Concatenate
Re: Concatenate
Hi emcodem
I'm testing the same trick i.e. modify s_source variable using the function $replace('%s_source%','\\?\UNC','[') - I have added it into the first Populate variables node of your wf. But foolowing node still works with the original s_source with \\?\UNC.
I'm missing something very stupid, but I don't know what it is
I'm testing the same trick i.e. modify s_source variable using the function $replace('%s_source%','\\?\UNC','[') - I have added it into the first Populate variables node of your wf. But foolowing node still works with the original s_source with \\?\UNC.
I'm missing something very stupid, but I don't know what it is
- Attachments
-
- emcodem_webuisubmit_concat_correctUNC.json
- (7.36 KiB) Downloaded 183 times
-
- modify s_source var.PNG (22.95 KiB) Viewed 3288 times
BR,
Silicon
--------
FFAStrans 1.3.0.2; WebInterface 1.3.0.0
Manager: VM: 2x Xeon E5-2630v3@2.4GHz, 8GB RAM
Workers: 3x HP DL360 G9 (2x Xeon E5-2643v3@3.4GHz,16GB RAM, nVidia M2000)+ 2x Lenovo SR665 (2x AMD EPYC730216C@3.0GHz,128GB RAM, nVidia P2200)
Silicon
--------
FFAStrans 1.3.0.2; WebInterface 1.3.0.0
Manager: VM: 2x Xeon E5-2630v3@2.4GHz, 8GB RAM
Workers: 3x HP DL360 G9 (2x Xeon E5-2643v3@3.4GHz,16GB RAM, nVidia M2000)+ 2x Lenovo SR665 (2x AMD EPYC730216C@3.0GHz,128GB RAM, nVidia P2200)
Re: Concatenate
OK, I have found a workarround, but still ... grrr
BR,
Silicon
--------
FFAStrans 1.3.0.2; WebInterface 1.3.0.0
Manager: VM: 2x Xeon E5-2630v3@2.4GHz, 8GB RAM
Workers: 3x HP DL360 G9 (2x Xeon E5-2643v3@3.4GHz,16GB RAM, nVidia M2000)+ 2x Lenovo SR665 (2x AMD EPYC730216C@3.0GHz,128GB RAM, nVidia P2200)
Silicon
--------
FFAStrans 1.3.0.2; WebInterface 1.3.0.0
Manager: VM: 2x Xeon E5-2630v3@2.4GHz, 8GB RAM
Workers: 3x HP DL360 G9 (2x Xeon E5-2643v3@3.4GHz,16GB RAM, nVidia M2000)+ 2x Lenovo SR665 (2x AMD EPYC730216C@3.0GHz,128GB RAM, nVidia P2200)
Re: Concatenate
Just to explain.... I'm mad about me not being able to make it work as required
BR,
Silicon
--------
FFAStrans 1.3.0.2; WebInterface 1.3.0.0
Manager: VM: 2x Xeon E5-2630v3@2.4GHz, 8GB RAM
Workers: 3x HP DL360 G9 (2x Xeon E5-2643v3@3.4GHz,16GB RAM, nVidia M2000)+ 2x Lenovo SR665 (2x AMD EPYC730216C@3.0GHz,128GB RAM, nVidia P2200)
Silicon
--------
FFAStrans 1.3.0.2; WebInterface 1.3.0.0
Manager: VM: 2x Xeon E5-2630v3@2.4GHz, 8GB RAM
Workers: 3x HP DL360 G9 (2x Xeon E5-2643v3@3.4GHz,16GB RAM, nVidia M2000)+ 2x Lenovo SR665 (2x AMD EPYC730216C@3.0GHz,128GB RAM, nVidia P2200)
Re: Concatenate
Hey @Silicon,
check out my updated workflow, it is using a new variable instead of s_source.
When you do
%s_source% = $replace('%s_source%','\\?\UNC','[')
Then ffastrans will immediately redo the changes internally.
Instead you need to drop the usage of s_source throughout the whole workflow and replace it by some user variable e.g.
%s_sanitized_source% = $replace('%s_source%','\\?\UNC','[')
From this moment, s_sanitized_source contains the interesting data and s_source should not be used anymore.
I wrote this in the post where i suggest the replace:
check out my updated workflow, it is using a new variable instead of s_source.
When you do
%s_source% = $replace('%s_source%','\\?\UNC','[')
Then ffastrans will immediately redo the changes internally.
Instead you need to drop the usage of s_source throughout the whole workflow and replace it by some user variable e.g.
