CLDR 49 Release Note

No.	Date	Rel. Note	Data	Charts	Spec	Delta	GitHub Tag	Delta DTD	CLDR JSON
49	2026-10-XX	v49	CLDR49	Charts49	CLDR49	Δ49	release-49-beta2	ΔDtd49	49.0.0-BETA1

Overview

Unicode CLDR provides key building blocks for software supporting the world’s languages. CLDR data is used by all major software systems (including all mobile phones) for their software internationalization and localization, adapting software to the conventions of different languages.

Changes

The most significant changes in this release are:

• Replaced the backronym for CLDR and unified the title of the CLDR specification
  • “Common Locale Data Repository (CLDR)” → “Unicode Common Locale Data & Rules (CLDR)”
  • “UTS #35: Unicode Locale Data Markup Language (LDML)” → “UTS #35: Unicode Common Locale Data & Rules (CLDR)”.
  • The former names and the acronym “LDML” led to many misunderstandings and have been changed for clarity.
  • Note that all references under the former names and the LDML acronym remain valid; such references on the Unicode site will be updated over time.
• Updated for Unicode 18 including annotations for the new emoji, changes to sorting order, etc.
• Added new date & time formatting features including:
  • Localized patterns for gluing date & timezoneAppend Items — e.g., Sept 3, EST
  • Ordinal days in dates — e.g., Sept 3rd
  • Customizing numeric datetime separators in patterns: 3-10-2031 → 3/10/2031
  • Dual format for zones with two offsets (summer and winter) — eg, GMT-8/-7
  • Deprecated Japanese era data before Meiji
  • Structure for preventing digit-digit merges — e.g., ‘2026/1/29 GMT-817时’ → ‘2026/1/29 GMT-8 17时’
  • Additional skeleton-patterns added for flexible and interval date formats
• Added new units (conversions and formatting)
  • 3 new units: poundal, syne, and milliinch (US mil)
  • 14 new display names (only in English: poundal and mil are only US units, while dyne is obsolete)
• Added nested bracket replacement — for constructing locale names with parts that have parentheses, eg, ”birmanês (Mianmar [Birmânia])”
• Added many additional localized locale option names for use in menus, such as calendar and number-system names
• Added client coverage data — coverage data in a format that is easy for clients to use and relatively small: see pathCoverage
  • This data allows implementations to determine precisely what the coverage level for each path in a given locale.
  • Note: This is not the same as the data for which locales have reached particular coverage levels. That is in properties/coverageLevels.txt
• Updated to the most recent versions of external standards and data sources, such as the language subtag registry, UN M49 macro regions, ISO 4217 currencies, etc.

Note: Some features are in Technical Preview, while others were included in dot releases of 48 and may be new for implementers that didn’t update beyond version 48.0.

For more details, see below.

Locale Coverage Status

Coverage Levels

Count	Level	Usage	Examples
100	Modern	Suitable for full UI internationalization	Afrikaans, shqip, አማርኛ, ‫العربية‬, հայերեն, অসমীয়া, azərbaycan
12	Moderate	Suitable for “document content” internationalization, eg. in spreadsheet	Akan, Cebuano, Māori, тоҷикӣ
73	Basic	Suitable for locale selection, eg. choice of language on mobile phone	भोजपुरी, बर’, डोगरी, eʋegbe, Gã, हरियाणवी

Note: This includes just the base language and script. There are many more regional variants.

Coverage Changes

±	New Level	Locales
📈	Modern	Akan
📈	Moderate	Breton, Coptic‡
📈	Basic	Adyghe, Central Kurdish, Colognian, Kabardian, Kikuyu, Kʼicheʼ, Ladin, Prussian, Qʼeqchiʼ, Sunwar (Sunuwar)
📉	Moderate*	Romansh, Shan, Tigrinya
📉	Basic*	Bashkir, Faroese, Interlingua, Sardinian, Tajik, Venetian

Notes:

* For every release, the number of items needed for Modern and Moderate increases. Therefore, locales without active contributors may drop down in coverage level.

