Useful Little Hacks

Greetings!

A hacklet is, in my view, a dumb little script or utility that you could probably have created yourself but someone else has already done it so why bother spending an hour getting the nits correct, eh?

Here are a few small hacklets and utilities I hacked (or co-hacked) to make life easier for myself:

• countWords.awk -- Wondering if your latest paper fits under the 5000 word limit? Here's an awk script I wrote in about 10 minutes that count the number of words in a Latex file, ignoring document preambles, comments, and tokens beginning with a backslash.
  Usage: awk -f countWords.awk foo.tex

• countLOC.awk -- an awk script written in 1988 by Dan Kozak that counts lines of code but ignores comments and blank lines. I hacked it slightly to ignore C++/Java-style comments that begin with "//". This is very simple minded.
  Usage: awk -f countLOC.awk *.[ch]

  Vishal Goenka sent me a link to his little Unix scripts that are even more useful.

• colourizeSource.sh -- a (rather trivial) shell script that creates colourized syntax highlighted HTML versions of any file that the vim text editor has syntax highlighting support for (you can also roll you own for your favourite private notations, as I did for grok below). For each file foo.ext, it creates foo.ext.html. There are a large number of supported notations, including almost all programming languages of note, shell scripts, HTML files, Xdefault files, ... It is kinda slow tho. Requires gvim and that X windows be running. Here's an example of a C file that I ran this through. (Don't like the colour scheme? You can change it.)

• grok.vim -- a syntax file for the grok relational calculator for use with the vim text editor.

• make2acacia.awk -- a script that generates a ksh script to run an Acacia (CCia) extraction on a system from a makefile. Typical use:

      % ./configure           (if appropriate)
      % make -n > make.out    (or make > make.out)
      % awk -f make2acacia.awk make.out > make.acacia
      % chmod 755 make.acacia 
      % ./make.acacia 
      

  The script make have to be hacked at still, but it's a good starting point. This works for me out of the box for simple systems.

• make2cfx.awk -- a script that generates a /bin/sh script to run an cfx/fbgen (PBS toolkit) extraction on a system from a makefile, analogously to make2acacia.awk. Typical use:

      % ./configure           (if appropriate)
      % make -n > make.out    (or make > make.out)
      % awk -f make2cfx.awk make.out > make.cfx
      % chmod 755 make.cfx 
      % ./make.cfx 
      

• taxform.tgz -- some scripts that allow you to use the Acacia extractor "CCia" (as well as the older "cia" extractor that comes with Acacia), as a drop-in replacement for cfx/fbgen with the PBS tool suite. Note that while Acacia/CCia does handle C++ reasonably well, these scripts assume your system is written in C. The advantage of using CCia over PBS's cfx/fbgen is that the Acacia extractor seems to be a bit more robust and slightly more accurate (plus there is more detail extracted if you want to muck around). This is part of the TAXFORM project. You must have the following software installed: Acacia, ksh93 (comes with Acacia), PBS, grok 3.0 or better, gawk. It works for me, but I make no promises. The README file gives gory details on use, and the doc subdir outlines the data schemas used. A good example system to try it all out on is ctags-3.0.tgz.

Last modification: Monday March 8, 2010 at 07:00:42 (EST)