frameless frames and ASP Treeview - can I highlight in "second frame"?

[lakleeman]

I am using two "frameless frames" with the CSS technique found at http://www.cssplay.co.uk/layouts/body5.html

My sample site is at http://www.cityattorney.net/cc3/
and the search page is at http://www.cityattorney.net/cc3/search.asp

The idea for the site is an online book - with the first frameless frame being the table of contents and the second frame containing the main text/chapters
(only the first two chapters are set up, so ignore the remainder)

(I've also set up the highlight/jump-to feature with javascript, etc.)

Although I use Zoomstop and Zoomrestart to avoid indexing the ASP Treeview menu, the search results still highlight the terms found in the menu.
(example: search "dogs")

More important problem: the searched for word does is neither highlighted nor jumped to in the second "frameless frame" - example: search "rabbits"

Any ideas on how to make the highlight/jump-to feature work in the second CSS'd frameless frame??

Thanks for any help.

[Ray]

There's alot of Javascript on that site, so I can not be 100% sure that this will fix everything. However, the first and foremost problem is that you have linked to the "highlight.js" script incorrectly so the highlight function is not even being called at the moment.

On this page:
http://www.cityattorney.net/cc3/BasicCode/Ch02Art01.aspx?zoom_highlight=rabbits

You have the following script link at the top of the file:

<script type="text/javascript" src="highlight.js"></script>

However, note that "highlight.js" is not in the same folder as that ASPX file, so that filepath is incorrect. The file appears to be located in the "cc3" folder, one level up, so the path should be:

<script type="text/javascript" src="[B]../[/B]highlight.js"></script>

It may also be different for other files on your site if they are several folders deep and you may also consider using absolute paths to fix this.

[Ray]

As for the problem with the ZOOMSTOP and ZOOMRESTART not successfully excluding the navigation menu from being indexed, I suspect that you may be indexing from a cached copy of the page before you added those tags in. To fix this, go to the Configuration window, and check the box for "Reload all files (do not use cache)" and index again.

[lakleeman]

What am I doing differently?

Before I was able to index this site...

now the indexer stops after the looking at the base page and doesn't follow the links in the ASP treeview menu!

How did I do this before??

(I am now running Windows Vista RC1 - should that matter?)

[David]

What am I doing differently?

Hard for us to say from here. You're in the best position to answer that.

At the moment we have done almost zero testing with Vista RC1. So we don't know if that is the problem or not. We'll be doing Vista testing sometime in the next couple of months (well before the public release of Vista).

Can you turn on verbose mode in Zoom and post the log file (or at least the interesting parts of it).

[lakleeman]

Here is the verbose log file....
The default.aspx (all others) use the master page which includes the Treeview menu.
Shouldn't it follow links in the menu?


19:32:16 - Spider from: http://www.cityattorney.net/cc3/Default.aspx
19:32:16 - Web site URL: http://www.cityattorney.net/cc3/
19:32:16 - Estimated RAM required during index process: 8554 KB
19:32:16 - Initiating HTTP session (thread #1) ...
19:32:16 - DL Thread #1, got URL (http://www.cityattorney.net/cc3/Default.aspx) off queue
19:32:16 - [DOWNLOAD] Downloading file http://www.cityattorney.net/cc3/Default.aspx (38464 bytes)
19:32:16 - Index Thread got ready buffer for http://www.cityattorney.net/cc3/Default.aspx (Content-type: HTML text)
19:32:16 - Initiating HTTP session (thread #2) ...
19:32:16 - [SCANNED] Scanning http://www.cityattorney.net/cc3/Default.aspx
19:32:17 - [FILEIO] Writing index data for ASP search... (Please wait)
19:32:17 - [FILEIO] Created dictionary data file (zoom_dictionary.zdat)
19:32:17 - [FILEIO] Created wordmap data file (zoom_wordmap.zdat)
19:32:17 - [FILEIO] Created pagetext data file (zoom_pagetext.zdat)
19:32:17 - [FILEIO] Created pages data file (zoom_pages.zdat)
19:32:17 - [FILEIO] Created titles data file (zoom_titles.zdat)
19:32:17 - [FILEIO] Created descriptions data file (zoom_descriptions.zdat)
19:32:17 - [FILEIO] Created dates data file (zoom_datetime.zdat)
19:32:17 - [FILEIO] Created script settings file (settings.asp)
19:32:17 - Indexing completed at Mon Sep 11 19:32:17 2006
19:32:17 - INDEX SUMMARY
19:32:17 - Files scanned: 1
19:32:17 - Files skipped: 0
19:32:17 - Unique words found: 68
19:32:17 - Total words found: 8
19:32:17 - Avg. unique words per page: 68
19:32:17 - Avg. words per page: 8
19:32:17 - Errors: 0
19:32:17 - URLs visited by spider: 1
19:32:17 - URLs in spider queue: 0
19:32:17 - Total bytes scanned/downloaded: 38464
19:32:17 - File extensions:
19:32:17 - .htm scanned: 0
19:32:17 - .html scanned: 0
19:32:17 - .txt scanned: 0
19:32:17 - .php scanned: 0
19:32:17 - .asp scanned: 0
19:32:17 - .cgi scanned: 0
19:32:17 - .aspx scanned: 1
19:32:17 - .pl scanned: 0
19:32:17 - .php3 scanned: 0
19:32:17 - No extensions scanned: 0
19:32:17 - Cleaning up memory used for index data... please wait.
19:32:17 - Finished cleaning up memory.

[Ray]

I've done some tests. Your website is returning very different output depending on the client/browser user-agent ID from what I can tell.

When I retrieve the page via IE, I get all the code for the TreeView including links to the actual pages.

When the same page is retrieved via Zoom or wget (a simple utility that makes a HTTP request and saves the output), most of the TreeView code is missing, and most importantly, the non-Javascript HREF links to the other pages are also missing. This is why Zoom is not finding any links to follow - because it was never given any.

Possible causes:

- You have made some changes to your website and enabled some script/method of switching output based on the client user-agent string. Since your script fails to recognize "ZoomSpider (wrensoft.com)" or "wget", it is giving them different output without the links.

- Alternatively, this may have always been the case, and it only worked before because Zoom was indexing the cached copy of the pages (it shares the same cache as IE, so it may have been indexing IE's cached copy).

Either way, it would seem like you would need to fix your website's script so that it returns the links for the ZoomSpider client, or for browsers besides the small recognized list it is doing now. I should point out that as it is at the moment, the same would happen with other internet-wide spiders such as Google etc, not being able to spider the rest of your site.

[lakleeman]

Thanks. I messed with the ASP Treeview and discovered a solution.
The "ExpandDepth" property of Treeview needs to be set to fully expanded, so the spider reads the full menu of links. After indexing, then you can set the ExpandDepth back to how you like it.

See http://www.cityattorney.net/cc4/
and the search page at
http://www.cityattorney.net/cc4/search.asp

Ray, from your previous posts, I now have highlights in my second frame!

However, I notice that it doesn't jump to the highlight.
Example: search "vaccine"

Any thoughts on that?

Thanks for the help so far.

[Ray]

The jump to highlighting script only attempts to scroll the window object.

This is probably different in your CSS layout where there are multiple DIV elements with independent scroll bars. The script will not be able to determine which element to adjust on its own and it may require some custom development. I'm not entirely sure if it is possible at all having not played around with such a CSS layout, but let us know if you have any luck.