longest_uniform_substring_after_replacements in Sliding Windows incorrectly tracks highest_freq

[harvv]

On page 97 in the book, an error is pointed out for an initial attempt at a solution, in that it doesn't track the highest_freq variable correctly as it runs and shrinks the window.
The solution given, which slides the window instead of shrinking in however, also has this problem if run with the parameters as:
longest_uniform_substring_after_replacements("aabcdeca", 2)

(whereas the book's test string is "aabcdcca")

Once the sliding window is on substring "abcde", the highest_freq value in the while loop is still 2 (which is incorrect).

(The overall function still seems to return the correct max_len value in this case, however.)

[shaungunaw]

Hi Harvv,

Thanks for pointing this out! Yeah, we were already aware of this. The reason highest_freq doesn't need to be updated in this case is because we only care about updating it when we find a new observed "highest frequency" that's greater than the current highest_freq.

This holds true because we're sliding the window instead of shrinking it, which means we're only ever updating the window to a larger size (or the same size). As a result, we don't need to bother with windows that have a lower "highest frequency" value.

I hope this makes sense! If you'd like further clarification, feel free to send a message on our Discord server (link on page 2).

[harvv]

Yeah, it isn't fundamental to the code working. I only opened the issue because the book text points out and makes a big deal out of the previous solution that it presented as not accurately tracking this variable. It might confuse less experienced coders when they realize the final solution also doesn't track it accurately - albeit this time with no serious consequence to the overall accuracy.