NAME

       hxref - generate cross-references inside and between HTML files

SYNOPSIS

       hxref [ -x ] [ -l ] [ -b base ] [ -i index ] [ -- ] [ input [ output ] ]

DESCRIPTION

       The hxref command links inline elements to DFN elements with the same content.  It adds ID
       attributes where necessary. If the content of a DFN or other element isn't  suitable,  the
       TITLE attribute may be used to provide the term to use for comparisons.

       Here is an example:

           <p>A <dfn>b-box</dfn> is defined to be...
           <p>For every b there is a <em>b-box</em>...

       The output of hxref will be similar to this:

           <p>A <dfn id="b-box">b-box</dfn> is defined to be...
           <p>For every b there is a <a href="#b-box"><em>b-box</em></a>...

OPTIONS

       The following options are supported:

       -x        Use XML conventions: empty elements are written with a slash at the end: <IMG />

       -b base   Sets  the  prefix  for  the  generated  URLs.  By  default  base is empty, which
                 generates URLs like "#b-box" above. If base is set to, e.g., "http://xyz/",  the
                 URLs will look like "http://xyz/#b-box".

       -i index  Directs  hxref to read terms from a database file before looking for them in the
                 document and afterwards store the terms that were found in the  same  file.  DFN
                 element  in the document override terms found in index.  This allows hxref to be
                 run multiple times on different files, to make the files refer to each other. It
                 may be necessary to run the commands twice, to resolve all cross-references.

       -l        Try   to   use   language-specific  information  to  match  instances  to  their
                 definitions. Currently, only English is supported and the only rules applied are
                 to  search  without  a  final "s" ("trees" matches "tree"), without a final "es"
                 ("bosses" matches "boss") and with a "y" replacing  a  final  "ies"  ("bounties"
                 matches  "bounty").   hxref  determines  the language from "lang" and "xml:lang"
                 attributes in the document.

OPERANDS

       The following operands are supported:

       input     The name of an HTML file. If  absent,  standard  input  is  read  instead.   The
                 special name "-" also indicates standard input. The input may be an URL.

       output    The file to write to. If absent, standard output is used. This may not be a URL.

DIAGNOSTICS

       The following exit values are returned:

       0         Successful completion.

       > 0       An  error  occurred  in the parsing of the HTML file.  hxref will try to correct
                 the error and produce output anyway.

SEE ALSO

       asc2xml(1), hxindex(1), hxnormalize(1), hxnum(1), hxtoc(1), xml2asc(1)

BUGS

       The error recovery for incorrect HTML is primitive.

       The program generates ID attributes, but doesn't generate <a name=...> tags, so the  links
       only work in browsers that recognize ID attributes.

       The  rules for matching English plurals are primitive. No dictionary is used. E.g., "bees"
       will be considered a plural of "be".

       There is currently no way to set the default language for a document  for  when  the  root
       element has no "lang" or "xml:lang" attribute.

       hxref  tries  first  to  generate  "readable"  ID attributes, by forming the ID out of the
       letters and digits found in the content of  the  element  and  falls  back  to  generating
       arbitrary  IDs  if  it  doesn't  find  enough.  However,  the algorithm in this version is
       primitive and only gives reasonable results for ASCII letters and digits.