1.1 --- /dev/null Thu Jan 01 00:00:00 1970 +0000
1.2 +++ b/intl/hyphenation/src/README.compound Wed Dec 31 06:09:35 2014 +0100
1.3 @@ -0,0 +1,87 @@
1.4 +New option of Libhyphen 2.7: NOHYPHEN
1.5 +
1.6 +Hyphen, apostrophe and other characters may be word boundary characters,
1.7 +but they don't need (extra) hyphenation. With NOHYPHEN option
1.8 +it's possible to hyphenate the words parts correctly.
1.9 +
1.10 +Example:
1.11 +
1.12 +ISO8859-1
1.13 +NOHYPHEN -,'
1.14 +1-1
1.15 +1'1
1.16 +NEXTLEVEL
1.17 +
1.18 +Description:
1.19 +
1.20 +1-1 and 1'1 declare hyphen and apostrophe as word boundary characters
1.21 +and NOHYPHEN with the comma separated character (or character sequence)
1.22 +list forbid the (extra) hyphens at the hyphen and apostrophe characters.
1.23 +
1.24 +Implicite NOHYPHEN declaration
1.25 +
1.26 +Without explicite NEXTLEVEL declaration, Hyphen 2.8 uses the
1.27 +previous settings, plus in UTF-8 encoding, endash (U+2013) and
1.28 +typographical apostrophe (U+2019) are NOHYPHEN characters, too.
1.29 +
1.30 +It's possible to enlarge the hyphenation distance from these
1.31 +NOHYPHEN characters by using COMPOUNDLEFTHYPHENMIN and
1.32 +COMPOUNDRIGHTHYPHENMIN attributes.
1.33 +
1.34 +Compound word hyphenation
1.35 +
1.36 +Hyphen library supports better compound word hyphenation and special
1.37 +rules of compound word hyphenation of German languages and other
1.38 +languages with arbitrary number of compound words. The new options,
1.39 +COMPOUNDLEFTHYPHENMIN and COMPOUNDRIGHTHYPHENMIN help to set the right
1.40 +style for the hyphenation of compound words.
1.41 +
1.42 +Algorithm
1.43 +
1.44 +The algorithm is an extension of the original pattern based hyphenation
1.45 +algorithm. It uses two hyphenation pattern sets, defined in the same
1.46 +pattern file and separated by the NEXTLEVEL keyword. First pattern
1.47 +set is for hyphenation only at compound word boundaries, the second one
1.48 +is for hyphenation within words or word parts.
1.49 +
1.50 +Recursive compound level hyphenation
1.51 +
1.52 +The algorithm is recursive: every word parts of a successful
1.53 +first (compound) level hyphenation will be rehyphenated
1.54 +by the same (first) pattern set.
1.55 +
1.56 +Finally, when first level hyphenation is not possible, Hyphen uses
1.57 +the second level hyphenation for the word or the word parts.
1.58 +
1.59 +Word endings and word parts
1.60 +
1.61 +Patterns for word endings (patterns with ellipses) match the
1.62 +word parts, too.
1.63 +
1.64 +Options
1.65 +
1.66 +COMPOUNDLEFTHYPHENMIN: min. hyph. dist. from the left compound word boundary
1.67 +COMPOUNDRIGHTHYPHENMIN: min. hyph. dist. from the right comp. word boundary
1.68 +NEXTLEVEL: sign second level hyphenation patterns
1.69 +
1.70 +Default hyphenmin values
1.71 +
1.72 +Default values of COMPOUNDLEFTHYPHENMIN and COMPOUNDRIGHTHYPHENMIN are 0,
1.73 +and 0 under the hyphenation, too. ("0" values of
1.74 +LEFTHYPHENMIN and RIGHTHYPHENMIN mean the default "2" under the hyphenation.)
1.75 +
1.76 +Examples
1.77 +
1.78 +See tests/compound* test files.
1.79 +
1.80 +Preparation of hyphenation patterns
1.81 +
1.82 +It hasn't been special pattern generator tool for compound hyphenation
1.83 +patterns, yet. It is possible to use PATGEN to generate both of
1.84 +pattern sets, concatenate it manually and set the requested HYPHENMIN values.
1.85 +(But don't forget the preprocessing steps by substrings.pl before
1.86 +concatenation.) One of the disadvantage of this method, that PATGEN
1.87 +doesn't know recursive compound hyphenation of Hyphen.
1.88 +
1.89 +László Németh
1.90 +<nemeth (at) openoffice.org>
