Fix encoding errors with strings returned by string ports

String ports, being 8-bit, store strings using the character encoding
of the port.  This fixes a bug where the default character encoding, and
not the port's encoding, was being used to convert the string port data
back to a string.

* libguile/strports.c: extra comments
  (scm_strport_to_string):  use port's encoding when converting port data
  to a string

* libguile/strings.c (scm_i_from_stringn): renamed from scm_from_stringn
  and made internal.  All callers changed.
  (scm_from_stringn): renamed to scm_i_from_stringn.

* libguile/strings.h: declaration for scm_i_from_stringn

diff --git a/libguile/strings.c b/libguile/strings.c
index 4a8390d..3b8d15d 100644 (file)
--- a/libguile/strings.c
+++ b/libguile/strings.c
@@ -1477,15 +1477,18 @@ scm_is_string (SCM obj)
   return IS_STRING (obj);
 }
 
-static SCM
-scm_from_stringn (const char *str, size_t len, const char *encoding,
-                  scm_t_string_failed_conversion_handler handler)
+SCM
+scm_i_from_stringn (const char *str, size_t len, const char *encoding,
+                    scm_t_string_failed_conversion_handler handler)
 {
   size_t u32len, i;
   scm_t_wchar *u32;
   int wide = 0;
   SCM res;
 
+  if (len == 0)
+    return scm_nullstr;
+
   if (encoding == NULL)
     {
       /* If encoding is null, use Latin-1.  */
@@ -1575,7 +1578,7 @@ scm_from_locale_stringn (const char *str, size_t len)
       hndl = SCM_FAILED_CONVERSION_ERROR;
     }
 
-  return scm_from_stringn (str, len, enc, hndl);
+  return scm_i_from_stringn (str, len, enc, hndl);
 }
 
 SCM
@@ -1590,7 +1593,7 @@ scm_from_locale_string (const char *str)
 SCM
 scm_i_from_utf8_string (const scm_t_uint8 *str)
 {
-  return scm_from_stringn ((const char *) str,
+  return scm_i_from_stringn ((const char *) str,
                            strlen ((char *) str), "UTF-8",
                            SCM_FAILED_CONVERSION_ERROR);
 }
@@ -1681,7 +1684,7 @@ unistring_escapes_to_guile_escapes (char **bufp, size_t *lenp)
 }
 
 char *
-scm_to_locale_stringn (SCM str, size_t * lenp)
+scm_to_locale_stringn (SCM str, size_t *lenp)
 {
   SCM outport;
   scm_t_port *pt;

diff --git a/libguile/strings.h b/libguile/strings.h
index 2393aae..c521926 100644 (file)
--- a/libguile/strings.h
+++ b/libguile/strings.h
@@ -111,6 +111,10 @@ SCM_API SCM scm_substring_shared (SCM str, SCM start, SCM end);
 SCM_API SCM scm_substring_copy (SCM str, SCM start, SCM end);
 SCM_API SCM scm_string_append (SCM args);
 
+SCM_INTERNAL SCM scm_i_from_stringn (const char *str, size_t len, 
+                                     const char *encoding,
+                                     scm_t_string_failed_conversion_handler 
+                                     handler);
 SCM_API SCM scm_c_make_string (size_t len, SCM chr);
 SCM_API size_t scm_c_string_length (SCM str);
 SCM_API size_t scm_c_symbol_length (SCM sym);

diff --git a/libguile/strports.c b/libguile/strports.c
index 490a15f..5bfeaad 100644 (file)
--- a/libguile/strports.c
+++ b/libguile/strports.c
@@ -301,9 +301,9 @@ scm_i_mkstrport (SCM pos, const char *locale_str, size_t str_len, long modes, co
      to a locale representation for storage.  But, since string ports
      rely on string functionality for their memory management, we need
      to create a new string that has the 8-bit locale representation
-     of the underlying string.  This violates the guideline that the
-     internal encoding of characters in strings is in unicode
-     codepoints. */
+     of the underlying string.  
+
+     locale_str is already in the locale of the port.  */
   str = scm_i_make_string (str_len, &buf);
   memcpy (buf, locale_str, str_len);
 
@@ -348,13 +348,18 @@ scm_mkstrport (SCM pos, SCM str, long modes, const char *caller)
      of the underlying string.  This violates the guideline that the
      internal encoding of characters in strings is in unicode
      codepoints. */
+
+  /* Ports are initialized with the thread-default values for encoding and
+     invalid sequence handling.  */
   buf = scm_to_locale_stringn (str, &str_len);
   z = scm_i_mkstrport (pos, buf, str_len, modes, caller);
   free (buf);
   return z;
 }
 
-/* create a new string from a string port's buffer.  */
+/* Create a new string from a string port's buffer, converting from
+   the port's 8-bit locale-specific representation to the standard
+   string representation.  */
 SCM scm_strport_to_string (SCM port)
 {
   scm_t_port *pt = SCM_PTAB_ENTRY (port);
@@ -363,7 +368,18 @@ SCM scm_strport_to_string (SCM port)
   if (pt->rw_active == SCM_PORT_WRITE)
     st_flush (port);
 
-  str = scm_from_locale_stringn ((char *)pt->read_buf, pt->read_buf_size);
+  if (pt->read_buf_size == 0)
+    return scm_nullstr;
+
+  if (pt->encoding == NULL)
+    {
+      char *buf;
+      str = scm_i_make_string (pt->read_buf_size, &buf);
+      memcpy (buf, pt->read_buf, pt->read_buf_size);
+    }
+  else
+    str = scm_i_from_stringn ((char *)pt->read_buf, pt->read_buf_size,
+                              pt->encoding, pt->ilseq_handler);
   scm_remember_upto_here_1 (port);
   return str;
 }