%s_sanitized_source% = $replace('%s_source%','\\?\UNC','[')
From this moment, s_sanitized_source contains the interesting data and s_source should not be used anymore.
I wrote this in the post where i suggest the replace:
Sometimes the stuff that i write contains vital information without highlighting it, sorry for that.in another variable than s_source because ffastrans would immediately redo the change.
emcodem, wrapping since 2009 you got the rhyme?
Re: Concatenate
Hi @ecmodem
I've missed this cruccial info that one have to use another variable than s_source because ffastrans would immediately redo the change
Thanks a lot indeed
That's exactly the workarround I've applied - use a new variable instead of s_source. So in fact it is not a workarroud but a proper way ... OKemcodem wrote: ↑Mon Nov 14, 2022 8:06 am When you do
%s_source% = $replace('%s_source%','\\?\UNC','[')
Then ffastrans will immediately redo the changes internally.
Instead you need to drop the usage of s_source throughout the whole workflow and replace it by some user variable e.g.
%s_sanitized_source% = $replace('%s_source%','\\?\UNC','[')
I've missed this cruccial info that one have to use another variable than s_source because ffastrans would immediately redo the change
Thanks a lot indeed
BR,
Silicon
--------
FFAStrans 1.3.0.2; WebInterface 1.3.0.0
Manager: VM: 2x Xeon E5-2630v3@2.4GHz, 8GB RAM
Workers: 3x HP DL360 G9 (2x Xeon E5-2643v3@3.4GHz,16GB RAM, nVidia M2000)+ 2x Lenovo SR665 (2x AMD EPYC730216C@3.0GHz,128GB RAM, nVidia P2200)
Silicon
--------
FFAStrans 1.3.0.2; WebInterface 1.3.0.0
Manager: VM: 2x Xeon E5-2630v3@2.4GHz, 8GB RAM
Workers: 3x HP DL360 G9 (2x Xeon E5-2643v3@3.4GHz,16GB RAM, nVidia M2000)+ 2x Lenovo SR665 (2x AMD EPYC730216C@3.0GHz,128GB RAM, nVidia P2200)
Re: Concatenate
Well all of it is a workaround until the issue is fixed in a new ffastrans version
Basically i could also hide the error by not using s_source for array job submission from webinterface but the only alternative is to use a user_variable but it would have to be "created" by the user before and i would not know an easy way how to tell webinterface in which userinput the filelist goes...
Basically i could also hide the error by not using s_source for array job submission from webinterface but the only alternative is to use a user_variable but it would have to be "created" by the user before and i would not know an easy way how to tell webinterface in which userinput the filelist goes...
emcodem, wrapping since 2009 you got the rhyme?
Re: Concatenate
It's fine emcodem, II have reached my point.
You should probably consider changing the text on help page https://ffastrans.com/wiki/doku.php?id= ... _variables saying %s_source% represent the input file for any processor. It can be altered using the “Populate variables”-node ... in fact it can't be altered
Anyway thanks again for this new feature!
You should probably consider changing the text on help page https://ffastrans.com/wiki/doku.php?id= ... _variables saying %s_source% represent the input file for any processor. It can be altered using the “Populate variables”-node ... in fact it can't be altered
Anyway thanks again for this new feature!
BR,
Silicon
--------
FFAStrans 1.3.0.2; WebInterface 1.3.0.0
Manager: VM: 2x Xeon E5-2630v3@2.4GHz, 8GB RAM
Workers: 3x HP DL360 G9 (2x Xeon E5-2643v3@3.4GHz,16GB RAM, nVidia M2000)+ 2x Lenovo SR665 (2x AMD EPYC730216C@3.0GHz,128GB RAM, nVidia P2200)
Silicon
--------
FFAStrans 1.3.0.2; WebInterface 1.3.0.0
Manager: VM: 2x Xeon E5-2630v3@2.4GHz, 8GB RAM
Workers: 3x HP DL360 G9 (2x Xeon E5-2643v3@3.4GHz,16GB RAM, nVidia M2000)+ 2x Lenovo SR665 (2x AMD EPYC730216C@3.0GHz,128GB RAM, nVidia P2200)
Re: Concatenate
@Silicon s_source can be altered to your desire at any point in the workflow. In this case, we just found out in ffastrans Version <= 1.3:
If the value is longer than 256 characters and these chars contain \\ or X:, ffastrans will replace the first character to \\?\UNC
That should be the exact description of what happens internally to s_source in our case here.
Looking forward to hear if your concat usecase finally worked I believe we need to work on the sorting because it should be sorted exactly as the user sorted it on webui but the files find will sort the transcoded files by name actually...
