Re: [PATCH] fix locale string reading

[Ludovic Courtès]

Hi Mark,

Mark H Weaver <address@hidden> skribis:

> address@hidden (Ludovic Courtès) writes:
>> Nala Ginrut <address@hidden> skribis:
>>> +  setlocale (LC_ALL, "");
>>
>> Currently, like in C, it’s the programmer’s responsibility to install
>> the locale with such a call, and none of Guile’s business.
>
> Unfortunately, I don't see a way for the user to call setlocale before a
> Guile script converts the command-line arguments to Scheme strings, at
> least not without providing their own `main' function in C.

Hmm, very good point.

[...]

> I think we should consider decoding the command-line arguments using the
> locale specified by the environment variables, at least in cases like
> this where there's no way for the user to call setlocale before the
> conversion happens.

Below is a patch that does roughly that (we should get ‘locale_encoding’
reviewed and perhaps added to Gnulib.)

It solves the problem:

--8<---------------cut here---------------start------------->8---
# With the patch.
$ ./meta/guile -c '(setlocale LC_ALL "en_US.UTF8")(display (command-line))' -- λ
(/home/ludo/src/guile/libguile/.libs/guile -- λ)

# Previously.
$ guile -c '(setlocale LC_ALL "en_US.UTF8")(display (command-line))' -- λ
(guile -- ??)
--8<---------------cut here---------------end--------------->8---

(Note that the ‘setlocale’ call here is just so that the output port is
correctly set up.)

WDYT?

Thanks,
Ludo’.

diff --git a/libguile/script.c b/libguile/script.c
index 5e0685a..20d7b9e 100644
--- a/libguile/script.c
+++ b/libguile/script.c
@@ -26,6 +26,7 @@
 #include <stdio.h>
 #include <errno.h>
 #include <ctype.h>
+#include <uniconv.h>
 
 #include "libguile/_scm.h"
 #include "libguile/eval.h"
@@ -368,6 +369,74 @@ scm_shell_usage (int fatal, char *message)
                : SCM_BOOL_F));
 }
 
+/* Return the name of the locale encoding suggested by environment
+   variables, even if it's not current, or NULL if no encoding is
+   defined.  Based on Gnulib's `localcharset.c'.  */
+static const char *
+locale_encoding (void)
+{
+  const char *locale, *codeset = NULL;
+
+  /* Allow user to override the codeset, as set in the operating system,
+     with standard language environment variables.  */
+  locale = getenv ("LC_ALL");
+  if (locale == NULL || locale[0] == '\0')
+    {
+      locale = getenv ("LC_CTYPE");
+      if (locale == NULL || locale[0] == '\0')
+        locale = getenv ("LANG");
+    }
+  if (locale != NULL && locale[0] != '\0')
+    {
+      /* If the locale name contains an encoding after the dot, return it.  */
+      const char *dot = strchr (locale, '.');
+
+      if (dot != NULL)
+        {
+         static char buf[2 + 10 + 1];
+          const char *modifier;
+
+          dot++;
+          /* Look for the possible @... trailer and remove it, if any.  */
+          modifier = strchr (dot, '@');
+          if (modifier == NULL)
+            return dot;
+          if (modifier - dot < sizeof (buf))
+            {
+              memcpy (buf, dot, modifier - dot);
+              buf [modifier - dot] = '\0';
+              return buf;
+            }
+        }
+
+      /* Resolve through the charset.alias file.  */
+      codeset = locale;
+    }
+
+  return codeset;
+}
+
+/* Return a list of strings from ARGV, which contains ARGC strings
+   assumed to be encoded in the current locale.  Use `locale_charset'
+   instead of relying on `scm_from_locale_string' because the user
+   hasn't had a change to call (setlocale LC_ALL "") yet.  */
+static SCM
+locale_arguments_to_string_list (int argc, char **const argv)
+{
+  int i;
+  SCM lst;
+  const char *encoding;
+
+  encoding = locale_encoding ();
+  for (i = argc - 1, lst = SCM_EOL;
+       i >= 0;
+       i--)
+    lst = scm_cons (scm_from_stringn (argv[i], (size_t) -1, encoding,
+                                     SCM_FAILED_CONVERSION_ESCAPE_SEQUENCE),
+                   lst);
+
+  return lst;
+}
 
 /* Given an array of command-line switches, return a Scheme expression
    to carry out the actions specified by the switches.
@@ -378,7 +447,7 @@ scm_compile_shell_switches (int argc, char **argv)
 {
   return scm_call_2 (scm_c_public_ref ("ice-9 command-line",
                                        "compile-shell-switches"),
-                     scm_makfromstrs (argc, argv),
+                    locale_arguments_to_string_list (argc, argv),
                      (scm_usage_name
                       ? scm_from_locale_string (scm_usage_name)
                       : scm_from_latin1_string ("guile")));