• The forum software that supports hummy.tv has been upgraded to XenForo 2.3!

    Please bear with us as we continue to tweak things, and feel free to post any questions, issues or suggestions in the upgrade thread.

[webif] Web interface

I've been looking at the diagnostics logs (unencrypt in particular), and they would be a lot easier to read if there were a few line breaks in them!

Are the logs created by appending all on one line, or are the appends new lines but the WebIF is presenting them without line breaks?
 
I have a specific HiDef recording, not decrypted yet, which the WebIF media browser is flagging as StDef. What could be the cause?
 
Black Hole said:
I have a specific HiDef recording, not decrypted yet, which the WebIF media browser is flagging as StDef. What could be the cause?
Click to expand...
Most likely a problem reading or parsing the .hmt sidecar file. Strange that it would happen to just one recording though. Does it show with the correct flag in the on-screen media browser?

I'd like a copy of the associated .hmt if you can grab it - it may be that the hmt utility needs a tweak.
 
Thanks for this.
This thread has become long and wide spread, so sorry if this has been asked for before.
Could the De-dup function, add the series name to the end of the file. I can understand that it might get in the way at the beginning but at the end it would be readable in the file browers etc. Of-cause to be able to set file rename rules would be top.
Thanks again.
 
Black Hole said:
I have a specific HiDef recording, not decrypted yet, which the WebIF media browser is flagging as StDef. What could be the cause?
Click to expand...
Now I know the cause - it really is StDef!

False alarm, sorry. :oops:

Next question: with my box doing a lot of decrypting, I expected to see "watching ..." coming up in the media browser - but it's not. Am I wrong?
 
I'm getting odd results when trying to use the Deduplicate/tidy function on a folder containing files that have been cropped. It seems that webif tries to get the proposed filename and the proposed title from the synopsis field and, not surprisingly, decides that it cannot process it. It looks as if the crop process is not leaving the sidecar files in a state that dedup understands. Is this easy to resolve?

dedup.jpg
 
I don't think that's related to them being cropped. The dedup code uses a variety of methods to try and extract a reasonable title from the EPG data and if all the methods fail then it says cannot process.

This is the current logic:
  • Remove 'new series.' from the start of synopsis if present;
  • Remove 'cbeebies.' from the start of synopsis if present;
  • Remove anything after a colon character in the synopsis;
  • If there are more than 40 characters left then:
    • Look for a literal . (dot) and take everything to the left hand side;
    • If the left hand side is less than 6 characters and the right hand side is less than 6 characters then use both sides but separate them with a _;
  • If the synopsis is now less than 10 characters then:
    • If the synopsis starts with the title then use the title;
    • otherwise append the title.
  • If the synopsis is > 40 characters then Cannot Process.
This works really well for children's programmes and things like Family Guy but not so well for your events. What would you suggest it choose as the programme title in your case? It could just take the first 40 characters of the synopsis...
 
Thanks, that's interesting - I hadn't seen that logic before.

I've seen this happen three times: with Whitechapel (ITV HD) and with 2 documentary series on C4 HD - Stephen Hawking's Brave New World and Tony Robinson's Gods and Monsters. Looking at the contents of the synopsis field of these programmes and comparing it to BBC series - all of which dedup renames quite correctly - it seems that ITV and C4 do not use the field in the same structured way as the BBC. Using the current logic, it seems unlikely that dedup will be able to rename many ITV/C4 series correctly. My linking of the problem to cropping is clearly wrong - as I only crop ITV/C4 programmes, these were the only ones where the glitch showed up and I picked on the wrong cause.

I guess there are 3 options, now:
- persuade the errant TV channels to do things properly, like the BBC (unlikely to succeed...)
- accept that some channels do not provide information in a format that dedup can use, or
- find some alternative logic based on the program name and the broadcast date/time in order to construct series information

I can anticipate some of the problems the last of these might throw up! Anyway, I'll start paying more attention to the way non-BBC channels use the synopsis field and see if anything occurs to me.

