Project of C++ atleast 1000 lines of code.

[nabeelkhanjadoon10]

I am first semester student and studiyg CS. I need a c++ project atleast 1000 lines of code. but the project is too simple. I have studied yet Arrays..

[Taywin]

Hmm... I am not sure what you mean by your post. You want a C++ program source code that contains more than a 1000 lines of code? But what does "the project is too simple" mean??? You haven't even studied array and you think there would be a reasonable simple program which contains more than a 1000 lines of code without using any array or pointer?

Anyway, if you want to find some source code, you should google it. There are tons of codes left on google code site or any opensource project.

[Schol-R-LEA]

Can you please describe this 'too simple' project for us? We can only give advice when we know what we're dealing with...

[rubberman]

I recently wrote a C++ program to take sar data from a linux system (system activity reports) and convert it into time-series metrics. The code is about 1500 lines. The original code took a few days to write. Getting it into production in a major corporate environment to monitor all of our world-wide servers (2000+ servers in 4 data centers) has taken another 9 months... most of which is politics and smoothing the rough edges (reliability, availability, failure-tolerance).

The basic project (capture system performance data into a time series database) was not difficult. Less than a week of effort to prove the concept and get valuable data from production systems. Just a lot of string parsing and filtering.

[ddanbe]

What is your definition for "a line of code"?
Does it really matter how many lines of "code" a program contains?
Is a program with 100000 lines of code better that one with 1000 lines of code?

[rubberman]

ddanbe - good point! A good engineer can write in 1000 lines what another may not accomplish in 10,000. A great one will reduce that to 100 lines that do no more, and no less, than needed... :-)

This is why most "enlightened" organization measure productivity in "function points" and not lines-of-code. However, that still begs the question as to whether or not those function points are relevant to the comapany's goals - 10 appropriate FPs may be more valuable than 1000 that don't address the organization's needs.

[Smeagel13]

1500+ lines in a few (2) days...

Either that includes a hell of lot of blank lines, copying and pasting, or there was no testing because that's just daft!

[mike_2000_17]

1500+ lines in a few (2) days...

Either that includes a hell of lot of blank lines, copying and pasting, or there was no testing because that's just daft!

First, it certainly doesn't include blank lines or comment lines, nobody includes those in code-line counts. Second, rubberman did mention that the code only got to production code about 9 months later, after thorough testing / QA / debates. Third, counting only the time for writing the code, starting from nothing to having a fully working (somewhat tested) piece of code of about 1500 LOCs is easily done in a few days if the task is relatively straight-forward.

[rubberman]

Re. Smeagel13:

1500+ lines in a few (2) days...
Either that includes a hell of lot of blank lines, copying and pasting, or there was no testing because that's just daft!

Well, after 30+ years of writing code that runs a lot of the biggest manufacturing enterprises in the world, a first cut of 1500 lines to solve a problem is, for me, about 2-3 days effort. One day of that was to design/model it in a top-end UML tool. FWIW, I wrote a complete TCP/IP stack for an embedded real-time operating system in the early 1990's, and that was 6 months from start to first formal system-level test. Every line of code is original. If you are interested, check out US Patent # 7185325. I am the sole inventor of it (although Applied Materials owns it): http://www.google.com/patents/US7185325

FWIW, the SAR project is now about 6000 lines of C++ code (started about 9 months ago), not including support scripts, rpm file specs, etc, and that is only a part of what I have been doing, such as mastering the Amazon cloud, Hadoop, developing algorithms (including all the math) for predictive analytics to determine when systems will fail using the data that the tool captures. We are collecting with this tool in real-time, data from a number of our world-wide data centers (about 1 billion data points per DC per day), and soon will be analyzing it, also in real-time. Do you know what a Kalman filter is? :-)

What did I do today? I spun up a 15 node hadoop cluster in the Amazon cloud, installed Hadoop and OpenTSDB, started collecting data to test the latest versions of Hadoop, OpenTSDB, our new secure cloud servers, and our latest code that adds another 350M data points per day from each data center when online - all between 9am and 5:30pm. What did you do?

Ok. I'll stop beating my chest now... :-)

[DoRight]

Rubberman wins. [grn]