Henrik Storner's patch to support mimedecode.

svn path=/trunk/; revision=1715

5 files changed, 82 insertions, 37 deletions

diff --git a/NEWS b/NEWS
index 0a8f41d3..4c5090ae 100644
--- a/NEWS
+++ b/NEWS
@@ -5,7 +5,6 @@
 * Allow an explicit "append" option to specify what rewrite appends?  (For
   when user's pop3 host isn't the same as his SMTP host.)
 * Allow -c -F to work to flush messages without retrieval.
-* Read from /etc/fetchmailrc before ~/.fetchmailrc?
 * Do is_host_alias checks by comparing IP addresses rather than names?
 * Handle multi-homed hosts correctly.
 * Add an "ident" per-user option for debugging that produces an 
@@ -18,8 +17,10 @@
 ------------------------------------------------------------------------------
 fetchmail-4.4.1 ():
 * We now properly shroud IMAP passwords containing ", \, and SP.
+* Checked for Y2K safety.  No 2-digit dates anywhere, so we're safe.
+* Added mimedecode option by Henrik Storner.
 
-There are 273 people on fetchmail-friends and 160 on fetchmail-announce.
+There are 274 people on fetchmail-friends and 164 on fetchmail-announce.
 
 ------------------------------------------------------------------------------
 fetchmail-4.4.0 (Mon Mar 16 14:57:38 EST 1998):
diff --git a/README b/README
index 473aa4f6..89b69a69 100644
--- a/README
+++ b/README
@@ -25,6 +25,8 @@ under AmigaOS.
 A beta OS/2 port of fetchmail is available from Jason F. McBrayer at 
 http://studentweb.tulane.edu/%7Ejmcbray/os2.
 
+Fetchmail is Y2K safe.
+
 See the distribution files FEATURES for a full list of features, NEWS
 for detailed information on recent changes and NOTES for design notes.
 
diff --git a/driver.c b/driver.c
index aa908ecd..a3e8a633 100644
--- a/driver.c
+++ b/driver.c
@@ -108,6 +108,7 @@ static int tagnum;
 static char shroud[PASSWORDLEN];	/* string to shroud in debug output */
 static int mytimeout;			/* value of nonreponse timeout */
 static int msglen;			/* actual message length */
+static int mimemsg;			/* bitmask indicating MIME body-type */
 
 /* use these to track what was happening when the nonresponse timer fired */
 #define GENERAL_WAIT	0	/* unknown wait type */
@@ -639,7 +640,7 @@ int num;		/* index of message */
     char		return_path[HOSTLEN + USERNAMELEN + 4]; 
     int			from_offs, reply_to_offs, resent_from_offs;
     int			app_from_offs, sender_offs, resent_sender_offs;
-    int			ctt_offs, env_offs;
+    int			env_offs;
     char		*headers, *received_for, *destaddr, *rcv, *cp;
     int 		n, linelen, oldlen, ch, remaining, skipcount;
     struct idlist 	*idp, *xmit_names;
@@ -656,10 +657,11 @@ int num;		/* index of message */
     /* read message headers */
     headers = received_for = NULL;
     from_offs = reply_to_offs = resent_from_offs = app_from_offs = 
-	sender_offs = resent_sender_offs = ctt_offs = env_offs = -1;
+	sender_offs = resent_sender_offs = env_offs = -1;
     oldlen = 0;
     msglen = 0;
     skipcount = 0;
+    mimemsg = 0;
 
     for (remaining = fetchlen; remaining > 0 || protocol->delimited; remaining -= linelen)
     {
@@ -880,8 +882,6 @@ int num;		/* index of message */
 	else if (!strncasecmp("Resent_Sender:", line, 14))
 	    resent_sender_offs = (line - headers);
 
-	else if (!strncasecmp("Content-Transfer-Encoding:", line, 26))
-	    ctt_offs = (line - headers);
  	else if (!strncasecmp("Message-Id:", buf, 11))
 	{
 	    if (ctl->server.uidl)
@@ -982,6 +982,16 @@ int num;		/* index of message */
      * In fact we have to, as this will tell us where to forward to.
      */
 
+    /* Decode MIME encoded headers. We MUST do this before
+     * looking at the Content-Type / Content-Transfer-Encoding
+     * headers (RFC 2046).
+     */
+    if (ctl->mimedecode) {
+	UnMimeHeader(headers);
+    }
+    /* Check for MIME headers indicating possible 8-bit data */
+    mimemsg = MimeBodyType(headers);
+
     /*
      * If there is a Return-Path address on the message, this was
      * almost certainly the MAIL FROM address given the originating
@@ -1197,7 +1207,7 @@ int num;		/* index of message */
     }
     else
     {
-	char	*ap, *ctt, options[MSGBUFSIZE], addr[128];
+	char	*ap, options[MSGBUFSIZE], addr[128];
 
 	/* build a connection to the SMTP listener */
 	if ((smtp_open(ctl) == -1))
@@ -1210,22 +1220,16 @@ int num;		/* index of message */
 	}
 
