Where do I begin?

[JG867]

Hi everyone! I've been lurking around the message boards for a while, but I'm still confused! I hope someone can help!

Ok, this is the first time I'm attempting to "officially" add a search button to some html user documentation. Months ago, I successfully (not counting minor tweaks) added the Zoom (free version) to a few test html pages on my desktop and it seemed to work well. Now, I need to really start working and focusing on this project. Then, convince management to spend the extra money on purchasing Zoom software. So, I really need to understand what I am doing! such pressure!

I'm now confused as to what type of search I need. The documentation is delivered as a fully enclosed html document, but is uploaded to a general webpage for users. They click on the documentation zip file, then launch the index page and start using the document. In my opinion, it functions much like a CD-ROM opened in a browser.

Please excuse my newness at this, but would I still be able to use the (easiest) javascript search capability? We deliver the documentation in its entirety, at each release, and the entire file is completely replaced onto a secured webpage for users to access. Can I treat it as if it's being delivered as a CD-ROM?


My thoughts are to place the search function onto only the index.html page (initially) then, with later releases, expand it to perhaps each html page.

If you need more info to answer this please let me know.

Thank you! I will post my other questions, but this is the most important one right now!

[Ray]

I'm now confused as to what type of search I need. The documentation is delivered as a fully enclosed html document, but is uploaded to a general webpage for users. They click on the documentation zip file, then launch the index page and start using the document. In my opinion, it functions much like a CD-ROM opened in a browser.

I'm not quite sure if I understand this. It sounds like the user will be downloading a ZIP file from your website. But how are they launching the index page after this step? Are they instructed to unzip the file, which presumably contains a collection of HTML files, and then they are asked to open the "index.html" page?

If this is the case, then yes, I can see what you mean by it needing to function like a CD-ROM. It would mean that the HTML pages that make up your documentation (and downloaded as a ZIP file) will be served offline, on their local computer, without a web server. This means that no server-side scripting will be available, and you will have to use the Javascript option. Please see chapter 4 of the Users Guide for information on this.

Now there are some limitations with what Javascript can do. Javascript is scripting that runs within your web browser. More info here. There are memory limitations which mean that you may run against a limit in the number of pages you can search through. And some functionality such as context descriptions and exact phrase matching are not available.

But by all means, your first attempt should be with the JS version, given that it would be the easiest to get working for your given scenario. If you have more than a thousand pages, or very large files to index however, you will have to consider alternative, more complicated setups. Let us know if this is the case.

[JG867]

Well, thank you so much for your quick reply.

I've been corrected on a few things, so I think your answer may be different now, unfortunately.

The doc is delivered in two sections. The first section (installation guide) is launched from a secure website. Once you click the "install" link, it brings you to the index.html page, then the users clicks from there. So, I guess this will complicate things for me! For the second section, I'm still working on getting the answer to how its packaged and delivered to the customer.

And, another question, how do I determine the size? Are you asking for the actual file size...the number of pages it references? For instance, I can tell you that the installation index.html page links to 19-20 different pages. I'm sorry to have to ask such basic questions...

Thanks in advance....

[Ray]

The doc is delivered in two sections. The first section (installation guide) is launched from a secure website. Once you click the "install" link, it brings you to the index.html page, then the users clicks from there.

You need to elaborate on what the "install" link does. What or how will it be installed? Is this an ActiveX control? A Flash or Java component? How does it get from the "install" link to the "index.html" page?

And, another question, how do I determine the size? Are you asking for the actual file size...the number of pages it references? For instance, I can tell you that the installation index.html page links to 19-20 different pages. I'm sorry to have to ask such basic questions...

It would roughly be the number of files you wish to index, for example, if they are all html pages, then it'll be the number of HTML files that make up your site. BUT if some of your pages are gigantic (for example, you may have a PDF document which contains the equivalent of a phone book of names and numbers, and is over 10 MB in size), then this would need to be taken into consideration as well. 20 or so pages would definitely not be a problem for the Javascript option.

Some more information in this FAQ:
Q. Does Zoom support very large sites? What are the practical limits of the size of sites that Zoom can be used with?

[JG867]

Hello again,

I'm sorry, I think you may have misunderstood because I didn't explain myself correctly. The "install" link just refers to the title of the document (the document is an installation guide). To use simple terms, the user would click on a "Read This Document" link and it would display the index.html page viewed in the users' browser. Do you still think javascript would work in this case? Or I need something more complex?

Thank you for the information on the size, I have always wondered how it's determined. I agree, it is a relatively small document, just html and screenshots. The second section I mentioned in an earlier post is very large (it's the users guide), but I will tackle that search problem when I get past this problem!

I really appreciate all the advice!

[Ray]

I'm sorry, I think you may have misunderstood because I didn't explain myself correctly. The "install" link just refers to the title of the document (the document is an installation guide). To use simple terms, the user would click on a "Read This Document" link and it would display the index.html page viewed in the users' browser. Do you still think javascript would work in this case? Or I need something more complex?

OK, I see now. For the record, the text that is used for a link makes no difference whatsoever, and so people usually wouldn't make reference to it, and that's why I interpreted it to mean that it was actually an "install" process.

From this, it would mean that the pages you are indexing here are actually hosted on the web server, and the user will be accessing them over the web. This means that you can potentially use a server-side scripting option such as PHP or ASP, which are definitely better than Javascript. However, it's not necessarily required. JS is just less ideal for the job. You would have to check with your host though, if PHP or ASP is supported. I would recommend looking up some of this stuff if you are completely new to web development, as it is something you would eventually need to familiarise yourself with. This is explained in better detail in the Users Guide chapters and support pages referenced before, so I would recommend starting there.