Robert Donahue writes:
> What is $next_pair_pr_rx and $any_next_pair_pr_rx3 ???
>
> Clearly they're global variables that get somewhere before
> d&do_cmd_ion gets called and they're there for my benefit.
There are a few things like that, but I've found very little that I
understand how to use other than the $next_pair_... and
$any_next_pair_... variables; I use my own regex for picking off
optional parameters. I also found that there aren't many of them;
when it comes to picking off the second, third, fourth parameters,
you're on your own. Here are some definitions I use (note that some
replace pre-defined definitions, but I find these work a little better
and don't cause anything to break in my documents):
my $cmark = "(?:$comment_mark)?";
# <<2>>...<<2>>
$next_pair_rx = "^[\\s\n%]*$O(\\d+)$C([\\s\\S]*)$O\\1$C";
$any_next_pair_rx3 = "[\\s\n%]*$O(\\d+)$C([\\s\\S]*)$O\\3$C";
$any_next_pair_rx5 = "[\\s\n%]*$O(\\d+)$C([\\s\\S]*)$O\\5$C";
$any_next_pair_rx7 = "[\\s\n%]*$O(\\d+)$C([\\s\\S]*)$O\\7$C";
$any_next_pair_rx9 = "[\\s\n%]*$O(\\d+)$C([\\s\\S]*)$O\\9$C";
# <#2#>...<#2#>
$next_pair_pr_rx = "^[\\s\n%]*$OP(\\d+)$CP([\\s\\S]*)$OP\\1$CP$cmark";
$any_next_pair_pr_rx3 = "[\\s\n%]*$OP(\\d+)$CP([\\s\\S]*)$OP\\3$CP$cmark";
$any_next_pair_pr_rx5 = "[\\s\n%]*$OP(\\d+)$CP([\\s\\S]*)$OP\\5$CP$cmark";
$any_next_pair_pr_rx7 = "[\\s\n%]*$OP(\\d+)$CP([\\s\\S]*)$OP\\7$CP$cmark";
$any_next_pair_pr_rx9 = "[\\s\n%]*$OP(\\d+)$CP([\\s\\S]*)$OP\\9$CP$cmark";
When forming the regex to pick off arguments, there are two
approaches: Pick off each parameter individually and set them aside,
or pick them off all at once. I tend to use the later myself:
sub do_cmd_something {
local($_) = @_;
# create the RX to use:
my $rx = "$next_pair_pr_rx$any_next_pair_pr_rx3";
# strip the parameters and assign to variables:
s/$rx//;
my($first,$second) = ($2, $4);
# -or- modify in place:
# (this reverses them and inserts a space)
s/$rx/\4 \2/;
...
If you modify the input string in place, remember to return $_.
Otherwise, remember to include $_ at the end of the return value.
The numbering of regex groups is significant. For the first
parameter, use $next_pair_pr_rx, and for subsequent parameters use
$any_next_pair_pr_rxN, where N is (param# * 2) + 1. The parameter
text is in group (param# + 1) * 2 and the "br_id" (bracket ID; each
pair of curly braces in the document gets a unique ID #) is (param# *
2) + 1. The br_id values can be useful for generating index entries;
look for make_index_entry and do_cmd_index in the main latex2html
script.
I use the regex "^\\s*(\\[([^]]*)\\])?" to detect optional
parameters in my do_cmd_... functions; this isn't fully general but
works for my documents. (Except for \item[...], but I didn't write my
own there!) I use this instead of $next_pair_... in my do_cmd_... and
do_env_... functions; the optional arg is param. 1, and the required
args start at 2. So it might look like this:
sub do_env_methoddesc{
local($_) = @_;
# Predefined $opt_arg_rx & $optional_arg_rx don't work because they
# require the argument to be there.
my $opt_arg_rx = "^\\s*(\\[([^]]*)\\])?";
my $rx = "$opt_arg_rx$any_next_pair_rx3$any_next_pair_rx5";
...
Ok, ok, this is getting pretty long....
-Fred
--
Fred L. Drake, Jr.
[EMAIL PROTECTED]
Corporation for National Research Initiatives
1895 Preston White Drive Reston, VA 20191
