Unifoundry.com

The original Unibetacode package contained standalone programs to convert between the Beta Code encoding method for polytonic Greek and UTF-8 Unicode. Version 2.0 added the libunibetacode library, with functions for converting Greek, Coptic, and Hebrew Beta Code strings to and from UTF-8 strings.

The standalone programs are designed to accept textual input from the Beta Code specification as implemented by the Thesaurus Linguae Graecae (TLG) Project at the University of California, Irvine, and also by the Perseus Project of Tufts University. The programs only handle character encodings, not the formatting codes (superscripts, font size changes, etc.) of the full TLG Beta Code specification.

The focus of Beta Code is an ASCII encoding of classical Greek, and thus the default encoding of the standalone programs is ASCII Beta Code for conversion to Greek UTF-8 Unicode. This package can be of use to those wishing to type polytonic Greek who are already good typists of ASCII characters, as well as to those wishing to convert documents in the TLG, Perseus Project, or similar corpus from Beta Code to Unicode.

The three standalone programs in this package are:

• unibetaprep — Performs preparatory transformation of special numbered code sequences in the full TLG Beta Code specification (for example the Unicode Byzantine Musical Symbols range), converting them to ASCII-specified Unicode code points. These are written as the sequence "\u" followed by the hexadecimal digits corresponding to the Unicode code point. Such Unicode code points can be enclosed in curly brackets ("{…}"). This full Unicode capability is an extension to Beta Code that favors Unicode over the original TLG encoding.
• beta2uni — Converts a Beta Code-encoded file to UTF-8. This program can handle the ordinary ASCII Beta Code encoding of polytonic Greek characters. It also handles Bohairic Coptic letters with the jinma (grave) accent, and Hebrew letters. Unicode character support extends this capability of the Beta Code specification to provide complete Unicode coverage with "\u…" sequences.
• uni2beta — Converts a UTF-8 document that contains polytonic Greek, Bohairic Coptic, and Hebrew letters to Beta Code.

The libunibetacode library contains these top-level functions:

• Conversion from Beta Code to UTF-8:
  • ub_beta2greek converts Greek Beta Code input strings to UTF-8 output strings.
  • ub_beta2coptic converts Coptic Beta Code input strings to UTF-8 output strings.
  • ub_beta2hebrew converts Hebrew Beta Code input strings to UTF-8 output strings.
  • ub_codept2utf8 converts a single Unicode code point input to a UTF-8 output string.
• Conversion from UTF-8 to Beta Code:
  • ub_greek2beta converts Greek UTF-8 input strings to Greek Beta Code output strings.
  • ub_coptic2beta converts Coptic UTF-8 input strings to Coptic Beta Code output strings.
  • ub_hebrew2beta converts Hebrew UTF-8 input strings to Hebrew Beta Code output strings.
  • ub_utf82codept converts a single UTF-8 code point input to a Unicode code point.

To use this library once installed on your system, simply compile a C program with the -lunibetacode flag. There are no additional header files to include. For example, the test program ublibcheck.c located in the test directory in the source distribution can be compiled as follows:

cc ublibcheck.c -o ublibcheck -lunibetacode

The program ublibcheck.c converts Greek, Coptic, and Hebrew Beta Code strings into UTF-8 Unicode, and then converts them back to Beta Code, verifying the round-trip conversion. In addition to being a test program, it thus provides a practical example of how to use the library functions.

The source package contains Beta Code examples in the "examples" directory, which are also used to test the standalone programs once installed on a system.

As a brief introduction to Beta Code, three short Greek Beta Code examples appear below. Examples in Hebrew and Coptic then follow.

Example — Interactive Terminal Input

The commands can be run interactively at a terminal. In the example below, the first line after invoking beta2uni at a terminal shell prompt ("$ ") is the user's typed Beta Code input and the second line is the generated polytonic Greek, which is output directly to the terminal after pressing the Enter or Return key. End the input by typing Control-D on a line by itself, followed by the Enter or Return key. The result can be copied and pasted into a document:

     $ beta2uni
     *gnw=qi seauto/n
     Γνῶθι σεαυτόν
     ^D
     $

      Know thyself —First Delphic maxim

The input ASCII Beta Code letters are lowercase in this example, following the convention of the Perseus Project. Had uppercase letters been used instead (as is the convention of the TLG Project), the Greek output would have been identical.

