It would probably be easier to encode the information in some way other than a bar code, such as a comment near the top or bottom of the PDF file.

Bar codes are a useful way to get digital information out of a physical document, but they're generally not the best way to transmit digital information between two computers.

The PDF is created automaticaly by the scanner. I never have a chance to encode it with anything. :-(

The bar code would be printed as a seperate process. I was also thinking of just haveing one barcode per location. Then there would be just a standard cover sheet they could use to send the documents.

Update: I can get the document scanned as tiff files, one page per file. I don't know if maybe tiff is easier to scan.

___________
Eric Hodges

The first thing I would ask myself if I was in your shoes would be why I was introducing an extra PC, some code, and an extra step for the sorting room people, when they system they have at the moment probably works OK as it is.....

Assuming that you have a good answer to this, then you may be interested to know that PDF is derived from PostScript. Postscript is a fully fledged programming language - which can be written in text format. It also supports comments :)

To see what I mean, paste the following output into a file called "test.ps", and open it in your favourite PostScript viewer.

%!PS-Adobe-3.0 EPSF-1.2
%%Title: (test1.ps)
%%LanguageLevel: 1
%%Creator: DTR
%%CreationDate: Sun Dec  8 18:25:51 2002
%%For: Console
%%DocumentMedia: A4 595.27559 841.88976 0 ( ) ( )
%%Orientation: Portrait
%%Pages: 1
%%BoundingBox: 0 0 595 841
%%EndComments
%%BeginProlog
%%BeginResource: PostScript::Simple
/u {} def
/STARTDIFFENC { mark } bind def
/ENDDIFFENC { 

% /NewEnc BaseEnc STARTDIFFENC number or glyphname ... ENDDIFFENC -
    counttomark 2 add -1 roll 256 array copy
    /TempEncode exch def
    
    % pointer for sequential encodings
    /EncodePointer 0 def
    {
        % Get the bottom object
        counttomark -1 roll
        % Is it a mark?
        dup type dup /marktype eq {
            % End of encoding
            pop pop exit
        } {
            /nametype eq {
            % Insert the name at EncodePointer 

            % and increment the pointer.
            TempEncode EncodePointer 3 -1 roll put
            /EncodePointer EncodePointer 1 add def
            } {
            % Set the EncodePointer to the number
            /EncodePointer exch def
            } ifelse
        } ifelse
    } loop    
    TempEncode def
} bind def

% Define ISO Latin1 encoding if it doesnt exist
/ISOLatin1Encoding where {
%    (ISOLatin1 exists!) =
    pop
} {
    (ISOLatin1 does not exist, creating...) =
    /ISOLatin1Encoding StandardEncoding STARTDIFFENC
        144 /dotlessi /grave /acute /circumflex /tilde 
        /macron /breve /dotaccent /dieresis /.notdef /ring 
        /cedilla /.notdef /hungarumlaut /ogonek /caron /space 
        /exclamdown /cent /sterling /currency /yen /brokenbar 
        /section /dieresis /copyright /ordfeminine 
        /guillemotleft /logicalnot /hyphen /registered 
        /macron /degree /plusminus /twosuperior 
        /threesuperior /acute /mu /paragraph /periodcentered 
        /cedilla /onesuperior /ordmasculine /guillemotright 
        /onequarter /onehalf /threequarters /questiondown 
        /Agrave /Aacute /Acircumflex /Atilde /Adieresis 
        /Aring /AE /Ccedilla /Egrave /Eacute /Ecircumflex 
        /Edieresis /Igrave /Iacute /Icircumflex /Idieresis 
        /Eth /Ntilde /Ograve /Oacute /Ocircumflex /Otilde 
        /Odieresis /multiply /Oslash /Ugrave /Uacute 
        /Ucircumflex /Udieresis /Yacute /Thorn /germandbls 
        /agrave /aacute /acircumflex /atilde /adieresis 
        /aring /ae /ccedilla /egrave /eacute /ecircumflex 
        /edieresis /igrave /iacute /icircumflex /idieresis 
        /eth /ntilde /ograve /oacute /ocircumflex /otilde 
        /odieresis /divide /oslash /ugrave /uacute 
        /ucircumflex /udieresis /yacute /thorn /ydieresis
    ENDDIFFENC
} ifelse

% Name: Re-encode Font
% Description: Creates a new font using the named encoding. 

/REENCODEFONT { % /Newfont NewEncoding /Oldfont
    findfont dup length 4 add dict
    begin
        { % forall
            1 index /FID ne 
            2 index /UniqueID ne and
            2 index /XUID ne and
            { def } { pop pop } ifelse
        } forall
        /Encoding exch def
        % defs for DPS
        /BitmapWidths false def
        /ExactSize 0 def
        /InBetweenSize 0 def
        /TransformedChar 0 def
        currentdict
    end
    definefont pop
} bind def

