq "0"} exit

whilst others like webiflog and rslog don't have a check and generate the invalid command name message when they first attempt to use it

What is strange that the files were accessible later in the decrypt process - it doesn't appear that the recording has been moved or renamed.

 

[af123]

With pleasure - see attached pdf

Thanks. I've found the problem and it will be fixed for the release version. You're right, it's due to the services changing between the backup and the retune.

 

[peterworks]

17:05 is close to but a bit later than I would

I do have 5 minutes padding at the end of all my recordings.

it doesn't appear that the recording has been moved or renamed

Sorry I didn't mention it but it was moved to a folder - it was the first action in the process before the 'trouble' occurred.

 

[peterworks]

I have found the the Queue page is now my first port of call when logging on as it provides a 'snapshot' of how the auto-process has performed.
Would it be possible to have an option to have this on the front page ? Also is there a 'magic' extra.css line to shrink the icons as per those in the Diag page please ?

 

[MymsMan]

I do have 5 minutes padding at the end of all my recordings.


Sorry I didn't mention it but it was moved to a folder - it was the first action in the process before the 'trouble' occurred.

So the hypothesis is that auto-trigger started running before auto-unprotect and moved the recording before auto-unprotect (and other tasks) had a chance to do their thing.

Renaming autotrigger to z-autotrigger and/or introducing a short sleep before auto processing might reduce the likelihood of the problem occurring

 

[jaybr]

I have found a possible bug in Flexview

When trying to download a video file via opt+ I get error 500, the webif-error.txt is as follows:-

1 /mod/webif/html/browse/download.jim:20: Error: invalid command name "0"
2 at file "/mod/webif/html/browse/download.jim", line 20

However if I select an mp3 file then download works.

Hopefully the fix mentioned will cover this issue as well.

 

[MymsMan]

I have found a possible bug in Flexview

When trying to download a video file via opt+ I get error 500, the webif-error.txt is as follows:-

1 /mod/webif/html/browse/download.jim:20: Error: invalid command name "0"
2 at file "/mod/webif/html/browse/download.jim", line 20

However if I select an mp3 file then download works.

Hopefully the fix mentioned will cover this issue as well.

FlexView has not yet been updated to support the Beta but that should affect downloading

Can you successfully download the file via the standard webif Browse menus - this would help narrow down whether the problem is within FlexView or in the common function that is shared with Browse.

 

[jaybr]

Hi.

I have no issues with downloading from the Browse screen.

I have reloaded the package in case there was a glitch on install.

 

[peterworks]

So the hypothesis is that auto-trigger started running before auto-unprotect and moved the recording before auto-unprotect (and other tasks) had a chance to do their thing.

Could this be caused by the insertion of a delay for DLNA in auto-processing ?
PS - I have now replicated it if that helps...

 

[af123]

I have found the the Queue page is now my first port of call when logging on as it provides a 'snapshot' of how the auto-process has performed.
Would it be possible to have an option to have this on the front page ?

Yes, I'll add that to the release version.

Also is there a 'magic' extra.css line to shrink the icons as per those in the Diag page please ?

If I add a class to each image there should be a way to do it in the next version.

 

[af123]

I have found the the Queue page is now my first port of call when logging on as it provides a 'snapshot' of how the auto-process has performed.
Would it be possible to have an option to have this on the front page ? Also is there a 'magic' extra.css line to shrink the icons as per those in the Diag page please ?

Here you go - you will need the beta version I'll be uploading shortly and a couple of lines in EXTRA.css (see below).

img.mmenu
{
    width: 120px !important;
    height: 125px !important;
}

 

[af123]

Webif 1.4.0 second beta now available. Changes since the last one:

• Show correct queue scan interval on queue page instead of 0;
• Option to show queue icon on main menu;
• Main menu icon sizes can be changed through EXTRA.css;
• Main menu icons will be put side-by side up to available window width instead of always wrapping at 4;
• Automatic decryption will no longer fail if a recording has an Enc flag set;
• Fix restoration of favourites.

 

[prpr]

Show correct queue scan interval on queue page instead of 0;

It's very difficult to read because of the lack of contrast between the text (80% grey) and the background (100% white).

 

[af123]

It's very difficult to read because of the lack of contrast between the text (80% grey) and the background (100% white).

Highlight it? I'll make it darker...

 

[peterworks]

Sorry af but restoration of favourites still failing. I did the same process as before and no favourites were restored although the schedule restored okay.
Attached is the backup and restore screen outputs:

 

[peterworks]

Not sure if important but the boot log shows this:



Line 92 shows "1.03.12". Is this the current (non beta) version ? If so, should the beta, if installed, be shown ?

 

[af123]

Sorry af but restoration of favourites still failing. I did the same process as before and no favourites were restored although the schedule restored okay.
Attached is the backup and restore screen outputs:

Did you reboot after the restore? That's when the favourites are added and diagnostic output will appear in xinit.log.

Line 92 shows "1.03.12". Is this the current (non beta) version ? If so, should the beta, if installed, be shown ?

That's the version of the underlying Humax firmware.

 

[peterworks]

Did you reboot after the restore?

Yes I did reboot. The xinit.log for that time is no longer there so will retry tomorrow and, if the same result, capture the log.

 

[peterworks]

The xinit.log for that time is no longer there so will retry tomorrow and, if the same result, capture the log.

Have retried using the same process, ie:
1. Schedule backup
2. Retune
3. Restore backup
4. Restart from WebIf (no favourite groups or favourites restored)
5. Restore backup
6. Restart from remote (no favourite groups or favourites restored)

Attached are xinit logs from after 4. and after 6.