Add addUsernameFromFilePathStep to support moving chrome_login_data_emails to KATC #1832
Conversation
| } | ||
|
|
||
| const ( | ||
| snappyDecodeTransformStep = "snappy" | ||
| deserializeFirefoxTransformStep = "deserialize_firefox" | ||
| deserializeChromeTransformStep = "deserialize_chrome" | ||
| camelToSnakeTransformStep = "camel_to_snake" | ||
| addUsernameFromFilePathStep = "add_username_from_filepath" |
There was a problem hiding this comment.
I'm somewhat unclear if we want this. The alternate pattern is to join using the home directory instead of username
There was a problem hiding this comment.
The source path is already included in the KATC table data, so I imagined we could do something like that with the preexisting data instead! I added this because I figured it would be better for the katc_chrome_login_data_emails to provide the exact same columns as the kolide_ table does.
There was a problem hiding this comment.
We don't have to keep the same table semantics.
There was a problem hiding this comment.
And the assumptions in homeDirLocations have edges cases that will break unexpectedly :|
There was a problem hiding this comment.
edges cases that will break unexpectedly :|
In that case, the row will still be returned, just without the username field included! And we'd still have the path column available.
I think my preference is to keep this for parity with the old table, just in case we do want it. But if you feel strongly about getting rid of the username column, I'm happy to close this PR too!
There was a problem hiding this comment.
My hunch is we should drop it. I'm not quite sure why. Unless we cannot avoid it, I'd rather not return potentially wrong data.
|
Decided we don't want the username column, so I'm closing this PR |
To support moving the
kolide_chrome_login_data_emailstable to KATC, we need to be able to add ausernamecolumn that extracts the username from the source file path. This PR adds a transform step that does that.