gettext returns UTF-8 characters as question marks

[Babak Razmjoo]

I certainly know this is not a bug in gettext, only my ignorance. I have a sample C++ progrem here:

```

#include <iostream>

#include <locale.h>

#include <libintl.h>

void printMessages()

{

const char* msg1 = gettext("Spiders are the only web developers who enjoy finding bugs");

const char* msg2 = gettext("A man, A horse, and a Gun");

std::cout << "Joke : "<< msg1 << "\nMovie : " << msg2 << "\n";

}

int main(int argc, const char *argv[])

{

std::cout << "فارسی با استفاده از استریم\n";

if (setlocale(LC_MESSAGES, "fa_IR.UTF8")) {

bindtextdomain("testgettext", ".");

    textdomain("testgettext");

printMessages();

return 0;

}

else

return 1;

}

```

and have xgettext to extract a messages.po file. I have translated this file in Persian like this:

```

# Laziness sucks.

# Copyright (C) 2020 Babak

# This file is distributed under the same license as the PACKAGE package.

# FIRST AUTHOR <EMAIL@ADDRESS>, YEAR.

#

#, fuzzy

msgid ""

msgstr ""

"Project-Id-Version: testgettext 0.1\n"

"Report-Msgid-Bugs-To: nowhere\n"

"POT-Creation-Date: 2020-11-25 21:31+0330\n"

"PO-Revision-Date: 2020-11-26 23:17+3:30\n"

"Last-Translator: Babak <EMAIL@ADDRESS>\n"

"Language-Team: Persian <LL@li.org>\n"

"Language: fa_IR\n"

"MIME-Version: 1.0\n"

"Content-Type: text/plain; charset=UTF-8\n"

"Content-Transfer-Encoding: 8bit\n"

#: main.cc:7

msgid "Spiders are the only web developers who enjoy finding bugs"

msgstr "عنکبوتها تنها توسعه دهندگان وبی هستند که از یافتن باگ لذت می برند"

#: main.cc:8

msgid "A man, A horse, and a Gun"

msgstr "یک مرد، یک اسب، یک اسلحه"

```

Also I have msgfmt to compile a `testgettext.mo`. This is the directory layout of all these files:

```

.

├── fa_IR

│   └── LC_MESSAGES

│       └── testgettext.mo

├── fa.po

├── main.cc

├── messages.po

└── testgettext

```

Now whenever I run the compiled program `testgettext`, I see this output:

فارسی با استفاده از استریم

Joke : ???????? ???? ????? ??????? ??? ????? ?? ?? ????? ??? ??? ?? ????

Movie : ?? ???? ?? ???? ?? ?????

As you can see, all characters of translated messages are displayed as '?'. But I can see correct results using gettext program:

$ export TEXTDOMAINDIR=.

$ export LANG=fa_IR.UTF-8

$ gettext testgettext "Spiders are the only web developers who enjoy finding bugs"

عنکبوتها تنها توسعه دهندگان وبی هستند که از یافتن باگ لذت می برند

What have i done wrong, or what should i do to get right output in my sample program?

Sent with ProtonMail Secure Email.