London | 25-SDC-July | Fatma Arslantas | Sprint 2 | Improve code with precomputing

[AFatmaa]

Learners, PR Template

Self checklist

• I have titled my PR with Region | Cohort | FirstName LastName | Sprint | Assignment Title
• My changes meet the requirements of the task
• I have tested my changes
• My changes follow the style guide

Changelist

I optimised common_prefix and count_letters using Precomputing techniques.

• Common Prefix:
  • Old: Compared every string with every other string.
  • New: I sorted the list first. Similar strings became neighbours. I only compared neighbours.
• Count Letters:
  • Old: Scanned the huge string repeatedly for every uppercase letter.
  • New: I created a Set of all characters first. Checking if a lowercase letter exists is now instant.

Questions

I don’t have any questions, thank you!

[cjyuan]

How do you know your solutions are better?

[AFatmaa]

Hi @cjyuan,

Thank you for the feedback!

For your question, I thought my solution is better for two reasons:

First, I analysed the old version that used nested loops (loops inside loops). I tried to improve this by using Sets and Sorting so the code doesn't need to check the same things repeatedly. And second, I ran the tests locally, and the heavy tests finished in 8 seconds. The old code would take much longer or freeze with that amount of data.

Do you think this logic is right?

[cjyuan]

You were in the right direction. A more convincing proof would be to compare their complexity using the big-O notations.

[illicitonion]

Closing PR because the SDC run has finished. Feel free to re-open if you're still working on it.