Manual Reference Pages - FONT::TTF::NAME (3)
Font::TTF::Name - String table for a TTF font
Strings are held by number, platform, encoding and language. Strings are
Notice that the language is held in an associative array due to its sparse
nature on some platforms such as Microsoft ($pid = 3). Notice also that the
array order is different from the stored array order (platform, encoding,
language, number) to allow for easy manipulation of strings by number (which is
what I guess most people will want to do).
By default, $Font::TTF::Name::utf8 is set to 1, and strings will be stored as UTF8 wherever
possible. The method is_utf8 can be used to find out if a string in a particular
platform and encoding will be returned as UTF8. Unicode strings are always
converted if utf8 is requested. Otherwise, strings are stored according to platform:
You now have to set <$Font::TTF::Name::utf8> to 0 to get the old behaviour.
Apple Unicode (platform id = 0)
Data is stored as network ordered UCS2. There is no encoding id for this platform
but there are language ids as per Mac language ids.
Mac (platform id = 1)
Data is stored as 8-bit binary data, leaving the interpretation to the user
according to encoding id.
Unicode (platform id = 2)
Currently stored as 16-bit network ordered UCS2. Upon release of Perl 5.005 this
will change to utf8 assuming current UCS2 semantics for all encoding ids.
Windows (platform id = 3)
As per Unicode, the data is currently stored as 16-bit network ordered UCS2. Upon
release of Perl 5.005 this will change to utf8 assuming current UCS2 semantics for
all encoding ids.
An array of arrays, etc.
Reads all the names into memory
Writes out all the strings
CW$t->XML_element($context, CW$depth, CW$key, CW$value)
Outputs the string element in nice XML (which is all the table really!)
CW$t->XML_end($context, CW$tag, CW%attrs)
Store strings in the right place
Returns the minimum size this table can be. If it is smaller than this, then the table
must be bad and should be deleted or whatever.
Returns whether a string of a given platform and encoding is going to be in UTF8
Hunts down a name in all the standard places and returns the string and for an
array context the pid, eid & lid as well
Removes all strings with the given name id from the table.
set_name($nid, CW$str[, CW$lang[, CW@cover]])
Sets the given name id string to $str for all platforms and encodings that
this module can handle. If $lang is set, it is interpretted as a language
tag and if the particular language of a string is found to match, then
that string is changed, otherwise no change occurs.
If supplied, @cover should be a list of references to two-element arrays
containing pid,eid pairs that should be added to the name table if not already present.
This function does not add any names to the table unless @cover is supplied.
Font::TTF::Name->match_lang($pid, CW$lid, CW$lang)
Compares the language associated to the string of given platform and language
with the given language tag. If the language matches the tag (i.e. is equal
or more defined than the given language tag) returns true. This is calculated
by finding whether the associated language tag starts with the given language
Returns the language tag associated with a particular platform and language id
Looks up the language name and returns a lang id if one exists
Returns an array of references to two-element arrays
containing pid,eid pairs that already exist in this name table.
Useful for creating @cover parameter to set_name().
Unicode type strings will be stored in utf8 for all known platforms,
once Perl 5.6 has been released and I can find all the mapping tables, etc.
Martin Hosken <http://scripts.sil.org/FontUtils>.
Copyright (c) 1998-2014, SIL International (http://www.sil.org)
This module is released under the terms of the Artistic License 2.0.
For details, see the full text of the license in the file LICENSE.
|perl v5.20.3 ||FONT::TTF::NAME (3) ||2014-12-10 |
Visit the GSP FreeBSD Man Page Interface.
Output converted with manServer 1.07.