 	/*
-	 * Compute ESMTP options.  It's a kluge to use nxtaddr()
-	 * here because the contents of the Content-Transfer-Encoding
-	 * headers isn't semantically an address.  But it has the
-	 * desired tokenizing effect.
+	 * Compute ESMTP options.
 	 */
 	options[0] = '\0';
-	if (ctl->server.esmtp_options & ESMTP_8BITMIME)
-	    if (ctl->pass8bits)
+	if (ctl->server.esmtp_options & ESMTP_8BITMIME) {
+             if (ctl->pass8bits || (mimemsg & MSG_IS_8BIT))
 		strcpy(options, " BODY=8BITMIME");
-	    else if ((ctt_offs >= 0) && (ctt = nxtaddr(headers + ctt_offs)))
-	    {
-		if (!strcasecmp(ctt,"7BIT"))
-		    strcpy(options, " BODY=7BIT");
-		else if (!strcasecmp(ctt,"8BIT"))
-		    strcpy(options, " BODY=8BITMIME");
-	    }
+             else if (mimemsg & MSG_IS_7BIT)
+		strcpy(options, " BODY=7BIT");
+        }
+
 	if ((ctl->server.esmtp_options & ESMTP_SIZE) && reallen > 0)
 	    sprintf(options + strlen(options), " SIZE=%ld", reallen);
 
@@ -1258,12 +1262,13 @@ int num;		/* index of message */
 		 * coming from this address, probably due to an
 		 * anti-spam domain exclusion.  Respect this.  Don't
 		 * try to ship the message, and don't prevent it from
-		 * being deleted.  Typical values.
+		 * being deleted.  Typical values:
 		 *
 		 * 571 = sendmail's "unsolicited email refused"
 		 * 501 = exim's old antispam response
 		 * 550 = exim's new antispam response (temporary)
 		 */
+		SMTP_rset(ctl->smtp_socket);	/* required by RFC1870 */
 		free(headers);
 		free_str_list(&xmit_names);
 		return(PS_REFUSED);
@@ -1292,6 +1297,7 @@ int num;		/* index of message */
 		return(PS_TRANSIENT);
 
 	    case 552: /* message exceeds fixed maximum message size */
+	    case 553: /* invalid sending domain */
 		/*
 		 * Permanent no-go condition on the
 		 * ESMTP server.  Don't try to ship the message, 
@@ -1543,13 +1549,15 @@ int len;		/* length of message */
 flag forward;		/* TRUE to forward */
 {
     int	linelen;
-    char buf[MSGBUFSIZE+1];
+    unsigned char buf[MSGBUFSIZE+1];
+    unsigned char *inbufp = buf;
+    flag issoftline = FALSE;
 
     /* pass through the text lines */
     while (protocol->delimited || len > 0)
     {
 	set_timeout(ctl->server.timeout);
-	if ((linelen = SockRead(sock, buf, sizeof(buf)-1)) == -1)
+	if ((linelen = SockRead(sock, inbufp, sizeof(buf)-1-(inbufp-buf)))==-1)
 	{
 	    set_timeout(0);
 	    if (ctl->mda)
@@ -1576,20 +1584,35 @@ flag forward;		/* TRUE to forward */
 	len -= linelen;
 
 	/* check for end of message */
-	if (protocol->delimited && *buf == '.')
-	    if (buf[1] == '\r' && buf[2] == '\n' && buf[3] == '\0')
+	if (protocol->delimited && *inbufp == '.')
+	    if (inbufp[1] == '\r' && inbufp[2] == '\n' && inbufp[3] == '\0')
 		break;
-	    else if (buf[1] == '\n' && buf[2] == '\0')
+	    else if (inbufp[1] == '\n' && inbufp[2] == '\0')
 		break;
 	    else
 		msglen--;	/* subtract the size of the dot escape */
 
 	msglen += linelen;
 
+	if (ctl->mimedecode && (mimemsg & MSG_NEEDS_DECODE)) {
+	    issoftline = UnMimeBodyline(&inbufp, (protocol->delimited && issoftline));
+	    if (issoftline && (sizeof(buf)-1-(inbufp-buf) < 200)) {
+		/* Soft linebreak, but less than 200 bytes left in
+		 * input buffer. Rather than doing a buffer overrun,
+		 * ignore the soft linebreak, NL-terminate data and
+		 * deliver what we have now.
+		 * (Who writes lines longer than 2K anyway?)
+		 */
+		*inbufp = '\n'; *(inbufp+1) = '\0';
+		issoftline = 0;
+	    }
+	}
+
 	/* ship out the text line */
-	if (forward)
+	if (forward && (!issoftline))
 	{
 	    int	n = stuffline(ctl, buf);
+	    inbufp = buf;
 
 	    if (n < 0)
 	    {
diff --git a/fetchmail-features.html b/fetchmail-features.html
index 42cbcf87..2e6a7ccf 100644
--- a/fetchmail-features.html
+++ b/fetchmail-features.html
@@ -10,17 +10,18 @@
 <table width="100%" cellpadding=0><tr>
 <td width="30%">Back to <a href="index.html">Fetchmail Home Page</a>
 <td width="30%" align=center>To <a href="/~esr/sitemap.html">Site Map</a>
-<td width="30%" align=right>$Date: 1998/03/03 21:22:31 $
+<td width="30%" align=right>$Date: 1998/03/20 16:37:36 $
 </table>
 <HR>
 
 <H1 ALIGN=CENTER>Fetchmail Feature List</H1>
 
-Here are fetchmail's main features.  Those unique to fetchmail
-are listed first. <P>
-
 <H2>Since 4.0:</H2>
 <UL>
+<LI> Automatically decodes armored 7-bit MIME into 8 bits.
+
+<LI> You can specify which SMTP error is recognized as a spam block.
+
 <LI> Support for Kerberos V authentication.
 
