Advanced Client-side Search Engine
Hi, |
You can define how I am looking into my database. You can
| compare your keyword within a keyword of the database | |
| compare both keywords beginning with the first char ([x] Begin of word) | |
| compare case-sensitive ([x] Case-sensitive) | |
| use boolean operators or, and, not and parenthesis to detail your search |
The following table shows a list of keywords, for example, from the HTML pages member, logo, aboutus:
| member.htm | logo.htm | aboutus.htm |
| Member | Quality | OASE |
| Fee | Member | Shareware |
| Offer | OASE | Aims |
| Service | OASE Service | service oriented |
Here are some search inputs and the corresponding hitlists (without looking on [x] Begin of word, [x] Case-sensitive):
| Input | Hitlist | Remark |
| Member | member, logo | normal keyword |
| not Member | aboutus | all keywords not equal with the input (inverted) |
| Member OASE | member, logo, aboutus | all keywords equal with keyword one or keyword two; we have a hit if only keyword one exists, or only keyword two exists, or both exist |
| Member or OASE | member, logo, aboutus | like the input above |
| Member and OASE | logo | we have a hit if both exist |
| not (Member and OASE) | member, aboutus | give all hits that not have both words in it (parenthesis show which part is calculated first) |
| not Member and OASE | aboutus | without parenthesis we calculate every entry that has not
the first word in it and in the result take only those that have the
second word in it (priority is not, and, or) the input not Member and OASE or Quality corresponds to ((not Member) and OASE) or Quality |
If we're looking on [x] Begin of word your input has to be equal with keywords from the first character on (in parts or full length). We have no look on [x] Case-sensitive:
| Input | [x] Begin of word hitlist |
[ ] Begin of word hitlist |
| Service | member, aboutus | member, logo, aboutus |
| not Service | logo | empty |
| Service OASE | member, logo, aboutus | member, logo, aboutus |
| Service or OASE | member, logo, aboutus | member, logo, aboutus |
| Service and OASE | aboutus | logo, aboutus |
| not (Service and OASE) | member, logo | member |
| not Service and OASE | logo | empty |
If your keyword matches a keyword in the database I create a hitlist with a link to the corresponding link to the web page. Here you can
| produce a new hitlist every time you start to search (old lists are deleted) | |
| append new entries (links) at the end of the hitlist ([x] Keep hitlist). Maybe you have to scoll down to see the new entries. | |
| hide the input form during selection of a link in the hitlist ([x] No input form). Try the back-button of your browser to get back to the input form. |
The hitlist, for example, looks like this:
| Hit per cent | Keyword | Description | |
| 2. | OASE OASE-Service |
Free information for members | |
| 1. | OASE-Logos | Quality standards according to the Euroware standard |
| Position | |
| How many characters of your keyword and the keyword in the database are equal (percentage)? | |
| Shows the keyword in the database (and equal characters of your input in bold) | |
| Short description of the topic of the web page (normaly the title) |
The hitlist is sorted descending to the percentage.
Keep in mind, that the hitlist is created during run time, so it is not persistent over time. If you klick on a link in the list you normally should get back the list after klicking on the back button of your browser. Alternatively you can create the list through klicking on the "Search"-Button again (maybe you have to put in the desired keyword again).
Standardly [x] Begin of word is activated. So you get links where your keyword and the keyword in the database are equal from the first character on. If the result is not what you expect you should deactivate the checkbox. If the hitlist shows wrong links again you should try another keyword (try a synonym).
It is possible that you get to many or wrong links in the hitlist.Try to activate [x] Case-sensitive for a better result.
The links in the hitlist can be loaded through klicking on the (underlined) keywords. You get the begin of the web page, although the keyword could be in the middle or at the end of the page. It is not possible to use HTML for a direct jump to a keyword yet. So if you want to see all positions in the web page where you can found the keyword you have to use the (local) web page search mechanism of your browser. Use the clipboard to get the keyword you have clicked on into the search dialog of your browser:
| Select the keyword in the hitlist with your mouse and copy it to the clipboard. If the keyword in the hitlist is equal to the keyword you put into the input form you can also select the whole string in the input form to copy. | |
| Click on the keyword in your hitlist to get the corresponding web page (wait a moment until the page is fully loaded). | |
| Call your browser's search dialog (for pages) and copy the clipboard into the input form (<Shift>-<Insert>). When you klick on "OK" you get the first keyword position in the web page. Maybe you have to repeat this until the right position is found. |
Produced by re.se@rch.suite: Copyright © 1997-99 by Rainer Eschen, OASE.