PostScript Japanese Font File Organization Overview Adobe Developer Support Technical Note #5132 25 December 1992 Adobe Systems Incorporated Adobe Developer Technologies 345 Park Avenue San Jose, CA 95110 http://partners.adobe.com/ PN LPS5132
Copyright 1992 by Adobe Systems Incorporated. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior written consent of the publisher. Any software referred to herein is furnished under license and may only be used or copied in accordance with the terms of such license. PostScript is a registered trademark of Adobe Systems Incorporated. All instances of the name PostScript in the text are references to the PostScript language as defined by Adobe Systems Incorporated unless otherwise stated. The name PostScript also is used as a product trademark for Adobe Systems implementation of the PostScript language interpreter. Any references to a PostScript printer, a PostScript file, or a PostScript driver refer to printers, files, and driver programs (respectively) which are written in or support the PostScript language. The sentences in this book that use PostScript language as an adjective phrase are so constructed to reinforce that the name refers to the standard language definition as set forth by Adobe Systems Incorporated. Adobe, the Adobe logo, Lasertalk, PostScript, the PostScript logo are trademarks of Adobe Systems Incorporated which may be registered in certain jurisdictions. Agfa Gevaert, Agfa and ProSet are registered trademarks and StarPlus is a trademark of Agfa Division, Miles, Inc. Apple and LaserWriter are registered trademarks of Apple Computer, Inc. Canon is a registered trademark of Canon, Inc. DEC is a trademark of Digital Equipment Corporation. ITC Zapf Dingbats is a registered trademark of International Typeface Corporation. Linotype-Hell is a trademark of Linotype-Hell AG. *Helvetica, Linotronic and Times are trademarks of Linotype-Hell AG and/or its subsidiaries. Monotype is a trademark of The Monotype Corporation plc registered in certain countries. BBB, Futo Min A101, Futo Go B101, Jun 101 and Ryumin are trademarks of Morisawa Company, Ltd. OKI and Microline are registered trademarks of OKI Electric Industry Company. NEC is a registered trademark of NEC. ColorScript and QMS are registered trademarks of QMS Corporation. Varityper is a registered trademark of Varityper, Inc. UNIX is a registered trademark of UNIX Laboratories. All other brand and product names are trademarks or registered trademarks of their respective holders. This publication and the information herein is furnished AS IS, is subject to change without notice, and should not be construed as a commitment by Adobe Systems Incorporated. Adobe Systems Incorporated assumes no responsibility or liability for any errors or inaccuracies, makes no warranty of any kind (express, implied or statutory) with respect to this publication, and expressly disclaims any and all warranties of merchantability, fitness for particular purposes and noninfringement of third party rights.
Contents List of Figures vii List of Tables ix PostScript Japanese Font File Organization Overview 11 1 Introduction 11 Design Goals of Japanese Font Organization 12 Japanese Font Organization Versions 13 2 Font Organization Version 1.0 14 System Files 14 User Files 15 File System Support Files 16 Encoding Arrays 17 Charstrings 19 PG Fonts 20 Base Fonts 22 Composite Fonts 23 3 Font Organization Version 2.0 28 System Files 30 User Files 30 File System Support Files 30 Encoding Arrays 31 Charstrings 34 PG Fonts 34 Base Fonts 37 Composite Fonts 39 4 Font Organization Version 3.0 42 System Files 44 File System Support Files 44 User Files 45 Encoding Arrays 46 Charstrings 47 PG Fonts 48 Base Fonts 49 Composite Fonts 51 5 Compatibility Issues 53 iii
Testing for the Language Level and the Disk Version 53 Accessing a Base Font 56 Translating Base Font Names 57 Providing Encoding and Charstring Compatibility 60 Creating Zeroslash Variants 62 Adjusting Mapping Differences 63 Appendix A: Comparison between Versions 1 and 2 65 A.1 Character Collections 65 A.2 Base Fonts 67 Adobe Standard Base Fonts 67 Additional Base Fonts 67 JIS 83 Base Fonts 67 JIS 78 Base Fonts 68 Ext Base Fonts 69 Add Base Fonts 70 Zeroslash Base Fonts 71 A.3 Composite Fonts 72 Composite Font Files 72 Composite Font Support Files 73 A.4 System Files 76 PostScript Printer System Files 76 Additional System Files 76 A.5 File System Support Files 77 A.6 User Files 78 A.7 Encoding Files 79 A.8 Charstring files 81 Appendix B: Comparison between Versions 2 and 3 83 B.1 Character Collections 83 B.2 Base Fonts 85 Adobe Standard Base Fonts 85 Additional Base Fonts 85 JIS 83 Base Fonts 85 JIS 78 Base Fonts 86 Ext Base Fonts 86 Add Base Fonts 86 Zeroslash Base Fonts 86 B.3 Composite Fonts 87 Composite Font Files 87 Composite Font Support Files 87 B.4 System Files 89 PostScript Printer System Files 89 Additional System Files 89 iv Contents (25 Dec 92)
B.5 File System Support Files 90 B.6 User Files 91 B.7 Encoding Files 92 B.8 Charstring files 94 Appendix C: Differences in Descendant Fonts 95 C.1 From <face>-h 95 C.2 From <face>-v 96 C.3 From <face>-euc-h 96 C.4 From <face>-euc-v 97 C.5 From <face>-suppa-h 97 C.6 From <face>-suppa-v 97 C.7 From <face>-suppb-hv 98 C.8 From <face>-ext-h 98 C.9 From <face>-ext-v 100 C.10 From <face>-ext-suppa-h 100 C.11 From <face>-ext-suppa-v 101 C.12 From <face>-ext-suppb-hv 101 C.13 From <face>-nwp-h 102 C.14 From <face>-nwp-v 104 C.15 From <face>-add-h 104 C.16 From <face>-add-v 106 C.17 From <face>-add-suppa-h 106 C.18 From <face>-add-suppa-v 107 C.19 From <face>-add-suppb-hv 107 Appendix D: File System Support Utilities Summary 109 D.1 Introduction 109 D.2 Summary Of Commands 110 D.3 Disk Maintenance Commands 112 D.4 Other Disk Maintenance Utility Programs 112 Appendix E: Changes Since Earlier Versions 115 Index 117 Contents v
vi Contents (25 Dec 92)
List of Figures Figure 1 File System Structure of Version 1.0 Organization. 14 Figure 2 File System Structure of Version 2.0 Organization 28 Figure 3 File System Structure of Version 3.0 Organization. 42 Figure 4 Sample Base Font Obtained by getdescendant 57 Figure 5 Zeroslash Characters 62 vii
viii List of Figures (25 Dec 92)
