tex2xindy
?
tex2xindy
is a filter that parses `.idx
' or similar files
and converts the \indexentry
macros into a form readable by
The parser of makeindex
can be configured to recognize different
quoting characters, etc. (see the man-page for makeindex
, section
input style specifiers for further details). We have tried to
extract the parser from makeindex
but due to several probems we
have finally rewritten the parser using lex
. Scanners written
with lex
are usually fixed to a specific character set used in
the regular expressions. Our parser, tex2xindy
is therefore not
configurable. If one uses a different configuration of the
makeindex
input style specifiers, one can change the source
(tex2xindy.l
) to generate a completely new parser. From our
personal experience we have rarely used more than two different
parsers in practice so we have written tex2xindy
in a form that
is easily maintainable. The input specifiers are stored symbolically
in the source. The definiton section looks like this:
KEYWORD \\indexentry
ENCAP \|
ACTUAL @
ESCAPE \\
LEVEL !
QUOTE \"
ROPEN \(
RCLOSE \)
ARGOPEN \{
ARGCLOSE \}
These definitions are essentially the input style specifiers as can be
found in the man-page of makeindex
. Changing this section
according to your needs and recompiling tex2xindy
should be an
easy task. Maybe we will include more pre-defined parsers in future
releases if necessary.
We strongly recommend using the LaTeX2e-package index
written
by David M. Jones, which is available at CTAN. It supports multiple
indexes as well as several shortcuts to easily index terms in a
document. Multiple indexes support the generation of several indexes
for one document. For instance, one can make an author or command
index in addition to a global index.
Another option is to use the xindy.sty
from Andreas Schlechte
that comes with the contrib
directory that should contain a version.