Hall of Fame

home *** CD-ROM | disk | FTP | other *** search

/ Hall of Fame / HallofFameCDROM.cdr / open / medlit31.lzh / READ_ME.1ST < prev next >

Wrap

Text File | 1986-12-08 | 15KB | 340 lines

///////////////////////////////////////////////////////////////////////////// M E D L I T / Medical Literature Database Version 3.0 \\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\ For the IBM PC and Compatibles C O P Y R I G H T This program is copyright (c) 1986 by Anescomp Systems, James A. Mihalcik ,, M.D. All rights reserved. Users are granted a limited license to copy this program for trial use by others subject to the following limitations : 1) The program is distributed in an unmodified form. 2) No fee or other consideration is accepted. 3) The program is not distributed in conjunction with any other product. If you intend to use MEDLIT3 on a regular basis, please register and show your support with the author. Commercial use by non-registered users is prohibited. Interested in multiple copies for use at work? Site licenses are available and we'd be happy to work with you to solve your business database needs. Author : Dr. James A. Mihalcik M.D. Data : N.Y.U. Medical Center BBS Fido / Mednet Node (107/18) / (901/3) Phone : ( 212 ) 254 - 3190 Comm parms : 300/1200/2400 Null, 8, 1 Mail: 305 2nd Avenue Suite 318 New York, N. Y. 10003 MEDLIT3 is a Clipper compiled Dbase III .PRG file. The files needed to run MEDLIT3 are : MEDLIT3D.EXE The main executable demo program Requires 196K RAM Allows 150 records in DBF file MEDLIT3.DBF Main Database in Dbase 3 format KEY.DBF Database of Keywords created by KEYDEX.NTX MEDLIT3. TEMP.DBF Temp holding file while importing ASCII text files from MEDLINE Can grow very large during import but is deleted at end of operation READ_ME.1st This file. >>>>>>>>> Starting To run MEDLIT3 simply drop everything (above) onto DOS bootable disk or a directory on a hard drive or RAM drive (hard or RAM recommended) and type MEDLIT3 at the DOS prompt. MEDLIT3 is a disk space eating program. Using the import program allows your database to grow rapidly. Its not uncommon to have a DBF file greater than 2 megabytes. Use the configuration options to place operational files on one disk and the dbf files on another if you try a floppy only system. Guarranteed the performance will be poor because large DBF files are disk intensive. A hard drive is almost certainly a necessity if you plan serious database operations. >>>>>>>> Opening Menu The menu system is based on the Lotus type menu system. Commands are arrayed across on line one. The active command is hi-lited. Choose between commands either by using cursor control keys or by selecting the first letter of the command. For example to see the next record in the database type " S N ". The "S" activates the Search menu choice and the "N" selects the NEXT option on the sub menu. The ESC key can be used to abort choices and return to the base command line. The key combo "ALT+C" will interrupt any MEDLIT3 operation. You will be given a query of Q/A/I ? Q is for quit, this attempts to close all DBF and index files. This ensures no data loss. The A simply aborts not closing files. This maybe needed if you hit a disk full error. I ignore the ALT+C command and returns to program execution. Several error messages may appear on the upper command line. if you notice one please contact me and I will remedy it. >>>>>>>> APPEND screen Each entry you make into the Database are called GETS in Dbase parlance. You may move back and forth between entries editing and changing until pressing RETURN in the last data field (line 4 of Abstract) or by pressing PgDn key at any point. After editing the GETS (info you are putting into this record) a READ is performed to physically place the data into the disk file. Other editing function keys are implemented as follows : o The HOME key takes you to the beginning of the current GET. o ^HOME takes you to the beginning of the first GET. o The END key takes you to the last character of the current GET. o ^END takes you to the beginning of the last GET. o The up arrow key takes you to the previous GET, and will not complete the READ when pressed at the first GET. o The down arrow key takes you to the next get and will not complete a READ when pressed at the last GET. o The right arrow key takes you one character to the right, will continue to the next GET when pressed at the end of a GET, and will not complete a READ when pressed at the last character of the last GET. o ^-> or ^F moves the cursor to the beginning of the next word. o The left arrow key will take you back one character in the current GET, but will not take you to a previous GET. o ^<- or ^A moves the cursor to the beginning of the current word. o ^T deletes the word to the right of the cursor. o ^Y deletes the rest of the GET field. o ^C or ^W or PgUp or PgDn terminates a READ and saves current values. o Esc terminates a READ without saving the current GET. o ^U - the UNDO key - returns a GET to its original contents. Suggestions : Journal: Set up a standard list for Journal abbreviations to keep the database consistent (i.e NEJM = New England Journal of Medicine). Why bother with this? It serves several purposes. Journal is a frequently searched field, searching long strings ( 40 or 50 characters) necessarily takes longer so efficiency is cut. Also if you are like me you read a limited number of journals so there isn't a super long list your secretary has to remember making data entry an easier chore. Also the fewer characters that are entered reduces the chances of a misspelling. Personally I recommend used a keyboard enhancer like PROKEY, SUPERKEY, or my personal favorite KEYWORKS to use various function key combo's for frequently used entries (i.e. Program the F1 key to be "NEMJ<enter>" then data entry can be reduced to a single button stroke. Major/Minor Keywords: To allow MEDLIT3 to manipulate keywords appropriately each Keyword entry must be terminated with a PERIOD (.) >>>>>>>>> Searching Type the first letter of the function you desire. N and P take you to nest and previous records respectively. E allows editing of the record. See above on keyboard