List of Tables Table 1 Glyph Sets and Encodings Supported by Version 1.0 17 Table 2 Glyph Set and Encodings Supported by Version 2.0 31 Table 3 Glyph Set and Encodings Not Supported by Version 2.0 32 Table 4 Glyph Set and Encodings Supported by Version 3.0 46 Table A.1 Comparison of pgfont Files 65 Table A.2 Comparison of Adobe Standard Base Fonts 67 Table A.3 Comparison of Additional Base Fonts 67 Table A.4 Comparison of JIS 83 Base Fonts 67 Table A.5 Comparison of JIS 78 Base Fonts 68 Table A.6 Comparison of Ext Base Fonts 69 Table A.7 Comparison of Add Base Fonts 70 Table A.8 Comparison of Zeroslash Base Fonts 71 Table A.9 Comparison of Composite Font Files 72 Table A.10 Comparison of Composite Font Support Files 73 Table A.11 Comparison of System Files 76 Table A.12 Comparison of Additional System Files 76 Table A.13 Comparison of File System Support Files 77 Table A.14 Comparison of User Files 78 Table A.15 Comparison of Encoding Files 79 Table A.16 Comparison of Charstring Files 81 Table B.1 Comparison of pgfont Files 83 Table B.2 Comparison of Adobe Standard Base Fonts 85 Table B.3 Comparison of Additional Base Fonts 85 Table B.4 Comparison of JIS 83 Base Fonts 85 Table B.5 Comparison of JIS 78 Base Fonts 86 Table B.6 Comparison of Ext Base Fonts 86 Table B.7 Comparison of Add Base Fonts 86 Table B.8 Comparison of Zeroslash Base Fonts 86 Table B.9 Comparison of Composite Font Files 87 Table B.10 Comparison of Composite Font Support Files 87 Table B.11 Comparison of System Files 89 Table B.12 Comparison of Additional System Files 89 Table B.13 Comparison of File System Support Files 90 Table B.14 Comparison of User Files 91 Table B.15 Comparison of Encoding Files 92 Table B.16 Comparison of Charstring Files 94 Table C.1 Descendant Fonts of <face>-h 95 Table C.2 Descendant Fonts of <face>-v 96 ix
Table C.3 Descendant Fonts of <face>-euc-h 96 Table C.4 Descendant Fonts of <face>-euc-v 97 Table C.5 Descendant Fonts of <face>-suppa-h 97 Table C.6 Descendant Fonts of <face>-suppa-v 97 Table C.7 Descendant Fonts of <face>-suppb-hv 98 Table C.8 Descendant Fonts of <face>-ext-h 98 Table C.9 Descendant Fonts of <face>-ext-v 100 Table C.10 Descendant Fonts of <face>-ext-suppa-h 100 Table C.11 Descendant Fonts of <face>-ext-suppa-v 101 Table C.12 Descendant Fonts of <face>-ext-suppb-hv 101 Table C.13 Descendant Fonts of <face>-nwp-h 102 Table C.14 Descendant Fonts of <face>-nwp-v 104 Table C.15 Descendant Fonts of <face>-add-h 104 Table C.16 Descendant Fonts of <face>-add-v 106 Table C.17 Descendant Fonts of <face>-add-suppa-h 106 Table C.18 Descendant Fonts of <face>-add-suppa-v 107 Table C.19 Descendant Fonts of <face>-add-suppb-hv 107 x List of Tables (25 Dec 92)
PostScript Japanese Font File Organization Overview 1 Introduction This document describes the structure, contents, and use of the files in the file system, and the glyph sets and the encodings for versions 1.0, 2.0 and 3.0 of the Japanese font organization used by Adobe Systems, Inc. on PostScript printers. These descriptions include comparisons between the Japanese font contents in all versions, indicating the differences. Potential incompatibilities between different versions are discussed, and significant issues are addressed by presenting a recommended solution for each case with sample programs. Because of their size, the Japanese fonts are stored on hard disks on most PostScript printers, although the ability to store the PostScript language outline data in ROM has been introduced with the Level 2 PostScript printer products. The file organization of the Japanese fonts has been evolving, both on the printer hard disk and in ROM, as Adobe continues to provide better performance, more flexibility, and smaller storage requirements. In the process, versions 1.0, 2.0 and 3.0 of the font file organization have evolved. All versions currently coexist in the market and are used by various end users. It is assumed that readers of this document are software or font developers. The objective is to provide sufficient information for those developers to access the fonts stored on the storage media, as well as to store and erase other font data files on the disk, independent of other font files. It is suggested that all readers become familiar with overall evolution of the organization from version 1.0 to version 3.0. Sections 2 through 4 describe each organization in detail. Application software developers may wish to refer to Section 5, Compatibility Issues, as well as Appendicies A through C. These parts of the document describe differences between the organizations by comparative lists of font file names and PostScript codes. In addition, information as to how campatibility issues can be resolved in application software and PostScript drivers is provided, including how to avoid performing findfont on base fonts. Appendix C would be of interest to developers whose application software supported the version 1.0 organization and would like to support newer versions. Font developers may wish to note subsections of Section 5, so that after-market font products will achieve 11
the compatibility between Level 1 and Level 2 PostScript printer products. Appendix D is added for those who wish to interactively work with Japanese composite fonts using various tools such as Lasertalk. This document is published to provide current information on the contents of the Japanese font disks, up to version 3.0. However, the Japanese font organization may be revised in the future in order to provide additional benefits. Furthermore, since hard disks are optional on some new Level 2 PostScript printers, there are diskless printers in the market with fonts stored in ROM. (Currently, all printers have a SCSI port for optionally connecting an external PostScript-formatted hard disk.) Therefore, it is recommended that software and font developers design their software and fonts carefully so as not to depend on the existence or contents of any particular files in a particular Japanese font organization. It is advantageous for developers to be independent of the font organization designed and provided by Adobe. Please note that the printer hard disk is an integral part of many PostScript printers, regardless of which version. It should not be used as a general storage device or for other purposes beyond those discussed in this document. Developers should not modify or erase