An asterisk ('*') denotes an uppercase Greek letter; if present, it always appears first, before the letter and any associated breathing mark and accents.

Breathing marks and then accents precede an associated uppercase letter but follow after an associated lowercase letter, reflecting the pre-Unicode typesetting conventions for printed Greek. Here is a longer example, showing more polytonic encoding combinations:

     $ beta2uni
     *(o bi/os braxu/s h( de\ te/xnh makrh/
     Ὁ βίος βραχύς ἡ δὲ τέχνη μακρή
     ^D
     $

      Life [is] short and art long. —Hippocrates

Example — Genesis 1:1

An ampersand ('&') switches to Latin mode and a dollar sign ('$') switches back to the default Greek mode. An iota subscript associated with a long vowel, entered as a verical bar ('|'), always appears last after the vowel and any associated accents. This rule holds whether the vowel is uppercase or lowercase. The next example demonstrates use of this symbol.

Beta Code Encoding Input to beta2uni:

     &Koine Greek (Septuagint):$
     *)en a)rxh=| e)poi/hsen o( *qeo\s to\n ou)rano\n kai\ th\n gh=n.

UTF-8 Unicode Output from beta2uni:

     Koine Greek (Septuagint):
     Ἐν ἀρχῇ ἐποίησεν ὁ Θεὸς τὸν οὐρανὸν καὶ τὴν γῆν.

Note the context-dependent conversion of the ASCII letter 's' in the Beta Code input to small medial (middle) sigma or small final sigma in the UTF-8 Unicode output.

Example — Byzantine Musical Symbols

The TLG Beta Code specification includes special numeric codes in the range "#2000" through "#2245" that map to Unicode Supplementary Multilingual Plane Byzantine Musical Symbols, as just one example of its special numeric codes. The unibetaprep program converts those codes to the Unicode extension to Beta Code that is unique to this Unibetacode package. However, Unicode has become the standard for exchange of textual information. There is no corresponding program to convert a file back to using the special numeric codes of the TLG Beta Code specification, and there are no plans to create such a program.

TLG-specific Special Numeric Code Input to unibetaprep:

     #2070 &(U+1D046)$ *)/ison
     #2071 &(U+1D047)$ *)oli/gon
     #2078 &(U+1D04E)$ *kenth/mata
     #2073 &(U+1D049)$ *petasth/
     #2081 &(U+1D051)$ *)apo/strofos

Unibetacode's Unicode Extension to Beta Code; Output from unibetaprep, Input to beta2uni:

     {\u1D046} &(U+1D046)$ *)/ison
     {\u1D047} &(U+1D047)$ *)oli/gon
     {\u1D04E} &(U+1D04E)$ *kenth/mata
     {\u1D049} &(U+1D049)$ *petasth/
     {\u1D051} &(U+1D051)$ *)apo/strofos

Beta Code Conversion to UTF-8 Unicode; Output from beta2uni:

     𝁆 (U+1D046) Ἴσον
     𝁇 (U+1D047) Ὀλίγον
     𝁎 (U+1D04E) Κεντήματα
     𝁉 (U+1D049) Πεταστή
     𝁑 (U+1D051) Ἀπόστροφος

Example — Genesis 1:1 — Multilingual

The following example shows Genesis 1:1 in several languages. The unencoded and encoded text below is contained in the files "examples/genesis.beta" and "examples/genesis.utf8" in the software package, respectively.

Beta Code Encoding Input to beta2uni:

     &Genesis 1:1$

     &Koine Greek (Septuagint):$
     *)en a)rxh=| e)poi/hsen o( *qeo\s to\n ou)rano\n kai\ th\n gh=n.

