Best way to parse HTML table

With lynx I can do: $ lynx -dump http://www.cityofames.org/ftp/routes/Fall/wdreds\&w.html 6:25 6:31 6:36 6:41 ----- 6:46 6:50 6:56 7:02 7:08 7:14 7:20 ----- 7:26 7:30 7:36 ----- ----- ----- ----- 7:38 7:43 7:47 7:53 1A 7:28 7:35 7:42 7:48 ----- 7:56 8:00 8:06 ----- ----- ----- ----- 7:58 8:03 8:07 8:13 1A ... becomes very easy to parse with scripting language of choice, html2text may also work(never used it). You could also play around with grep/sed to format it.

HTML is too difficult to be understood by any parser. You need to first convert this to a reasonably close XML format(for wellformedness- means tags that are matched) like XHTML using a program like tidy(http://tidy.sourceforge.net/). You can then use a XML/XHTML parser to parse the wellformed XML. Note that you will have to process your data based on font styles and convert the tags based on font styles to an array of times. Here is what you can do when parsing start TR element --Create Array start b element -- Add One time end b element start b element -- Add second time end b element end TR element