searchTHIS!

Beta 2 Release!!

• Meta Tag insertion - it appears as if this was the most problemsome peice of code, so I removed it, and also the meta tag insertion capability. If you don't have description and keyword tags in your html files you can still enter them in, they will not be added to the appropriate html file.
• Option command line argument - I have added a command line argument called option. The value for this is:
  -s -- Skip Errors, do not index files that have no meta tags.
  -k -- Keep Errors, add files that have no meta tags, the user will be prompted for missing information.
  
• Frame applet Parameter - The applet has a new parameter called frame. This will dictate the location of the search results from the searchTHIS Applet.
• NETSCAPE POWER!! - The latest release generates search.html, which searches for keywords for those web visitors that are using Microsoft Internet Explorer. search.html will load up searchit.html automatically if your web site visitor is using Netscape Navigator. searchit.html has some awesome NETSCAPE POWER! It will enable your visitors to search on not only the keywords, but also the actual text of the html files for results, AND return them in order of relevance. SHEEW! That is pretty darn cool if you ask me. One of the most powerful serverless java script search engines available... And it is All FREE!! Click Here to try out the search engine.

• How it works...
  Download the files into your web site's root directory and unzip them. Next type: java indexTHIS <score> -<option> where score is a number that designates the amount of times a word must be found in a file in order for it to be considered "important." This number is usually around 3 or 4, but could be more or less, read more about that specific topic below. The indexer will go through all of your html files and index them. It will go through and create indicies, one based on the meta description and keywords tags. It will then go through and index all of the text in the body of the html files, but not the html tags. It will score them, and they will be included in the index file based on whether the score is greater than the score you entered when you first ran the program. After all of the files are indexed, there will be 5 files that are created, search.html, searchit.html, netscape.js, db.txt, and db.js. The first two are javascript/jscript search engines. The first one, search.html, just searches keywords, because it is the IE compatible version. The second, searchit.html is one bad to the bone javascript search engine. It reads all of it's data from netscape.js, and returns results based on the keywords, text of the file, and score, or the most relevant results first. That is pretty awesome. Due to limitations with IE, the really good version works ontly under netscape. Netscape.js is used by searchit.html javascript. The other two are used by the applet to search. When the applet loads up, the files are loaded on the local machine. Then, the applet searches them based on text entered by the users. The results are ordered by the score of the word in the file(THIS IS WHERE THE SCORE COMES IN). This way, the results are returned in order of relevance...the first in the list being the most relevant(having matched the word with the highest score) and getting less as the list goes down.
• It generates a db.txt file that contains a list of all of the words in your html files AND scores them, then the applet uses this file, and the db.js file, to search your web pages.
• IT TAKES LONGER TO INDEX!!!....very important, it takes longer to index your web site...the larger your site and files, the longer it will take. Indexing time could range from anywhere between 10 seconds, to 10 Minutes, depending on you computer, file size and file count.
• Instructions: Download and unzip into your root directory...IT MUST BE IN YOUR ROOT DIRECTORY TO WORK!!!
  Then type:
  java indexTHIS <score> -<option>
  where the score is the minimum number of occurrences of a word that will allow it into your db.txt file. The higher this number, the more relevant and shorter your db.txt file is, and the shorter the load up time for the applet. THe lower this number, the more words in the db.txt file, and the longer the load up time for the applet.
• the applet:
  You can customize just about everything on the applet!! The font size, type, style, foreground, background, when the files are downloaded, whether or not to include the descriptions in the result list box, the text on the button, the text of the Enter Search Words message, etc. If you have problems customizing your applet, use the settings indicated above.
• CGI Friendly -- the db.txt file that is created is VERY friendly to cgi scripts if you choose to use them, it is VERY easily parseable!
• Download:
  Download the update here!
  If you still want the old version that may cause some problems, get it forget it, it's GONE!!
  
• If you have any questions or concerns ,email me at: clear@user1.netcarrier.com