Assignments #

Created Saturday 21 December 2019

The remove duplicate and compress strings algos can be done in a single scan. The idea: A writing head, borrowed from COA. For remove duplicate: Make a left and a rigth pointer, the left will indicate the current (which has not yet been written). Find a diffrent letter(it could be EOS), just write the *left to wr left =right, coz we are done with the currnt letter, let’s start with the different letter found(EOS case has been handled). Move wr by 1, for a fresh writing. The next iteration begins, if we had reached EOS, we stop, else we continue;

For compression(in 1 go): make a count variable. Set left = head, right = head, wr = head; Look for a number or change of character. add count. If count > 2; Just write *wr = *left, wr ++; *wr = count+48, wr++; If count < 1, i.e we found an a different chararacter *wr = *left;

left = right; we are done with the present charcter let’s look at the difference. The catch, as we have already noted down everything neatly of the previous part, so no need to think of that now. So no number checking. As and when we catch EOS as difference, we do the needful for the present char+number or char. The EOS is taken care of in the while loop. No worries.

• Lessons learnt, do the problem in a go.
• Don’t process in steps, as we have to keep track of everything we did, which is painstaking and costly too.
• Use what is efficient and working, rather than going for ideals.
• Make a special case seem a general case by including it in the condition, but after the current process.