New function to decode an IMAP mailbox name from modified UTF-7 encoding

2002-01-18  Jeffrey Stedfast  <fejj@ximian.com>

	* providers/imap/camel-imap-utils.c (imap_mailbox_decode): New
	function to decode an IMAP mailbox name from modified UTF-7
	encoding to UTF-8.
	(imap_mailbox_encode): New function to convert a mailbox name from
	UTF-8 to IMAP's modified UTF-7 encoding.

svn path=/trunk/; revision=15361

2 files changed, 266 insertions, 1 deletions

diff --git a/camel/providers/imap/camel-imap-utils.c b/camel/providers/imap/camel-imap-utils.c
index 7d2073142b..746aec2167 100644
--- a/camel/providers/imap/camel-imap-utils.c
+++ b/camel/providers/imap/camel-imap-utils.c
@@ -24,6 +24,7 @@
 #include <stdio.h>
 #include <string.h>
 #include <time.h>
+#include <iconv.h>
 
 #include "camel-imap-utils.h"
 #include "camel-imap-summary.h"
@@ -173,8 +174,10 @@ imap_parse_folder_name (CamelImapStore *store, const char *folder_name)
 			continue;
 		}
 		
-		if (*p == store->dir_sep)
+		if (*p == store->dir_sep) {
+			/* FIXME: decode mailbox */
 			g_ptr_array_add (heirarchy, g_strndup (folder_name, p - folder_name));
+		}
 		
 		p++;
 	}
@@ -841,3 +844,262 @@ imap_namespace_concat (CamelImapStore *store, const char *name)
 
