class PDF::Reader::GlyphHash
:nodoc:
The mapping is read from a data file on disk the first time it’s needed.
A Hash-like object that can convert glyph names into a unicode codepoint.
def initialize
The mapping is read from a data file on disk the first time it's needed.
A Hash-like object that can convert glyph names into a unicode codepoint.
def initialize @@by_codepoint_cache ||= nil @@by_name_cache ||= nil # only parse the glyph list once, and cache the results (for performance) if @@by_codepoint_cache != nil && @@by_name_cache != nil @by_name = @@by_name_cache @by_codepoint = @@by_codepoint_cache else by_name, by_codepoint = load_adobe_glyph_mapping @by_name = @@by_name_cache ||= by_name @by_codepoint = @@by_codepoint_cache ||= by_codepoint end end
def load_adobe_glyph_mapping
a text file supplied by Adobe at:
returns a hash that maps glyph names to unicode codepoints. The mapping is based on
def load_adobe_glyph_mapping keyed_by_name = {} keyed_by_codepoint = {} paths = [ File.dirname(__FILE__) + "/glyphlist.txt", File.dirname(__FILE__) + "/glyphlist-zapfdingbats.txt", ] paths.each do |path| File.open(path, "r:BINARY") do |f| f.each do |l| _m, name, code = *l.match(/([0-9A-Za-z]+);([0-9A-F]{4})/) if name && code cp = "0x#{code}".hex keyed_by_name[name.to_sym] = cp keyed_by_codepoint[cp] ||= [] keyed_by_codepoint[cp] << name.to_sym end end end end return keyed_by_name.freeze, keyed_by_codepoint.freeze end
def name_to_unicode(name)
=> 34
h.name_to_unicode(:34)
=> 48
h.name_to_unicode(:G30)
=> 74
h.name_to_unicode(:X4A)
=> 8364
h.name_to_unicode(:Euro)
=> 65
h.name_to_unicode(:A)
h = GlyphHash.new
if no conversion is possible.
attempt to convert a PDF Name to a unicode codepoint. Returns nil
def name_to_unicode(name) return nil unless name.is_a?(Symbol) name = name.to_s.gsub('_', '').intern str = name.to_s if @by_name.has_key?(name) @by_name[name] elsif str.match(/\AX[0-9a-fA-F]{2,4}\Z/) "0x#{str[1,4]}".hex elsif str.match(/\Auni[A-F\d]{4}\Z/) "0x#{str[3,4]}".hex elsif str.match(/\Au[A-F\d]{4,6}\Z/) "0x#{str[1,6]}".hex elsif str.match(/\A[A-Za-z]\d{1,5}\Z/) str[1,5].to_i elsif str.match(/\A[A-Za-z]{2}\d{2,5}\Z/) str[2,5].to_i else nil end end
def unicode_to_name(codepoint)
=> [:34]
h.unicode_to_name(34)
=> [:Euro]
h.unicode_to_name(8364)
=> [:A]
h.unicode_to_name(65)
h = GlyphHash.new
if no conversion is possible.
attempt to convert a Unicode code point to the equivilant PDF Name. Returns nil
def unicode_to_name(codepoint) @by_codepoint[codepoint.to_i] || [] end