Lab46 Wiki

This page is read only. You can view the source, but not change it. Ask your administrator if you think this is wrong.
======SYSPROG PROJECT: Writing Unix Tools (wut1)======

=====OBJECTIVE=====
To apply your burgeoning sysprog skills in the writing of familiar and described UNIX tools.

=====GRABIT=====
I have rigged up a grabit for **wut1**, which contains a directory structure, some useful Makefiles, and a header file to assist you in the development process.

=====PROGRAM=====
It is your task to implement getopt(3)-style command-line arguments and ensure functionality for the following tools (pick one, but your one must be unique; each person is doing a different one). You may choose to do others for potential bonus point considerations.

The programs are (edit the appropriate section and record your name to claim it; once someone has claimed a program, it is considered unavailable for others in the class to do for core credit):

====command-line arguments====
All of the programs should accept the following command-line arguments (check for program-specific additional ones as well in the appropriate section):

  * **-h** display usage information and exit
  * **-V** display version information and exit
  * **-s "STRING"** specify STRING as value to process
  * **-4** set nibble as processing unit/word size
  * **-7** set byte/word as 7-bits
  * **-8** set byte/word as 8-bits (default)
  * **-n** no delimiter between processing units
  * **-d 'CHAR'** use CHAR as delimiter between processing units (space is default)
  * **-q** quiet, do not display anything to STDOUT
  * **-v** verbose, display more information to STDOUT 

====general program processing/usage====

**__usage__:** the program takes delimiter-separated numeric values via specified filename, or as a string communicated via the command-line, or from STDIN, converts them and then displays the result as appropriate for the particular tool (considering delimited output as well; delimiter is set to a single space by default). The program terminates upon exhausting all of its input.

====compiling===

The Makefiles are set up for this, but you are basically to compile each program with the following compiler options:

  * **-Wall** increased warning pickiness/verbosity
  * **-Werror** treat warnings as errors
  * **--std=c99** compile with C99 language standards

====binary.c====

<code>

 binary.c - program to display hex numbers in their
            binary representation.

  synopsis: binary [OPTION]... [FILE]...

   execute: binary -s "de ad be ef 00 01 5a"
        or: echo "de ad be ef 00 01 5a" | binary
</code>

Additional Command-line arguments to implement:

  * **-B** input data is to be considered as binary (basically a passthrough)
  * **-O** input data is to be considered as octal
  * **-D** input data is to be considered as decimal
  * **-H** input data is to be considered as hexadecimal (default)

Claimed by: Christian Cattell (ccattell)

====parity.c====

<code>
 parity.c - program to check the parity (odd, even, none)
            of input binary values.

  synopsis: parity PARITYOPTION [OPTION]... [FILE]...

   execute: parity -O -s "10110010"
        or: echo "10110010" | parity -E
</code>

Additional Command-line arguments to implement:

  * **-O** odd parity
  * **-E** even parity
  * **-N** no parity

Lack of one of these options should generate an error.

Claimed by:Jeff Jansen

====endian.c====

<code>
endian.c  - program to encode value(s) according to
            an indicated endianness.

  synopsis: endian [OPTION]... [FILE]...

   execute: endian -s "dead beef" -e -w 16 -b 16
        or: echo "de ad be ef" | endian -E -b 16
</code>

Additional Command-line arguments to implement:

  * **-w #** set wordsize (power of 2 between 4 and 128, 8 is default)
  * **-u** also consider underlying bytes in endianness encoding
  * **-E** encode as big endian
  * **-e** encode as little endian (default)
  * **-b BASE** input/output base (of 2, 8, 10, or 16; base 2 is default)

Your program can indicate a "setting not supported" message (and exit) if the **-7** argument is specified.

The **-4** and **-8** arguments are equivalent to **-w 4** and **-w 8**, respectively.