If the value is longer than 256 characters and these chars contain \\ or X:, ffastrans will replace the first character to \\?\UNC
That should be the exact description of what happens internally to s_source in our case here.
Looking forward to hear if your concat usecase finally worked I believe we need to work on the sorting because it should be sorted exactly as the user sorted it on webui but the files find will sort the transcoded files by name actually...
emcodem, wrapping since 2009 you got the rhyme?
Re: Concatenate
@ecmodem. Thank you explanation.
As for your question: for my usecase the sorting of source files based on filnames is OK, because I need to merge individal files originated from XDCAM disc or SxS card, where the naming convention is "sequential". But you are right that another usecase could require user-defined sorting.
Another topic to think about: should the user be advised then his list of files contains mixed resolution (SD and HD) or even mixed framerates?
BTW You've mentioned Find Files plugin: When I set Include filter to *.mxf,*.mp4 the main branch of workflow fails, becasue Find Files is not able to create Concat file properly.
So the question follows: how to modify the workflow properly to proccess just MXF and MP4 files (because BFU will certianly submit also the sidecar XML files )
As for your question: for my usecase the sorting of source files based on filnames is OK, because I need to merge individal files originated from XDCAM disc or SxS card, where the naming convention is "sequential". But you are right that another usecase could require user-defined sorting.
Another topic to think about: should the user be advised then his list of files contains mixed resolution (SD and HD) or even mixed framerates?
BTW You've mentioned Find Files plugin: When I set Include filter to *.mxf,*.mp4 the main branch of workflow fails, becasue Find Files is not able to create Concat file properly.
So the question follows: how to modify the workflow properly to proccess just MXF and MP4 files (because BFU will certianly submit also the sidecar XML files )
BR,
Silicon
--------
FFAStrans 1.3.0.2; WebInterface 1.3.0.0
Manager: VM: 2x Xeon E5-2630v3@2.4GHz, 8GB RAM
Workers: 3x HP DL360 G9 (2x Xeon E5-2643v3@3.4GHz,16GB RAM, nVidia M2000)+ 2x Lenovo SR665 (2x AMD EPYC730216C@3.0GHz,128GB RAM, nVidia P2200)
Silicon
--------
FFAStrans 1.3.0.2; WebInterface 1.3.0.0
Manager: VM: 2x Xeon E5-2630v3@2.4GHz, 8GB RAM
Workers: 3x HP DL360 G9 (2x Xeon E5-2643v3@3.4GHz,16GB RAM, nVidia M2000)+ 2x Lenovo SR665 (2x AMD EPYC730216C@3.0GHz,128GB RAM, nVidia P2200)
Re: Concatenate
@Silicon
You are right, the include filter was broken when used with multiple include patterns, i fixed that now and uploaded a new version (also added a size filter)
https://ffastrans.com/wiki/doku.php?id= ... processors
But in the example workflow "emcodem_webuisubmit_concat_different_formats", it does actually not really make sense to set the files find to multiple extensions. It only lists the files that the lower branch transcoded so there is no way you need to include multiple extensions. Not to mention that multiple extensions would imply multiple different formats which would automatically lead to fubar output file at the concat.
About the topic "warning the user about mixed framerates and such", i suspect usually users are not interested in format specific warnings. Also, the "different_formats" workflow examples that i posted here do not care about format specifics, they just transcode everything. If you want to change that, you could for example use the "transwrap when possible" option in the xdcam encoder or add a decision and a second branch in the "per file" branch.
You know what i mean?
I'm interested in your xdcam/sxs workflow, but let's talk about that in PM's so we don't spam this concat thread with it
You are right, the include filter was broken when used with multiple include patterns, i fixed that now and uploaded a new version (also added a size filter)
https://ffastrans.com/wiki/doku.php?id= ... processors
But in the example workflow "emcodem_webuisubmit_concat_different_formats", it does actually not really make sense to set the files find to multiple extensions. It only lists the files that the lower branch transcoded so there is no way you need to include multiple extensions. Not to mention that multiple extensions would imply multiple different formats which would automatically lead to fubar output file at the concat.
About the topic "warning the user about mixed framerates and such", i suspect usually users are not interested in format specific warnings. Also, the "different_formats" workflow examples that i posted here do not care about format specifics, they just transcode everything. If you want to change that, you could for example use the "transwrap when possible" option in the xdcam encoder or add a decision and a second branch in the "per file" branch.
You know what i mean?
I'm interested in your xdcam/sxs workflow, but let's talk about that in PM's so we don't spam this concat thread with it
emcodem, wrapping since 2009 you got the rhyme?