Checking the Stock Market from the Command Line

Andrew Watts — Thu, 09 Oct 2008 01:51:57 +0000

With all the financial trouble in the country I have been checking the market a lot lately, and I find it a little annoying to constantly switch to my web browser and refresh or check my dashboard or whatever. Since I spend a lot of time in a terminal, I put a function together, which will always display the latest market conditions for the Dow Jones Industrial Average, the Nasdaq and the S&P 500 as well as any stocks you feed the function.

For example:

$ market
Name Price Change Pct Day Lo Day Hi
Dow Jones Industr 9258.10 -189.01 -2.00% 9194.78 9628.07
NASDAQ COMPOSITE 1740.33 -14.55 -0.83% 1706.86 1806.89
S&P 500 INDEX,RTH 984.94 -11.29 -1.13% 970.97 1021.06

$ market GE XOM
Name Price Change Pct Day Lo Day Hi
Dow Jones Industr 9258.10 -189.01 -2.00% 9194.78 9628.07
NASDAQ COMPOSITE 1740.33 -14.55 -0.83% 1706.86 1806.89
S&P 500 INDEX,RTH 984.94 -11.29 -1.13% 970.97 1021.06
GEN ELECTRIC CO 20.65 0.35 +1.72% 19.90 21.99
EXXON MOBIL CP 77.00 0.93 +1.22% 74.00 79.39

The key to making this work is a csv generated by a http request to Yahoo. The request takes 2 arguments, the stocks and a string of tags.

For Example:
http://download.finance.yahoo.com/d/quotes.csv?s=^DJI+GE&f=nk1k2m2

Which says grab the name, last trade price, change percent, and the days range (all real time) for all the stocks defined by s. There is a whole slew of tags that you can grab, the best resource of this information that I found is located at Gummy Stuff’s Yahoo Data page.

After some magic with sed and awk you get the above result.

And finally Here is the code, if anyone wants to add it to their bash profile.

function market {
stocks="^DJI+^IXIC+^GSPC"
for arg in "$@"
do
stocks+="+${arg}"
done
# Note the sed commands below do the following:
# 1. Strip any html
# 2. Replace the '","' delimeter with a #
# 3. Replace the ' – ' delimeter with a #
# 4. Strip the leading and trailing " characters
# The awk command does some pretty printing on each record with # used for a delimeter
curl -s "http://download.finance.yahoo.com/d/quotes.csv?s=$stocks&f=nk1k2m2" | \
sed 's/< [^>]*>//g' | \
sed 's/","/#/g' | \
sed 's/ – /#/g' | \
sed 's/"$.*$"/\1/' | \
awk 'BEGIN{FS="#"; printf("%-20s %-8s %-7s %-6s %-8s %-9s\n", "Name", "Price", "Change", "Pct", "Day Lo", "Day Hi")} { printf("%-20s %8s %7s %6s %8s %9s\n", $1, $3, $4, $5, $6, $7) }'
}

I think some of the sed commands could be improved to reduce the number of them, but this is simple and works. So I’m happy for now, or would be if I started see some numbers from the economy that looked better.

Handling whitespace in bash for loops

Andrew Watts — Sat, 28 Jun 2008 03:31:24 +0000

Tonight I was trying to cleanup my macports installation by uninstalling the packages that are no longer active. I decided to write a script to do this, instead of manually uninstalling them one by one. But in writing the script I came across something I had never seen and the google didn’t give me a clear answer so I thought I would share.

A typical for loop might look like:

$ for file in `ls`
> do
> echo $file
> done

And this assumes there are no spaces in any of the files returned by ls. I did not know this. So I started doing something similar with this:

$ for package in `port installed | awk 'NF==2{print $0}' `
> do
> echo $package
> done

But I got output that looked like this:

curl
@7.17.1_0
gettext
@0.17_2

And wanted output like this:

curl @7.17.1_0
gettext @0.17_2

The problem is the awk output had spaces in it which were split by the for loop processing. And well to handle the spaces as desired you need to pipe the output to a while loop like so:

$ port installed | awk 'NF==2{print $0}' | while read pakage
> do
> sudo port uninstall $package
> done

And now $package is what I expect including the spaces, and my macports installation is a little cleaner.

thirstymind.org » bash

Checking the Stock Market from the Command Line

Handling whitespace in bash for loops