Navigate version history on a per-file basis

[rangermauve]

Navigate version history on a per-file basis

[RangerMauve]

Hmm, not sure if HEAD would make sense here.

Let me think about diffs for a week or two.

[josephmturner]

To accomplish this, we could extrapolate on the ideas discussed here and here.

If the latest version of the drive is version 100 and user wants to checkout the version of /example.txt before the last change, send a GET request to hyper://public-key/$/previous/100/example.txt. Assuming that the most recent different version of /example.txt is version 42, this would have the same effect as GET hyper://public-key/$/version/42/example.txt. A HEAD request should return just the headers, allowing for a quick overview of the version history of particular file, including file sizes, modification dates, and version numbers.

Going forward through version history, GET hyper://public-key/$/previous/42/example.txt should give version 43. GET hyper://public-key/$/previous/42/example.txt should give version 100 (current).

[RangerMauve]

To get a diff we can use the ETag versins returned on a HEAD for a file.

• Get the current version of the drive
• Loop
• HEAD on the hyper://key/$/version/${version}/${file_path}
• If 404, break the loop
• Get whatever file info you need from the HEAD response and log them
• Set version to the ETag