Update speedate and truncate microseconds by default (#762)

Author: adriangb

Cargo.lock

@@ -1,6 +1,6 @@
 [package]
 name = "pydantic-core"
-version = "2.2.0"
+version = "2.3.0"
 edition = "2021"
 license = "MIT"
 homepage = "https://github.com/pydantic/pydantic-core"
@@ -35,7 +35,7 @@ enum_dispatch = "0.3.8"
 serde = { version = "1.0.147", features = ["derive"] }
 # disabled for benchmarks since it makes microbenchmark performance more flakey
 mimalloc = { version = "0.1.30", optional = true, default-features = false, features = ["local_dynamic_tls"] }
-speedate = "0.9.1"
+speedate = "0.10.0"
 ahash = "0.8.0"
 url = "2.3.1"
 # idna is already required by url, added here to be explicit

Cargo.toml

@@ -848,6 +848,7 @@ class TimeSchema(TypedDict, total=False):
     lt: time
     gt: time
     tz_constraint: Union[Literal['aware', 'naive'], int]
+    microseconds_precision: Literal['truncate', 'error']
     ref: str
     metadata: Any
     serialization: SerSchema
@@ -861,6 +862,7 @@ def time_schema(
     lt: time | None = None,
     gt: time | None = None,
     tz_constraint: Literal['aware', 'naive'] | int | None = None,
+    microseconds_precision: Literal['truncate', 'error'] = 'truncate',
     ref: str | None = None,
     metadata: Any = None,
     serialization: SerSchema | None = None,
@@ -884,6 +886,7 @@ def time_schema(
         lt: The value must be strictly less than this time
         gt: The value must be strictly greater than this time
         tz_constraint: The value must be timezone aware or naive, or an int to indicate required tz offset
+        microseconds_precision: The behavior when seconds have more than 6 digits or microseconds is too large
         ref: optional unique identifier of the schema, used to reference the schema in other places
         metadata: Any other information you want to include with the schema, not used by pydantic-core
         serialization: Custom serialization schema
@@ -896,6 +899,7 @@ def time_schema(
         lt=lt,
         gt=gt,
         tz_constraint=tz_constraint,
+        microseconds_precision=microseconds_precision,
         ref=ref,
         metadata=metadata,
         serialization=serialization,
@@ -914,6 +918,7 @@ class DatetimeSchema(TypedDict, total=False):
     # defaults to current local utc offset from `time.localtime().tm_gmtoff`
     # value is restricted to -86_400 < offset < 86_400 by bounds in generate_self_schema.py
     now_utc_offset: int
+    microseconds_precision: Literal['truncate', 'error'] = ('truncate',)
     ref: str
     metadata: Any
     serialization: SerSchema
@@ -929,6 +934,7 @@ def datetime_schema(
     now_op: Literal['past', 'future'] | None = None,
     tz_constraint: Literal['aware', 'naive'] | int | None = None,
     now_utc_offset: int | None = None,
+    microseconds_precision: Literal['truncate', 'error'] = 'truncate',
     ref: str | None = None,
     metadata: Any = None,
     serialization: SerSchema | None = None,
@@ -956,6 +962,7 @@ def datetime_schema(
         tz_constraint: The value must be timezone aware or naive, or an int to indicate required tz offset
             TODO: use of a tzinfo where offset changes based on the datetime is not yet supported
         now_utc_offset: The value must be in the past or future relative to the current datetime with this utc offset
+        microseconds_precision: The behavior when seconds have more than 6 digits or microseconds is too large
         ref: optional unique identifier of the schema, used to reference the schema in other places
         metadata: Any other information you want to include with the schema, not used by pydantic-core
         serialization: Custom serialization schema
@@ -970,6 +977,7 @@ def datetime_schema(
         now_op=now_op,
         tz_constraint=tz_constraint,
         now_utc_offset=now_utc_offset,
+        microseconds_precision=microseconds_precision,
         ref=ref,
         metadata=metadata,
         serialization=serialization,
@@ -983,6 +991,7 @@ class TimedeltaSchema(TypedDict, total=False):
     ge: timedelta
     lt: timedelta
     gt: timedelta
+    microseconds_precision: Literal['truncate', 'error']
     ref: str
     metadata: Any
     serialization: SerSchema
@@ -995,6 +1004,7 @@ def timedelta_schema(
     ge: timedelta | None = None,
     lt: timedelta | None = None,
     gt: timedelta | None = None,
+    microseconds_precision: Literal['truncate', 'error'] = 'truncate',
     ref: str | None = None,
     metadata: Any = None,
     serialization: SerSchema | None = None,
@@ -1017,6 +1027,7 @@ def timedelta_schema(
         ge: The value must be greater than or equal to this timedelta
         lt: The value must be strictly less than this timedelta
         gt: The value must be strictly greater than this timedelta
+        microseconds_precision: The behavior when seconds have more than 6 digits or microseconds is too large
         ref: optional unique identifier of the schema, used to reference the schema in other places
         metadata: Any other information you want to include with the schema, not used by pydantic-core
         serialization: Custom serialization schema
@@ -1028,6 +1039,7 @@ def timedelta_schema(
         ge=ge,
         lt=lt,
         gt=gt,
+        microseconds_precision=microseconds_precision,
         ref=ref,
         metadata=metadata,
         serialization=serialization,

