London | 25-SDC-Nov | Aida Eslamimoghadam | Sprint 2 | Use precomputing for better performance

[aydaeslami]

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

Introduced precomputing to reduce repeated work and improve performance of common_prefix and count_letters functions.

[cjyuan]

Code looks good.

Can you use complexity to explain how the new implementation is better than the original implementation?

[aydaeslami]

Thank you, @cjyuan, for your help and support.

For " find_longest_common_prefix" :
The previous implementation compared the common prefix of every pair of strings, resulting in a time complexity of O(n² × m). The new implementation sorts the strings first and only compares adjacent strings, reducing the time complexity to O(n log n + n × m).

And

For " count_letters" :
The previous implementation searched the string for each uppercase letter, resulting in a time complexity of O(n²). The new implementation uses sets for faster lookups, reducing the time complexity to O(n).

[cjyuan]

Sorting strings involve comparing strings character by character. If we take into account, $m$, the average length of the strings in the given array, then the time complexity of sorting would also affected by $m$.

[aydaeslami]

Thanks @cjyuan, Yes, you are right. i missed the cost of string comparison during sorting.
ِThe overal complexity is :O(n log n × m + n × m)
simplifies to : O(n log n × m)

This part is a bit complicated for me. Please let me know if I need to change anything.

Many thanks.

[cjyuan]

All good. Nothing needs to be changed.