‡ Indicates locales that have gone up or down by more than one level.

For a full listing, see Coverage Levels

Specification Changes

The following are the most significant changes to the CLDR specification, aside from those covered under DTD changes below.

• Replaced the title of the specification “UTS #35: Unicode Locale Data Markup Language (LDML)” with “UTS #35: Unicode Common Locale Data & Rules (CLDR)”.
  • Note that all references under the former names and the LDML acronym remain valid; such references on the Unicode site will be updated over time.
• Clarified the process of selecting the best dateFormatItem when there is no exact match, and how to use appendItems to add missing fields.
  • This includes a clarification of what are date fields and what are time fields, and a note that appendItems for date and time fields should be appended before combining them.
• Added more information to the Key/Type Description table to make clear which key/types use constructed values and supply a brief description of the typeValue element.
  • Also changed the table from HTML to Markdown, with each Key description (such as co for collation) having its own subheader.
• Updated plural rules to make it clear that they are evaluated in semantic order (zero, then one,…).
• Updated Part 5: Collation has been updated to reflect the changes that were upstreamed into UTS #10 as part of Unicode 18.0. As a result, the order of all characters in the CLDR root collation is now the same as in the UCA DUCET.
• Revised the numberFormat description.
• For Units, made the formatting and phrasing more internally consistent.
• For MessageFormat
  • The :currency and :percent functions are now Stable, with the same implementations as previously.
  • The u:locale option (previously in Draft) has been dropped from the specification.
  • Clarified the computation of the exemplar city for non-location zones.
• Added additional subsections for reference.

See the Modifications section of the specification for details.

DTD Changes

Added:

• placeholderBoundarySpacing to allow characters to be inserted where necessary, such as between digits in Chinese dates.
• dayOfMonth (and parent elements) for days that are not purely numeric. This is used for ordinal dates, like “Sept 13th, 2026”. Technical Preview
• Date-Timezone and Time-Day-Of-Week as fallback options for appendItem in dates.
• intervalFormatRange for constructing ranges (used internally for consistency checks).
• numericDateSeparator and numericTimeSeparator for customization of numeric dates and times (3-10-2031 → 3/10/2031). Technical Preview
• alt (alternative) forms of gmtFormat and gmtUnknownFormat to request use of localized equivalents of the term “UTC” instead of “GMT” (limited locales).
• dualOffsetFormat so that the Localized GMT format can express differences more clearly (eg, Los Angeles → GMT-8/-7; Phoenix → GMT-7; Denver → GMT-7/-6).
• typeValue for localized menus with locale ID options.
• numberSystem in currency formats to allow for different formats for locales with multiple number systems.
• nestedBracketReplacement (48.2) for constructing locale names with parts that have parentheses, eg, ”birmanês (Mianmar [Birmânia])”.

For a full listing, see Delta DTDs.

Data Changes

Supplemental Data Changes

Updates for:

• BCP47 codes
• ISO 636 changes
• Unit conversions
• Likely subtags (eg, abj → abj_Latn_IN)
• Language matches
• Language population data
• Language group data (from Wikidata)
• Language script data
• Metazone and timezone data
• Plural/Ordinal rules (af, bg, es, fr, gl, nn, tg, vi — see Language Plural Rules chart for more information)
  • Fixed bug so Norwegian Nynorsk (nn) inherits plural rules from Norwegian (no)
• Currencies
• Deprecation of Japanese eras (pre Meiji)
• Time preferences (12 vs 24, day periods)
  • Updated AR, CL, PY, UY, and ZM to prefer 24 hour time
• Coverage: fixed issue with cross-language inheritance which was giving Haitian Creole an artificially high coverage level.
• Number Spellout: Added or improved RBNF rules for many locales including Catalan, Italian, Croatian, Greek, Romanian, Ukrainian and more. See RBNF tickets for full list.
• Client coverage data — coverage data in a format that is easy for clients to use and relatively small: see pathCoverage. Technical Preview