python/pydantic_core/core_schema.py

@@ -1,7 +1,8 @@
 use pyo3::intern;
 use pyo3::prelude::*;
 use pyo3::types::{PyDate, PyDateTime, PyDelta, PyDeltaAccess, PyDict, PyTime, PyTzInfo};
-use speedate::{Date, DateTime, Duration, ParseError, Time};
+use speedate::MicrosecondsPrecisionOverflowBehavior;
+use speedate::{Date, DateTime, Duration, ParseError, Time, TimeConfig};
 use std::borrow::Cow;
 use strum::EnumMessage;
 
@@ -264,8 +265,17 @@ pub fn bytes_as_date<'a>(input: &'a impl Input<'a>, bytes: &[u8]) -> ValResult<'
     }
 }
 
-pub fn bytes_as_time<'a>(input: &'a impl Input<'a>, bytes: &[u8]) -> ValResult<'a, EitherTime<'a>> {
-    match Time::parse_bytes(bytes) {
+pub fn bytes_as_time<'a>(
+    input: &'a impl Input<'a>,
+    bytes: &[u8],
+    microseconds_overflow_behavior: MicrosecondsPrecisionOverflowBehavior,
+) -> ValResult<'a, EitherTime<'a>> {
+    match Time::parse_bytes_with_config(
+        bytes,
+        TimeConfig {
+            microseconds_precision_overflow_behavior: microseconds_overflow_behavior,
+        },
+    ) {
         Ok(date) => Ok(date.into()),
         Err(err) => Err(ValError::new(
             ErrorType::TimeParsing {
@@ -276,8 +286,17 @@ pub fn bytes_as_time<'a>(input: &'a impl Input<'a>, bytes: &[u8]) -> ValResult<'
     }
 }
 
-pub fn bytes_as_datetime<'a, 'b>(input: &'a impl Input<'a>, bytes: &'b [u8]) -> ValResult<'a, EitherDateTime<'a>> {
-    match DateTime::parse_bytes(bytes) {
+pub fn bytes_as_datetime<'a, 'b>(
+    input: &'a impl Input<'a>,
+    bytes: &'b [u8],
+    microseconds_overflow_behavior: MicrosecondsPrecisionOverflowBehavior,
+) -> ValResult<'a, EitherDateTime<'a>> {
+    match DateTime::parse_bytes_with_config(
+        bytes,
+        TimeConfig {
+            microseconds_precision_overflow_behavior: microseconds_overflow_behavior,
+        },
+    ) {
         Ok(dt) => Ok(dt.into()),
         Err(err) => Err(ValError::new(
             ErrorType::DatetimeParsing {
@@ -389,8 +408,17 @@ fn map_timedelta_err<'a>(input: &'a impl Input<'a>, err: ParseError) -> ValError
     )
 }
 
-pub fn bytes_as_timedelta<'a, 'b>(input: &'a impl Input<'a>, bytes: &'b [u8]) -> ValResult<'a, EitherTimedelta<'a>> {
-    match Duration::parse_bytes(bytes) {
+pub fn bytes_as_timedelta<'a, 'b>(
+    input: &'a impl Input<'a>,
+    bytes: &'b [u8],
+    microseconds_overflow_behavior: MicrosecondsPrecisionOverflowBehavior,
+) -> ValResult<'a, EitherTimedelta<'a>> {
+    match Duration::parse_bytes_with_config(
+        bytes,
+        TimeConfig {
+            microseconds_precision_overflow_behavior: microseconds_overflow_behavior,
+        },
+    ) {
         Ok(dt) => Ok(dt.into()),
         Err(err) => Err(map_timedelta_err(input, err)),
     }

src/input/datetime.rs

@@ -238,42 +238,72 @@ pub trait Input<'a>: fmt::Debug + ToPyObject {
         self.strict_date()
     }
 
