Uniscribe
Uniscribe is the Microsoft Windows set of services for rendering Unicode-encoded text, especially complex text layout. They are implemented in the DLL USP10.DLL. USP10.dll became available to the public with Windows 2000 and Internet Explorer 5.0. In addition, the Windows CE platform has supported Uniscribe since version 5.0.
Although Uniscribe continues to be maintained, with Windows 7 its intended replacement DirectWrite was introduced, which has more features.
USP10.dll
USP is an initialism for Unicode Scripts Processor. The main purpose of Uniscribe includes the following:
- arranging input text from the input sequence to visual sequence.
- substituting glyphs according to context (e.g. different forms of Arabic characters)
- ordering displayed text based on text flow direction (e.g. LTR vs RTL, Horizontal vs Vertical).
Below are listed some common versions of usp10.dll, as well as the methods by which they are distributed.
Features are added according only the "major.minor" part of the version number, the third part in the full version number is used for system target identification numbers for which the DLL was ported by Microsoft, and the last part is the build number on each target system version (which may change within regular system/software updates). Some hotfixes provide upgrades only for specific applications (notably in the Office installation directory), and not suited for use in the Windows system directory (whose version of the DLL should never be updated and is often protected by the system):
File sizes may vary depending on specific localizations of the DLL (depending on the target system or application for which it was compiled); those given here are for the US-English localization.
Version number | File Size | File Date | Software Bundled by | New feature of this version |
---|---|---|---|---|
10.0.14393.0 | 79,360 bytes (x64) | 2016-08-02 | Windows 10 Anniversary Update | ? |
1.626.7601.23259 | 627,712 bytes (x86) | 2015-11-03 | ? | ? |
1.626.7601.19054 | 627,712 bytes (x86) | 2015-11-03 | ? | ? |
10.0.10240.16834 | 626,688 bytes (x64) | 2015-07-10 | Windows 10 | Universal Shaping Engine support for Unicode 7.0 complex scripts[1][2][3] |
1.626.7601.18454 (Windows 7 GDR) | 626,688 bytes (x86) 801,280 bytes (x64) |
2014-04-25 | KB2957509 – Windows 7, Windows Server 2008 R2, Windows Vista, Windows Server 2008, Windows Server 2003 | Security update |
1.626.7601.22666 (Windows 7 LDR) | 626,688 bytes (x86) 801,792 bytes (x64) | |||
6.3.9431.0 | 76,288 bytes | 2013-06-16 | Windows 8.1 (preview) | ? |
6.2.9200.16384 | 74 KiB | 2012-07-26 | Windows 8 | ? |
1.626.7601.22171 | 626,688 bytes (x86) | 2012-11-22 | ? | ? |
1.626.7601.18009 | 612 KiB | 2012-11-22 | KB2786400 | An update is available that changes the default settings of the shaping behavior for Arabic text rendering in Windows 7 and Windows Server 2008 R2. |
1.626.7601.17514 | 781 KiB | 2010-11-20 | Windows 7 SP1 x64 (RTM) | ? |
611 KiB | Windows 7 SP1 x86 (RTM) | |||
1.626.7601.17105 | 611 KiB | 2010-09-30 | Windows 7 SP1 (RC) | Restore support for unassigned code points[4] |
1.626.7600.20796 | 625 KiB | 2010-09-21 | PowerPoint Viewer 2010 (SP1) KB2460050 | ? |
1.626.7600.20602 | 623 KiB | 2010-01-07 | Microsoft Office 2010 (RTM) | Support PR-37: Clarification of the Use of Zero Width Joiner in Indic Scripts;[5] This version supports OpenType on Windows XP. |
