Merge branch 'main' into anypayloadprovider

components/collator/src/comparison.rs

@@ -14,12 +14,13 @@ use crate::elements::{
     NO_CE_SECONDARY, NO_CE_TERTIARY, OPTIMIZED_DIACRITICS_MAX_COUNT, QUATERNARY_MASK,
 };
 use crate::options::CollatorOptionsBitField;
-use crate::provider::CollationDataV1Marker;
 use crate::provider::CollationDiacriticsV1Marker;
 use crate::provider::CollationJamoV1Marker;
 use crate::provider::CollationMetadataV1Marker;
 use crate::provider::CollationReorderingV1Marker;
+use crate::provider::CollationRootV1Marker;
 use crate::provider::CollationSpecialPrimariesV1Marker;
+use crate::provider::CollationTailoringV1Marker;
 use crate::{AlternateHandling, CollatorOptions, MaxVariable, ResolvedCollatorOptions, Strength};
 use core::cmp::Ordering;
 use core::convert::TryFrom;
@@ -55,8 +56,8 @@ impl AnyQuaternaryAccumulator {
 #[derive(Debug)]
 pub struct Collator {
     special_primaries: Option<DataPayload<CollationSpecialPrimariesV1Marker>>,
-    root: DataPayload<CollationDataV1Marker>,
-    tailoring: Option<DataPayload<CollationDataV1Marker>>,
+    root: DataPayload<CollationRootV1Marker>,
+    tailoring: Option<DataPayload<CollationTailoringV1Marker>>,
     jamo: DataPayload<CollationJamoV1Marker>,
     diacritics: DataPayload<CollationDiacriticsV1Marker>,
     options: CollatorOptionsBitField,
@@ -72,6 +73,9 @@ impl Collator {
     pub fn try_new(locale: &DataLocale, options: CollatorOptions) -> Result<Self, DataError> {
         Self::try_new_unstable_internal(
             &crate::provider::Baked,
+            DataPayload::from_static_ref(
+                crate::provider::Baked::SINGLETON_COLLATION_ROOT_V1_MARKER,
+            ),
             DataPayload::from_static_ref(
                 icu_normalizer::provider::Baked::SINGLETON_CANONICAL_DECOMPOSITION_DATA_V1_MARKER,
             ),
@@ -110,7 +114,8 @@ impl Collator {
     ) -> Result<Self, DataError>
     where
         D: DataProvider<CollationSpecialPrimariesV1Marker>
-            + DataProvider<CollationDataV1Marker>
+            + DataProvider<CollationRootV1Marker>
+            + DataProvider<CollationTailoringV1Marker>
             + DataProvider<CollationDiacriticsV1Marker>
             + DataProvider<CollationJamoV1Marker>
             + DataProvider<CollationMetadataV1Marker>
@@ -124,14 +129,17 @@ impl Collator {
             provider.load(Default::default())?.payload,
             provider.load(Default::default())?.payload,
             provider.load(Default::default())?.payload,
+            provider.load(Default::default())?.payload,
             || provider.load(Default::default()).map(|r| r.payload),
             locale,
             options,
         )
     }
 
+    #[allow(clippy::too_many_arguments)]
     fn try_new_unstable_internal<D>(
         provider: &D,
+        root: DataPayload<CollationRootV1Marker>,
         decompositions: DataPayload<CanonicalDecompositionDataV1Marker>,
         tables: DataPayload<CanonicalDecompositionTablesV1Marker>,
         jamo: DataPayload<CollationJamoV1Marker>,
@@ -143,7 +151,8 @@ impl Collator {
         options: CollatorOptions,
     ) -> Result<Self, DataError>
     where
-        D: DataProvider<CollationDataV1Marker>
+        D: DataProvider<CollationRootV1Marker>
+            + DataProvider<CollationTailoringV1Marker>
             + DataProvider<CollationDiacriticsV1Marker>
             + DataProvider<CollationMetadataV1Marker>
             + DataProvider<CollationReorderingV1Marker>
@@ -177,7 +186,7 @@ impl Collator {
 
         let metadata = metadata_payload.get();
 
-        let tailoring: Option<DataPayload<crate::provider::CollationDataV1Marker>> =
+        let tailoring: Option<DataPayload<crate::provider::CollationTailoringV1Marker>> =
             if metadata.tailored() {
                 Some(
                     provider
@@ -209,8 +218,6 @@ impl Collator {
             }
         }
 
