Full Text Index
Query System
    This is an initial entry page for a prototype image file search system currently under development as a research project of the Texas A&M Bioinformatics Working Group. The system provides 'string query' access to a digital library of vascular plant images that is also available, via family-level browsing, from the Vascular Plant Image Gallery. It is based on a full-text index of a file generated by the program imawwwo1.ipf, which also produces the family-level browse pages.

     The system will return, as an HTML page, all 'documents' (linked image references) that carry a string of characters matching the string inserted in the box below:


    The search/return system under development employs a technology similar to that used for global network queries (Alta-Vista, HotBot, etc.) in that a database/SQL interface is not involved.  It provides a relatively simple, efficent approach for WWW access to large text files and a rapid response.  However, as is the case with network'search engines, there is a 'trade-off' for for speed and simplicity and this involves spurious returns.  Distinct query strings ('Cucurbita', 'Asteraceae', etc.) will usually produce expected results, but use of less specific query strings will force the user to 'cull' returns for those that are relevant.

    It is also possible to query the index directly (not using this page) via URL and, as a result, the system can be used as a WWW content resource.  If, for instance, you are developing a web page relating to the Cucurbitaceae, the construct: 

http://www.csdl.tamu.edu/FLORA/cgi/gallery_query?q=cucurbitaceae
can be inserted into the page to return image references that carry 'cucurbitaceae'. In addition, this can be refined by using the '+' character to obtain image file 'documents' that carry two strings. The construct: 
http://www.csdl.tamu.edu/FLORA/cgi/gallery_query?q=cucurbitaceae+mexico
will return only those 'documents' that carry both query strings 'cucurbitaceae' and 'mexico'. It is also possible to exclude image files using the '!' character in the URL. The URL: 
http://www.csdl.tamu.edu/FLORA/cgi/gallery_query?q=cucurbitaceae+!pepo
will return all 'documents' carrying the string 'cucurbitaceae' except those with the string 'pepo'. These 'tags' can also be used for queries entered into the input box (above), using a '&' instead of a '+'.

    Once a useful query URL is established for a given topic using this interface, it can be 'cut' from your browser's 'location' box and pasted into local HTML pages or bookmarked for later reference. 

This system is a research product of the Texas A&M Bioinformatics Working Group supported by Advanced Research Program funding from the Texas Higher Education Coordinating Board.  Content of this site is provided for personal use of Texas A&M systematic botany students and the internet community.  The site is not designed for commercial use or publication and images accessed via this system are the property of the photographers.  Data processing for this system is accomplished by the program, imawwwo1.ipf, written by Hugh Wilson, TAMU Department of Biology Herbarium.  File indexing and CGI code for this prototype were produced by Erich Schneider during his tenure as a grad student with the TAMU Department of Computer Science.
This page last updated: 18 October 2001.