[libcxx-commits] [PATCH] D120317: [libcxx] [test] Fix get/put long_double_ru_RU on Glibc, FreeBSD and Windows

[Martin Storsjö via Phabricator via libcxx-commits]

mstorsjo added a comment.

In D120317#3359301 <https://reviews.llvm.org/D120317#3359301>, @uabelho wrote:

>> Alternatively, even better, can you provide the output of this test program:
>>
>>   #include <locale.h>
>>   #include <stdio.h>
>>   
>>   int main(int argc, char **argv) {
>>       const char *locale = "ru_RU.UTF-8";
>>       if (argc > 1)
>>           locale = argv[1];
>>       if (!setlocale(LC_ALL, locale)) {
>>           printf("setlocale(%s) failed\n", locale);
>>           return 1;
>>       }
>>       struct lconv *lconv = localeconv();
>>       printf("mon_thousands_sep: \"%s\",", lconv->mon_thousands_sep);
>>       for (const char *c = lconv->mon_thousands_sep; *c; c++)
>>           printf(" %02x", (unsigned char)*c);
>>       printf("\n");
>>       printf("thousands_sep: \"%s\",", lconv->thousands_sep);
>>       for (const char *c = lconv->thousands_sep; *c; c++)
>>           printf(" %02x", (unsigned char)*c);
>>       printf("\n");
>>       printf("int_curr_symbol: \"%s\"\n", lconv->int_curr_symbol);
>>       printf("currency_symbol: \"%s\",", lconv->currency_symbol);
>>       for (const char *c = lconv->currency_symbol; *c; c++)
>>           printf(" %02x", (unsigned char)*c);
>>       printf("\n");
>>       printf("decimal_point: \"%s\",", lconv->decimal_point);
>>       for (const char *c = lconv->decimal_point; *c; c++)
>>           printf(" %02x", (unsigned char)*c);
>>       printf("\n");
>>       printf("mon_decimal_point: \"%s\",", lconv->mon_decimal_point);
>>       for (const char *c = lconv->mon_decimal_point; *c; c++)
>>           printf(" %02x", (unsigned char)*c);
>>       printf("\n");
>>       printf("negative_sign: \"%s\"\n", lconv->negative_sign);
>>       printf("positive_sign: \"%s\"\n", lconv->positive_sign);
>>       printf("n_sign_posn: %d\n", lconv->n_sign_posn);
>>       printf("p_sign_posn: %d\n", lconv->p_sign_posn);
>>   #ifndef _WIN32
>>       printf("int_n_sign_posn: %d\n", lconv->int_n_sign_posn);
>>       printf("int_p_sign_posn: %d\n", lconv->int_p_sign_posn);
>>   #endif
>>       return 0;
>>   }
>
> Yep
>
>   mon_thousands_sep: " ", c2 a0
>   thousands_sep: " ", c2 a0
>   int_curr_symbol: "RUB "
>   currency_symbol: "руб", d1 80 d1 83 d0 b1
>   decimal_point: ",", 2c
>   mon_decimal_point: ".", 2e
>   negative_sign: "-"
>   positive_sign: ""
>   n_sign_posn: 1
>   p_sign_posn: 1
>   int_n_sign_posn: 1
>   int_p_sign_posn: 1

Thanks! I get this:

  mon_thousands_sep: " ", e2 80 af
  thousands_sep: " ", e2 80 af
  int_curr_symbol: "RUB "
  currency_symbol: "₽", e2 82 bd
  decimal_point: ",", 2c
  mon_decimal_point: ",", 2c
  negative_sign: "-"
  positive_sign: ""
  n_sign_posn: 1
  p_sign_posn: 1
  int_n_sign_posn: 1
  int_p_sign_posn: 1

What glibc version do you have? As you're on RHEL, I would expect an older one (mine is on Ubuntu 20.04, with glibc 2.31). There seems to be differences in most fields, in particular in `mon_decimal_point` here. We already have some conditionals due to the fact that `currency_symbol` differs for ru_RU.UTF-8 in various glibc versions (apparently with the new form appearing in 2.24), but we never had any glibc conditionals for the differing `mon_decimal_point`. As that probably would require quite intrusive changes to the test, I guess it's probably best to make the test bail out on such older versions (as it just has been XFAILed forever on that platform before anyway).


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D120317/new/

https://reviews.llvm.org/D120317