mintty 终端、python、curses 和 utf8：显示不正确

发布于 2024-10-14 09:37:06 字数 1939 浏览 5 评论 0原文

我已询问这里，但我很确定我不会得到答案。

复制 - 粘贴此 python 代码：
执行它
在经典的 utf8 shell 中，它可以正常工作并显示“Coordonnées”，但在我家的 mintty 终端中它显示“CoordonnM-CM-)es”。知道如何让它工作，或者我应该在哪里查看（mintty 没有字符编码选项）？

代码：

#!/usr/bin/env python
# -*- coding: utf-8 -*-

import os, sys, time, random, copy, pprint
import termios, tty, fcntl
import thread, curses
import locale

def curses_main(stdscr, code):
    curses.curs_set(0)
    stdscr.addstr(2, 0, "Coordonnées")
    stdscr.refresh()
    # (1) = délai d'attente de 1/10 ème de seconde :
    curses.halfdelay(1)
    while True:
        try:
            c=stdscr.getch()
        except:
            c=-1
        if 0<c<256:
            if chr(c) in 'Qq':
                break
def main():
    locale.setlocale(locale.LC_ALL, '')
    code = locale.getpreferredencoding()
    curses.wrapper(curses_main, code)

if __name__ == '__main__':
    main()

更新

错误：我认为找到了它：它到处都是从该代码开始的：

locale.setlocale(locale.LC_ALL, '')

所以我尝试了它并且它起作用了：

locale.setlocale(locale.LC_ALL, 'fr_FR')

=>强制其为法语

... 它有效，但仅适用于口音。如果我尝试打印线条画字符：

stdscr.addstr(20,20, "─ │ ┌ ┘ ┐ └ ├ ┤ ┬ ┴ ┼ ═ ║ ╔ ╝ ╗ ╚ ╠ ╣ ╦ ╩ ╬")

给我：

M-b~T~@ M-b~T~B M-b~T~L M-b~T~X M-...

什么是头撞是vim正确显示这些字符。所以它看起来像是来自Python？从我的 Windows/cygwin/mintty 可以正常工作：

env | grep -i lang
LANG=C.UTF-8

如果我在家里强制在我的 Windows/cygwin/mintty 中进行编码，那么不起作用：

locale.setlocale(locale.LC_ALL, 'C.UTF-8')

这不起作用。

知道我可以去哪里看吗？ python 似乎正确地覆盖了区域设置...控制台配置正确，并且字体可以显示线条画字符，因为我用 vim 看到它们。所以仔细想想，可能是python的问题，也可能是curses的问题……

原文

I've asked here but I'm pretty sure I won't get an answer.

copy - paste this python code :
execute it
in a classical utf8 shell it work properly and displays "Coordonnées" but in my mintty terminal at home it displays "CoordonnM-CM-)es". Any idea how I could make it work, or where I should look (mintty has no character encoding options)?

code:

#!/usr/bin/env python
# -*- coding: utf-8 -*-

import os, sys, time, random, copy, pprint
import termios, tty, fcntl
import thread, curses
import locale

def curses_main(stdscr, code):
    curses.curs_set(0)
    stdscr.addstr(2, 0, "Coordonnées")
    stdscr.refresh()
    # (1) = délai d'attente de 1/10 ème de seconde :
    curses.halfdelay(1)
    while True:
        try:
            c=stdscr.getch()
        except:
            c=-1
        if 0<c<256:
            if chr(c) in 'Qq':
                break
def main():
    locale.setlocale(locale.LC_ALL, '')
    code = locale.getpreferredencoding()
    curses.wrapper(curses_main, code)

if __name__ == '__main__':
    main()

Update

Error: I thought found it:
it's written everywhere to begin with that code:

locale.setlocale(locale.LC_ALL, '')

so i tried with that and it worked:

locale.setlocale(locale.LC_ALL, 'fr_FR')

=> force it to French language

...
it worked but only for the accents.
If i try to print line drawing chars:

stdscr.addstr(20,20, "─ │ ┌ ┘ ┐ └ ├ ┤ ┬ ┴ ┼ ═ ║ ╔ ╝ ╗ ╚ ╠ ╣ ╦ ╩ ╬")

gives me:

M-b~T~@ M-b~T~B M-b~T~L M-b~T~X M-...

what is head banging is that vim displays properly those chars. So it seems like it comes... from python?
From my Windows/cygwin/mintty that works:

env | grep -i lang
LANG=C.UTF-8

if I force encoding at home in my Windows/cygwin/mintty that doesn't work:

locale.setlocale(locale.LC_ALL, 'C.UTF-8')

this doesn't work.

Any idea where I could look? python seems to override properly the locale settings... the console is properly configured, and the fonts can display line drawing chars, because I see them with vim. So thinking more about it, it may be either a problem of python or a problem of curses...

分享到QQ

分享到微博