internal package
Foswiki::Tables::Parser Re-usable sequential access event-based parser for TML tables.
A sequential access event-based parser works by parsing content and calling back to "event listeners" when syntactic constructs are recognised.
StaticMethod
parse( $text, \&dispatch ) This is a sequential event-based parser. As each line in the text is read, it is analysed and if it meets the criteria for an event, it is fired.
In keeping with the line-oriented nature of TML, the parser works on a line-by-line basis.<verbatim>
and <literal>
blocks are
respected.
Events are fired by a call to $dispatch( … ). The following events are fired:
open_table($line)
close_table()
line($line)
open_tr($before)
$before
- leading content (spaces and |)
th($pre, $data, $post)
$pre
- preamble (spaces)
$data
- real content
$post
- postamble (spaces)
td($pre, $data, $post)
$pre
- preamble (spaces)
$data
- real content
$post
- postamble (spaces)
close_tr($after)
$after
- trailing content (| and spaces)
end_of_input()
An additional event is provided for those seeking to perform special processing of certain lines, including rewriting them.
early_line($line) → $integer
If early_line returns a positive result, then the parser will open a table on the next line, whether or not it is a table line, *BUT ONLY IF* the early_line handler for that next line returns 0. Any non-whitespace left in $line will be inserted as text.
If it returns a negative result, then any non-whitespace left in $line will be inserted as text, but no other processing will be performed.
Otherwise the line is processed normally.
Note thatearly_line
operates on the internal representation of the
line in the parser. Certain constructs, such as verbatim blocks, are
specially marked in this content. early_line
can be used to rewrite
the $line in place, but only with great care. Caveat emptor.
The early_line
handler is fired for all lines that may be part of
a table (i.e. not verbatim or literal lines).
StaticMethod
split_cell($cell) → ($pre, $main, $post, $ish) Given a table cell datum with significant leading and trailing space, split the cell data into pre-, main-, and post- text, and set $ish if it is a header cell.