You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Shelf files were a later addition to the PdsFile class. Re-factoring the code to use them for this purpose would not be a bad idea. Somebody would have to implement a glob.glob replacement that uses shelf files. Feel free!
Note that, if the match pattern crosses volume boundaries, it could actually be slower than a direct call to glob.glob. That's because each volume has its own shelf files.
It seems like PdsFile maintains these enormous "shelve" files (or pickle
files) that contain all sorts of useful information about every file in
pdsdata. But then when you call opus_products(), it uses glob() on the
real file system instead of looking into these files. Likewise when you
create a ViewSet for preview images, it has "must_exist=True", which
checks the actual filesystem. Same for the "exists" property.
Shouldn't we be using the shelve files for all of these cases?
.
The text was updated successfully, but these errors were encountered:
Shelf files were a later addition to the PdsFile class. Re-factoring the code to use them for this purpose would not be a bad idea. Somebody would have to implement a glob.glob replacement that uses shelf files. Feel free!
Note that, if the match pattern crosses volume boundaries, it could actually be slower than a direct call to glob.glob. That's because each volume has its own shelf files.
.
The text was updated successfully, but these errors were encountered: