I'm new to greasemonkey and javascript stuff, so bear with me. I've looked through 'dive into greasemonkey' but couldn't find an example of what I need. below is an snippet of a page's html that I am trying to make useful for me. I figured out how to find each link by searching for TITLE's that equal "Document Details..." but now I would like to grab the numbers for the Book and Page (book 22, page 4 in this example). They are in there twice (in the href and as normal page text), but I can't find an example out how get these numbers into variables. Is this possible, or is the coding to old-school? Is there a different way to go about it?

< A HREF="/CGI/WWB012/WWM506/R?D=19960112&B=00022&P=004" TITLE="Document Details ...">< SMALL>Book 22< BR>< SMALL>Page< SMALL> 4< R>< /A>

Thanks.
(man, a preview-before-posting option would be useful for this forum)

This is probably easiest to do with regular expressions. In this example, "thelink" is the link with the title "Document Details ..." in your snippet.

rematches = thelinks.textContent.match(/\s*Book\s*(\d+)\s*Page\s*(\d+)/);
book = rematches[1];
page = rematches[2];

I agree that a preview function would be useful. Is there a way to contribute to this site?

Jasper: Sure: http://userscripts.devjavu.com/projects/userscr...

This forum is based around Beast, so you'd contribute to that, too.

I don't feel a need for a preview since posts can be edited.

Thanks, that works. Is there a online tutorial that explains the code within the 'match' function? I don't understand it fully.

before I get to my next question I need to know if it is possible to link to a file on your own local harddrive? I got the link made with the book and page, but when I click the link, it doesn't open the file. After reading around, I get the impression local file access was removed from greasemonkey for security reasons?

or maybe I wrote it out wrong:
newElement.setAttribute('href', 'file:///C:/ElDoradoMaps/Parcel/Book 11/011-001.tif');

(i tried it without the space between 'Book' and '11' and it still didn't work)

match documentation can be found at developer.mozilla.org: http://developer.mozilla.org/en/docs/Core_JavaS...

Greasemonkey scripts can't read local files but they can create HTML that links to a local file.

actually, Firefox itself doesn't link to local files. There is an extension called 'Local Link' you can get that enables it. So I got that part figured out....

ok, whats the best way to get text that occurs on the page before this link that I am processing. basically, it looks like there are going to be a few < td> 's and and < span>'s between this bit of text and the link I'm currently evaluating. This bit of text is inbetween basic bold tags and I can get all the bold tags on the page into an array, but there are more bold tags then link tags - its not 1 to 1. most of these other bold tags have class attributes, so i've been trying to figure out how to just get the plain bold tags so it is 1 to 1 with the links. I think I'll be able to figure that out after more debugging (or you can just tell how to do it ;) ), but it doesn't seem to be a good method and if there is a better way to get this bit of text, please let me know.

Well. I have a hard time visualizing the problem from your description without any code snippets, or maybe a link to an example. If you worry about various tags inside the link-tags, don't. Just use the textContent property of the link, and there you have it, all the text inside the link with all the tags stripped out.

http://main.co.el-dorado.ca.us/CGI/WWB012/WWM501/R

paste this into the search box and submit it: T12N R18E S31
and look at the page source.

the book and page link is what I am working with. I am trying to grab the 'PARCEL MAP' and 'REC SURVEY' text on the left side each row, as there is book 1 and page 1 (and so forth) for both rec surveys and parcel maps. (the maps themselves are in seperate folders on my hard drive, and I'm trying to add a link to open these files from the webpage). The link itself doesn't have any useful info to differentiate the two.
thanks. hopefully this is enough info to explain what I'm trying to do.