Base URL Problems

[gthomson]

For some reason I am having problems updating my search file and when I create a new Index, the results always point to the files on my hard drive, regardless of what I enter for a base URL.

Initially I was not having this problem and would make changes to my help files (as mentioned in previous posts, I am using Robohelp to create a Webhelp file with the Zoom Search attached) and after updating my help content, I would then re-index my site to pick up changes made to the help. Not sure why, but now this is not working.

I have created several configuration files to match the different environments the help runs off. I start with a QA version that points to internal servers where we test and then a production configuration file with the base URL changed to the production site.

When I index the search and run my help file and enter a search, the URL for the search results is always pointing to the directory on my hard drive (c:\\mydirectory\help\etc....) instead of the URL I have entered in the BASE URL. I go back and index the configuration file for the QA URL and the search results link to the correct directory. For some reason, it won't index correctly for my production location.

What am I doing wrong? I haven't made any other changes to the configuration file. I manually go in and change the BASE URL but it won't update? Does the system only recognize a certain # of configuration files created?

I need to get this resolved asap because the search on my production server is not working and I've had to roll back the search file to a release when it did work.

**In Addition**

I am looking at the log information in Verbose display and I see a lot of skipped files due to "blocked by extension list" what does this mean?

**Addition-Addition**

I updated recently to 5.1 and then moved my config files from the old folder to the new. Did this cause the problem?

[David]

Moving your config file from one folder to another should have no impact.

There is no limit to the number of configuration files you can create.

"blocked by extension list" means the file is of a type that you have decided not to index. e.g. it is a JPG image file, but you are not indexing images.

You didn't mention if you are using spider mode or offline mode?
You also didn't mention what URL or path you are using as a base URL?

If you are using offline mode, and the URLs in the search results don't reflect the base URL you have used, and never change even if you change the base URL, then i would suspect that you are using an old set of index files.

For example, the search function continues to use an old set, and the new set are being created in a folder that you are not expecting and so they are not used. This is just a guess however, as we don't have access to you site, or search function nor a copy of your configuration files.

[Ray]

In addition to the above, if you are not sure what to enter for your base URL for an offline distribution of your search files (eg. CD-ROM, DVD, or anywhere where a web server's not involved), you should refer to Appendix 9.2, "Base URL for CD-ROM distribution (absolute and relative paths)" in the Users Guide.

[gthomson]

thanks for the response:

I'm using offline mode and the base URL is my site address (www.pointclickcare.com/online/help/....) which includes the diretory where my help files and the search files are kept in the application.

Our company saves all files in the application to a source directory tool and then uploads all the files to production server when deploying a release.

I'm not sure how an older version of the index files would be deployed unless the files I upload are not over writing the files on the production server? the strange thing is, when I open the zoom_pageinfo file with note pad and scroll down to the page data= section, half the pages list as "http://www.pointclickcare.com/help/online/etc...." and then change to read the file directory on my computer????

the Start directory and the Output directory have not changed and are correct. When I go into the output directory after Indexing, the files show the correct time and date but there appears to be no change with the file, no matter what I enter in the base URL. As a test, I just entered the word "nothing" in the base URL, started indexing, and then opened the file zoom_pageinfo to look at the information in pagedata=section and the URL did not change. It still continued to list my directory on my system?

[Ray]

You might want to check if there are any error messages at the end of indexing (if you have turned off certain log messages, make sure to turn them back on, on the "Index Log" tab of the Configuration window).

If you have changed the base URL to the word "nothing", this should definitely cause all your URLs to be preceded with that word. I can only think of a couple of exceptions:

(a) You have multiple start points that you have forgotten about. Click on the "More" button next to the start directory, and you may see that the other start points have different base URLs defined.

(b) If you are using Vista, and you are writing to the "Program Files" folder, then you may be confused as to where your files are really written to. Generally, Vista does not encourage you to write to the "Program Files" folder or other similarly secured "system" folders. It will sometimes pretend to write to the folder, when in fact it has written to a folder elsewhere on the system. These are known as "Compatibility Files" and can be very confusing. See this article for more information.

Best way to avoid the latter is to use an Output Directory somewhere else besides what Vista considers to be restricted folders.

If you're still have problems, you might want to send us a copy of your ZCFG file, and we can see if there's any other obivous reasons for what you're describing. You can e-mail us from here.