|Title||»||Enable joining partial .skyd files when the filenames do not have a hash|
The following scenario does not work:
Create partial .skyd files using the --import-no-join option in Folder_1 and Folder_2
Copy the partial .skyd files to a third folder:
and attempt to join the partial .skyd file with this command:
--in=file.sky --import-file="Folder_1/file1.raw" --import-file="Folder_2/file2.raw" --save
This fails with a "File does not exist" error since Skyline expects a hash in the .skyd file names based on the path of the .raw file given in the --import-file arguments.
This fails too:
--in=file.sky --import-file="file1.raw" --import-file="file2.raw" --save
Here Skyline will not expect a hash in the .skyd filenames since the raw file path is the same as the location of the .sky file. But it still fails. I think the path to the raw file that you want to import has to be the same as the path stored in the partial .skyd file.
From Brendan's email:
"Partial .skyd files are always used, but may not last beyond the life of the Skyline process if you do not use --import-no-join. It seems like we should probably make that option simpler. Maybe something like the option:
If you use "none", then no hash is used (requires all file names to be unique) and no path checking is ever done. If there is a .skyd file with the same basename as the raw data file in the Skyline document folder it is used, similar to what Nick was hoping would already work. If you use "all" then a hash is always used and as long as the original full-path is use again, the appropriate .skyd file will be found and used, similar to what Vagisha and I discussed last week."