With AWS S3, I can upload a file test.png to any directory I like, regardless of whether or not it exists... because S3 will automatically generate the full path & directories.
For example, if I when I upload to S3, I use the path this/is/a/new/home/for/test.png, S3 will create directories this, is, a, ... and upload test.png to the correct folder.
I am migrating over to Azure, and I am looking to use their file storage. However, it seems that I must manually create EVERY directory... I could obviously do it programmatically by checking to see if the folder exists and if not, create it... but wow...why should I work so hard?
I did try:
file_service.create_file_from_path('testshare', 'some/long/path', 'test.png', 'path/to/local/location/of/test.png')
However, that complains that the directory does not exist... and will only work if I either manually create the directories or replace some/long/path with None.
Is it possible to just hand Azure a path and have it create the directories?
Azure Files closely mimics OS File System and thus in order to push a file in a directory, that directory must exist. What that means is if you need to create a file in a nested directory structure, that directory structure must exist. Azure File service will not create that structure for you.
A better option in your scenario would be to use Azure Blob Storage. It closely mimics Amazon S3 behavior that you mentioned above. You can create a Container (similar to Bucket in S3) and then upload a file with a name like this/is/a/new/home/for/test.png.
However please note that the folders are virtual in Blob Storage (same as S3) and not the real one. Essentially the name with which the blob (similar to Object in S3) will be saved is this/is/a/new/home/for/test.png.