EDIT: It does occur to me that when used on 'non-compliant' channels there is a definite risk of dedup mis-identifying duplicates (i.e. where programmes within a series all share the same synopsis). Users need to be alert before hitting the 'Process Folder' button. I've never actually done this, but if you process a folder, are files flagged as duplicates automatically deleted?
I'm a little uneasy about doing all this by parsing the synopsis field unless there is some certainty that there actually are rules about how this field is authored, even within the BBC. Could a change of staff result in a change of grammar - semi-colon instead of colon or whatever? Clearly, if there are rules, they apply only to a given broadcaster, not across the industry.
 
The only other option in the Dedup process (granted you'd need the webif rather than just running it in Telnet) is to have something in the view of what Dedup proposes to do to which would allow the user to enter some text of their own? You could still show everything it does now but have another field which is ignored if empty or the contents used if something has been typed in? Don't know how easy that would be but it would help solve Fenlanders problem with the less than useful ITV/C4 data?
 
Yeah. Some sort of global replace function might work, such as you find in decent file managers. FreeCommander does this very well. It would need to be global, otherwise it would offer no advantage over editing each file individually via the Opt+ button in webif. It would be much more complex in programming terms than the present functionality, though, and I doubt that it would be justifiable in cost/benefit terms.
 
I second the idea of a "type your own" pop-up - the only time I have tried dedup it came up with a right dog's breakfast, no doubt for the reasons above.
 
fenlander said:
- find some alternative logic based on the program name and the broadcast date/time in order to construct series information
Click to expand...

I suppose that if all else fails, it could at least put the date and time into the name that's shown in the media title.

Whitechapel 30/1/12 21:00
Whitechapel 6/2/12 21:00

I'm open to suggestions and if someone can describe how a type your own popup would work then I'll look at implementing it.
 
Black Hole said:
I second the idea of a "type your own" pop-up - the only time I have tried dedup it came up with a right dog's breakfast, no doubt for the reasons above.
Click to expand...

I think it could work on selected files within a folder by prompting for:
- Filename
- Display Name (leave blank=same as filename)
- Series number (default=1)
- No. in series (default=no of files selected)
- Start index (default=1)
From this could be constructed filenames like Whitechapel.S02E01.ts (as usenet) and displaynames like Whitechapel Series 2 Part 1/6. The files would presumably be numbered in date/time order, though there would be an issue if episodes had been saved out of sequence for some reason. There would also have to be some checking to avoid duplication of existing filenames.

Of course, this has ceased to be a 'de-duplicator' and become a multi-file renamer.
 
For the type your own popup I had in mind using the existing Dedup proposal grid with the extra column for me to fill in. I would type something that would be used in all the episodes and paste it into the boxes and edit something on the end to make it unique (or perhaps the Dedup process could incrementally number items that were already duplicates?). Granted this would work for a few episodes of a series (i.e. 6 or 8 epsiodes maybe) but not for af123's now legendary Octonaughts collection :)

Now my mind wanders...

We still have some additional columns next to the standard dedup screen proposals (or it might need another screen to make space) but at the top of these we have empty fields titled, lets say, Prog Name & Episode ID to start with. I can enter (using the example above) 'Whitechapel Murders' into the Prog Name, and in episode ID I can select an option to ID each episode. I can choose from an option list of Number or Date. If I choose number I could enter a starting number somewhere (hopefully 1 but in case I missed the start...). (Maybe if we can spot a / character in the description it might imply an episode count so it could look for those to try?) If I choose date then it works on af123's suggestion from date in the the media title. I may need to check off the episodes down the side to say which ones I want to include in the rename or a master button to check all, then I hit 'Next' or whatever we want to call it. The program then uses the info just entered to cycle down the list and name and increment each episode and show me what it will look like. If I like it I click 'Do it' and if I don't I change the stuff at the top and try it again. If I really get myself messed up or bottle it I just come out of the screen and go back in to start again.

I have no idea how feasible that would be but perhaps it sparks some ideas that would fall into the realms of possible.

[EDIT] - thought too much and typed too slow!! Looks like FL had the same sort of idea
 
You must log in or register to reply here.
Back
Top