View Bug Details
ID | Project | Category | View Status | Date Submitted | Last Update |
---|---|---|---|---|---|
0001369 | DCP-o-matic | Bugs | public | 2018-09-18 18:02 | 2021-01-08 00:45 |
Reporter | overlookmotel | Assigned To | carl | ||
Priority | normal | Severity | crash | Reproducibility | always |
Status | closed | Resolution | fixed | ||
Product Version | 2.12.0 | ||||
Summary | 0001369: Error when trying to open a certain DCP in DCP-o-matic | ||||
Description | I have come across what I think is a bug. I was trying to re-wrap a DCP from an external source (made with OpenDCP) but getting an error when try to add DCP to the project. Running via command line:
DCP-o-matic GUI produces same error. The log file reveals nothing more than the same error message. I am not sure what it unusual about this DCP (have successfully run the above command on hundreds of DCPs and this was the first problem). The only thing noticeably odd about it is that the CPL title is very long and contains unusual characters (":" and "+"). However, I edited the DCP XML to remove the strange chars and it didn't solve the errors. So I'm stumped. The DCP passes Carl, I'm emailing you a copy of the DCP. | ||||
Tags | No tags attached. | ||||
Branch | |||||
Estimated weeks required | |||||
Estimated work required | Unknown | ||||
|
This is DCP-o-matic 2.12.10, by the way. |
|
I'm not completely sure it's a bug in DCP-o-matic rather than a fault in the input DCP, but certainly the error message DCP-o-matic gives is a little opaque! |
|
It's an empty <ContentKind> tag. I think the standard says that this is not allowed, but it's not very clear. I think DCP-o-matic should refuse to load this (but obviously give a better error...) |
|
You are quite right! I filled in that tag in the XML as "short" and changed the hash in CPL + ASSETMAP and it passes Thanks so much for investigating so swiftly, and sorry for falsely accusing DOM of being at fault. I'm not sure if DCP-o-matic should refuse to load DCPs which are strictly speaking faulty, but which DOM can process. Personally, I think it might be better if it gave a warning and asked "do you want to proceed?". One use for DOM might be to rewrap faulty DCPs to solve exactly these kinds of problems. |
|
Hash in PKL I mean... |
|
Yup. I guess we'd prefer to categorize errors in critical and formal only, so DCP-o-matic can be used to repair things. While we can usually try to load audio and video MXFs individually, it would still be desirable to load a DCP as far as possible according to it's CPL.
|
Date Modified | Username | Field | Change |
---|---|---|---|
2018-09-18 18:02 | overlookmotel | New Bug | |
2018-09-18 18:03 | overlookmotel | Note Added: 0002686 | |
2018-09-18 18:06 | overlookmotel | Note Added: 0002687 | |
2018-09-18 20:02 | carl | Note Added: 0002688 | |
2018-09-18 21:40 | overlookmotel | Note Added: 0002689 | |
2018-09-18 21:40 | overlookmotel | Note Added: 0002690 | |
2018-09-18 23:44 | Carsten | Note Added: 0002691 | |
2020-09-18 21:55 | carl | Assigned To | => carl |
2020-09-18 21:55 | carl | Status | new => resolved |
2020-09-18 21:55 | carl | Resolution | open => fixed |
2021-01-08 00:45 | carl | Status | resolved => closed |