1.626.7600.16385 | 628 KiB | 2009-08-04 | Microsoft Office 2010 (Beta) | ? |
612 KiB | 2009-07-14 | Windows 7 (RTM) | ? | |
1.626.7100.0 | 612 KiB | 2009-04-22 | Windows 7 RC | ? |
1.626.6002.22402 | 491 KiB | 2010-07-11 | Microsoft Office 2007 (SP2) Security Update KB2288621 | MS10-063: Description of the security update for the 2007 Office suites: September 14, 2010 |
1.626.6002.18244 | 607 KiB | 2010-04-16 | Windows Server 2008 SP2, Windows Vista SP2 Security Update KB981322 | MS10-063: Description of the security update for Windows Unicode Scripts Processor: September 14, 2010 |
1.626.6002.18005 | 607 KiB | 2009-04-11 | Windows Server 2008 SP2, Windows Vista SP2 | ? |
1.626.6001.18000 | 490 KiB | 2008-01-19 | Windows Server 2008 (RTM), Windows Vista SP1, VOLTSupplementalFiles (July 2008), MS VOLT 1.3 (released July 31, 2008), SIL FieldWorks, Keyman Desktop 7.1 | Supports newer OpenType table features; improved support for ligatures and contextual shapes in Indic scripts with simpler fonts |
1.626.6001.16510 | 491 KiB | 2007-04-18 | Windows Server "Longhorn" (Beta 3) | ? |
1.626.6000.20581 | 491 KiB (502,784 bytes) | 2007-04-19 02:15:55 UTC | Windows Vista Hotfix KB936176 | FIX: incorrect rendering of fully decomposed characters in NFD form encoded in sequences with more than 2 characters (affects 114 Latin characters decomposed in sequences of 3 codes) |
1.626.6000.16386 | 491 KiB (502,784 bytes) | 2006-11-02 09:44:03 UTC | Windows Vista Ultimate (RTM) | DirectX 10? WDDM? |
1.626.5756.0 | 491 KiB | 2006-10-13 | Microsoft Office 2007 Ultimate (RTM) | Supports Oriya and Kannada scripts |
1.615.5384.4 | 484 KiB | 2006-06-17 | Windows Vista (Beta 2) | DirectX 9? |
1.614.5315.0 | 454 KiB | 2006-03-13 | Microsoft Office 2007 (Beta 2) | OpenType typographic features? |
1.613.5291.0 | 481 KiB (492,544 bytes) | 2006-01-04 | Windows Vista (Beta 1) | Microsoft VOLT 1.2 |
1.609.5219.0 | 469 KiB | 2005-08-17 | Microsoft Office 12 Professional (Beta 1) | OpenType typographic features? |
1.606.5065.1 | ? | ? | Third-party software | Supports Mongolian script |
1.606.5078.0 | ? | ? | Hotfix for SIL fonts | FIX : Combining mark sequences in Latin script may be incorrectly rendered |
1.601.5022.8 | 428 KiB | 2005-01-07 | Microsoft Sinhala Enabling Pack for XP 0.42 | Supports Sinhala script |
1.473.4067.15 | 415 KiB (424,960 bytes) | 2004-10-22 | Third-party software | Microsoft Visual OpenType Layout Tool (VOLT) 1.1.225 update |
1.473.4067.0 | 415 KiB (424,960 bytes) | 2004-10-22 | Third-party software | Supports Malayalam script; VOLTSupplementalFiles (Nov 2004); rendering improvement up to 3 diacritics with CGJ |
1.471.4063.0 | 415 KiB (424,960 bytes) | 2004-02-04 | Microsoft Office 2003 (fix), SIL FieldWorks (Speech Tools Phonology Assistant 3.0.1) | ? |
1.471.4030.0 | 404 KiB (413,184 bytes) | 2004-04-15 | Microsoft Office 2003, Google Earth | |
1.468.4015.0 | ? KiB (? bytes) | ? | Paratext 6 | Supports Tibetan script. |
1.468.4011.0 | ? KiB (? bytes) | ? | ? | |
1.460.3707.0 | ? KiB (? bytes) | ? | MS VOLT 1.1.206 | Supports Khmer script |
1.453.3665.0 | ? KiB (? bytes) | ? | VOLTSupplementalFiles (Aug 2002) | ? |
1.422.3790.1830 | 355 KiB (364,032 bytes) | 2005-03-30 | Windows Server 2003 (SP1) | FIX? |
1.421.3790.0 | 345 KiB (353,280 bytes) | 2003-03-25 | Windows Server 2003 | Support for DirectX, GDI+ and newer display driver model |
