Lâmôlabs

Background

Before my wife and I had kids we thought it would be nice to collect the United States quarters that were released during 1999 through 2008 to commemorate each of the 50 states. Seemed like something simple to do and would be a nice gift for the kids when they got older. So we bought a couple of the blue books which you can fill up with quarters as you find them. Each book contains 100 slots, 2 for each state. One slot is for the Philadelphia minted version of the quarter, and the other slot is for the Denver mint.

Problem

Well we ended up having 3 kids so we have to collect 300 quarters. The task of finding the quarters has been more of a dad task so when I have a chance, I’ll put a $5 dollar bill in various soda machines at work and go quarter fishing. This approach has been working fairly well and we’ve collected ~130 of the 300 quarters thus far.

However I’ve started getting to the point where I’m netting a lot of duplicates and the job of having to bring them home to weed through them is starting to get old. Having a list of which quarters we already have would sure be nice, so I could quickly nix any duplicates.

Solution

Of course I wanted a low tech solution, i.e. a piece of paper in my wallet would do the job, but how to do it?

The answer? A text file that I could maintain would suffice. No need for a bloated spreadsheet or some fancy handheld app. So I created a file, quarters.txt, like so:

1
2
3
4
5
6
7
8
9
10
11
12
13
14

State           Year  Cnt (P/D)
=====           ====  =========
Alabama         2003  2/1
Alaska          2008  3/1
Arizona         2008  3/0
Arkansas        2003  3/0
California      2005  3/0
Colorado        2006  3/0
Connecticut     1999  3/2
Delaware        1999  0/0
Florida         2004  2/0
Georgia         1999  2/1
Hawaii          2008  3/0
...

It’s about as simple a text file as you can get. 3 columns, State, Year, and Counts. The 3rd column shows how many P and D quarters I have for a given state. So for example, for Alabama, I’ve got 2 Philadelphia minted quarters, and 1 Denver minted.

So you’re probably wondering, “why the hell is this guy writing up this in a blog post?”

We’ll the interesting bit to this low tech solution is how I print this list out. For this task I make use of a pretty powerful UNIX command called enscript, which lets you do all kinds of nifty things to a text file to augment how it looks when it gets printed.

About the only thing enscript doesn’t do for you, is give you the ability to preview your text file prior to printing. To accomplish this bit, I made use of another powerful UNIX command called ps2pdf. This command will take a postscript file (ps) and convert it to a pdf file.

So putting all the pieces together I came up with the following command:

1

enscript --fancy-header -U 4 quarters.txt -o - | ps2pdf - quarters_sm.pdf

The first part of this command, will call enscript instructing it to convert the file quarters.txt, and print it to standard out -o -. The printout will include some fancy headers and enscript will print the text file out in what is called 4 UP. This means that 4 pages will be printed on a single piece of paper. You could also print the page out in 2 UP, 8 UP, etc. It only needs to be a power of 2. BTW, 2 and 4 are the most commonly used, 8 is pretty hard to read.

The second part of this command passes the postscript generated by enscript through a UNIX pipe which gets picked up by ps2pdf, and converts it into a PDF file, quarters_sm.pdf.

From here you can check what the page would look like using your favorite PDF viewer, such as evince or xpdf. Once you’re comfortable with the page you can actually print it out from the PDF reader, or via the command-line.

…. Continue reading → [one-liner]: Previewing a Pretty Printed Text File using enscript & ps2pdf »»

Background

From time to time I need to move and/or copy a subset of files from one directory to another. I typically would use something like one of these to do the task:

1
2
3
4
5

#-- COPY
find . -type f -ctime -1 | xargs -I {} cp {} /some/other/directory
 
#-- MOVE
find . -type f -ctime -1 | xargs -I {} mv {} /some/other/directory

NOTE: The 1st command finds all the files in the current directory that are less than 24 hours old, and copies them to /some/other/directory. The 2nd command finds all the files in the current directory that are less than 24 hours old, and moves them to /some/other/directory.

But then I realized that by using xargs’ –I switch I was basically limiting xargs to doing a single file at a time. According to xargs’ man page, when you use the –I switch you’re implying the –x switch AND the –L 1 switch. The L switch is what tells xargs how many lines of input to process at a time, so we’re basically telling it to only handle one file at a time. This made no sense. I was unintentially limiting xargs’ ability to optimize the command-line. So I found a better way.

New Approach

By utilizing 2 little used switches (–t | ––target-directory) on cp and mv I could un-tie xargs’ hands.

copy

1
2
3
4
5

# long form
find . -type f -ctime -1 | xargs -0 cp --target-directory=/some/other/directory
 
# short form
find . -type f -ctime -1 | xargs -0 cp -t /some/other/directory

move

1
2
3
4
5

# long form
find . -type f -ctime -1 | xargs -0 mv --target-directory=/some/other/directory
 
# short form
find . -type f -ctime -1 | xargs -0 mv -t /some/other/directory

NOTE: For further details regarding my one-liner blog posts, check out my one-liner style guide primer.