/database July 2001 This subdirectory contains the IDL procedures, mostly written by Don Lindler, (ACC/Goddard) to create and access an IDL database. These database procedures are currently used by the HST-ACS, HST-STIS, HST-WFC3 and SOHO science teams. The LaTex file database.tex in the /text directory describes the use of the database procedures with an emphasis on astronomical applications. Further information on the database software is available in Chapter 3 and Appendix J of "A User's Guide to the GHRS Software" available as a postscript file or as .html help from http://hrssun.gsfc.nasa.gov/daf/sw_docs.html The database procedures require the non-standard system variables !PRIV, !TEXTOUT, and !TEXTUNIT. These can be added to the IDL session by using the procedure ASTROLIB. Sample astronomical catalogs, formatted as IDL databases, are available in the /zdbase directory. These include the Yale Bright Star Catalog, the RC3 Galaxy catalog, and a catalog of HST observations. Very large (> 15 million entries) databases may require the use of 64 bit integers - and at least V5.2 -- internally. Updated versions of DBINDEX, DBINDEX_DBF, DBFIND_SORT, and DBEXT_IND are available in ftp://idlastro.gsfc.nasa.gov/pub/v53. 23-Jul-2001: DBCREATE, DBFIND, DBFIND_SORT, DBXVAL, DBINDEX_BLK, DBBUILD - Now allow user to specify items with 64 bit or unsigned integer datatypes. 2-Dec-2000: DBINDEX, DBEXT_IND - Now allow multiple-valued index items Minor updates to database.tex file 17-Nov-2000: DBFIND, DBFIND_ENTRY, DBFIND_SORT, DBSEARCH, IMDBASE - Now have a Count keyword; deprecate use of !ERR 11-Oct-2000: DBXVAL - Work for multiple-valued strings 08-Aug-2000: DBPRINT - Change a loop index to type LONG (sigh...) 5-Feb-2000: DBPRINT - Only byteswap when necessary, for much improved speed. 1-Nov-1999: DBINDEX, DBEXT_DBF - Now allow a maximum of 18 instead of 15 items 21-Sep-1999: DBPRINT - Fixed occasional problem when /NOHEADER supplied 04-Sep-1999: DBCIRCLE - Fixed bug that could cause some entries to be missed 05-Aug-1999: DBMATCH - Removed loop (faster) for substring searches DBMATCH() - Change some loop indicies to type LONG DBCIRCLE() - Leave output DIS vector unchanged if no entries found