Claimed by: Dillon Vargeson(dvarges3)

====checksum.c====

<code>
checksum.c - program to calculate checksum
             of provided values.

   synopsis: checksum [OPTION]... [FILE]...

    execute: checksum -b 16 -s "de ad be ef"
         or: echo "de ad be ef" | checksum -b 16
</code>

Additional Command-line arguments to implement:

  * **-b BASE** input/output base (of 2, 8, 10, or 16; base 2 is default)
  * **-P** perform parity word checksum
  * **-S** perform word sum checksum (default)

Claimed by:Patrick (phastin1)

====twoscomp.c====

<code>
twoscomp.c - program to calculate two's complement
             of provided values.

   synopsis: twoscomp [OPTION]... [FILE]...

    compile: gcc -o twoscomp twoscomp.c -Wall -Werror --std=c99
    execute: twoscomp -s "10011010"
         or: echo "11001011" | twoscomp
</code>

Additional Command-line arguments to implement:

  * **-w #** set wordsize (power of 2 between 4 and 128, 8 is default)
  * **-b BASE** input/output base (of 2, 8, 10, or 16; base 2 is default)

Claimed by: Ben Schultes (bschulte)

====addition.c====

<code>
addition.c - program to perform addition (without
             actually using any addition- only logic operations)
             on the input values.

   synopsis: addition [OPTION]... [FILE]...

    execute: addition -s "10011010 11010001"
         or: echo "11001011 00011101" | addition
</code>

Additional Command-line arguments to implement:

  * **-b BASE** input/output base (of 2, 8, 10, or 16; base 2 is default)
  * **-C** final carry out value is set to return value (otherwise return 0 on success)
  * **-I #** set an initial carry in value (defaults to 0)

Claimed by: Kris (kbeykirc)

====bitmask.c====

<code>
bitmask.c  - program to apply specified bitmask
             against set of input values.

   synopsis: bitmask -M BITMASK [OPTION]... [FILE]...

    execute: bitmask -M "10110111" -s "10011010 11010001"
         or: echo "11001011 00011101" | bitmask -M "10110101"
</code>

Additional Command-line arguments to implement:

  * **-w #** set wordsize (power of 2 between 4 and 128, 8 is default)
  * **-b BASE** input/output base (of 2, 8, 10, or 16; base 2 is default)
  * **-M "BITMASK"** specify bitmask value for processing (required)

Claimed by: Andrei Bratkovski (abratkov)

====rotate.c====

<code>
rotate.c   - program to perform a bit rotate
             against a set of input values.

   synopsis: rotate -L[#]|-R[#] [OPTION]... [FILE]...

    execute: rotate -L 3 -s "10011010 11010001"
         or: echo "1100101100011101" | rotate -W 16 -R
</code>

A bit rotate preserves bits, a bit shift discards bits.

Additional Command-line arguments to implement:

  * **-w #** set wordsize (power of 2 between 4 and 128)
  * **-L #** perform a left rotate operation (shift by # bits; 1 if not specified)
  * **-R #** perform a right rotate operation (shift by # bits; 1 if not specified)

Claimed by: Aaron Houghtaling (ahought2)

====invert.c====

<code>
invert.c   - program to perform a bit inversion
             against a set of input values (toggle all
             0s to 1s, and 1s to 0s).

   synopsis: invert [OPTION]... [FILE]...

    execute: invert -s "10011010 11010001"
         or: echo "1100101100011101" | invert
</code>

Word size doesn't really matter for invert. It is merely flipping bits, however many there are.

Additional Command-line arguments to implement:

  * **-B** input data is to be considered as binary (default)
  * **-O** input data is to be considered as octal
  * **-D** input data is to be considered as decimal
  * **-H** input data is to be considered as hexadecimal

Claimed by: Matthew Chon (mchon)

=====SPECIFICATIONS=====

You will be implementing in accordance to the following specifications:

  * well commented
    * top comment banner including compile and sample execution instructions (and description)
    * consistent presentation of comments throughout code identifying process particulars
  * consistently indented
  * getopt(3)-style command-line arguments
    * help/usage
  * implement described functionality
  * do reasonable error checking to ensure correct operation

=====SUBMISSION=====
Your standard submission logic applies:

<cli>
lab46:~/src/sysprog/wut#$ make submit
...
</cli>
Lab46 Wiki

User Tools

Site Tools

Page Tools
