Documentation of AnsiCompareText/AnsiCompareStr misleading
The documentation of AnsiCompareText and AnsiCompareStr says: "Compare two strings, ... , ignoring accents characters"
I think the latter part is misleading because it suggests that a comparison of 'A' with 'Á' (accented A) is true which is not observed (see also https://forum.lazarus.freepascal.org/index.php/topic,65872.msg502671).
I checked the Delphi documentation and it does not mention anything like this. Translated from German:
"Compares two strings taking care of the current locale without respecting upper/lower case. AnsiCompareText compares the two strings S1 and S2 without distinguishing between upper and lower case. The comparison is controlled by the current locate. The return value of AnsiCompareText is negative when S1 < S2. The return value is positive when S1 > S2. When S1 = S2, zero is returned."
My suggestion is to remove the "ignoring accents characters", but mention, like in Delphi, that the comparison is made based on the current system locale.