For a full listing, see ¤¤BCP47 Delta and ¤¤Supplemental Delta

Locale Data Changes

• Updated en-AU and en-NZ to include exemplar characters for Indigenous languages
• Added many localized names for locale ID options, for menus and formatting locale names (for example, c12 → “12-Stunden-Format”)
• Added many additional date patterns for more consistency across locales
• Additional timezone names (and removal of obsolete ones)
• Names and search keywords for the new Unicode 18.0 emoji
• Other spot fixes for specific locales

For a full listing, see Delta Data

Keyboard Changes

• Four new keyboard layouts: Egyptian Hieroglyphic, Gandhari, Sanskrit, Classical Tibetan.
• Spec clarifications were made for the display, layer, and backspace transform elements. For a full list of spec changes, see Keyboard Spec Changes in v49.

File Changes

• New locale files (28 files, e.g. brh.xml)
• New test data
  • Date/time, decimal, messageFormat (10 files)
  • Person name formatting (1 file, br.txt)
  • Rule-based number formatting (99 files, e.g. af.ssv)
• MessageFormat tests/functions (1 file, math.json)
• New keyboard files (4 files)
• Many new or modified readme files

JSON Data Changes

Package: cldr-localenames-full

• New file: typeValues.json. CLDR-19774
• localeDisplayNames.json now has core-scoped types in a subkey _core. CLDR-19774

Tooling Changes

Spec Tooling improvements

• Migrated the spec and site to generate previews to Cloudflare.
• Automated updates to the header for key elements such as Date.
• Added support for symbolic links (e.g. using the CLDR version such as 48 in the link will take you to the latest version for that major release, such as tr35-78 for 48.2.
• Added a check to require edits to modifications when the spec is updated.

Survey Tool improvements

• Improved the ability to search for items including: value, English value, code.
• Improved display and editing of hidden and ambiguous characters in the Survey Tool.
• Updated the Datetime and Numbers report so that vetters can see formats with both native digits and in the Latin script for locales in a script that has native digits that are still commonly used.
• Added a category of “New” in the Dashboard and Vetting participation to make it easier for vetters to see which items did not have a winning value in the previous release.
• Made it easier to see which items have forum posts in the voting view. 👁️‍🗨️ if there are any open discussions, and 💬 if there are discussions, but all are closed.
• Revised the Survey Tool to display a candidate item in the Winning column if it is currently winning, even if it has the status “missing” due to it not having enough recorded votes. Previously, such items were shown in the Others column. Items with the status “missing” may still be published in the final release.
• Automatically generated minimal pair items based on a locale’s plurals and ordinals so vetters can contribute data in the Survey Tool instead of requiring changes in XML.
• Updated voting for non-TC CLDR Organizations. We now calculate the winning vote for a non-TC organization to be the value with the most votes (instead of the most recent vote, as we do for TC organizations).
• Improved error detection:
  • Improved detection of non-date pattern characters in dates and times that are not marked as literals appropriately.
  • Enabled tests to catch collisions in emoji short names and metazone names.
  • Fixed an issue where collisions were only visible if 3 items were colliding.
  • Improved consistency of error visibility between Survey Tool components (e.g. Dashboard, Voting View & Info Panel).
• Fixed bugs causing bulk imports to fail.
• Updated the automatic messages sent following a Survey Tool push to production to include the list of changes included in the push.
• Added information about the creation date of Survey Tool accounts to make troubleshooting quicker.

Access changes based on Survey Tool phase

• Added the ability for the CLDR TC to allow specific locales to continue submitting new data items during the Vetting phase.
• Updated READONLY phase to only allow creating and responding to posts in the forum.

Migration

The following changes have been made in CLDR 49. Please plan accordingly to avoid disruption.