-        let root: DataPayload<CollationDataV1Marker> = provider.load(Default::default())?.payload;
-
         let tailored_diacritics = metadata.tailored_diacritics();
         let diacritics: DataPayload<CollationDiacriticsV1Marker> = provider
             .load(if tailored_diacritics {
@@ -339,7 +346,7 @@ impl Collator {
     }
 
     fn compare_impl<I: Iterator<Item = char>>(&self, left_chars: I, right_chars: I) -> Ordering {
-        let tailoring: &DataPayload<CollationDataV1Marker> =
+        let tailoring: &DataPayload<CollationTailoringV1Marker> =
             if let Some(tailoring) = &self.tailoring {
                 tailoring
             } else {
@@ -356,7 +363,7 @@ impl Collator {
                 // should we have a no-op tailoring that contains a
                 // specially-crafted CodePointTrie that always returns
                 // a FALLBACK_CE32 after a single branch?
-                &self.root
+                self.root.cast_ref()
             };
 
         // Sadly, it looks like variable CEs and backward second level

components/collator/src/provider.rs

@@ -59,7 +59,8 @@ const _: () = {
         pub use icu_collections as collections;
     }
     make_provider!(Baked);
-    impl_collation_data_v1_marker!(Baked);
+    impl_collation_root_v1_marker!(Baked);
+    impl_collation_tailoring_v1_marker!(Baked);
     impl_collation_diacritics_v1_marker!(Baked);
     impl_collation_jamo_v1_marker!(Baked);
     impl_collation_metadata_v1_marker!(Baked);
@@ -70,7 +71,8 @@ const _: () = {
 #[cfg(feature = "datagen")]
 /// The latest minimum set of markers required by this component.
 pub const MARKERS: &[DataMarkerInfo] = &[
-    CollationDataV1Marker::INFO,
+    CollationRootV1Marker::INFO,
+    CollationTailoringV1Marker::INFO,
     CollationDiacriticsV1Marker::INFO,
     CollationJamoV1Marker::INFO,
     CollationMetadataV1Marker::INFO,
@@ -114,13 +116,20 @@ fn data_ce_to_primary(data_ce: u64, c: char) -> u32 {
 /// including in SemVer minor releases. While the serde representation of data structs is guaranteed
 /// to be stable, their Rust representation might not be. Use with caution.
 /// </div>
-#[icu_provider::data_struct(marker(
-    CollationDataV1Marker,
-    "collator/data@1",
-    // TODO(#3867): Use script fallback
-    fallback_by = "language",
-    attributes_domain = "collator",
-))]
+#[icu_provider::data_struct(
+    marker(
+        CollationRootV1Marker,
+        "collator/root@1",
+        singleton,
+    ),
+    marker(
+        CollationTailoringV1Marker,
+        "collator/tailoring@1",
+        // TODO(#3867): Use script fallback
+        fallback_by = "language",
+        attributes_domain = "collator",
+    )
+)]
 #[derive(Debug, PartialEq, Clone)]
 #[cfg_attr(feature = "datagen", derive(serde::Serialize, databake::Bake))]
 #[cfg_attr(feature = "datagen", databake(path = icu_collator::provider))]

components/collator/tests/tests.rs

@@ -1286,7 +1286,7 @@ fn test_nb_nn_no() {
     strs.sort_by(|a, b| collator.compare(a, b));
     assert_eq!(strs, expected);
     assert_eq!(
-        DataProvider::<CollationDataV1Marker>::load(
+        DataProvider::<CollationTailoringV1Marker>::load(
             &icu_collator::provider::Baked,
             DataRequest {
                 id: DataIdentifierBorrowed::for_locale(&locale),
@@ -1306,7 +1306,7 @@ fn test_nb_nn_no() {
     strs.sort_by(|a, b| collator.compare(a, b));
     assert_eq!(strs, expected);
     assert_eq!(
-        DataProvider::<CollationDataV1Marker>::load(
+        DataProvider::<CollationTailoringV1Marker>::load(
             &icu_collator::provider::Baked,
             DataRequest {
                 id: DataIdentifierBorrowed::for_locale(&locale),
@@ -1326,7 +1326,7 @@ fn test_nb_nn_no() {
     strs.sort_by(|a, b| collator.compare(a, b));
     assert_eq!(strs, expected);
     assert_eq!(
-        DataProvider::<CollationDataV1Marker>::load(
+        DataProvider::<CollationTailoringV1Marker>::load(
             &icu_collator::provider::Baked,
             DataRequest {
                 id: DataIdentifierBorrowed::for_locale(&locale),

components/datetime/src/any/zoned_datetime.rs

@@ -26,7 +26,7 @@ use icu_decimal::provider::DecimalSymbolsV1Marker;
 use icu_plurals::provider::OrdinalV1Marker;
 use writeable::Writeable;
 