files on these hard disks except for those which they have installed or customized. Also note that previous descriptions of Japanese font-related PostScript printer products in some previous technical documents and books published by Adobe Systems are based on version 1.0 of the Japanese font organization scheme. A few PostScript language programs contained in these documents are incompatible with the version 2 and 3 organizations. The following is the list of such publications: PostScript Language Tutorial and Cookbook, Japanese version Technical Note #5076, Gaiji in PostScript Printers Technical Note #5077, Tutorial on Composite Fonts 1.1 Design Goals of Japanese Font Organization It is helpful to understand the design goals of Japanese font organization implemented on PostScript printers. Japanese fonts are used in a variety of computer environments. Since PostScript is a device-independent page description Language, its Japanese fonts contain several encodings ( Japanese Industrial Standard or JIS, Shift JIS, and EUC) and several glyph sets (JIS 83, NEC JIS, Fujitsu JIS and others) to facilitate compatibility and ease of use across a variety of platforms. 12 PostScript Japanese Font File Organization Overview (25 Dec 92)
1.2 Japanese Font Organization Versions To date there have been three major versions of the Japanese font organization. Sections 2 through 4 describe versions of the organizations in detail. The primary characteristics of each of these versions are as follows: Version 1.0 Version 1.0 was implemented on PostScript printer hard disks of NEC, Apple, and Fujitsu first, released between 1987 and 1990. (DEC KPS-4011 was the first PostScript printer which could process Japanese fonts, but it was a diskless printer and initially contained only bitmap fonts.) This version supports numerous combinations of glyph sets and encodings known to exist in the Japanese language environment. In its implementation, version 1.0 used pgfonts, a collection of character descriptions, and derived fonts, a logical collection of pointers to character descriptions that are stored in pgfonts. About 20 pgfonts and approximately 600 to 800 derived fonts were included on these disks. An intermediate version 1.0 organization was produced with improvements in the number of character descriptions, most notably improved half-width katakana and roman characters. Version 2.0 Version 2.0 was also implemented in the form of PostScript printer hard disks. Examples of the PostScript printer products with version 2.0 disks are Oki, Epson, and Canon. This version substantially reduced the number of derived fonts and simplified other files and file structures. Pgfonts were divided into two directories in the file system, pgfonts and pgfsupp. Unused combinations of glyph sets and encodings were eliminated. Several intermediate versions of this organization, including a version implemented on printer hard disks containing five typefaces, were produced. The JIS 78 glyph set and encodings were eliminated in this version. Version 3.0 This is the font organization for the current generation of Japanese PostScript Level 2 printers. Taking advantage of new Level 2 operators, various files in the organization, such as font and encoding files, are defined as Named Resource objects. Refer to PostScript Language Reference Manual, Second Edition, Section 3.9, Named Resources. Each of these three versions are described in more detail in the following sections. It must be emphasized again that there will inevitably be changes in future versions of the Japanese font organization and that we are striving for backward compatibility. Therefore, it is recommended that developers follow the guidelines that are provided by Adobe Systems to maintain compatibility as new versions are released and modifications are made in the future. 1 Introduction 13
2 Font Organization Version 1.0 Version 1.0 of the Japanese font organization was introduced in 1987. It was conceived to be a disk-based file system organization but was first implemented on DEC KPS-4011 which had bitmap-only fonts in RAM. It was followed by NEC PR-602 PS in late 1988 and Apple LaserWriter NTX-J in early 1989 with two Japanese typefaces, Ryumin -Light and GothicBBB -Medium. They supported both JIS 78 and JIS 83 glyph sets. Version 1.0 was also implemented on imagesetters, such as Agfa-Gevaert, Linotype-Hell, Monotype, and Varityper. The file system structure was formed as illustrated below: Figure 1 File System Structure of Version 1.0 Organization. root Sys Usr FS fonts fsupp pgfonts encodings charstrings The Sys directory contains system program files. Some of the files are required to support the disk-based file system. Others are specific to supporting the Japanese font organization scheme. The Usr directory has files which support a user s local environment. The FS directory contains various file system support files, which may or may not be essential to support the Japanese font organization. The fonts directory is where all font data files are stored. They include base fonts and composite fonts. The majority of the base fonts found in this directory are derived fonts which are fonts whose character descriptions reside in pgfonts. These pgfonts are contained in the pgfonts directory. The fsupp directory contains a collection of files that are part of the Japanese composite font extension. These files provide enhancements in representation of a PostScript language composite font. The files containing encoding arrays are stored in the encodings directory. Each file in this directory represents one encoding supported by the font organization. The charstrings directory contains files which have charstrings dictionaries. 2.1 System Files The following files contain system programs. Their functions are undocumented unless described below. Application software and font developers should not be concerned with any of these files. By referring to the system files, application software and font products will become dependent upon a 14 PostScript Japanese Font File Organization Overview (25 Dec 92)