 <LI> Support for IMAP-OTP authentication using Craig Metz's patches 
@@ -114,6 +115,8 @@ are listed first. <P>
 
 <LI> Strict conformance to relevant RFCs and good debugging options.
      You could use fetchmail to test and debug server implementatations.
+
+<LI> For anybody who cares, fetchmail is Y2K safe.
 </UL>
 
 <H2>Features in common with other remote-mail retrieval programs:</H2>
@@ -141,7 +144,7 @@ get-mail, gwpop, pimp-1.0, pop-perl5-1.2, popc, popmail-1.6 and upop.<P>
 <table width="100%" cellpadding=0><tr>
 <td width="30%">Back to <a href="index.html">Fetchmail Home Page</a>
 <td width="30%" align=center>To <a href="/~esr/sitemap.html">Site Map</a>
-<td width="30%" align=right>$Date: 1998/03/03 21:22:31 $
+<td width="30%" align=right>$Date: 1998/03/20 16:37:36 $
 </table>
 
 <P><ADDRESS>Eric S. Raymond <A HREF="mailto:esr@thyrsus.com">&lt;esr@snark.thyrsus.com&gt;</A></ADDRESS>
diff --git a/fetchmail.man b/fetchmail.man
index 50c947fd..e5b74ff2 100644
--- a/fetchmail.man
+++ b/fetchmail.man
@@ -418,7 +418,7 @@ Every mode except ETRN requires authentication of the client.
 Normal user authentication in 
 .I fetchmail
 is very much like the authentication mechanism of 
-.I ftp(1).
+.IR ftp (1).
 The correct user-id and password depend upon the underlying security
 system at the mailserver.  
 .PP
@@ -912,6 +912,9 @@ T}
 dropstatus	\&	T{
 Strip Status and X-Mozilla-Status lines out of incoming mail
 T}
+mimedecode	\&	T{
+Convert quoted-printable to 8-bit in MIME messages (default)
+T}
 no keep  	-K	T{
 Delete seen messages from server (default)
 T}
@@ -936,6 +939,9 @@ T}
 no dropstatus	\&	T{
 Don't drop Status headers (default)
 T}
+no mimedecode	\&	T{
+Don't convert quoted-printable to 8-bit in MIME messages
+T}
 limit   	-l	T{
 Set message size limit
 T}
@@ -971,7 +977,8 @@ All options correspond to the obvious command-line arguments, except
 the following: `via', `interval', `aka', `is', `to', `dns'/`no dns',
 \&`password', \&`preconnect', \&`postconnect', `localdomains',
 \&`stripcr'/`no stripcr', \&`forcecr'/`no forcecr', `pass8bits'/`no
-pass8bits' `dropstatus/no dropstatus', and `no envelope'.
+pass8bits' `dropstatus/no dropstatus', `mimedecode/no mimedecode',
+and `no envelope'.
 .PP
 The `via' option is for use with ssh, or if you want to have more
 than one configuration pointing at the same site.  If it is present,
@@ -1095,6 +1102,15 @@ confuse some new-mail notifiers, which assume that anything with a
 Status line in it has been seen.  (Note: the empty Status lines
 inserted by some buggy POP servers are unconditionally discarded.)
 .PP
+The `mimedecode' option controls whether MIME messages using the
+quoted-printable encoding are automatically converted into pure
+8-bit data. If you are delivering mail to an ESMTP-capable,
+8-bit-clean listener (that includes all of the major programs
+like sendmail), then this will automatically convert quoted-printable
+message headers and data into 8-bit data, making it easier to
+understand when reading mail. If your e-mail programs know how to
+deal with MIME messages, then this option is not needed.
+.PP
 .SS Miscellaneous Run Control Options
 The words `here' and `there' have useful English-like
 significance.  Normally `user eric is esr' would mean that 
@@ -1524,9 +1540,9 @@ in foreground while a background fetchmail is running will do
 whichever of these is appropriate to wake it up.
 
 .SH BUGS AND KNOWN PROBLEMS
-The RFC822 parser used in multidrop mode chokes on some @-addresses that
-are technically legal but bizarre.  Strange uses of quoting and
-embedded comments are likely to confuse it.
+The RFC822 address parser used in multidrop mode chokes on some
+@-addresses that are technically legal but bizarre.  Strange uses of
+quoting and embedded comments are likely to confuse it.
 .PP
 Use of any of the supported protocols other than POP3 with OTP or RPA, APOP,
 KPOP, IMAP-K4, IMAP-GSS, or ETRN requires that the program send unencrypted