1.420.2600.5969 | 397 KiB (406,016 bytes) | 2010-04-16 | Windows XP (SP3) Security Update KB981322 | MS10-063: Description of the security update for Windows Unicode Scripts Processor: September 14, 2010 |
1.420.2600.5512 | 397 KiB (406,016 bytes) | 2008-04-14 | Windows XP Build 5512 (SP3) | FIX? |
1.420.2600.3163 | 397 KiB (406,016 bytes) | 2007-06-26 | Windows XP (SP2) Hotfix KB939450 | FIX: The GetCharacterPlacement function unexpectedly returns zero |
1.420.2600.2791 | 397 KiB (406,016 bytes) | 2005-11-05 | Windows XP (SP2) Hotfix KB910466 | FIX: You may receive a "MEM_BAD_POINTER" error message on a computer that is running Windows XP with Service Pack 2 |
1.420.2600.2180 | 397 KiB (406,528 bytes) | 2004-08-12 | Windows XP Build 2180 (SP2) | Supports Bengali and Malayalam scripts; supports complex Latin (for Vietnamese); FIX: ligation of tone contours |
1.409.2600.1106 | 331 KiB (339,456 bytes) | 2002-08-29 | Windows XP Build 1106 (SP1) | ? |
1.408.2600.1020 | 331 KiB (339,456 bytes) | 2002-04-17 | Internet Explorer 6.0.2800.1106 (SP1) | ? |
1.407.2600.0 | 331 KiB (339,456 bytes) | 2001-08-17 | Windows XP | Supports Thaana, Gujarati, Kannada, Gurmukhi (Punjabi), Syriac and Telugu scripts |
1.405.2416.1 | 317 KiB (325,120 bytes) | 2001-01-15 | Microsoft Office XP (2002), Microsoft Word 2000 (SP1) | Fully supports Hebrew, Devanagari and Tamil scripts |
1.400.2411.1 | 316 KiB (323,072 bytes) | 2000-12-13 | Internet Explorer 6 | ? |
1.325.2195.6692 | 308 KiB (315,664 bytes) | 2003-06-19 | Windows 2000 SP4 (?) | ? |
1.325.2195.? | ? | 2001-09-25 | Microsoft Windows Installer 2.0 Redistributable for Windows 95/98/Me KB827763 | Needed for Microsoft viewers (which otherwise require 98SE at least) of Powerpoint 2003 (and later) presentations on Windows 95 and 98 |
1.325.2195.1340 | 308 KiB (315,664 bytes) | 2000-07-21 | Windows 2000 SP1 | Fully supports Arabic script |
1.325.2180.1 | 316 KiB (323,584 bytes) | 2000-06-08 | Windows Me | Improved API for scripts with bidirectional and complex layouts; minimum support for Arabic and Hebrew scripts (no contextual letter forms, requires compatibility character mappings in fonts) |
307 KiB (315,152 bytes) | 2000-04-26 | Microsoft Global IME for Microsoft Office XP | ||
307 KiB (315,152 bytes) | 1999-11-30 | Internet Explorer 5.5 release, SP1 & SP2 | ||
1.175.0.1 | 268 KiB (274,432 bytes) | 1999-04-05 | Windows 98 SE | First stable Uniscribe API; supports simple alphabetic scripts with decomposed diacritics (Latin, Greek, Cyrillic) |
258 KiB (264,976 bytes) | 1999-01-28 | Internet Explorer 5.01 | ||
1.163.1890.1 | 262 KiB (268,288 bytes) | 1998-09-22 | Third-party software | Multilanguage Text Layout and Complex Scripts (MTLCS) snapshot (early release for developers) |
Universal Shaping Engine
Scripts with complex text layout have contextual and non-linear requirements to correctly render their typography. These requirements include: ligatures, where two consecutive characters have to be combined into one shape (Latin, Devanagari); reordering, where some characters have to be displayed before the letter they follow in actual pronunciation (Bengali, Sinhala, and other Indic languages); and context-shaping, as in cursive scripts where some letters have to change shape depending on whether they occur in the beginning, middle, or the end of the word (Arabic, Mongolian).
