microsoft/cpprestsdk

utility::conversions::latin1_to_utf16 - Is there a bug for characters above 0x80?

Open

#101 ouverte le 23 mars 2016

Voir sur GitHub
 (1 commentaire) (0 réactions) (0 assignés)C++ (7 596 stars) (1 670 forks)batch import
bughelp wanted

Description

I've tried to use utility::conversions::latin1_to_utf16 to convert latin1 string to UTF-16. The string contained 'í' character (0xed). After converting it should become 0x00ed but it seems that converting method replaced it with 0xffed.

I think this may be result of casting signed char to utf16char and dest[i] = utf16char(s[i]); in asyncrt_utils.cpp should be replaced by dest[i] = utf16char(static_cast<unsigned char>(s[i])); to fix it. Can someone check if I'm right?

Guide contributeur

Stack technique
cpp
Domaine
apibackend
Type d'issue
bug
DifficultéDifficulté estimée pour un nouveau contributeur, de 1 pour un très petit changement à 5 pour un travail expert.
2
Temps estiméFourchette de temps approximative pour investiguer, implémenter, tester et préparer une pull request.
under 1 hour
Statut d'activitéDisponibilité apparente de l'issue : fraîche, active, ancienne, bloquée ou en attente d'un mainteneur.
stale
ClartéClarté avec laquelle l'issue explique le changement attendu, les critères d'acceptation et la prochaine étape.
clear
Prérequis
basic C++ knowledgeunderstanding of signed/unsigned char
Accessibilité débutantScore de 1 à 100 estimant l'accessibilité de cette issue pour un premier contributeur.
70
Direction de recherche
The issue identifies a potential bug in `asyncrt utils.cpp` in the `utility::conversions::latin1 to utf16` function. The problem is that casting a signed `char` to `utf16char` (likely `unsigned short`) causes sign extension for characters above 0x80, resulting in incorrect values like 0xFFED instead of 0x00ED. The suggested fix is to use `static cast<unsigned char>` before casting to `utf16char`. To confirm, examine the source file `asyncrt utils.cpp` in the repository and test the conversion with a Latin 1 character like 'í' (0xED). The maintainer should verify the fix and ensure it doesn't break other use cases.