From my point if view, it would be enough to handle a single table, since WebParser can easily capture everything between <table> and </table>, but hey, if you really want to put all eggs into one basket, go ahead, I don't think it will add too much to the complexity of the script besides an additional level of nesting for all tables, although those tables might be clearly unrelated to one another.Nah, as you said, it could be useful in certain situations. However, what would be the most useful output?
I would make the <th>...</th> row 0 in the Lua table (same header content for the same colspan, assuming header contents are different, or something similar? this is debatable), then the <tr>...</tr> the rows 1 to N in it, and <td>...</td> the columns 1 to M in that Lua table, I think a simple matter of incrementing the said indexes as you parse the table in the HTML.
One thing to consider would be the complementarity with the opposite function. Bar whatever visuals, a tabletohtml() followed by a htmltotable() cycle or viceversa should result in the same output as the original.
Statistics: Posted by Yincognito — Today, 10:51 am