Case Converter

Case conversion is mostly a cleanup job: normalising an all-caps import, fixing text that arrived from a system that lowercases everything, or making a set of headings consistent. The operations look trivial and mostly are, with one asymmetry worth knowing — lowercasing is safe and reversible in principle, while uppercasing destroys information. Once "iPhone" becomes "IPHONE", nothing can tell you it was not "Iphone".

That is why proper nouns need checking after any conversion. Personal names, place names, brands with deliberate internal capitals, and acronyms all get flattened by a blanket pass. Sentence case is the most destructive here, because it lowercases everything after the first character and there is no general rule that distinguishes a proper noun from an ordinary one without understanding the sentence.

Case rules are also less universal than they look. German capitalises all nouns, not just proper ones, so a sentence-case pass on German text is simply wrong. Turkish has dotted and dotless i, and converting between cases without applying Turkish locale rules produces the wrong letter — a well-known source of bugs. If content is multilingual, apply case conversion per language rather than across the whole corpus, and read the output before publishing it.