specific version of the Japanese font organization and the specific contents of the files. Such dependencies should be avoided in order to ensure compatibility with future PostScript output devices. PostScript Printer System Files The following files can be found on any hard drive formatted with the Post- Script language, regardless of whether the disk contains Japanese fonts. The PostScript interpreter requires these files in order to support the disk file system. Downloaders for the Japanese fonts will not download desired fonts onto a printer hard disk unless these files are present. Therefore, these files should not be overwritten, altered, or erased. Sys/Root0 Sys/Root1 Sys/AllocMap Sys/Directory Sys/Start Sys/Idle Sys/Start contains the standard file that will be executed as the first PostScript language job after the printer has been started or restarted. There may be a file called Sys/Bootlist on imagesetters. Depending on OEMs and their models, Sys/Bootlist may be the file which is executed first when an imagesetter is booted, instead of Sys/Start. It is possible that Sys/Bootlist invokes PostScript language programs contained in other files, including Sys/Start. There may be other system files on imagesetters. Those files are not discussed in this document, because they are OEM-specific files. Additional System Files These are additional system files that are specific to the Japanese font disk product. They should not be overwritten, altered, or erased. Sys/PGFInit Sys/PGFKeys Sys/Info 2.2 User Files The following files contain user programs for local system administration. Usr/Start Usr/Idle Usr/IdleHint 2 Font Organization Version 1.0 15
Usr/Start is provided, so that users may customize the behavior of the Post- Script interpreter. PostScript language code in this file is invoked by Sys/Start after it has performed its actions. The code in Usr/Start is executed outside server loop; consequently any changes to VM made by the code will stay until the printer is restarted. An example of this file s use would be to load encoding vectors and charstring dictionaries and to perform findfont on composite fonts that are likely to be used according to user s needs. The default Usr/Start contains the following PostScript language code: /Ryumin-Light-H findfont pop /Ryumin-Light-V findfont pop The functions of the other programs are undocumented. 2.3 File System Support Files The following files in the FS directory contain file system support programs. They are mostly used by advanced users to maintain the file system, including backup of the printer hard disk, checksum generation on the disk, listing files on the disk, and so forth. FS/FileUtils FS/CopyDisk FS/LoadDisk FS/MergeKeys FS/KeyMaker_Std FS/FileUtils contains various procedures which can be utilized as basic means to maintain the disk-based file system. To use these utilities, after connecting to a PostScript printer through utility software such as Lasertalk, execute: (FS/FileUtils) run then Unix-like commands can be invoked. Refer to Appendix D for detailed information about the utility procedures. FS/CopyDisk provides a means to copy all the files listed in this document from one disk to another disk that is daisy-chained through the SCSI port. It uses the FileUtils program to provide the operations it requires. This utility may be used after su command (See Appendix D.) is executed. Also note that the utility does not copy composite font support files. FS/LoadDisk allows downloading of files directly to a named file on disk. It takes a filename string as an argument, and reads the data from standard input expressed as %stdin. The functions of FS/MergeKeys and FS/KeyMaker_Std are undocumented and unsupported. 16 PostScript Japanese Font File Organization Overview (25 Dec 92)
2.4 Encoding Arrays The files containing encoding arrays are stored in the encodings directory, where they are accessed by the procedure findencoding. The complete JIS 83 glyph set is provided in Adobe Systems Japanese font products as well as the following common glyph sets:83pv (JIS 83-based partially vertical character set), NWP (New word processor character set used by NEC), and Ext (Extended character set used by NEC 9801 series personal computers). Another glyph set, Add (Additional character set used by Fujitsu FMR personal computers), was added in the later release of the same font organization. The encodings provided in any Japanese fonts are JIS, Shift-JIS, Roman-Kana Shift-JIS (RKSJ), and Extended Unix Code (EUC). The following combinations are supported by version 1.0: Table 1 Glyph Sets and Encodings Supported by Version 1.0 Glyph Set JIS 83 83pv JIS 78 Ext NWP Add Encoding JIS, Shift-JIS, RKSJ, EUC RKSJ JIS, Shift-JIS, RKSJ, EUC JIS, Shift-JIS, RKSJ, EUC JIS JIS, RKSJ The encoding files are used mostly in the implementation of the resident fonts supplied by Adobe Systems and are therefore considered implementation-dependent. The encodings provided on any font organization scheme can be accessed by the composite font extension operator findencoding. The encoding files supporting Japanese fonts include two types of PostScript language programs: one maps the character codes to the charstring names at the base font level, and one maps the row codes of double-byte glyph tables to the descendant fonts indicated by integers. 2 Font Organization Version 1.0 17
Base Font Level Encoding Files The following files refer to the character codes at the base font level. encodings/notdefencoding encodings/standardencoding encodings/isolatin1encoding encodings/hromanencoding encodings/katakanaencoding encodings/hiraganaencoding encodings/hankakuencoding encodings/jisencoding encodings/shiftjisencoding encodings/pchirakataencoding encodings/pcsymencoding Composite Font Level Encoding Files The following files are used at the composite font level. The word Encoding is preceded by a prefix CF in all encoding file names associated with composite fonts. encodings/jis1-88-cfencoding encodings/jis1-97-cfencoding encodings/jis12-88-cfencoding encodings/jis12-97-cfencoding encodings/shiftjis-a-cfencoding encodings/shiftjis-b-cfencoding encodings/shiftjis1-88-cfencoding encodings/shiftjis12-88-cfencoding encodings/extjis1-88-cfencoding encodings/extjis1-97-cfencoding encodings/extjis12-88-cfencoding encodings/extjis12-97-cfencoding encodings/extshiftjis-a-cfencoding encodings/extshiftjis1-88-cfencoding encodings/extshiftjis12-88-cfencoding The files listed above are commonly found on any printer hard disks on which the Japanese font organization version 1.0 has been implemented. Optional Base Font Level Encoding Files encodings/heurocharencoding encodings/hroman83pvencoding encodings/eucencoding encodings/sjencoding 18 PostScript Japanese Font File Organization Overview (25 Dec 92)
