Skip to content

[format] Support sep as fallback key of csv.field-delimiter - #9565

Merged
JingsongLi merged 1 commit into
apache:masterfrom
jackylee-ch:csv-sep-fallback-key
Sep 3, 2026
Merged

[format] Support sep as fallback key of csv.field-delimiter#9565
JingsongLi merged 1 commit into
apache:masterfrom
jackylee-ch:csv-sep-fallback-key

Conversation

@jackylee-ch

Copy link
Copy Markdown
Contributor

Purpose

csv.field-delimiter accepts field-delimiter, seq and delimiter as fallback keys, but not sep. Setting sep is silently ignored and the delimiter falls back to the default ,.

The other fallback keys in CsvOptions mirror Spark's CSV option names one for one (lineSep, quote, escape, header, nullvalue, mode); sep is the only one missing. Same reason as #9170, which added delimiter.

Tests

CsvFileFormatTest#testCsvFieldDelimiterFallbackKeys, extending the existing parameterized fallback-key list.

Written with Claude Code; reasoning and verification are mine.

sep is Spark's option name for the CSV field delimiter, and the other
fallback keys in CsvOptions already mirror Spark's CSV option names.
Without it, setting sep is silently ignored and the delimiter falls
back to the default comma.
@JingsongLi

Copy link
Copy Markdown
Contributor

+1

@JingsongLi
JingsongLi merged commit 14040b1 into apache:master Sep 3, 2026
13 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants