Converting a column of identifiers

Mixed input is the normal case#

Every line is read on its own, so a list can hold canonical UUIDs, bare hex, braced Windows values, base64, byte arrays, ULIDs and high/low pairs at the same time. The reader recognises each and converts it to the one format you chose. That is what makes the tool useful on the output of a query somebody else wrote.

Comments survive the trip#

A # opens a comment, and so does // at the start of a line or after whitespace. The text after it is kept next to the result instead of being thrown away, which means a list can carry its own labels:

f81d4fae-7dec-11d0-a765-00a0c91e6bf6   # the row that broke
01890a5d-ac96-774b-bcce-b302099a8057   // fixture, do not delete

Nothing disappears quietly#

A line the reader cannot make sense of is reported, not skipped, and a line that is already in the target format says so instead of pretending to convert. That matters when the list is long enough that you would not notice two missing rows.

Where the results go#

Converted lines land in a history you can search, tag and keep across reloads, because it lives in the browser's own storage. Nothing is uploaded: the whole pipeline runs in the tab, which is the same reason the pages of this reference can compute their examples in front of you.

RFC 9562 § 4 · UUID FormatThe text forms a reader has to accept, which is why mixed input works at all