Among intermediate releases of version 1.0 disks, any of the above files may optionally exist. It is emphasized again that developers should not depend upon the existence of any of the files listed above. Optional Composite Font Level Encoding Files encodings/add-jis12-88-cfencoding encodings/add-sj12-88-cfencoding encodings/add-sj-b-cfencoding encodings/euc12-88-cfencoding encodings/ext-jis12-88-cfencoding encodings/ext-sj-a-cfencoding encodings/ext-sj12-88-cfencoding encodings/jis12e-88-cfencoding encodings/jis12e-97-cfencoding encodings/nwp-jis12-88-cfencoding encodings/shiftjise-b-cfencoding encodings/shiftjis12e-88-cfencoding encodings/sj12-88-cfencoding encodings/sj-a-cfencoding encodings/sj-b-cfencoding 2.5 Charstrings The charstrings files are the dictionaries that define the key-value paires where the keys are the character names and the values are the corresponding character descriptions. These files are accessible by the PostScript language procedure findcharstrings. The files listed below are commonly found on any printer hard disks on which the Japanese font organization version 1.0 has been implemented. charstrings/charmap charstrings/hromancharstrings charstrings/hankakucharstrings charstrings/hiraganacharstrings charstrings/jischarstrings charstrings/jiszscharstrings charstrings/katakanacharstrings charstrings/pchirakatacharstrings charstrings/pcsymcharstrings charstrings/shiftjischarstrings charstrings/shiftjiszscharstrings 2 Font Organization Version 1.0 19
The following files may optionally exist on printers with the version 1.0 organization implemented on their hard disks. charstrings/euccharstrings charstrings/eucjischarstrings charstrings/heurocharstrings charstrings/hroman2charstrings charstrings/hroman83pvcharstrings charstrings/hroman83pv2charstrings charstrings/sjcharstrings 2.6 PG Fonts A pgfont, or primogenital font, is a PostScript language object which contains a set of character descriptions in the PostScript language. After a pgfont is created, a derived font can be established from it to address the needs of various Japanese language operating environments. Many glyph sets used today have common subsets. Using this as a basis, Adobe has organized a group of glyph collections that can be used to build these subsets and then build the more specialized sets required for various markets. Further, glyphs have been grouped according to their typographical characteristics and their logical relationships. This greatly reduces the number of duplicate glyphs among collections and makes each collection independent. A pgfont consists of three files: a file containing a pgfont dictionary which plays a similar role to a font dictionary, a file containing CharStrings array (with extensions _CSA) which is a collection of character descriptions for the characters in a pgfont, and a file containing CharOffsets dictionary (with _COD extensions) that has entries giving information about the positions of the character descriptions in the collection. The CharOffsets dictionary is required for generating derived fonts, but not for scan-conversion of glyphs. Therefore, some printer products may omit this file in order to save space. The following glyph collections have been organized into separate sets: Zenkaku (full-width) generic line-drawing glyphs Hankaku (half-width) generic line-drawing glyphs Zenkaku Japanese Industrial Standard (JIS) 83 Level 1 Kanji and symbols Zenkaku JIS 83 Level 2 Kanji Zenkaku Roman Hankaku Roman 20 PostScript Japanese Font File Organization Overview (25 Dec 92)
Hankaku Kana Zenkaku JIS 78 Kanji glyphs Extended JIS 78 zenkaku Kanji for compatibility with other common glyph sets Extended zenkaku Roman Extended hankaku Roman PGFont Files The Japanese font organization version 1.0 includes the following pgfont files. The structure and function of pgfonts are undocumented and proprietary. pgfonts/generic::stdwidth pgfonts/generic::hankaku pgfonts/<face>::jis-1 pgfonts/<face>::jis-2 pgfonts/<face>::roman pgfonts/<face>::hroman pgfonts/<face>::hkana pgfonts/<face>::jis78 pgfonts/<face>::ext pgfonts/<face>::roman-ext pgfonts/<face>::hroman-ext PGFont Support Files The following files are the pgfont support files which are included in version 1.0. pgfonts/generic::stdwidth_csa pgfonts/generic::stdwidth_cod pgfonts/generic::hankaku_csa pgfonts/generic::hankaku_cod pgfonts/<face>::jis-1_csa pgfonts/<face>::jis-1_cod pgfonts/<face>::jis-2_csa pgfonts/<face>::jis-2_cod pgfonts/<face>::roman_csa pgfonts/<face>::roman_cod pgfonts/<face>::hroman_csa pgfonts/<face>::hroman_cod pgfonts/<face>::hkana_csa pgfonts/<face>::hkana_cod pgfonts/<face>::jis78_csa pgfonts/<face>::jis78_cod 2 Font Organization Version 1.0 21
pgfonts/<face>::ext_csa pgfonts/<face>::ext_cod pgfonts/<face>::roman-ext_csa pgfonts/<face>::roman-ext_cod pgfonts/<face>::hroman-ext_csa pgfonts/<face>::hroman-ext_cod 2.7 Base Fonts The majority of the fonts in the Japanese font organization version 1.0 are derived fonts, whose character descriptions reside in pgfonts. The printer hard disk is delivered with these derived fonts already created and stored in the fonts directory. The exceptions to this in the Japanese font organization versions 1.0 are PCHelvetica*, PCTimes*-Roman and NotDefFont, which also reside in the fonts directory of the file system. In version 1.0, when a composite font is found by a program, the descendant base fonts are also found, and the files on the disk containing the base fonts are subsequently opened. It is possible to perform findfont on base fonts directly, although this method of accessing base fonts is strongly discouraged. Such use will create unnecessary dependencies between a PostScript language programs and the font organization version, and dependencies between software or font products and particular PostScript printers. Furthermore, if the above method is used in your software, it may not work with future font and VM organizations. It is recommended to access fonts at the composite font level. See section 2.8. Adobe Standard Base Fonts The following files contain base fonts that are part of the standard Adobe Japanese font set. fonts/notdeffont fonts/pchelvetica fonts/pctimes-roman fonts/<face>.hankaku fonts/<face>.hiragana fonts/<face>.katakana fonts/<face>.roman fonts/<face>.wp-symbol The later releases of printer hard disks under the version 1.0 organization have the following additional standard base fonts. fonts/<face>.oubun fonts/<face>.oubun-add 22 PostScript Japanese Font File Organization Overview (25 Dec 92)
