Homework & Coursework Questions

1. The problem statement, all variables and given/known data:

I'm fairly confident I can brute force this assignment, but let's not do that ;-).

Basically I'm required to support input such as ps aux | grep blah >> blah.txt& echo 'slslslsl'

My question is what is the best way to parse that line? I am required to use a linux based system and pure c code.

2. Relevant commands, code, scripts, algorithms:

I use strtok_r to handle breaking up the command.

3. The attempts at a solution (include all code and scripts):

I am considering building a syntax tree, but I realize it wouldn't be easy to parse because of c's lack of function overloading, no objects, etc. Perhaps I should read it token by token then?

4. Complete Name of School (University), City (State), Country, Name of Professor, and Course Number (Link to Course):

Name of School: University of Delaware
City: Newark (DE)
Country: US
Name of Professor: Dr. Swany
Course Number: CISC663 (no course link)

Edit:

This is a theoretical question~

I can handle the coding; just need to figure out the best/most flexible way of handling the input string.

I have to support "; & >> > << |".

Alright! Brute force try 1.

Lesse, I'm thinking tokenize by ";".

Tokenize again if string contains "&". Add property to command to be background command.

Now I'm not so sure what next. First off, I know that I have to read forward to detect redirection/piping. But I also have to be able to support a variable number of redirecting/pipes.

This would be much simpler if I had an OO language.

strtok can do the job fairly easily and does support mutiple tokens. check out flex. it's built for tasks like this if your allowed to use it. what is your desired output?

flex: The Fast Lexical Analyzer

I don't think I'm actually allowed to use it~

Right now, I'm attempting to build an AST tree, starting simple.

First I parse ';'. Easy enough.

I'm stuck on '&' with the tokenizer though. I can't tokenize using '&' because something like 'blah blah &' won't work. I can't depend on the first occurrence because something like 'blah & blah&' won't work. So, I'm thinking of parsing it manually.

>.>

Personally, I'm going to avoid regexps since I'm not too good with them.

which -- flex or strtok? strtok's a simple C function for splitting apart strings on delimiters. Beware that it actually modifies the input string.

Oh, strtok is fine.

Sorry, I've been really busy this semester.

I ended up brute forcing it in the end

Well, don't have enough time to actually think it through.