R: reduce redundency

[zekrom-vale]

There is some redundancy in the code. It should use utility and helper functions to streamline and standardize the process.

• For each 'span.explain' element, extract the 'title' attribute and replace the element's text with the title

spans|>
	walk(~{
		# Due to pointers this modifies the original HTML source
		title = rvest::html_attr(.x, "title")  # Extract the 'title' attribute
		xml2::xml_text(.x) <- paste0(":!:", title)  # Replace the element's text with the title
	})

• Rename the columns that match the specified patterns to "name"

rename_with(~ "name", .cols = matches("^(Name|Pok\u00e9mon)$"))

• Mutate the 'ndex' column by attempting to extract integer values from it, and if that fails (returns NA), use the original 'ndex' values

mutate(
	ndex = coalesce(as.integer(str_extract(ndex, "[\\d+.]")), suppressWarnings(as.integer(ndex)))
)

• The level inverse functions that use uniroot

ranges <- list(c(0, 2), c(2, 10^256))

# Find the range where e falls
for (range in ranges) {
	if (e >= exp$medium_slow(range[1]) && e <= exp$medium_slow(range[2])) {
		# Use uniroot to find the root of the function medium_slow(n) - e
		root <- uniroot(function(n) exp$medium_slow(n) - e, range)
		return(floor(root$root))
	}
}
# Return NA if no solution was found
return(NA)

• requireNamespace functions
• EXTRACT_REGEX = "([^()]+)(?:\\(([^()]+)\\))?"
• Mutate across characters if "" replace with NA_character_

mutate(across(
	where(is.character),
	~ if_else(.=="", NA_character_, ., NA_character_)
))

• Standardized column name function
• Standard mega processing
• Standard regional/form/name/mega processing
• Move away from useless variables IE URL (May keep HTML or pass it to a cashing function)