%!

% PS WEBSITE SEARCH ENGINE QUERY LOG EXTRACTOR GRABSRCH.PS
% ========================================================
% Copyright c 1999 by Don Lancaster and Synergetics, Box 809, Thatcher, AZ, 85552
% (520) 428-4073 don@tinaja.com http://www.tinaja.com
% Consulting services available per http://www.tinaja.com/info01.html 

% All commercial rights and all electronic media rights fully reserved.
% Personal use permitted provided header and entire file remains intact.
% Linking welcome. Reposting expressly forbidden.

% version 1.2

% This specialized PostScript-as-language routine reads a specified file line by
% line, scans each line for ABSENCE a key phrase, and saves only those lines containing 
% the phrase to a new file. In this example, log files are searched for the absence
% of a self-reference and the presence of a search engine marker. Leaving only valid
% query strings from external search engines.

% GRABREFS.PS is a similar program that extracts ONLY the other referral queries.
% More detail in http://www.tinaja.com/glib/muse138.pdf

% To use this program, enter the full path sourcefile and target file names 
% and keyphrase below and resave. Then send to Acrobat Distiller. 

% Note that a NO FILE PRODUCED message is normal and expected.

% IMPORTANT: Be sure to use "\\" when you mean "\" in any PostScript string!
%            Be sure to remove any ending nulls in your referral log file.

/sourcefilename (C:\\medocs\\logfile.txt) def    % the name of the input file
/targetfilename (C:\\medocs\\searchfile.txt) def % extracted search output file

/searchphrase (tinaja) def                       % exclusion term to filter on

/workstring 20000 string def

% checkline tests to see if the referral is from your own site...

/checkline {addtooutfile1}  def                 % further tests go here

% addtooutfile1 filters for valid search engine query...

/addtooutfile1 {(http://) search                % extract referral string
                  {pop pop 
                   dup (?) search               % but only when an actual
                      {addtooutfile }if         % search engine query  
                   }if 
                } def

% /startoutfile creates an outputfile object...

/ws {writefile exch writestring} def            % disk write utility

/startoutfile {targetfilename (w+) file 
               /writefile exch def              % make a file to write

   (\n\nLines containing a search engine) ws    % label the file and criteria
   ("?" in document ) ws
    sourcefilename ws
   (:\n\n) ws   

               } def

% /addtooutfile adds the current url line to the report file...

/addtooutfile {ws (\n) ws ws ws (\n\n) ws} def

% /endoutfile finishes up and closes the output file...

/endoutfile {(\n\n) ws                           % outdent
             writefile closefile
            } def

% this is the main loop. It reads one line of the source pdf file at a time
% for processing...

/grabphrase {sourcefilename (r) file
             /workfile exch def                  % make a file to read
             startoutfile                        % start output file 
              {mark
               workfile workstring readline      % read one line at a time
               {checkline}{exit} ifelse          % test lines till done
               cleartomark                       % just in case sloppy
              } loop

              endoutfile                         % complete output file
              pop 
             } def

% This actually does it...

grabphrase                                       % extract search engine queries

%% EOF