• The pre-Meiji Japanese eras were removed:
  • There was too much uncertainty in the exact values and we received feedback that the general practice for referencing exact dates for the pre-Meiji era is to use the Gregorian calendar.
• A new zip file named cldr-exemplars-delta-49.zip includes data files for CLDR locales that are below basic coverage, including from the exemplars directory. These are not recommended for general use, but are available especially implementations that need more extensive exemplar data.
  • To use this file, first unpack the regular zip file, and then unpack the delta file in the same location.
  • Any feedback on this new file is welcome, and can be submitted as a new ticket. CLDR-19806
• The core.zip / cldr-common-49.zip data file (production data) no longer includes data for locales not yet at Basic level.
  • These are 94 locales in Locale Coverage that do not have basic in the Computed Level column nor ICU in the ICU column.
  • As noted below, locales are slated for removal from CLDR in version 50 if they do not reach Basic level. If you are able to help contribute data to bring a language up to Basic coverage in a future release, see CLDR Organization for how to contribute.

Advanced warnings CLDR 50 and beyond

The following changes are planned for CLDR 50. Please plan accordingly to avoid disruption.

• Locales which do not have Core data will be removed if still missing core data by alpha. CLDR-16004
• Montenegrin cnr will be considered a separate locale and will no longer alias to sr_Latn_ME. CLDR-10769
• The default week numbering system will change to follow ISO (where weeks are numbered based on Thursday), instead of being based on the start of the calendar week. The calendar week will be more clearly targeted at matching usage in displayed month calendars. CLDR-18275
• The major components in supplementalData.xml and supplementalMetadata.xml files will be organized more logically and moved into separate files. - This will make it easier for implementations to filter out data that they don’t need, and make internal maintenance easier. This will not affect the data, just which file it is located in. Please plan to update XML and JSON parsers accordingly.
• For line breaking (and the other segmentation types: grapheme cluster, word, and sentence breaks), the UTC is planning new data files.
  • CLDR will adopt this tailoring mechanism once the UTC has approved its publication, because it will drastically reduce the size and complexity of segmentation tailorings. CLDR-18624
  • See PRI#555 for the proposal. The data files will:
    • allow implementations other than ICU/ICU4X to achieve both conformant behavior and high performance
    • provide for tailorings (e.g., for CSS profiles) without data duplication.
• For UnicodeSet, the UTC is developing a new specification for all Unicode encoding specifications (as well as in the higher levels: CLDR, ICU4C/J/X).
  • It is a much more complete and rigorous specification than what is in the CLDR specification UTS #35. See UTS #61: Unicode Set Notation (currently in draft).
  • Following UTC approval of UTS #61, CLDR plans to deprecate the section in UTS #35, and redirect users to the new UTS #61. UTS #35 will retain a short description, and make sure that all the links redirect correctly. CLDR-18624
• The emoji /properties/labels.txt is not currently maintained and may be deprecated in CLDR 50.
  • If you rely on this data please comment on CLDR-19752 to let us know your use case.
• Date/time fields only intended for use in patterns will be deprecated in skeletons.
  • These include stand-alone fields (L, q, c) and non-canonical hour fields (K, k) — as well as day period symbols (a, b, B) without an accompanying 12-hour field (h, j) or standalone usage.
  • Skeletons should use the vanilla symbols M, Q, E/e, and h/H instead. CLDR-19757
• Redundant plural cases, and possibly other redundant data, will be removed from production data. This is where inheritance (including Lateral Inheritance) produces identical results. CLDR-19193

Known Issues

• ISO 3166-2 subdivision codes for Iran changed in 2020, and there are not yet new equivalent stable codes. See CLDR-19060 for more details.
• Keyboards
  • Normalization-safe segments definition does not cover all normalization cases. CLDR-19218
  • minDeviceWidth attribute is inadequately speified. CLDR-19754
  • Some keyboard files have build errors under kmc. CLDR-19483

Acknowledgments

Many people have made significant contributions to CLDR (structure, data, specification, tooling, and process); see the Acknowledgments page for a full listing.

The Unicode Terms of Use apply to CLDR data; in particular, see Exhibit 1.

For web pages with different views of CLDR data, see http://cldr.unicode.org/index/charts.