In the latest tidyverse dplyr blog post, first example explanation:
Filter for rows where “US” and “CA” have a score between 200-300, or rows where “PR” and “RU”
Should be
Filter for rows where “US” or “CA” have a score between 200-300, or rows where “PR” or “RU”
since it's impossible to have rows where US AND CA, or PR AND RU have anything - it's 1 row per country.
Bot the biggest deal, but an ironic confusion given the subject.
IDK if this is the right place to post this.
Also since it's the same blog post: for recode_values, case_match is ostensible the same as recode_values and that's addressed but at the end - could y'all link to that section after:
But
score ==is repeated so many times!
Since folks who use case_match will be wondering about it, and then you go to another function, before coming back to the case_match similarity.
One small wording concern: calling recode_values() a drop-in replacement for case_match() is a bit misleading in practice. While the overall logic is the same, the change from .default to default means existing code cannot be updated by a simple search-and-replace. If .default is left unchanged, code errors immediately.
In larger projects with substantial data wrangling, this becomes non-trivial, especially when case_match() and case_when() are used side by side and need to be preserved. You cannot safely replace all .default arguments globally without breaking other code.
So while the functions are conceptually very close, I would hesitate to describe this as a true drop-in replacement from a practical maintenance perspective.
Apart from that, great work — thanks for pushing this forward.