-size_test!(ZonedDateTimeFormatter, zoned_date_time_formatter_size, 6224);
+size_test!(ZonedDateTimeFormatter, zoned_date_time_formatter_size, 6200);
 
 /// [`ZonedDateTimeFormatter`] is a formatter capable of formatting
 /// date/times with time zones from any calendar, selected at runtime. For the difference between this and [`TypedZonedDateTimeFormatter`](crate::TypedZonedDateTimeFormatter),
@@ -49,7 +49,7 @@ size_test!(ZonedDateTimeFormatter, zoned_date_time_formatter_size, 6224);
 ///
 /// # Examples
 ///
-/// Using a GMT time zone:
+/// Using the UTC time zone:
 ///
 /// ```
 /// use icu::calendar::DateTime;
@@ -82,13 +82,13 @@ size_test!(ZonedDateTimeFormatter, zoned_date_time_formatter_size, 6224);
 /// );
 /// ```
 ///
-/// Using a non-GMT time zone, specified by id:
+/// Using a custom time zone, specified by id:
 ///
 /// ```
 /// use icu::calendar::DateTime;
 /// use icu::datetime::{options::length, ZonedDateTimeFormatter};
 /// use icu::locale::locale;
-/// use icu::timezone::{CustomTimeZone, GmtOffset, MetazoneCalculator, ZoneVariant, TimeZoneBcp47Id};
+/// use icu::timezone::{CustomTimeZone, UtcOffset, MetazoneCalculator, ZoneVariant, TimeZoneBcp47Id};
 /// use tinystr::tinystr;
 /// use writeable::assert_writeable_eq;
 ///
@@ -108,9 +108,9 @@ size_test!(ZonedDateTimeFormatter, zoned_date_time_formatter_size, 6224);
 ///     .expect("Failed to construct DateTime.");
 /// let any_datetime = datetime.to_any();
 ///
-/// // Create a time zone for America/Chicago at GMT-6:
+/// // Create a time zone for America/Chicago at UTC-6:
 /// let mut time_zone = CustomTimeZone::new_empty();
-/// time_zone.gmt_offset = "-06:00".parse::<GmtOffset>().ok();
+/// time_zone.offset = "-06:00".parse().ok();
 /// time_zone.time_zone_id = Some(TimeZoneBcp47Id(tinystr!(8, "uschi")));
 /// time_zone.zone_variant = Some(ZoneVariant::daylight());
 ///
@@ -162,7 +162,7 @@ impl ZonedDateTimeFormatter {
     /// options.month = Some(components::Month::Long);
     /// options.hour = Some(components::Numeric::Numeric);
     /// options.minute = Some(components::Numeric::Numeric);
-    /// options.time_zone_name = Some(components::TimeZoneName::GmtOffset);
+    /// options.time_zone_name = Some(components::TimeZoneName::Offset);
     ///
     /// let zdtf = ZonedDateTimeFormatter::try_new_experimental(
     ///     &locale!("en-u-ca-gregory").into(),

components/datetime/src/fields/symbols.rs

@@ -680,9 +680,9 @@ field_type!(
         /// For example: "Pacific Standard Time"
         'z' => LowerZ = 0,
         /// Field symbol for any of: the ISO8601 basic format with hours, minutes and optional seconds fields, the
-        /// long localized GMT format, or the ISO8601 extended format with hours, minutes and optional seconds fields.
+        /// long localized offset format, or the ISO8601 extended format with hours, minutes and optional seconds fields.
         'Z' => UpperZ = 1,
-        /// Field symbol for the localized GMT format of a time zone.
+        /// Field symbol for the localized offset format of a time zone.
         ///
         /// For example: "GMT-07:00"
         'O' => UpperO = 2,

