Been hammering at this a little. Not much progress yet but a few initial thoughts and observations...
Sadly its not as simple as the drive being marked as removable - the 1st usb device on the hd, mapped to sda is flagged as non-removable anyway. cat /sys/block/sda/removable gives 0 also.
I...
Just also tried mapping the whole video folder, and that worked fine too, so I think it would be perfectly possible to temporarily (or permanently) map the video folder to an external device. As BH pointed out the recordings would be encrypted, but should be picked up by the DLNA decrypt in...
Just tried a quick experiment of using mount to map the folder that a series record was about start, to a folder on a different drive, and that worked fine. This is on an HD (mapping the usb drive to a network drive), but I suspect it would work on an HDR also.
Tim, there is a new version (1.5) of nicesplice is uploaded, with better memory usage and no file length limit. Let me know if it works with your recording in case if you still have it.
Thanks!
Steve
New version (1.5) of nicesplice is uploaded, with better memory usage and no file length limit. I'll try and take a look fiddle the timestamps for fussy players issue at some point.
Apologies for only just picking up this thread. Magic folders produces a log in tmp/editMonitor.log which should show any problems it encountered. You should be able to view this through the webif (diagnostics page I think).
I notice above it looks like you may have copied the file to crop...
The sidecar files are only used by the humax, and the times in these are all fixed up. I don' t do anything with the timestamps in the ts stream chuncks which are trickier to fix up, and didn't seem to cause problem with most external players, but I can certainly see how it may confuse some is...
Yup - it was just lazy static array. No longer though.
Still better to fix the cause rather than the symptom. Its something I've been meaning look at for ages, as the maximum video size / length limitaions have also caused a few problems over the summer (reading festival, Wimbledon etc)...
I've spent the last few evenings working on a caching system for nicesplice to reduce its memory footprint, and turn it back into a sweeter fruit.
For anyone interested in the technical details, the issue is that it loads the entire nts (the videos time/position index) into memory so it can...
Nice one - it had crossed my mind to try this, mainly to check out the new iplayer, but haven't had the time, and I must admit was mildly worried it might brick it :eek: . I think the reason DLNA doesn't work is more down to the fact I could never persuade it to believe it had an internal hard...
It's been a fair old while since I've looked at the nicesplice code, but I'll try and find some time soon to see if I can recreate this problem. It certainly looks like it's down to the size of the recording - it was quite tricky getting support for files of > 4 gigs in, and there may be some...
I'll have a play and see if I can replicate it, but any more clues appreciated (eg is it only if you remove stuff from the start, is it definitely only HD recordings, does the problem occur if the recording is shrunk beforehand?)
No, I'm afraid I did make a start on that, but never got it working. Should only be a rare problem, as recordings you want to join are nearly always from the same source, but I might look at it again though if there is demand though...
No, it will have just have (almost) 2 minutes of the 2nd channel, linked with the .hmt from the current (ie first) channel.
This won't actually play, as the program/channel ids in the hmt won't match those in the stream. This is why you have to wait at least as long back on the first channel...
This is expected - when you switch channels it starts overwriting the previous buffer, so how much you get depends on how long you were on the other channel.
Yup - obviously make sure you've not left any recordings in there before you do ;)
Worth pointing out as I didn't mention it explicitly above: One of the most useful features it gives to the HD is a limited ability to record 2 channels at once (as long as they are on the same mux), as you can record one, watch another, then grab the tsr. Its good that the 4 main HD channels...
It is possible, but I noticed the word "mac"... If you are using osx Lion or later, smb to the Humax doesn't work, as Apple's new implementation only supports the latest version of the protocol, and samba on the box doesn't.
However you can use the nfs protocol instead - there are some details...
Been hammering at this a little. Not much progress yet but a few initial thoughts and observations...
Sadly its not as simple as the drive being marked as removable - the 1st usb device on the hd, mapped to sda is flagged as non-removable anyway. cat /sys/block/sda/removable gives 0 also.
I...
Just also tried mapping the whole video folder, and that worked fine too, so I think it would be perfectly possible to temporarily (or permanently) map the video folder to an external device. As BH pointed out the recordings would be encrypted, but should be picked up by the DLNA decrypt in...
Just tried a quick experiment of using mount to map the folder that a series record was about start, to a folder on a different drive, and that worked fine. This is on an HD (mapping the usb drive to a network drive), but I suspect it would work on an HDR also.
Tim, there is a new version (1.5) of nicesplice is uploaded, with better memory usage and no file length limit. Let me know if it works with your recording in case if you still have it.
Thanks!
Steve
New version (1.5) of nicesplice is uploaded, with better memory usage and no file length limit. I'll try and take a look fiddle the timestamps for fussy players issue at some point.
Apologies for only just picking up this thread. Magic folders produces a log in tmp/editMonitor.log which should show any problems it encountered. You should be able to view this through the webif (diagnostics page I think).
I notice above it looks like you may have copied the file to crop...
The sidecar files are only used by the humax, and the times in these are all fixed up. I don' t do anything with the timestamps in the ts stream chuncks which are trickier to fix up, and didn't seem to cause problem with most external players, but I can certainly see how it may confuse some is...
Yup - it was just lazy static array. No longer though.
Still better to fix the cause rather than the symptom. Its something I've been meaning look at for ages, as the maximum video size / length limitaions have also caused a few problems over the summer (reading festival, Wimbledon etc)...
I've spent the last few evenings working on a caching system for nicesplice to reduce its memory footprint, and turn it back into a sweeter fruit.
For anyone interested in the technical details, the issue is that it loads the entire nts (the videos time/position index) into memory so it can...
Nice one - it had crossed my mind to try this, mainly to check out the new iplayer, but haven't had the time, and I must admit was mildly worried it might brick it :eek: . I think the reason DLNA doesn't work is more down to the fact I could never persuade it to believe it had an internal hard...
It's been a fair old while since I've looked at the nicesplice code, but I'll try and find some time soon to see if I can recreate this problem. It certainly looks like it's down to the size of the recording - it was quite tricky getting support for files of > 4 gigs in, and there may be some...
I'll have a play and see if I can replicate it, but any more clues appreciated (eg is it only if you remove stuff from the start, is it definitely only HD recordings, does the problem occur if the recording is shrunk beforehand?)
No, I'm afraid I did make a start on that, but never got it working. Should only be a rare problem, as recordings you want to join are nearly always from the same source, but I might look at it again though if there is demand though...
No, it will have just have (almost) 2 minutes of the 2nd channel, linked with the .hmt from the current (ie first) channel.
This won't actually play, as the program/channel ids in the hmt won't match those in the stream. This is why you have to wait at least as long back on the first channel...
This is expected - when you switch channels it starts overwriting the previous buffer, so how much you get depends on how long you were on the other channel.
Yup - obviously make sure you've not left any recordings in there before you do ;)
Worth pointing out as I didn't mention it explicitly above: One of the most useful features it gives to the HD is a limited ability to record 2 channels at once (as long as they are on the same mux), as you can record one, watch another, then grab the tsr. Its good that the 4 main HD channels...
It is possible, but I noticed the word "mac"... If you are using osx Lion or later, smb to the Humax doesn't work, as Apple's new implementation only supports the latest version of the protocol, and samba on the box doesn't.
However you can use the nfs protocol instead - there are some details...
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.