 	return imap_concat (store, store->namespace, name);
 }
+
+
+#define UTF8_TO_UTF7_LEN(len)  ((len * 3) + 8)
+#define UTF7_TO_UTF8_LEN(len)  (len)
+
+enum {
+	MODE_USASCII,
+	MODE_AMPERSAND,
+	MODE_MODUTF7
+};
+
+#define is_usascii(c)  (((c) >= 0x20 && (c) <= 0x25) || ((c) >= 0x27 && (c) <= 0x7e))
+#define encode_mode(c) (is_usascii (c) ? MODE_USASCII : (c) == '&' ? MODE_AMPERSAND : MODE_MODUTF7)
+
+char *
+imap_mailbox_encode (const unsigned char *in, int inlen)
+{
+	const unsigned char *start, *inptr, *inend;
+	unsigned char *mailbox, *m, *mend;
+	size_t inleft, outleft, conv;
+	char *inbuf, *outbuf;
+	iconv_t cd;
+	int mode;
+	
+	cd = (iconv_t) -1;
+	m = mailbox = g_malloc (UTF8_TO_UTF7_LEN (inlen) + 1);
+	mend = mailbox + UTF8_TO_UTF7_LEN (inlen);
+	
+	start = inptr = in;
+	inend = in + inlen;
+	mode = MODE_USASCII;
+	
+	while (inptr < inend) {
+		int new_mode;
+		
+		new_mode = encode_mode (*inptr);
+		
+		if (new_mode != mode) {
+			switch (mode) {
+			case MODE_USASCII:
+				memcpy (m, start, inptr - start);
+				m += (inptr - start);
+				break;
+			case MODE_AMPERSAND:
+				while (start < inptr) {
+					*m++ = '&';
+					*m++ = '-';
+					start++;
+				}
+				break;
+			case MODE_MODUTF7:
+				inbuf = (char *) start;
+				inleft = inptr - start;
+				outbuf = (char *) m;
+				outleft = mend - m;
+				
+				if (cd == (iconv_t) -1)
+					cd = iconv_open ("UTF-7", "UTF-8");
+				
+				conv = iconv (cd, &inbuf, &inleft, &outbuf, &outleft);
+				if (conv == (size_t) -1) {
+					g_warning ("error converting mailbox to UTF-7!");
+				}
+				iconv (cd, NULL, NULL, &outbuf, &outleft);
+				
+				/* shift into modified UTF-7 mode (overwrite UTF-7's '+' shift)... */
+				*m++ = '&';
+				
+				while (m < (unsigned char *) outbuf) {
+					/* replace '/' with ',' */
+					if (*m == '/')
+						*m = ',';
+					
+					m++;
+				}
+				
+				break;
+			}
+			
+			mode = new_mode;
+			start = inptr;
+		}
+		
+		inptr++;
+	}
+	
+	switch (mode) {
+	case MODE_USASCII:
+		memcpy (m, start, inptr - start);
+		m += (inptr - start);
+		break;
+	case MODE_AMPERSAND:
+		while (start < inptr) {
+			*m++ = '&';
+			*m++ = '-';
+			start++;
+		}
+		break;
+	case MODE_MODUTF7:
+		inbuf = (char *) start;
+		inleft = inptr - start;
+		outbuf = (char *) m;
+		outleft = mend - m;
+		
+		if (cd == (iconv_t) -1)
+			cd = iconv_open ("UTF-7", "UTF-8");
+		
+		conv = iconv (cd, &inbuf, &inleft, &outbuf, &outleft);
+		if (conv == (size_t) -1) {
+			g_warning ("error converting mailbox to UTF-7!");
+		}
+		iconv (cd, NULL, NULL, &outbuf, &outleft);
+		
+		/* shift into modified UTF-7 mode (overwrite UTF-7's '+' shift)... */
+		*m++ = '&';
+		
+		while (m < (unsigned char *) outbuf) {
+			/* replace '/' with ',' */
+			if (*m == '/')
+				*m = ',';
+			
+			m++;
+		}
+		
+		break;
+	}
+	
+	*m = '\0';
+	
+	if (cd != (iconv_t) -1)
+		iconv_close (cd);
+	
+	return mailbox;
+}
+
+
+char *
+imap_mailbox_decode (const unsigned char *in, int inlen)
+{
+	const unsigned char *start, *inptr, *inend;
+	unsigned char *mailbox, *m, *mend;
+	unsigned char mode_switch;
+	iconv_t cd;
+	
+	cd = (iconv_t) -1;
+	m = mailbox = g_malloc (UTF7_TO_UTF8_LEN (inlen) + 1);
+	mend = mailbox + UTF7_TO_UTF8_LEN (inlen);
+	
+	start = inptr = in;
+	inend = in + inlen;
+	mode_switch = '&';
+	
+	while (inptr < inend) {
+		if (*inptr == mode_switch) {
+			if (mode_switch == '&') {
+				/* mode switch from US-ASCII to UTF-7 */
+				mode_switch = '-';
+				memcpy (m, start, inptr - start);
+				m += (inptr - start);
+				start = inptr;
+			} else if (mode_switch == '-') {
+				/* mode switch from UTF-7 to US-ASCII or an ampersand (&) */
+				mode_switch = '&';
+				start++;
+				if (start == inptr) {
+					/* we had the sequence "&-" which becomes "&" when decoded */
+					*m++ = '&';
+				} else {
+					char *buffer, *inbuf, *outbuf;
+					size_t buflen, outleft, conv;
+					
+					buflen = (inptr - start) + 2;
+					inbuf = buffer = alloca (buflen);
+					*inbuf++ = '+';
+					while (start < inptr) {
+						*inbuf++ = *start == ',' ? '/' : *start;
+						start++;
+					}
+					*inbuf = '-';
+					
+					inbuf = buffer;
+					outbuf = (char *) m;
+					outleft = mend - m;
+					
+					if (cd == (iconv_t) -1)
+						cd = iconv_open ("UTF-8", "UTF-7");
+					
+					conv = iconv (cd, &inbuf, &buflen, &outbuf, &outleft);
+					if (conv == (size_t) -1) {
+						g_warning ("error decoding mailbox from UTF-7!");
+					}
+					iconv (cd, NULL, NULL, NULL, NULL);
+					
+					m = (unsigned char *) outbuf;
+				}
+				
+				/* point to the char after the '-' */
+				start = inptr + 1;
+			}
+		}
+		
+		inptr++;
+	}
+	
+	if (*inptr == mode_switch) {
+		if (mode_switch == '&') {
+			/* the remaining text is US-ASCII */
+			memcpy (m, start, inptr - start);
+			m += (inptr - start);
+			start = inptr;
+		} else if (mode_switch == '-') {
+			/* We've got encoded UTF-7 or else an ampersand */
+			start++;
+			if (start == inptr) {
+				/* we had the sequence "&-" which becomes "&" when decoded */
+				*m++ = '&';
+			} else {
+				char *buffer, *inbuf, *outbuf;
+				size_t buflen, outleft, conv;
+				
+				buflen = (inptr - start) + 2;
+				inbuf = buffer = alloca (buflen);
+				*inbuf++ = '+';
+				while (start < inptr) {
+					*inbuf++ = *start == ',' ? '/' : *start;
+					start++;
+				}
+				*inbuf = '-';
+				
+				inbuf = buffer;
+				outbuf = (char *) m;
+				outleft = mend - m;
+				
+				if (cd == (iconv_t) -1)
+					cd = iconv_open ("UTF-8", "UTF-7");
+				
+				conv = iconv (cd, &inbuf, &buflen, &outbuf, &outleft);
+				if (conv == (size_t) -1) {
+					g_warning ("error decoding mailbox from UTF-7!");
+				}
+				iconv (cd, NULL, NULL, NULL, NULL);
+				
+				m = (unsigned char *) outbuf;
+			}
+		}
+	} else {
+		/* illegal encoded mailbox... */
+		g_warning ("illegal mailbox name encountered!");
+		memcpy (m, start, inptr - start);
+		m += (inptr - start);
+	}
+	
+	*m = '\0';
+	
+	if (cd != (iconv_t) -1)
+		iconv_close (cd);
+	
+	return mailbox;
+}
diff --git a/camel/providers/imap/camel-imap-utils.h b/camel/providers/imap/camel-imap-utils.h
index f64d56743a..bedc604116 100644
--- a/camel/providers/imap/camel-imap-utils.h
+++ b/camel/providers/imap/camel-imap-utils.h
@@ -72,6 +72,9 @@ void     imap_uid_array_free       (GPtrArray *arr);
 char *imap_concat (CamelImapStore *imap_store, const char *prefix, const char *suffix);
 char *imap_namespace_concat (CamelImapStore *store, const char *name);
 
+char *imap_mailbox_encode (const unsigned char *in, int inlen);
+char *imap_mailbox_decode (const unsigned char *in, int inlen);
+
 #ifdef __cplusplus
 }
 #endif /* __cplusplus */