PUNCTUATION COMMON TO ALL MODES¶

QUOTATION MARK STYLES¶

Double quotes in the unibetacode package just use ASCII quotation marks in a Beta Code source file; the quotation style is determined by the language mode. This greatly simplifies round-trip conversion between Beta Code and a Unicode UTF-8 document. Double quotation marks must be balanced. The first quotation mark encountered will be interpreted as a left quotation mark for the chosen style, the second will be interpreted as a right quotation mark, and so on. Double quotation marks are converted as follows:

Single quotes are specified explicitly. For Latin and Hebrew, they are specified with a Grave Accent (U+0060) for an opening single quote and an apostrophe (U+0027) for a closing single quote. For Greek and Coptic, they are specified with a "<" for an opening single quote and a ">" for a closing single quote. The use of "<" and ">" for Greek and Coptic was a compromise so that an ASCII apostrophe in Greek mode would render as U+02BC, MODIFIER LETTER APOSTROPHE. As with double quotation marks, the rendering of these characters is dependent on the language mode. Single quotation marks are converted as follows:

ASCII STRING INSERTION¶

This is an extension to standard Beta Code; the TLG specification assigns a different use to '{'. By itself, the use of '{' not followed by a decimal number is deprecated in the TLG specification, so this should avoid some conflict. Curly brackets are not allowed in the string.

SPECIAL UNICODE CHARACTER INSERTION¶

The unibetacode package allows any Unicode code point to be specified in hexadecimal (which is how The Unicode Standard provides them) as a string inside an ASCII '{'...'}' escape sequence. The Unicode hexadecimal code point of one to six digits is preceded by "\u", taking the form "\ux...x". For example, strings such as

can be used to insert the Unicode character U+03D8, GREEK LETTER ARCHAIC KOPPA, which does not have an associated Beta Code letter assignment. Such Unicode code point strings can be mixed with other characters in the same string, as long as any character that follows the Unicode code point is not a hexadecimal digit of '0'-'9', 'A'-'F', or 'a'-'f'.

NUMERIC DIGITS¶

LETTERS¶

Small letters are given in Beta Code in this order: (1) letter, (2) breathing marks, (3) accents, and (4) iota subscript. This follows the traditional typed appearance of small polytonic Greek letters, where breathing marks and then accent marks appear on top of the small letter, and iota subscripts appear below small long vowels.

Capital letters are given in Beta Code in this order: (1) asterisk (which denotes a Capital letter), (2) breathing marks, (3) accents, (4) letter, and (5) iota subscript. This follows the traditional typed appearance of capital polytonic Greek letters, where breathing marks and then accent marks appear to the left of the capital letter, and iota subscripts appear to the right of capital long vowels.

The letter mapping is as follows, in Greek alphabetical order. Letters can be capital or small; generally speaking, small is easier to read, so it is the default output from uni2beta(1):

Example: "*to fws", "the light" (without accent marks). This could also be written as "*TO FWS"; both capital and small letters give the same conversion into UTF-8.

BREATHING MARKS AND ACCENTS¶

Example: "*to\ fw=s", "the light", with a grave accent, or varia, over the omicron and a circumflex accent, or perispomeni, over the omega. This could also be written as "*TO\ FW=S". N.B.: Note that the case of the Latin letter does not matter for accent placement; it is only the case of the Greek letter that matters. Greek capital letters are encoded with a preceding asterisk, so in this example, "O\" and "W=" will appear as small UTF-8.

IOTA SUBSCRIPT¶

GREEK PUNCTUATION¶

UNICODE GREEK¶

Also in the Greek Extended Unicode range, the TLG Project considers U+1FBF to be the equivalent of a smooth breathing mark, and uni2beta(1) will convert it as such.