Additional Base Fonts The following files contain fonts that are used as components of composite fonts for the Japanese fonts. fonts/<face>.ext.hiragana fonts/<face>.ext.katakana fonts/<face>.suppk JIS 83 Base Fonts Refer to Appendix A for the list of base font files for this category. JIS 78 Base Fonts Refer to Appendix A for the list of base font files for this category. Ext Base Fonts Refer to Appendix A for the list of base font files for this category. Add Base Fonts These base fonts have been added in the later releases of printer hard disks under version 1.0 of the Japanese font organization. Refer to Appendix A for the list of base font files for this group. 2.8 Composite Fonts Composite fonts provide the final step in the mapping between multi-byte (double-byte in case of Japanese fonts) character codes and glyphs. The glyph set is determined by the derived fonts which are generated from pgfonts, and the composite font determines the font and glyph that will be selected from scanned character codes in the show string. It is strongly recommended that developers access fonts at the composite level by performing findfont on any desired composite fonts for current and future compatibility between software or fonts and PostScript printers. 2 Font Organization Version 1.0 23
Composite Fonts Files The following are composite font files found on disks under organization version 1.0. These files are executed when the corresponding composite font is not found in FontDirectory by the findfont operator. fonts/mincho-pc-hiragana fonts/mincho-pc-katakana fonts/usergaiji fonts/<face>-h fonts/<face>-v fonts/<face>-euc-h fonts/<face>-euc-v fonts/<face>-sj-h fonts/<face>-sj-v fonts/<face>-rksj-h fonts/<face>-rksj-v fonts/<face>-78-h fonts/<face>-78-v fonts/<face>-78-euc-h fonts/<face>-78-euc-v fonts/<face>-78-sj-h fonts/<face>-78-sj-v fonts/<face>-78-rksj-h fonts/<face>-78-rksj-v fonts/<face>-83pv-rksj-h fonts/<face>-ext-h fonts/<face>-ext-v fonts/<face>-ext-euc-h fonts/<face>-ext-euc-v fonts/<face>-ext-sj-h fonts/<face>-ext-sj-v fonts/<face>-ext-rksj-h fonts/<face>-ext-rksj-v fonts/<face>-nwp-h fonts/<face>-nwp-v 24 PostScript Japanese Font File Organization Overview (25 Dec 92)
Optional Composite Font Files The later releases of printer hard disks under the Japanese font organization version 1.0 include the following additional composite fonts. These composite fonts were not supported in the earlier releases. Developers should be cautious in using these composite fonts and not depend on their existence. fonts/<face>-add-h fonts/<face>-add-v fonts/<face>-add-euc-h fonts/<face>-add-euc-v fonts/<face>-add-sj-h fonts/<face>-add-sj-v fonts/<face>-add-rksj-h fonts/<face>-add-rksj-v fonts/<face>-83pv-rksj-v Japanese font organization version 1.0 provides an almost complete collection of composite fonts with many combinations of glyph sets and encodings. For any particular character set (JIS 83, JIS78, etc.), several mappings are possible. For each possible character set and mapping, both horizontal and vertical fonts are provided. A standard version of the composite font and a fast-loading version are provided in all cases. The standard representation of a PostScript composite font consists of a dictionary object. This dictionary has several entries including FDepVector which consists of an array where several calls of the findfont procedure occur. It takes considerable time to perform a findfont on a composite font such as Ryumin-Light-H with about 80 occurrences of findfont in its FDepVector, because each findfont has to open a file, read the file, and close it. The solution to this overhead is to transform the representation of the composite font, when the descendant fonts are derived fonts. All the inner findfont calls are done in-line and no files need to be opened in this representation. The transformation is accomplished by combining the descriptions for both the horizontal and vertical versions of a composite font into one file, and accessing it by both. 2 Font Organization Version 1.0 25
Composite Font Support Files These fast composite font files are stored in the fsupp directory. All the composite font files referred to should be in the fonts directory. The following composite font support files are provided: fsupp/<face> fsupp/<face>-h fsupp/<face>-v fsupp/<face>-euc fsupp/<face>-euc-h fsupp/<face>-euc-v fsupp/<face>-sj fsupp/<face>-sj-h fsupp/<face>-sj-v fsupp/<face>-rksj-hv fsupp/<face>-78 fsupp/<face>-78-h fsupp/<face>-78-v fsupp/<face>-78-euc fsupp/<face>-78-euc-h fsupp/<face>-78-euc-v fsupp/<face>-78-sj fsupp/<face>-78-sj-h fsupp/<face>-78-sj-v fsupp/<face>-78-rksj-hv fsupp/<face>-ext fsupp/<face>-ext-h fsupp/<face>-ext-v fsupp/<face>-ext-euc fsupp/<face>-ext-euc-h fsupp/<face>-ext-euc-v fsupp/<face>-ext-sj fsupp/<face>-ext-sj-h fsupp/<face>-ext-sj-v fsupp/<face>-ext-rksj-hv fsupp/<face>-nwp fsupp/<face>-nwp-h fsupp/<face>-nwp-v 26 PostScript Japanese Font File Organization Overview (25 Dec 92)
Optional Composite Font Support Files The later releases of printer hard disks under the Japanese font organization version 1.0 include the following additional composite font support files. fsupp/<face>-add fsupp/<face>-add-h fsupp/<face>-add-v fsupp/<face>-add-euc fsupp/<face>-add-euc-h fsupp/<face>-add-euc-v fsupp/<face>-add-sj fsupp/<face>-add-sj-h fsupp/<face>-add-sj-v fsupp/<face>-add-rksj-hv fsupp/<face>-83pv-rksj-hv 2 Font Organization Version 1.0 27
