Pin Arrow CSV Boolean alias support for DataFusion

Author: osipovartem

Cargo.lock

@@ -89,30 +89,30 @@ version = "55.1.0"
 #
 # See for more details: https://github.com/rust-lang/cargo/issues/11329
 apache-avro = { version = "0.21", default-features = false }
-arrow = { git = "https://github.com/Embucket/arrow-rs.git", rev = "25a9c23d7b9706238b00b1288d27f036c186cd8e", features = [
+arrow = { git = "https://github.com/Embucket/arrow-rs.git", rev = "0e9c0b17db31ff260254d7dbfeecffe535675ed5", features = [
     "prettyprint",
     "chrono-tz",
 ] }
-arrow-avro = { git = "https://github.com/Embucket/arrow-rs.git", rev = "25a9c23d7b9706238b00b1288d27f036c186cd8e", default-features = false, features = [
+arrow-avro = { git = "https://github.com/Embucket/arrow-rs.git", rev = "0e9c0b17db31ff260254d7dbfeecffe535675ed5", default-features = false, features = [
     "deflate",
     "snappy",
     "zstd",
     "bzip2",
     "xz",
 ] }
-arrow-buffer = { git = "https://github.com/Embucket/arrow-rs.git", rev = "25a9c23d7b9706238b00b1288d27f036c186cd8e", default-features = false }
-arrow-data = { git = "https://github.com/Embucket/arrow-rs.git", rev = "25a9c23d7b9706238b00b1288d27f036c186cd8e", default-features = false }
-arrow-flight = { git = "https://github.com/Embucket/arrow-rs.git", rev = "25a9c23d7b9706238b00b1288d27f036c186cd8e", features = [
+arrow-buffer = { git = "https://github.com/Embucket/arrow-rs.git", rev = "0e9c0b17db31ff260254d7dbfeecffe535675ed5", default-features = false }
+arrow-data = { git = "https://github.com/Embucket/arrow-rs.git", rev = "0e9c0b17db31ff260254d7dbfeecffe535675ed5", default-features = false }
+arrow-flight = { git = "https://github.com/Embucket/arrow-rs.git", rev = "0e9c0b17db31ff260254d7dbfeecffe535675ed5", features = [
     "flight-sql-experimental",
 ] }
 # Both codecs are required here to make sure that code paths like
 # file-spilling have access to all compression codecs.
-arrow-ipc = { git = "https://github.com/Embucket/arrow-rs.git", rev = "25a9c23d7b9706238b00b1288d27f036c186cd8e", default-features = false, features = [
+arrow-ipc = { git = "https://github.com/Embucket/arrow-rs.git", rev = "0e9c0b17db31ff260254d7dbfeecffe535675ed5", default-features = false, features = [
     "lz4",
     "zstd",
 ] }
-arrow-ord = { git = "https://github.com/Embucket/arrow-rs.git", rev = "25a9c23d7b9706238b00b1288d27f036c186cd8e", default-features = false }
-arrow-schema = { git = "https://github.com/Embucket/arrow-rs.git", rev = "25a9c23d7b9706238b00b1288d27f036c186cd8e", default-features = false }
+arrow-ord = { git = "https://github.com/Embucket/arrow-rs.git", rev = "0e9c0b17db31ff260254d7dbfeecffe535675ed5", default-features = false }
+arrow-schema = { git = "https://github.com/Embucket/arrow-rs.git", rev = "0e9c0b17db31ff260254d7dbfeecffe535675ed5", default-features = false }
 async-trait = "0.1.89"
 bigdecimal = "0.4.8"
 bytes = "1.11"
@@ -178,7 +178,7 @@ memchr = "2.8.1"
 num-traits = { version = "0.2" }
 object_store = { version = "0.13.2", default-features = false }
 parking_lot = "0.12"
-parquet = { git = "https://github.com/Embucket/arrow-rs.git", rev = "25a9c23d7b9706238b00b1288d27f036c186cd8e", default-features = false, features = [
+parquet = { git = "https://github.com/Embucket/arrow-rs.git", rev = "0e9c0b17db31ff260254d7dbfeecffe535675ed5", default-features = false, features = [
     "arrow",
     "async",
     "object_store",

Cargo.toml

@@ -203,7 +203,8 @@ impl CsvFormat {
         self
     }
 
-    /// Accept exact `0` and `1` values in Boolean CSV columns.
+    /// Accept `0`/`1` and Snowflake-style text aliases (`t`/`f`, `yes`/`no`,
+    /// `y`/`n`, `on`/`off`) in Boolean CSV columns.
     pub fn with_numeric_boolean_values(mut self, allow: bool) -> Self {
         self.numeric_boolean_values = allow;
         self

datafusion/datasource-csv/src/file_format.rs

@@ -116,13 +116,14 @@ impl CsvSource {
         self
     }
 
-    /// Accept exact numeric Boolean literals (`0` and `1`) while scanning CSV.
+    /// Accept `0`/`1` and Snowflake-style text aliases (`t`/`f`, `yes`/`no`,
+    /// `y`/`n`, `on`/`off`) in Boolean CSV columns.
     pub fn with_numeric_boolean_values(mut self, allow: bool) -> Self {
         self.numeric_boolean_values = allow;
         self
     }
 
-    /// Whether exact `0` and `1` values are accepted in Boolean CSV columns.
+    /// Whether numeric and Snowflake-style text aliases are accepted in Boolean CSV columns.
     pub fn numeric_boolean_values(&self) -> bool {
         self.numeric_boolean_values
     }