Download ICU 4.6 Release
Description
ICU is the premier library for software internationalization. ICU 4.6 is a major release of ICU, with new features, new APIs, and many bug fixes in data and code.
Major changes in ICU 4.6 since ICU 4.4 include the following:
Common Changes
Supports final version of Unicode 6.0
New UCA data for collation/sorting (#7264), with refinements from CLDR; this data is revamped for more effective use of collation weights, and noncharacters are now handled (#8007)
Support for 2,088 new characters, including the new emoji and Indian Rupee sign
Fully updated properties
Support for UTS #46 Unicode IDNA Compatibility Processing (#7144).
Alternate number symbols based on numbering system (#7599).
Compact collation tailoring syntax for reduced memory and disk footprint (#7015).
New collation [import] rule for reduced footprint and improved maintenance (#7023).
Fast string BiDi direction detection (#7772).
ICU4C Specific Changes
ICU 4.6 requires compiler RTTI to be turned on. Please see the ICU4C readme for more details (#7455).
pkg-config files for a standard way of linking against ICU (#6981).
Promotion to @draft (from @internal) for most regex functions that provide access via UText (#8102).
Regex support for a "find progress" callback (#7666).
Enhance regex APIs to support full 64-bit offsets and indices (#7813).
New regex API to set match and start position independently (#7949).
ICU4J Specific Changes
Common Updates in ICU 4.6.1
Common Locale Data Repository (CLDR) 1.9.1
Update timezone data support to Olson 2011c
#8271 UCOL_RUNTIME_VERSION should be updated for 4.6
#8277 Collation Reordering Use Of USCRIPT_UNKNOWN
#8290 Can't find Hangul with search coll (usearch doesn't handle CE iter behavior)
#8303 ULocale#toLanguageTag() should not supply "und" as language when the locale has only private use
#8341 USpoof uses NFKD, should be NFD
ICU4C Download
Version: 4.6.1
Release Date: 2011-03-11
No API changes since ICU4C 4.6, see API Changes since ICU4C 4.4
Known issues from ICU4C 4.6, still present in ICU4C 4.6.1:
A collation rule using the new star notation “<*” in which the list of characters following “<*” includes two successive APOSTROPHE characters ('') is malformed, but may not be detected and will produce erroneous results (#8204).
On AIX xlC v9, having certain patterns in the source directory path will cause intltest to not build (see #8211 for simple workaround as well as patch) - this is already merged into the maintenance branch for future 4.6.x releases.
#8571 (C only) U_HIDE_DRAFT_API does not operate on the correct set of APIs; might also affect U_HIDE_DEPRECATED_API, U_HIDE_INTERNAL_API and U_HIDE_SYSTEM_API
#8563 Wrong primary collation weights for some characters in some language tailorings (e.g., bn & my)
ICU4C 4.6 issues fixed in the 4.6.1 release:
#8199 ICU misparses numbers in scientific notation
#8200 Layout of telugu rutvam
#8211 test dep generation fails on xlC v9 with a numeric source path
#8225 Build errors on C code on HPUX PA-risc on nested namespaces/definitions
#8267 ucstrTextExtract does not pin start param or handle >2GB text
#8281 pkgdata failure with large CFLAGS
#8363 trap in usearch_search()
#8364 detect out of memory issue for Hashtable in low memory situations
#8378 Assertion failure in intltest spoof/testConfData
#8388 uloc_minimizeSubtag and uloc_addLikelySubtag have extraneous or missing character
(If the list of files does not appear above, see ICU4C Binaries.)
(If the list of files does not appear above, see ICU4C Source.)
Version: 4.6
Release Date: 2010-12-02
Known issues:
A collation rule using the new star notation “<*” in which the list of characters following “<*” includes two successive APOSTROPHE characters ('') is malformed, but may not be detected and will produce erroneous results (#8204).
On AIX xlC v9, having certain patterns in the source directory path will cause intltest to not build (see #8211 for simple workaround as well as patch) - this is already merged into the maintenance branch for future 4.6.x releases.
ICU4C Supported Platforms
Key to testing frequency
Frequently tested
ICU will work on these platforms with these compilers
Rarely tested
ICU has been ported to these platforms but may not have been tested there recently
ICU4J Download
Version: 4.6.1.1
Release Date: 2011-08-24
Back ported Java 7 Locale support features from the trunk
ICU4J 4.6.1 issues fixed in the 4.6.1.1 release:
#6408 DecimalFormat fraction digit problem with rounding BigDecimal
#8419 In IndianCalendar, the add operation performs like roll
#8484 endless loop in RuleBasedCollator.getSortKeyBytes
#8549 UTF-7 error handling consumes too many valid subsequent chars
#8569 UTF-7 missing terminating shift character
#8596 GregorianCalendar.getActualMaximum(Calendar.WEEK_OF_YEAR) returns 52 when it should return 53
#8624 ucol_getSortKey strncmp and ucol_strcoll give different comparison result
Known issues in ICU4J 4.6, still present in ICU4J 4.6.1:
A collation rule using the new star notation “<*” in which the list of characters following “<*” includes two successive APOSTROPHE characters ('') is malformed, but may not be detected and will produce erroneous results (#8204).
An exception may occur when using AlphabeticIndex with ULocale.ROOT (#8189)
#8563 Wrong primary collation weights for some characters in some language tailorings (e.g., bn & my)
(If the list of files does not appear above, see ICU4J Source.)
Version: 4.6.1
Release Date: 2011-03-11
No API changes since ICU4J 4.6, see API Changes since ICU4J 4.4
ICU4J 4.6 issues fixed in the 4.6.1 release:
#7919 ICU4J plug-in for Eclipse 3.7.x
#8258 Avoid UnicodeSet constructors taking String pattern when all code points are known
#8259 Use Java StringTokenizer instead of ICU in ICUResourceBundle and VTimeZone
#8275 ICU4J Locale service returns incorrect result with unsupported class loader protocol
#8283 ICU4J MessageFormat should use varargs in format method
#8284 Resource Bundle lookup for locale root@numbers=latn falls back to en_US
#8391 NPE from StringSearch with 4.6
Version: 4.6
Release Date: 2010-12-02
Known issues:
A collation rule using the new star notation “<*” in which the list of characters following “<*” includes two successive APOSTROPHE characters ('') is malformed, but may not be detected and will produce erroneous results (#8204).
An exception may occur when using AlphabeticIndex with ULocale.ROOT (#8189)