3 Font Organization Version 2.0 Version 2.0 of the Japanese font organization was released in July 1990. It was implemented as a disk-based file system organization on printer hard disks of such products as Oki Microline 801PS, Epson LP7500-PS, QMS ColorScript 100 model 10J and 40J, and Canon PS-IPU with two standard Japanese typefaces, Ryumin-Light and GothicBBB-Medium. Disks for some models of the above printers are shipped with three additional typefaces: FutoMinA101 -Bold, FutoGoB101 -Bold, and Jun101 -Light. Furthermore, Additional glyph sets, as well as accompanying encodings, composite fonts, and base fonts are included as a standard subset of the organization. Disks on some of the imagesetters and stand-alone RIP products were also upgraded to version 2.0. There are various models with the version 2.0 disks in use including ones that contain the above five typefaces, such as Agfa-Gevaert s ProSet series and Star Plus SX/J, Linotype-Hell s Linotronic* series and PS RIP 40-J, Monotype s ExpressMaster series and PS4000, Varityper s 5500 series and RIP Series 4000, and others. The file system was restructured as illustrated below: Figure 2 File System Structure of Version 2.0 Organization root Sys Usr FS fonts fsupp pgfonts pgfsupp encodings charstrings Alterations The disk organization of version 2.0 was the result of extensive consultation with Japanese ISVs and OEMs regarding font usage in the desktop publishing (DTP) and word processing (WP) markets. The version 2.0 organization is designed to meet the requirements of the Japanese market more completely and efficiently while reducing the amount of disk storage used. There are a number of important market requirements that the version 2.0 disk organization is designed to accommodate. There is a great need for aftermarket, downloadable fonts. Version 2.0 provides a more logical organization of glyph collections, which makes the process of creating downloadable fonts for different markets easier. The disk structure has also been simplified to meet the need for large numbers of fonts required by the market. This simplification increases the efficiency of the disk's operation. 28 PostScript Japanese Font File Organization Overview (25 Dec 92)
The experience of Adobe's ISVs and OEMs has shown that some of the combinations of glyph sets and encodings provided by the original disk were not used. These combinations have been eliminated, again resulting in greater simplicity, while still meeting market requirements. Impact of Changes None of the changes affect user programs that use the extensions according to Adobe s recommended programming practices. The composite font extensions to the PostScript language do not change in version 2.0. PostScript language programs that access individual base fonts or the files that contain these fonts are affected by changes in organization. They may fail because the fonts have been renamed or eliminated. Many test programs used by Adobe's OEM customers accessed base fonts in this way. These programs need to be rewritten. In general, programs should not access base fonts by referring to their names. See Sections 5.2 and 5.3 for the proper methods of accessing a base font. Adobe Systems is not aware of any commercial applications using the eliminated encodings. The few affected ISVs have the updates to their software that do not require the old encodings. The Japanese version of PostScript Language Tutorial and Cookbook contains examples that use the encodings eliminated in version 2.0. Eventually this book will be supplemented by a document which provides appropriate information consistent with the latest set of encodings provided by Adobe Systems. The changes made in the Japanese font disk structure are the results of reevaluating the structure from many aspects. Changes were not made for system files, utility programs, or other support files. The changes are described in detail in the sections below. They can be summarized as follows: The organization of glyph collections has been changed to provide more logically structured collections which better satisfy the requirements of various markets. The pgfont organization has been changed. Some pgfonts have been added, renamed, or deleted. A large number of base fonts have been eliminated, due to applying an adaptation of Fast Composite Font technique to the entire Japanese composite font. (See Section 2.5.) The names of remaining base fonts have been changed to reflect the new glyph organization, while duplicate base fonts have been eliminated from the disk. 3 Font Organization Version 2.0 29
Composite font files for unused combinations of glyph sets and encodings have been eliminated. Some encoding and charstring files have been renamed or eliminated. Some of the font data from the original disk has been improved. The following sections describe the version 2.0 organization of the Japanese font disk. For complete comparisons of the disk contents between versions 1.0 and 2.0, see the Appendices. 3.1 System Files The following files contain system programs. Application software developers should not be concerned with the system files. It is recommended that users and developers not modify or erase any of these files. PostScript Printer System Files No changes have been made to these files from version 1.0. Sys/Root0 Sys/Root1 Sys/AllocMap Sys/Directory Sys/Start Sys/Idle Additional System Files No changes have been made to these files from version 1.0. Sys/PGFInit Sys/PGFKeys Sys/Info 3.2 User Files No changes have been made to these files from version 1.0. Usr/Start Usr/Idle Usr/IdleHint 3.3 File System Support Files Some changes were made from version 1.0, but there are no differences in the effects or functionality of these files. 30 PostScript Japanese Font File Organization Overview (25 Dec 92)
FS/FileUtils FS/CopyDisk FS/LoadDisk FS/MergeKeys FS/KeyMaker_Std 3.4 Encoding Arrays The version 2.0 font organization supports the complete JIS 83 glyph set in its Japanese font products as well as the common glyph sets; 83pv, NWP, Ext, and Add. The encodings provided in any Japanese fonts with version 2.0 are JIS, Roman-Kana Shift-JIS (RKSJ), and Extended Unix Code (EUC). The Shift-JIS (SJ) encoding is not provided in any of the fonts with version 2.0 organization, although it is provided on the disk and is therefore available to the findencoding operator. See Section 5.4 for a description of how compatibility can be obtained. As noted above, composite fonts using certain encodings such as Shift-JIS have been removed; however, there are still separate encodings provided on disk as files in the encodings directory. Some encodings from version 1.0 have been eliminated in version 2.0, others have been renamed, while others have been added. Consult Appendix A for a comparison of existing files between version 1.0 and 2.0. The following combinations are supported by version 2.0: Table 2 Glyph Set and Encodings Supported by Version 2.0 Glyph Set JIS 83 83pv NWP Ext Add Encoding JIS, EUC, RKSJ RKSJ JIS JIS, RKSJ JIS, RKSJ 3 Font Organization Version 2.0 31