UniScribe uses several script-specific shaping engines for handling typography in supported complex scripts; these are implemented in addition to a generic engine for non-complex scripts (such as Cyrillic, Greek, Latin, etc.). The currently used engines include[2] Indic (Bengali, Devanagari, Gujurati, Gurmukhi, Kannada, etc.), Arabic, Hangul, Hebrew, Khmer, Myanmar, and Thai/Lao variants.
The complexity of the Unicode standard and ambiguities in OpenType specification often result in incomplete or erroneous implementations of complex text layout. Script-specific shaping engines work on a case-by-case basis and do not consistently handle common features of OpenType fonts, which makes it difficult for OS programmers and font developers to support new scripts. Implementation errors are very hard or impossible to correct at a later stage without breaking up backward compatibility for existing documents and fonts, often requiring new OpenType layout features and a redesign of existing fonts and typography rendering engines.[2][6][7][8]
In Windows 10, major refactoring work was done to implement a generalized shaping model, the Universal Shaping Engine (USE). This engine is directly based on glyph properties defined in the Unicode standard, in the hope that any complex script with a suitable font would be supported without the time and effort required to create a dedicated shaping engine.[3]
USE builds on a generalized "universal cluster model" developed for the Indic scripts, which models a superset of human writing systems. The engine classifies each character of a complex script into several categories, base classes and subclasses. For example, a provisional Indic classification includes general, syllabic and positional categories, further divided them into base (number, consonant, tone letter, dependent vowel, etc.), base vowel (independent vowel), number (Brahmi joining number), final, medial, and modifier consonants, medial consonants, as well as top, bottom, left and right consonants and vowels. Unicode symbol strings are converted into collection of USE classes using well-defined rules, making glyph composition a standard procedure and allowing inter-character interactions not possible with current language features defined in OpenType specifications.[2]
The Universal Shaping Engine was presented at the OpenType Developer Meeting in 2014; a compatible approach has also been implemented by the open source HarfBuzz text shaper. In Windows 10, the USE handles a total of 45 complex scripts: Balinese, Batak, Brahmi, Buginese, Buhid, Chakma, Cham, Duployan, Egyptian Hieroglyphs, Grantha, Hanunoo, Javanese, Kaithi, Kayah Li, Kharoshthi, Khojki, Khudawadi, Lepcha, Limbu, Mahajani, Mandaic, Manichaean, Meitei Mayek, Modi, Mongolian, N’Ko, Pahawh Hmong, Phags-pa, Psalter Pahlavi, Rejang, Saurashtra, Sharada, Siddham, Sinhala, Sundanese, Syloti Nagri, Tagalog, Tagbanwa, Tai Le, Tai Tham (extremely imperfectly), Tai Viet, Takri, Tibetan, Tifinagh and Tirhuta.
Versions
Although Uniscribe has been available since Windows 2000, new versions of Uniscribe provided more functions to the system, namely, support for other writing systems. An earlier update of it supports the display of Arabic and Hebrew, then Thai and Vietnamese. Since Windows XP, more South Asian and Assyrian alphabets are supported.
See also
References
- https://msdn.microsoft.com/en-us/goglobal/bb688099.aspx#W10>
- John Hudson (10 May 2016). "Making fonts for the Universal Shaping Engine" (PDF). Tiro Typeworks. Retrieved 2020-06-09.
- Windows shapes the world’s languages - Andrew Glass
- "RE: Symbola font (was: James Kass and Code2000 font)". Retrieved 2011-01-29.
- Public Review Issue #37, Proposal on Clarification and Consolidation of the Function of ZERO WIDTH JOINER in Indic Scripts, Unicode Consortium
- Problems for Indic typography in current OpenType Layout implementations - John Hudson
- Fixing Indic2 OpenType Layout - John Hudson
- Beyond shaping and towards a general model of OpenType typography - John Hudson
- Notes
- Uniscribe
- Microsoft Typography
- Uniscribe at MSDN
- Summer Institute of Linguistics. Uniscribe versions