Updates from code review.

Author: Omega359

datafusion/functions/Cargo.toml

@@ -40,7 +40,7 @@ workspace = true
 [features]
 crypto_expressions = ["md-5", "sha2", "blake2", "blake3"]
 # enable datetime functions
-datetime_expressions = []
+datetime_expressions = ["chrono-tz"]
 # Enable encoding by default so the doctests work. In general don't automatically enable all packages.
 default = [
     "datetime_expressions",
@@ -71,7 +71,7 @@ base64 = { version = "0.22", optional = true }
 blake2 = { version = "^0.10.2", optional = true }
 blake3 = { version = "1.8", optional = true }
 chrono = { workspace = true }
-chrono-tz = "0.10.4"
+chrono-tz = { version = "0.10.4", optional = true }
 datafusion-common = { workspace = true }
 datafusion-doc = { workspace = true }
 datafusion-execution = { workspace = true }

datafusion/functions/src/datetime/common.rs

@@ -116,20 +116,23 @@ pub(crate) fn string_to_datetime_formatted<T: TimeZone>(
     // custom parser (or switching to Jiff)
     let tz: Option<chrono_tz::Tz> = if format.ends_with(" %Z") {
         // grab the string after the last space as the named timezone
-        let parts: Vec<&str> = datetime_str.rsplitn(2, ' ').collect();
-        let timezone_name = parts[0];
-        datetime_str = parts[1];
-
-        // attempt to parse the timezone name
-        let result: Result<chrono_tz::Tz, chrono_tz::ParseError> = timezone_name.parse();
-        let Ok(tz) = result else {
-            return Err(err(&result.unwrap_err().to_string()));
-        };
+        if let Some((dt_str, timezone_name)) = datetime_str.rsplit_once(' ') {
+            datetime_str = dt_str;
+
+            // attempt to parse the timezone name
+            let result: Result<chrono_tz::Tz, chrono_tz::ParseError> =
+                timezone_name.parse();
+            let Ok(tz) = result else {
+                return Err(err(&result.unwrap_err().to_string()));
+            };
 
-        // successfully parsed the timezone name, remove the ' %Z' from the format
-        format = format.trim_end_matches(" %Z");
+            // successfully parsed the timezone name, remove the ' %Z' from the format
+            format = &format[..format.len() - 3];
 
-        Some(tz)
+            Some(tz)
+        } else {
+            None
+        }
     } else if format.contains("%Z") {
         return Err(err(
             "'%Z' is only supported at the end of the format string preceded by a space",

datafusion/functions/src/datetime/to_timestamp.rs

@@ -239,7 +239,7 @@ pub struct ToTimestampMicrosFunc {
 #[user_doc(
     doc_section(label = "Time and Date Functions"),
     description = r#"
-Converts a value to a timestamp (`YYYY-MM-DDT00:00:00.000000000Z<TZ>`) in the session time zone. Supports strings,
+Converts a value to a timestamp (`YYYY-MM-DDT00:00:00.000000000<TZ>`) in the session time zone. Supports strings,
 integer, unsigned integer, and double types as input. Strings are parsed as RFC3339 (e.g. '2023-07-20T05:44:00')
 if no [Chrono formats] are provided. Strings that parse without a time zone are treated as if they are in the
 session time zone. Integers, unsigned integers, and doubles are interpreted as nanoseconds since the unix epoch (`1970-01-01T00:00:00Z`).

datafusion/sqllogictest/test_files/to_timestamp_timezone.slt

@@ -63,7 +63,7 @@ SET datafusion.execution.time_zone = 'America/New_York';
 
 ## Test 7: Naive timestamp with configured timezone
 ## '2020-09-08T13:42:29' in America/New_York is EDT (UTC-4)
-## So this should become '2020-09-08T13:42:29Z-04:00'
+## So this should become '2020-09-08T13:42:29-04:00'
 query P
 SELECT to_timestamp('2020-09-08T13:42:29');
 ----

docs/source/user-guide/sql/scalar_functions.md

@@ -2909,7 +2909,7 @@ Additional examples can be found [here](https://github.com/apache/datafusion/blo
 
 ### `to_timestamp_nanos`
 
-Converts a value to a timestamp (`YYYY-MM-DDT00:00:00.000000000Z<TZ>`) in the session time zone. Supports strings,
+Converts a value to a timestamp (`YYYY-MM-DDT00:00:00.000000000<TZ>`) in the session time zone. Supports strings,
 integer, unsigned integer, and double types as input. Strings are parsed as RFC3339 (e.g. '2023-07-20T05:44:00')
 if no [Chrono formats] are provided. Strings that parse without a time zone are treated as if they are in the
 session time zone. Integers, unsigned integers, and doubles are interpreted as nanoseconds since the unix epoch (`1970-01-01T00:00:00Z`).