function keys in the ADD mode . S is the search mode ; enter the EXACT entries you want searched for. Search is a very literal function, it will find only exact matches, say if you added a series of articles your using in a paper all sequentially in the databases, set a keycode to RW (recent work) and the record pointer on the database will be shifted to that point in the Database rather than you N your way till you get where you want to be. The Filter option essentialy hides all records not meeting the filters exspression. By setting a filter, then executing a NEXT or PREVIOUS command, you can hone in on a sub group of records easily. Filter uses the Dbase "SET FILTER TO" command. You may use any Dbase language exspressions to create a filter. A few examples follow : Author = 'Swan' / This is not much better than the Search function. It will find records with the author entry of 'Swan' only. Since few if any papers come out with just one author this is of limited usefulness. AT ('Swan', Author) > 0 / Oh now stop booing out there, Ashton Tate made that syntax. The AT function searches for substrings, here it will search the Author field, you reveal all of the records with 'Swan' as all or part of their entry. Journal = 'CIRC' .AND. Year = 1973 .AND. AT ('Swan', Author) > 0 This mouthful with yield that classic article that heralded the Swan-Ganz catheter because it selects the Journal 'Circulation' (Circ for short) Year as 1973 (notice no quotes around number entries) and searches for Swan as a substring in the author field. For more info review the LOCATE function in dBASE as well as its basic logic syntax. Once you turn on FILTER, a '**' symbol appears in the lower left of the screen. This tells you FILTER is ON. When on FILTER filters all selection be it Printing, Editing, Deleting, or Next commands. How to turn FILTER off ? Several ways : If you enter Previous command on the main search menu FILTER is inactivated When the last record meeting the FILTER criteria are met FILTER is inactivated. You may use the FILTER feature with the Wordstar interface feature as well. >>>>>>>>> Report Print.outputs the present record to the printer File outputs the same data to an ASCII text file of your choosing, once a file name is selected, future file output is appendded onto the end of it. Keyword update causes an alphabetized listing of all keywords in MEDLIT3 to be formed. On a large DBF file this can take awhile. It takes 15 minutes on my system to create a keyword list of 6500 entries. Once again this happens from using the import function alot and getting an enourmous DBF file built up. After the above function has created a fresh KEY.DBF file, the KEY.DBF has many duplicate entries of keywords held in common amongst various articles. The Keyword Duplicate Delete will remove these but once again it is slow so go have a cup of coffee because on the same 6500 keywords this procedure takes 30 minutes. Once you have gone thru the above painful steps you are rewarded with an alphabetized listing of all KEYWORDS in MEDLIT3. This is an invaluable aid in performing searches. Included is a sample report form, it produces a condensed print out of the database contents. Other report forms can be added for any repetitious task contact the author for support . Since this is a dBASE 3 compatible database all dBASE utilities such as QUICKREPORT will work with the database. dBASE report forms are not compatible with MEDLIT3 )))))))))) Wordstar Interface This feature allows you to generate the files needed to print a document with MAILMERGE option of WORDSTAR. It produces two files, you give the first eight letters of the filename and MEDLIT3 adds the extensions '.DAT' and '.DOC'. The '.DAT' file is the MAILMERGE form of the data you desire. The '.DOC' file is a document header, it contains the appropriate MAILMERGE dot commands to link the '.DAT' file to your letter. For more info see your MAILMERGE manual. >>>>>>>>>> MEDLINE Import If you use the MEDLINE database alot you maybe in the habit like I am of openning a screen log file while performing searches on MEDLINE. By using the MEDLINE display command D, to scroll by references, you can capture the contents to an ASCII text file which MEDLIT3 will absorb into itself. This is a very quick way of creating a large dbf file. It also allows conservation of costly connect time with MEDLINE. The MEDLINE Import function expects the ASCII test file to have the exstension ".TXT"(i.e. VALIUM.TXT) The only editing that needs to be done to the file is to be sure that the name ends in ".TXT" MEDLIT3 is fairly unruffled by line noise or other stray garbage that can pop up in on line ASCII downloads. >>>>>>>>>> INDEX The index commands either updates or creates an index file. An index file is a reordering of the sequence of the files. To alphabetize by Title select title and wait. Subseqent searches or next previous commands will follow this ordering. >>>>>>>>>> Quit Quits to DOS and close all files apropriately. Best to use this method for exiting just to be sure all data buffers are flushed. >>>>>>>>>> Demo Version The demo version lacks the on line context sensitive help facility and is limited to 150 records in the DBF file. If you like MEDLIT3 and ant to become a registered user send $50 to : Dr. James Mihalcik, M.D. 300 East 39th Street N.Y. N.Y. 10016 Besides the addition of context sensitive help and removal of the DBF limit ( MEDLIT3 is operationaly limited to 2 Billion records) The program is written in more machine native code which is markedly faster on searches. A more complete set of printed documentation will be provided as well. You will also be granted access priveledges to my BBS system, which can give you 24 hour Q&A about MEDLIT3 useage. Only registered users will receive support. Registration gets you a free update and printed manual. Future enhancements : Context sensitive help with the F1 button as well Journal list and Keycode list can be brought up on screen quickly during data entry / editing to check codes. I would greatly appreciate if you could inform me of bugs or improvements you think could be used in MEDLIT3 to the above addresses. Thank You Jim Mihalcik, M.D.