% Reencode the std fonts: 
/Courier-iso ISOLatin1Encoding /Courier REENCODEFONT
/Courier-Bold-iso ISOLatin1Encoding /Courier-Bold REENCODEFONT
/Courier-BoldOblique-iso ISOLatin1Encoding /Courier-BoldOblique REENCO
+DEFONT
/Courier-Oblique-iso ISOLatin1Encoding /Courier-Oblique REENCODEFONT
/Helvetica-iso ISOLatin1Encoding /Helvetica REENCODEFONT
/Helvetica-Bold-iso ISOLatin1Encoding /Helvetica-Bold REENCODEFONT
/Helvetica-BoldOblique-iso ISOLatin1Encoding /Helvetica-BoldOblique RE
+ENCODEFONT
/Helvetica-Oblique-iso ISOLatin1Encoding /Helvetica-Oblique REENCODEFO
+NT
/Times-Roman-iso ISOLatin1Encoding /Times-Roman REENCODEFONT
/Times-Bold-iso ISOLatin1Encoding /Times-Bold REENCODEFONT
/Times-BoldItalic-iso ISOLatin1Encoding /Times-BoldItalic REENCODEFONT
/Times-Italic-iso ISOLatin1Encoding /Times-Italic REENCODEFONT
/Symbol-iso ISOLatin1Encoding /Symbol REENCODEFONT
/box {
  newpath 3 copy pop exch 4 copy pop pop
  8 copy pop pop pop pop exch pop exch
  3 copy pop pop exch moveto lineto
  lineto lineto pop pop pop pop closepath
} bind def
/circle {newpath 0 360 arc closepath} bind def
%%EndResource
%%EndProlog


% TRY CHANGING SOME OF THESE VALUES TO GET A FEEL 
% FOR WHAT HAPPENS 

0.6 0 0 setrgbcolor        %red
/Arial findfont 20 scalefont setfont

newpath
1 u 450 u moveto
(Hello world!) show stroke

newpath
1 u 200 u moveto
(This \(stuff\) was generated entirely from Perl) show stroke

0 0.8 0 setrgbcolor
50 u 50 u 150 u 150 u box stroke

1 1 0.2 setrgbcolor
100 u 100 u 50 u circle fill

0 0 0.8 setrgbcolor
newpath
100 u 100 u moveto
(This is a test) dup stringwidth pop 2 div neg 0 rmoveto show


% A simple arc
newpath
300 300 50 0 90 arc
closepath
stroke

% A arc between 2 lines
% giving the appearance of a rounded corner
newpath
400 400 moveto
400 410 lineto
400 420 410 420 10 arct
420 420 lineto
stroke





% An example of a box in a dropout colour
newpath
500 500 moveto
0 20 rlineto
20 0 rlineto
0 -20 rlineto
closepath
1 1 1 setrgbcolor
fill
/Arial findfont 20 scalefont setfont
510 500 moveto
0.6 0.6 0.4 setrgbcolor



0 0.8 0 setrgbcolor
0 u 180 u 20 u 160 u box stroke
22 u 180 u 42 u 160 u box stroke
44 u 180 u 64 u 160 u box stroke
/Arial findfont 18 scalefont setfont
0 0 0.8 setrgbcolor
newpath
10 u 163 u moveto
(B) dup stringwidth pop 2 div neg 0 rmoveto show
newpath
32 u 163 u moveto
(O) dup stringwidth pop 2 div neg 0 rmoveto show
newpath
54 u 163 u moveto
(X) dup stringwidth pop 2 div neg 0 rmoveto show
/Arial findfont 12 scalefont setfont
0.8 0 0 setrgbcolor
newpath
10 u 350 u moveto
(My address is:) show stroke
newpath
10 u 330 u moveto
(HELLO) show stroke
%%EOF
[download]

NOTE - where some of the "REENCODEFONT" words have been wrapped above with a "+" sign, you need to remove the "+" sign and put them onto one line for this to work

Credit is due to the PostScript::Simple module on CPAN for getting me started with this. Also - disclaimer - I know just enough PostScript to draw the circles, boxes, and text I drew on that page - no more.

Anyway, that should be enough to get you started :). You should be able to insert a few harmless comments of your own at the top of a PostScript file (you may also be able to do it with PDF), and use these to keep track of where the document should go.

I don't know anything about your ScanToEmail machine, but would there be any way of spitting out the actual barcode number as part of the scanning process? Say, put the number into the Subject header or something?

Cheers,
Ben.

-derby

Thanks for all the replies but i appear to have confused you all somewhere.

The documents all start in paper form. Not digital. The cover sheet would also need to be in paper form not digital so that they could join it with which ever set of documents they want to transfer.

So I start with all paper, and scan it to email (using an HP 4100mfp). There is no computer attatched to the MFP to attatch any form of digital document.

I've been searching and have found some Barcode recognition software that i may try, since barcodes still seem to be the only safe way to put the info on need on a peice of paper and have it read by the scanner. Currently they do send each batch to its own email address, but that has its own limitations and downfalls.

Thanks again!

To not sure if this would work for you, and I may be way off base but how about this.

Your current system is, sort->scan->send. And you are trying t simplify the send step. This appears to be the simplest step. Now if if you really want to get crazy how about this.

scan->PDF->Bayesian filters->send

Here as the paper docs arrive they are immediately scanned as PDF. Then using one of the PDF modules you pull the text out and send it through a series of Bayesian filters to determine where they should go. The PDF file is then attached into an email or loaded into a db based on the characteristics in the file. (The text version is only used for the Bayesian filter, and is discarded as soon as you deal with the PDF.)

This reduces the workload of the sorters to just scanning the files.

Would be great. Unfortunatly the sort part of the process includes more than just sorting. Often there are human actions (phone calls, scheduling, etc) that need to take place, and most of the documents are hardly legible (hand writting, doodled on, faxed many times, etc) to humans, let alone computers :-(

___________
Eric Hodges