Pages and Terminal output different results of word count

Because you are passing the raw RTF syntax in that document to wc, and Pages is showing only the actual words in your document. The RTF control words permeate throughout that .rtf document. Here is an example:

{\rtf1\ansi\ansicpg1252\cocoartf1265\cocoasubrtf210

{\fonttbl\f0\fnil\fcharset0 Baskerville;}

{\colortbl;\red255\green255\blue255;\red26\green26\blue26;\red164\green8\blue0;}

\margl1440\margr1440\vieww8820\viewh17160\viewkind0

\deftab720

\pard\pardeftab720\sa200\pardirnatural

{\header \pard\ql\b\f0\fs28 [ zig.rtf ] \par}

\f0\fs38 \cf2 Lorem ipsum dolor sit amet, ligula suspendisse nulla pretium, rhoncus tempor placerat fermentum, enim integer ad vestibulum volutpat. Nisl rhoncus turpis est, vel elit, congue wisi enim nunc ultricies sit, magna tincidunt. Maecenas aliquam maecenas \cf0 ligula\cf2 \cf0 nostra\cf2 , accumsan taciti. Sociis mauris in integer, a dolor netus non dui aliquet, sagittis felis sodales, dolor sociis mauris, vel eu libero cras. Interdum at. Eget habitasse elementum est, ipsum purus pede porttitor class, ut adipiscing, aliquet sed auctor, imperdiet arcu per diam dapibus libero duis. Enim eros in vel, volutpat nec pellentesque leo, {\field{\*\fldinst{HYPERLINK "http://www.hp.com"}}{\fldrslt temporibus}} scelerisque \cf3 nec\cf2 .\

and this is the text that Pages '09 shows you:

You can use textutil to convert to txt, output to stdout and pipe to wc

textutil -stdout -convert txt Untitled.rtf | wc

Tony's solution will work fine, though since you are interested in word count, I would suggest piping the textutil output to wc -w. This however gets you a word count value that is left padded with 7 spaces. To get your word count with leading and trailing spaces trimmed use the following:

$ textutil -stdout -convert txt five_words.rtf | wc -w | awk '$1=$1'

Result: 5

Here is an automator workflow that prompts for a PDF file, extracts text from it as RTF, passes that RTF through the above command sequence, and then pops a dialog displaying the word count. In my example, this was a one-page PDF.

There is the bc calculator as a UNIX tool. See man bc(1). Using the $words variable from my previous Automator example. You know how to make that display in AppleScript now from the command-line.

$ math_result=$(echo "$words * 10" | bc)

math_result = 3610.

You could do the math in the 'awk' segment

textutil -stdout -convert txt five_words.rtf | wc -w | awk '{print $1 * 10}'

If you need to pass in additional values for the math, you can do something like

textutil -stdout -convert txt five_words.rtf | wc -w | \

awk -v v1=${var1} -v v2=${var2} '{print ($1 * v1) + v2}'

You can have as man -v options as you need.

If you need to capture the output you can use

answer=$(textutil -stdout -convert txt five_words.rtf | wc -w | \

awk -v v1=${var1} -v v2=${var2} '{print ($1 * v1) + v2}')

echo $answer

Hi Bob,

Thanks for adding the awk calculations.

Screen capture (shift+command+4 and drag around the Extract PDF Text and Run Shell Script actions) your workflow, and use the camera icon in this editor to post the capture image here. You have a syntax error somewhere. You are correct: spaces in UNIX filesystem names are taboo, and must be escaped, or quoted.

Show me your code.

In your Run Shell Script action, change the solitary $f to "${f}" in the textutil command line, and it will handle filenames with spaces. Since I grew up on UNIX, I never, ever use filenames with spaces, or filenames that look like sentences.