CHAPTER 5
426
Text
quality typography or non-Latin writing systems. Different encodings can se-
lect different subsets of the same character set.
Latin-text font programs produced by Adobe Systems use the
Adobe standard
encoding,
often referred to as
StandardEncoding
. The name
StandardEncoding
has no special meaning in PDF, but this encoding does play a role as a default en-
coding (as shown in Table 5.11 below). The regular encodings used for Latin-text
fonts on Mac OS and Windows systems are named
MacRomanEncoding
and
WinAnsiEncoding
, respectively. An encoding named
MacExpertEncoding
is used
with “expert” fonts that contain additional characters useful for sophisticated ty-
pography. Complete details of these encodings and of the characters present in
typical fonts are provided in Appendix D.
In PDF, a font is classified as either
nonsymbolic
or
symbolic
according to whether
all of its characters are members of the Adobe standard Latin character set. This
is indicated by flags in the font descriptor; see Section 5.7.1, “Font Descriptor
tioned above ordinarily do not apply. Such font programs have built-in encodings
that are usually unique to each font. The standard 14 fonts include two symbolic
fonts,
Symbol
and
ZapfDingbats
, whose encodings and character sets are docu-
mented in Appendix D.
A font program’s built-in encoding can be overridden or altered by including an
Encoding
entry in the PDF font dictionary. The possible encoding modifications
depend on the font type, as discussed below. The value of the
Encoding
entry is
either a named encoding (the name of one of the predefined encodings
MacRomanEncoding
,
MacExpertEncoding
, or
WinAnsiEncoding
) or an
encoding
dictionary.
An encoding dictionary contains the entries listed in Table 5.11.