Parsing an HTML file in Arexx

[Gundam]

I want to extract information from <td> elements, eg:

<td>
A

</td>
<td>
B

</td>

...

How do I achieve that in Arexx ?
Thanks in advance.

[thomas]

How much do you know about these HTML files? For example, are <td> value </td> always on seperate lines? Are <td> </td> always written in lower case?

To write a generic HTML parser might be very difficult. But if you know that the files you read are always built up in the same way, it will get easier.

[Gundam]

Originally Posted by thomas

How much do you know about these HTML files? For example, are <td> value </td> always on seperate lines? Are <td> </td> always written in lower case?

To write a generic HTML parser might be very difficult. But if you know that the files you read are always built up in the same way, it will get easier.

The <td> elements (on separate lines) are within a <table class="tableclass">; the "class" attribute "tableclass" is a unique identifier.

[thomas]

This wasn't what I asked for.

Let's look at an example. I saved the source code of this page as board.htm. The ARexx program board.rexx extracts the posting time and user name from the table of posts. It's done in a quite easy way but this means that it only works on HTML files which look exactly like this.

3> rx board.rexx
Today, 12:14 Gundam
Today, 13:02 thomas
Today, 13:55 Gundam
3>

[Gundam]

Originally Posted by thomas

This wasn't what I asked for.

Let's look at an example. I saved the source code of this page as board.htm. The ARexx program board.rexx extracts the posting time and user name from the table of posts. It's done in a quite easy way but this means that it only works on HTML files which look exactly like this.
...

that's OK!
I don't need a generic parser; I just want to extract the contents of <td> elements which are in the <table> with THAT specific identifer.
The value is on a separate line, like this:

<td>
value
</td>

the <td> tags don't have any "id" attribute, so they can't be easily identified; thats why I asked for help.

[daxb]

Thats easy. Use Thomas script as template and change/reduce it to your needs. You just have to readln() and check if the string is <td>. If yes, read next line until </td> is reached.

[Gundam]

Originally Posted by daxb

Thats easy. Use Thomas script as template and change/reduce it to your needs. You just have to readln() and check if the string is <td>. If yes, read next line until </td> is reached.

thanks!

[Lonewolf10]

Originally Posted by Gundam

that's OK!
I don't need a generic parser; I just want to extract the contents of <td> elements which are in the <table> with THAT specific identifer.
The value is on a separate line, like this:

<td>
value
</td>

the <td> tags don't have any "id" attribute, so they can't be easily identified; thats why I asked for help.

Just be aware that not all HTML code and data are on separate lines. Code generated by programs usually are, but I personally hand generate my HTML code for my website as I find it more fun, and easier to read doing it my way. What I'm trying to say is that some data may be stored thusly:

<TABLE>
<TR><TH>Some data</TH></TR>
<TR><TD>More data</TD></TR>
<TR><TD>Even more data</TD></TR>
<!-- and the rest of the table follows... --->
</TABLE>

<TR> is table row start and </TR> marks it's end
<TH> is table (column/s) header and </TH> marks it's end

The above example is a 1 column table, it is entirely possible to have multiple entries of each tag on the same row.