components/datetime/src/format/datetime.rs

@@ -20,7 +20,7 @@ use crate::provider::date_time::{
 use crate::time_zone::{
     Bcp47IdFormat, ExemplarCityFormat, FallbackTimeZoneFormatterUnit, FormatTimeZone,
     FormatTimeZoneError, GenericLocationFormat, GenericNonLocationLongFormat,
-    GenericNonLocationShortFormat, Iso8601Format, LocalizedGmtFormat,
+    GenericNonLocationShortFormat, Iso8601Format, LocalizedOffsetFormat,
     SpecificNonLocationLongFormat, SpecificNonLocationShortFormat, TimeZoneDataPayloadsBorrowed,
     TimeZoneFormatterUnit,
 };
@@ -36,7 +36,7 @@ use icu_calendar::AnyCalendarKind;
 use icu_decimal::FixedDecimalFormatter;
 use icu_plurals::PluralRules;
 use icu_provider::DataPayload;
-use icu_timezone::{CustomTimeZone, GmtOffset};
+use icu_timezone::{CustomTimeZone, UtcOffset};
 use writeable::{Part, Writeable};
 
 /// [`FormattedDateTime`] is a intermediate structure which can be retrieved as
@@ -752,12 +752,12 @@ where
     ZS: ZoneSymbols<'data>,
 {
     fn write_time_zone_missing(
-        gmt_offset: Option<GmtOffset>,
+        offset: Option<UtcOffset>,
         w: &mut (impl writeable::PartsWrite + ?Sized),
     ) -> fmt::Result {
-        match gmt_offset {
-            Some(gmt_offset) => w.with_part(Part::ERROR, |w| {
-                Iso8601Format::default_for_fallback().format_infallible(w, gmt_offset)
+        match offset {
+            Some(offset) => w.with_part(Part::ERROR, |w| {
+                Iso8601Format::default_for_fallback().format_infallible(w, offset)
             }),
             None => w.with_part(Part::ERROR, |w| "{GMT+?}".write_to(w)),
         }
@@ -777,12 +777,12 @@ where
         }
         Some(custom_time_zone) => match zone_symbols {
             None => {
-                write_time_zone_missing(custom_time_zone.gmt_offset, w)?;
+                write_time_zone_missing(custom_time_zone.offset, w)?;
                 Err(DateTimeWriteError::MissingZoneSymbols)
             }
             Some(zs) => match ResolvedNeoTimeZoneSkeleton::from_field(field_symbol, field_length) {
                 None => {
-                    write_time_zone_missing(custom_time_zone.gmt_offset, w)?;
+                    write_time_zone_missing(custom_time_zone.offset, w)?;
                     Err(DateTimeWriteError::UnsupportedField(field))
                 }
                 Some(time_zone) => {
@@ -792,8 +792,8 @@ where
                     match do_write_zone(units, &zone_input, payloads, w)? {
                         Ok(()) => Ok(()),
                         Err(()) => {
-                            write_time_zone_missing(custom_time_zone.gmt_offset, w)?;
-                            // Return an error since GMT data was missing
+                            write_time_zone_missing(custom_time_zone.offset, w)?;
+                            // Return an error since offset data was missing
                             Err(DateTimeWriteError::MissingZoneSymbols)
                         }
                     }
@@ -809,9 +809,9 @@ fn select_zone_units(time_zone: ResolvedNeoTimeZoneSkeleton) -> [Option<TimeZone
     let mut formatters = (
         None,
         None,
-        // Friendly Localized GMT Format (requires "essentials" data)
+        // Friendly Localized offset Format (requires "essentials" data)
         Some(TimeZoneFormatterUnit::WithFallback(
-            FallbackTimeZoneFormatterUnit::LocalizedGmt(LocalizedGmtFormat {}),
+            FallbackTimeZoneFormatterUnit::LocalizedOffset(LocalizedOffsetFormat {}),
         )),
     );
     match time_zone {
@@ -856,11 +856,11 @@ fn select_zone_units(time_zone: ResolvedNeoTimeZoneSkeleton) -> [Option<TimeZone
             ));
         }
         // `O`
-        ResolvedNeoTimeZoneSkeleton::GmtShort => {
+        ResolvedNeoTimeZoneSkeleton::OffsetShort => {
             // TODO: For now, use the long format. This should be GMT-8
         }
         // `OOOO`, `ZZZZ`
-        ResolvedNeoTimeZoneSkeleton::GmtLong => {
+        ResolvedNeoTimeZoneSkeleton::OffsetLong => {
             // no-op
         }
         // 'V'