Duplicate --s3-storage-class flags should be combined

hi and thanks for the ever improving documentation.

https://rclone.org/s3/#s3-storage-class-3
should be merged into
https://rclone.org/s3/#s3-storage-class

Hmm, yes the docs for s3 are a bit mad aren't they. It is because there is duplicated docs for each provider.

What do you think the docs should do here - show the docs for one provider? Or make subsections for each provider? They are getting a bit long too.

yes, they are mad and getting madder all the time.

if we come up with a plan, i would be willing to revamp that page.

this is just a quick mock up, not perfect in terms of details

for the flag

--s3-storage-class
The storage class to use when storing new objects in S3.
Config: storage_class
Env Var: RCLONE_S3_STORAGE_CLASS
Type: string
Default: ""

Possible Values Definition
"" Default
STANDARD Standard storage class
REDUCED_REDUNDANCY Reduced redundancy
STANDARD_IA Standard Infrequent Access
ARCHIVE Archive storage mode.
ONEZONE_IA One Zone Infrequent Access
GLACIER Glacier
DEEP_ARCHIVE Glacier Deep Archive
INTELLIGENT_TIERING Intelligent-Tiering

for the providers

Provider Permited Values
AWS "", STANDARD, REDUCED_REDUNDANCY, STANDARD_IA, ONEZONE_IA, GLACIER, DEEP_ARCHIVE, INTELLIGENT_TIERING
Tencent COS "", STANDARD, ARCHIVE, STANDARD_IA

or perhaps something to look like https://rclone.org/overview/#features
i am not sure how to replicate that exact look here in the forum but this should give you an idea of what i suggest.

Provider "" STANDARD REDUCED_REDUNDANCY STANDARD_IA ONEZONE_IA GLACIER DEEP_ARCHIVE INTELLIGENT_TIERING
AWS Yes Yes Yes Yes Yes Yes Yes Yes
Tencent COS Yes Yes - - - - - -

I like your thinking.

These docs are auto generated from the code so changing them is going to need a bit of coding. If you type rclone help backend s3 rclone will generate the docs on the fly for you to see.

These are the duplicated regions with how many times they are duplicated

$ rclone help backend s3 | grep '^####' | sort | uniq -cd
      7 #### --s3-endpoint
      3 #### --s3-location-constraint
      3 #### --s3-region
      4 #### --s3-storage-class

So I guess the obvious thing to do would be to try to group the duplicated sections and have subsections for providers

You'll see that some of the help is different for the same options between providers which is annoying - maybe we can ignore that.

What would you think if we made a table like this

Possible Values Provider Definition
"" All Default
STANDARD AWS,Tencent,COS Standard storage class
REDUCED_REDUNDANCY AWS Reduced redundancy
STANDARD_IA AWS,Tencent,COS Standard Infrequent Access
ARCHIVE Tencent,COS Archive storage mode.
ONEZONE_IA AWS One Zone Infrequent Access
GLACIER AWS Glacier

Oops, press submit too soon

That would be fairly easy to do.

The Endpoint table would be really, really long though.

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.