-    fn validate_time(&self, strict: bool) -> ValResult<EitherTime> {
+    fn validate_time(
+        &self,
+        strict: bool,
+        microseconds_overflow_behavior: speedate::MicrosecondsPrecisionOverflowBehavior,
+    ) -> ValResult<EitherTime> {
         if strict {
-            self.strict_time()
+            self.strict_time(microseconds_overflow_behavior)
         } else {
-            self.lax_time()
+            self.lax_time(microseconds_overflow_behavior)
         }
     }
-    fn strict_time(&self) -> ValResult<EitherTime>;
+    fn strict_time(
+        &self,
+        microseconds_overflow_behavior: speedate::MicrosecondsPrecisionOverflowBehavior,
+    ) -> ValResult<EitherTime>;
     #[cfg_attr(has_no_coverage, no_coverage)]
-    fn lax_time(&self) -> ValResult<EitherTime> {
-        self.strict_time()
-    }
-
-    fn validate_datetime(&self, strict: bool) -> ValResult<EitherDateTime> {
+    fn lax_time(
+        &self,
+        microseconds_overflow_behavior: speedate::MicrosecondsPrecisionOverflowBehavior,
+    ) -> ValResult<EitherTime> {
+        self.strict_time(microseconds_overflow_behavior)
+    }
+
+    fn validate_datetime(
+        &self,
+        strict: bool,
+        microseconds_overflow_behavior: speedate::MicrosecondsPrecisionOverflowBehavior,
+    ) -> ValResult<EitherDateTime> {
         if strict {
-            self.strict_datetime()
+            self.strict_datetime(microseconds_overflow_behavior)
         } else {
-            self.lax_datetime()
+            self.lax_datetime(microseconds_overflow_behavior)
         }
     }
-    fn strict_datetime(&self) -> ValResult<EitherDateTime>;
+    fn strict_datetime(
+        &self,
+        microseconds_overflow_behavior: speedate::MicrosecondsPrecisionOverflowBehavior,
+    ) -> ValResult<EitherDateTime>;
     #[cfg_attr(has_no_coverage, no_coverage)]
-    fn lax_datetime(&self) -> ValResult<EitherDateTime> {
-        self.strict_datetime()
-    }
-
-    fn validate_timedelta(&self, strict: bool) -> ValResult<EitherTimedelta> {
+    fn lax_datetime(
+        &self,
+        microseconds_overflow_behavior: speedate::MicrosecondsPrecisionOverflowBehavior,
+    ) -> ValResult<EitherDateTime> {
+        self.strict_datetime(microseconds_overflow_behavior)
+    }
+
+    fn validate_timedelta(
+        &self,
+        strict: bool,
+        microseconds_overflow_behavior: speedate::MicrosecondsPrecisionOverflowBehavior,
+    ) -> ValResult<EitherTimedelta> {
         if strict {
-            self.strict_timedelta()
+            self.strict_timedelta(microseconds_overflow_behavior)
         } else {
-            self.lax_timedelta()
+            self.lax_timedelta(microseconds_overflow_behavior)
         }
     }
-    fn strict_timedelta(&self) -> ValResult<EitherTimedelta>;
+    fn strict_timedelta(
+        &self,
+        microseconds_overflow_behavior: speedate::MicrosecondsPrecisionOverflowBehavior,
+    ) -> ValResult<EitherTimedelta>;
     #[cfg_attr(has_no_coverage, no_coverage)]
-    fn lax_timedelta(&self) -> ValResult<EitherTimedelta> {
-        self.strict_timedelta()
+    fn lax_timedelta(
+        &self,
+        microseconds_overflow_behavior: speedate::MicrosecondsPrecisionOverflowBehavior,
+    ) -> ValResult<EitherTimedelta> {
+        self.strict_timedelta(microseconds_overflow_behavior)
     }
 }