These are the combinations that are not supported by version 2.0: Table 3 Glyph Set and Encodings Not Supported by Version 2.0 Glyph Set JIS 83 JIS 78 Ext Add Encoding SJ JIS, EUC, SJ, RKSJ EUC, SJ EUC, SJ The encoding files are used mostly in the implementation of the resident fonts supplied by Adobe Systems and are therefore considered implementation-dependent. The encodings on any disk can be accessed by the composite font extension operator findencoding (or by findresource in Level 2), but accessing the encodings in this manner will bind the PostScript language program to that particular version of the disk. See Section 5.4 for information on how to work around this potential problem. The encoding files supporting Japanese fonts include two types of PostScript language programs. One maps the row codes of the glyph tables to the descendant fonts at the composite font level, and the other maps the glyph codes to the charstring names at the base font level. The encoding files in version 2.0 have the following differences from those in version 1.0. The EUC fonts have their own base fonts in version 2.0, while they shared the same base fonts as the JIS encoding in version 1.0. This eliminates use of the 9/7 mapping for the EUC fonts. Instead, the 8/8 mapping is used. The encoding files for the 9/7 mapping were all eliminated, and two new encoding files, EUCEncoding and EUC12-88-CFEncoding have been added. All the encoding file names with ShiftJIS have been renamed simply SJ. All the encoding files for the JIS encoding have been combined to include both JIS level 1 and 2, for each of the glyph sets. HEuroCharEncoding has been eliminated. 32 PostScript Japanese Font File Organization Overview (25 Dec 92)
As a result of the changes described above, version 2.0 now has the following encoding files. Base Font Level Encoding Files encodings/notdefencoding encodings/standardencoding encodings/isolatin1encoding encodings/hromanencoding encodings/katakanaencoding encodings/hiraganaencoding encodings/hankakuencoding encodings/jisencoding encodings/sjencoding encodings/pchirakataencoding encodings/pcsymencoding Composite Font Level Encoding Files encodings/jis12-88-cfencoding encodings/sj12-88-cfencoding encodings/sj-a-cfencoding encodings/sj-b-cfencoding encodings/ext-jis12-88-cfencoding encodings/ext-sj-a-cfencoding encodings/ext-sj12-88-cfencoding encodings/add-jis12-88-cfencoding encodings/add-sj12-88-cfencoding encodings/add-sj-b-cfencoding encodings/euc12-88-cfencoding encodings/eucencoding encodings/nwp-jis12-88-cfencoding Optional Base Font Level Encoding Files encodings/hroman83pvencoding The above file may be present on hard disks under the version 2.0 organization with some of the printer products. 3 Font Organization Version 2.0 33
3.5 Charstrings The following changes have been made for version 2.0. The dictionaries which included zeroslash characters, JISzsCharStrings and ShiftJISzsCharStrings have been eliminated. The file name extensions ShiftJIS have been simplified to as SJ. As a result, ShiftJISCharStrings has been renamed SJCharStrings. EUCCharStrings file has been added. HEuroCharStrings file has been eliminated. As a result of the changes described above, version 2.0 now has the following charstrings files. charstrings/charmap charstrings/euccharstrings charstrings/hromancharstrings charstrings/hankakucharstrings charstrings/hiraganacharstrings charstrings/jischarstrings charstrings/katakanacharstrings charstrings/pchirakatacharstrings charstrings/pcsymcharstrings charstrings/sjcharstrings Optional Charstrings Files charstrings/hroman2charstrings charstrings/hroman83pvcharstrings charstrings/hroman83pv2charstrings The above file may be present on hard disks under the version 2.0 organization with some of the printer products. 3.6 PG Fonts Version 2.0 of the Japanese font organization logically grouped pgfonts into the following collections of glyph sets. Zenkaku generic line-drawing glyphs Hankaku generic line-drawing glyphs Zenkaku (Full-width) Japanese Industrial Standard (JIS) 83 Level 1 Kanji and symbols 34 PostScript Japanese Font File Organization Overview (25 Dec 92)
Zenkaku JIS 83 Level 2 Kanji Hankaku (Half-width) Roman glyphs Zenkaku Kana Hankaku Kana Zenkaku JIS 78 Kanji glyphs Zenkaku and hankaku face-specific symbols Additional zenkaku Kanji for compatibility with other common glyph sets This provides a number of distinct sets that can be effectively combined to create different glyph sets for different markets. For example, the first four collections above provide a minimal JIS 83 glyph set for applications that do not require compatibility with WP documents. Adding the next two collections produces the basic JIS 83 DTP glyph set, which satisfies the needs of many existing applications. Adding the other collections in the sequence above can produce glyph sets that meet the requirements of different systems vendors and their markets. The changes to the collections are of greatest concern to font designers. Font manufacturers can use this organization to build their fonts, which can in turn be used by font downloaders and utilities which understand this organization. The complete packaging of downloadable fonts is beyond the scope of this document. The pgfonts organization has been changed to accommodate the glyph collection reorganization. This implies that there are some files in the pgfonts directory of the Japanese font file system that have been renamed in order to indicate more clearly to which glyph collection each pgfont belongs. In addition, a new directory in the file system, pgfsupp, has been introduced. This directory holds pgfont support files. There are two types of pgfont support files: CharStringArray files and CharOffsetDirectory files. These files, which resided in pgfonts directory in version 1.0, have been moved to the pgfsupp directory in version 2.0. 3 Font Organization Version 2.0 35