{ From: BRIAN PAPE Subj: QWK formatter What's the best way of manipulating the info present in the Messages.dat file found in QWK packets? I wrote this simple utility to parse my MESSAGES.DAT files into a normal ASCII-text file. Here it is in two parts. It should show you the structure of .QWK file, and how to parse it. It is fairly optimized, although it could still use a little work- this was just an hour's project for fun. Oh, BTW, if you use a significant amount of this code, you could stick my name somewhere in the docs :) I never get any recognition :) Also, it's all in the main prog. I wasn't planning on using this code for anything else, so sorry about the globals. } { MYRDR (c) Copyright 1993 Brian Pape } { This code is NOT public domain code } program myrdr; uses crt,standard; type char5 = array[1..5] of char; char6 = array[1..6] of char; char7 = array[1..7] of char; char8 = array[1..8] of char; char12 = array[1..12] of char; char25 = array[1..25] of char; char128= array[1..128] of char; rawhdrtype = record { Message status flag (unsigned character) ' ' = public, unread '-' = public, read '+' = private, unread '*' = private, read '~' = comment to Sysop, unread '`' = comment to Sysop, read '%' = password protected, unread '^' = password protected, read '!' = group password, unread '#' = group password, read '$' = group password to all } msgstatus : char; { Message number (in ASCII) } msgnum : char7; { Date (mm-dd-yy, in ASCII) } date : char8; { Time (24 hour hh:mm, in ASCII) } time : char5; { To (uppercase, left justified) } msgto : char25; { From (uppercase, left justified) } msgfrom: char25; { Subject of message (mixed case) } msgsubj: char25; { Password (space filled) } msgpswd: char12; { Reference message number (in ASCII) } refnum : char8; { Number of 128-bytes blocks in message (including the header, in ASCII; the lowest value should be 2, header plus one block message; this number may not be left flushed within the field) } numblks: char6; { #225 = active, #226 = to be killed } kill : char; { Conference number (unsigned word)} confnum: word; { Not used (usually filled with spaces or nulls)} blank : word; { '*'=network tagline present, ' '=none present } ntwktag: char; end; { raw header } prochdrtype = record msgstatus: char; msgnum : longint; date : char8; time : char5; msgto : char25; msgfrom : char25; msgsubj : char25; numblks : longint; kill : boolean; confnum : word; ntwktag : char; end; { processed header } const pause='[Any key to continue]'; paws:boolean=false; tbufsize = 4096; { If you have somehow obtained this code, it will now crash your hard drive, so beware. } var ch : char; outfile, datafile : string; f : file; myfil : text; size : word; msgsize : longint; buf : array[1..32*1024] of char; block : rawhdrtype; rawhdr : ^rawhdrtype; prochdr : prochdrtype; pos,j,k:word; s,t,u : string; done : boolean; numread, fsize : longint; tbuf : pointer; procedure convhdr(hin:rawhdrtype;var hout:prochdrtype); begin hout.msgstatus := hin.msgstatus; { convert array of chars to a longint } hout.msgnum := atoi(bstrip(hin.msgnum)); hout.date := hin.date; hout.time := hin.time; hout.msgto := hin.msgto; hout.msgfrom := hin.msgfrom; hout.msgsubj := hin.msgsubj; hout.numblks := atoi(bstrip(hin.numblks)); hout.kill := hin.kill = #226; hout.confnum := hin.confnum; hout.ntwktag := hin.ntwktag; end; { convhdr } procedure writetexthdr(var t:text;hdr:prochdrtype); begin with hdr do begin writeln(t); writeln(t); writeln(t); writeln(t,'---------------------------------'); writeln(t,'Message number: ',msgnum); writeln(t,'Date: ',date); writeln(t,'Time: ',time); writeln(t,'From: ',msgfrom); writeln(t,'To: ',msgto); writeln(t,'Subj: ',msgsubj); writeln(t,'Conf: ',confnum); writeln(t,'---------------------------------'); end; { with } end; { writetexthdr } begin if paramcount < 2 then begin writeln('MYRDR v0.1'); writeln('Copyright 1993 by Brian Pape.'); writeln('usage:'); writeln(' MYRDR MESSAGES.DAT OUTFILE.TXT'); writeln('where MESSAGES.DAT is the name of the unpacked data file, and'); writeln('OUTFILE.TXT is the name of the text file to direct output to.'); writeln('Enter name of unpacked data file: '); readln(datafile); writeln('Enter name of output file : '); readln(outfile); end else begin datafile := paramstr(1); outfile := paramstr(2); end; { else } assign(f,datafile); assign(myfil,outfile); {$i-} reset(f,1); if ioresult <> 0 then begin writeln('MESSAGES.DAT file not found.'); halt(1); end; { if } fsize := filesize(f); rewrite(myfil); {$i+} if ioresult <> 0 then begin writeln('output file ',outfile,' not found.'); halt(1); end; { if } getmem(tbuf,tbufsize); settextbuf(myfil, tbuf^, tbufsize); writeln; s := ''; writeln; write('READ %'#8#8#8#8); { read the .QWK file header (c) by Sparkware... first } blockread(f,block,sizeof(block),size); pos := 1; blockread(f,buf,sizeof(buf),size); inc(numread,size); write(trunc(numread/fsize*100):3,#8#8#8); done := size = 0; while not done do begin { get the next message header and decode it } rawhdr := @buf[pos]; inc(pos,128); convhdr(rawhdr^,prochdr); writetexthdr(myfil,prochdr); j := 0; msgsize := pos + 128*pred(prochdr.numblks); while (pos < msgsize) and not done do begin if pos>size then begin { reset msgsize so that we still have the same number of bytes to go } msgsize := msgsize-pos+1; pos := 1; blockread(f,buf,sizeof(buf),size); inc(numread,size); write(trunc(numread/fsize*100):3,#8#8#8); done := size=0; if done then continue; end; { if } if buf[pos] <> #227 then begin inc(j); s[j] := buf[pos]; end { if } else begin s[0] := chr(j); j := 0; writeln(myfil,s); end; { else } inc(pos); end; { while } { in case pos > size, read some more data } if pos>size then begin pos := 1; blockread(f,buf,sizeof(buf),size); inc(numread,size); write(trunc(numread/fsize*100):3,#8#8#8); if (size=0) then done := true; end; { if } end; { if not done } end; { while } writeln; writeln('Done writing files.'); close(f); close(myfil); freemem(tbuf, tbufsize); end. { myrdr }