OfficeMax I've been thinking about getting an ADF scanner for a while, so that I can scan to PDF all of the print articles that currently occupy several boxes in the corner of my office. Dedicated ADF scanners tend to go for about $500+, so this seemed like an inexpensive way to go. The MP780 has duplex printing but not duplex scanning, which is about the only thing that the more expensive ones have.
My MP780 arrived yesterday. Here are some of my notes:
The build quality is very high. I'm very happy with it mechanically. It seems like an absolute bargain for $199. I know Canon wants to make it back with ink cartridge prices, but I'm not likely to use it for printing much (since I have a laser printer in my office).
The sheet feeder is fine, a little slow. About 30 second page with 200 dpi scanning. Vuescan doesn't work, so I have to use the Canon software which is a bit primitive.
Not being able to scan double-sided is the biggest pain since I'm scanning my old notes and files. What I'm doing right now is scanning both sides separately and then using a small UNIX shell script and pdftk to conjoin and collate the two files. Here's my little script, it's not much use unless you have pdftk installed and you know UNIX:
#!/bin/sh
if [ $# -ne 3 ]; then
echo 1>&2 Usage: $0 inputa.pdf inputb.pdf outputc.pdf
exit 127
fi
echo Bursting $1
pdftk $1 burst output _page_%02da.pdf
echo Bursting $2
pdftk $2 burst output _page_%02db.pdf
echo Conjoining to $3
pdftk _page*.pdf cat output $3
echo Removing extraneous files
rm _pag*.pdf
rm doc_data.txt
open $3
rm -i $1 $2
|
Since I'm not clever enough to program pdftk to collate the pages so that you can scan 1 3 5 7 9 .... flip over and scan 10 8 6 4 2 and put them together to make 1 2 3 4 5 6 7 8 9 10, what I'm doing is scanning 1 3 5 7 9 and then collating manually the physical pages to scan 2 4 6 8 10 and then using the short script above to conjoin the two to get 1 2 3 4 5 6 7 8 9 10. If anyone is clever enough to program pdftk so that I don't have to reshufle the pages, I'd appreciate it!
Update 2005.11.07Here's a revised script that I wrote. It's even uglier than the first one but it'll do reverse collating. Gosh it's ugly. Can some /bin/sh guru please rewrite it so that it's prettier? What a hack. With this script (up to a limit of 99 total pages), it'll do the reverse collating of the B-side, so you can scan 1 3 5 7 9 then flip over the stack and scan 8 6 4 2 and it'll conjoin them properly to give you 1 2 3 4 5 6 7 8 9.
#!/bin/sh
if [ $# -ne 3 ]; then
echo 1>&2 Usage: $0 inputa.pdf inputb.pdf outputc.pdf
exit 127
fi
echo Bursting $1
pdftk $1 burst output _page_%02da.pdf
echo Bursting $2
pdftk $2 burst output _page_%02db.pdf
total=`ls _page*b.pdf | wc -l`
count=$((total))
echo B-side has $count pages
while [ $count -gt 0 ]; do
dest=$((total - count + 1))
if [ $count -lt 10 ]; then
countspace="0"
else
countspace=""
fi
if [ $dest -lt 10 ]; then
destspace="0"
else
destspace=""
fi
echo Moving _page_${countspace}${count}b.pdf to _page_${destspace}${dest}c.pd
f
/bin/mv _page_${countspace}${count}b.pdf _page_${destspace}${dest}c.pdf
count=$((count - 1))
done
echo Conjoining to $3 -- this may take a while
pdftk _page*.pdf cat output $3
echo Removing extraneous files
rm _pag*.pdf
rm doc_data.txt
open $3
rm -i $1 $2
|
Are you using a Mac? If so, Automator's PDF actions will do what you want (using shuffle option).
Dirk -
I'm Automator ignorant. Can you really program it so that it can do a complex shuffle like that? It has a BASIC or C-like language? For some reason, I had thought of it as being very simple and useless. Can you give me any tips?
My 1-month ownership report is that the MP780 is fantastic. The front door doesn't want to flip down for me, so I have to prod it with my finger but that's the only mechanical "quirk". The OSX software that Canon provides is lousy but does the job (even if it hangs from time to time). The printer component is excellent although it does guzzle ink (like all inkjets). I'm switching to generic ink after I run out.
Hi Karen, I was Automator-ignorant until the day before yesterday. After half an hour I had a workflow that synchs my iBook with my iMac the way I like.
I think what you are trying to do needs just two actions, select two input files ("Get Files" under Finder actions) and do the PDF merge in action two (under PDF actions). In shuffle mode it would take p 1 from file A, then p 1 from file B, then p 2 from file A, p 2 from file B and so on to produce the end result.
I can send you the workflow if you like, but I think you would have no trouble doing it yourself.
Just realised that my proposed solution only does what you are already doing. I did not consider that pages are in reverse order after the first scan.
It seems that you can add third party actions to Automator, so a little research may find you an action that does a reverse shuffle. You may even be able to create one yourself. You just need to find a way to reverse the page order of the second scan... Good luck.
Yeah. If I was a better shell scripter, I bet I could do it in the shell script too since pdftk is a pretty powerful program. I guess I need to go pick up an O'Reilly's on the UNIX shell.
Karen
p.s. Dirk, please please please TypeKey register! That way I don't have to keep manually approving your posts. (I get sooooo much comment spam, that only TypeKey posts are automatically let in)
I might have help for you with concern to your double-sided scanning on the Canon and OSX. Instead of manually reordering the pages, scan the back-side pages in reverse order (last page first). You'll still have two scan files (1 with all the front pages in order and the other with all the back pages in reverse order).
Then open the back page file in a freeware program called Combine PDFs for OS X. I think I found it on Versiontracker. In any case, drag the back side file to Combine PDFs and click the Page column to have it sort by page number. You can click it to have it sort in reverse page order, then save it as a new file.
From there, use the Automator idea from above to combine the front document and the newly ordered back document.
One caveat to Combine PDFs is that it sorts the page numbers the old, archaic way (1 gets lumped in with 10, 11, 12; 2 gets lumped in with 20, 21, 22, etc), so you may have to move a few pages manually (no more than 9).
Hope that helps.
I studied a bit up on the /bin/sh and wrote an automated script that does what I want. It's extremely ugly, but works... I think. I'm scanning several THOUSAND articles, so I wanted something simple so I could batch scan as much as possible, which meant either drag-n-drop or a simple script. Phew, thanks for your help guys!
This is an old post which i stumbled upon using google so I'm not sure if you care anymore. You can use pdftk to reverse the page rather than bash. Anyway, here is a simpler script which should do what you want:
#!/bin/sh
if [ $# -ne 3 ]; then
echo 1>&2 Usage: $0 inputa.pdf inputb.pdf outputc.pdf
exit 127
fi
echo Bursting $1
pdftk $1 burst output _page_%02da.pdf
echo Reversing $2
pdftk A=$2 cat Aend-1 output $2_reverse.pdf
echo Bursting $2_reverse.pdf
pdftk $2_reverse.pdf burst output _page_%02db.pdf
echo Conjoining to $3
pdftk _page*.pdf cat output $3
echo Removing extraneous files
rm _pag*.pdf
rm doc_data.txt
rm -i $1 $2 $2_reverse.pdf