     &Hebrew, Letters Only (Standard Beta Code):$
     &300brAsyt brA Alhym2 At hsm1ym2 vAt hArT2$

     &Hebrew, Full Orthography (Unicode Extension to Beta Code):$
     &300b{\u05B0\u05BC}r{\u05B5}As{\u05B4\u05C1\u0596}yt b{\u05B8\u05BC}r{\u05B8\u05B3}A A{\u05B1}l{\u05B9}h{\u05B4\u0591}ym2$
     &300A{\u05B5\u05A5}t h{\u05B7}s{\u05B8\u05BC\u05C1}m1{\u05B7\u0596}y{\u05B4}m2 v{\u05B0}A{\u05B5\u05A5}t h{\u05B8}A{\u05B8\u05BD}r{\u05B6}T2{\u05C3}$

     &Bohairic Coptic:$
     &100*kEN OUARXH A\ F\NOUt QAMIO\ N\TFE NEM PHAOI$

UTF-8 Unicode Output from beta2uni:

     Genesis 1:1

     Koine Greek (Septuagint):
     Ἐν ἀρχῇ ἐποίησεν ὁ Θεὸς τὸν οὐρανὸν καὶ τὴν γῆν.

     Hebrew, Letters Only (Standard Beta Code):
     בראשית ברא אלהים את השמים ואת הארץ

     Hebrew, Full Orthography (Unicode Extension to Beta Code):
     בְּרֵאשִׁ֖ית בָּרֳָא אֱלֹהִ֑ים
     אֵ֥ת הַשָּׁמַ֖יִם וְאֵ֥ת הָאָֽרֶץ׃

     Bohairic Coptic:
     Ϧⲉⲛ ⲟⲩⲁⲣⲭⲏ ⲁ̀ ⲫ̀ⲛⲟⲩϯ ⲑⲁⲙⲓⲟ̀ ⲛ̀ⲧⲫⲉ ⲛⲉⲙ ⲡⲏⲁⲟⲓ

Reference Documents

Unix-style man (manual) pages are listed below. The unibetacode man page describes the Beta Code format that these utilities implement, so read it first. Examples in the "examples" directory in the source code demonstrate practical use.

More information is available in the following documents:

• unibetacode — Reference for Beta Code that this package implements; read this first.
• unibetaprep — Prepare a file containing special TLG numeric sequences for special text to Beta Code with an extension for Unicode code points.
• beta2uni — Beta Code to UTF-8 Unicode conversion program.
• uni2beta — UTF-8 Unicode to Beta Code conversion program.
• libunibetacode — Library of functions for converting between Beta Code and UTF-8 Unicode.

Unibetacode Download

The source tarball and GnuPG signature file can be downloaded at these links:

• unibetacode-2.3.tar.gz (538 kB)
• unibetacode-2.3.tar.gz.sig

Unibetacode Installation

Compiling the package requires a C compiler and the Unix make utility. To compile and install the Unibetacode package on a system with a Unix-style command line interface (GNU/Linux, BSD, macOS, Cygwin, etc.), type these commands in a terminal window:

     ./configure
     make
     make check
     make install
     make clean

The "make check" command will run six tests using sample text files that are in the "examples" directory, and then will run one test program that exercises the functions in the libunibetacode library.

The "make install" command might need to be run as "sudo make install" on your system.

By default, the standalone programs (unibetaprep, beta2uni, and uni2beta) will be installed in "/usr/local/bin", the libunibetacode library will be installed in "/usr/local/lib", and the man (Unix manual) pages will be installed in the appropriate subdirectories under "/usr/local/share/man".

The latest version of Unibetacode uses the GNU libtool package to build static and shared libraries, first introduced in Unibetacode 2.0. Both library files are built together during the "make" operation.

If the target system supports it, the "make install" command will install both the static and the shared versions of the libunibetacode library.

Technical Note: The source code for textual pattern matching is written using the flex lexical analyzer generator. This source code is in the *.l files in the src/progsrc directory. The initial build process converts these files to C sources in the same directory for the distribution tarball, so that the package will build with just a C compiler and the make utility whether or not flex (or a similar Unix/POSIX lex program) is installed on the build system.

License

The software on this site, unless otherwise noted, is released under the terms of the GNU General Public License (GNU GPL) version 2.0, or (at your option) a later version.