Github user dhunziker commented on the issue:
https://github.com/apache/spark/pull/16812
> This can be easily worked around, no?
It can be worked around, losing the boolean type by converting to string in
the process though. With any workaround it'll be better to
Github user dhunziker commented on the issue:
https://github.com/apache/spark/pull/16812
Oracle doesn't have boolean so it's usually modelled as char(1) with Y/N,
Sybase doesn't have boolean either but bit which is 1/0. PosgreSQL supports a
range of values
(https
Github user dhunziker commented on the issue:
https://github.com/apache/spark/pull/16812
That would remain a workaround though. The uniVocity parser for boolean
supports this as well:
https://github.com/uniVocity/univocity-parsers/blob/master/src/main/java/com/univocity/parsers
Github user dhunziker commented on a diff in the pull request:
https://github.com/apache/spark/pull/16812#discussion_r99953924
--- Diff:
sql/core/src/test/scala/org/apache/spark/sql/execution/datasources/csv/CSVSuite.scala
---
@@ -139,6 +140,20 @@ class CSVSuite extends QueryTest
Github user dhunziker commented on a diff in the pull request:
https://github.com/apache/spark/pull/16812#discussion_r99953321
--- Diff:
sql/core/src/test/scala/org/apache/spark/sql/execution/datasources/csv/CSVInferSchemaSuite.scala
---
@@ -73,6 +73,12 @@ class
Github user dhunziker commented on a diff in the pull request:
https://github.com/apache/spark/pull/16812#discussion_r99951876
--- Diff:
sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/csv/UnivocityParser.scala
---
@@ -110,7 +110,11 @@ private[csv] class
GitHub user dhunziker opened a pull request:
https://github.com/apache/spark/pull/16812
[SPARK-19465][SQL] Added options for custom boolean values in CSV
## What changes were proposed in this pull request?
It adds trueValue and falseValue options for customising the values
