mhbuild: Transfer-encode MIME parts of type application always with base64

Application files likely contain NUL bytes. scan_content(), however, uses
stdio to check the files to attach. NUL bytes break these checks. Instead
of improving the checks to allow e.g. application/x-tex files to be sent
using 7bit or QP encoding, it was chosen to take the easy way and send
application files base64-encoded.

This solves a bug when attaching tar files. (tar files begin with a
header mostly filled with NUL bytes). Thanks to <philipp@bureaucracy.de>
for discovering and reporting the bug.

diff --git a/uip/mhbuild.c b/uip/mhbuild.c
index 1b2afa5..fde4a1d 100644 (file)
--- a/uip/mhbuild.c
+++ b/uip/mhbuild.c
@@ -1239,10 +1239,10 @@ static int
 scan_content(CT ct)
 {
        int len;
-       int check8bit = 0, contains8bit = 0;  /* check if contains 8bit data */
-       int checklinelen = 0, linelen = 0;  /* check for long lines */
-       int checkboundary = 0, boundaryclash = 0; /* check if clashes with multipart boundary   */
-       int checklinespace = 0, linespace = 0;  /* check if any line ends with space */
+       int check8bit = 0, contains8bit = 0;
+       int checklinelen = 0, linelen = 0;
+       int checkboundary = 0, boundaryclash = 0;
+       int checklinespace = 0, linespace = 0;  /* trailing whitespace */
        unsigned char *cp = NULL, buffer[BUFSIZ];
        struct text *t = NULL;
        FILE *in = NULL;
@@ -1297,13 +1297,6 @@ scan_content(CT ct)
                }
                break;
 
-       case CT_APPLICATION:
-               check8bit = 1;
-               checklinelen = 1;
-               checklinespace = 1;
-               checkboundary = 1;
-               break;
-
        case CT_MESSAGE:
                check8bit = 0;
                checklinelen = 0;
@@ -1311,13 +1304,10 @@ scan_content(CT ct)
                checkboundary = 1;
                break;
 
+       case CT_APPLICATION:
        case CT_AUDIO:
        case CT_IMAGE:
        case CT_VIDEO:
-               /*
-               ** Don't check anything for these types,
-               ** since we are forcing use of base64.
-               */
                check8bit = 0;
                checklinelen = 0;
                checklinespace = 0;
@@ -1334,9 +1324,6 @@ scan_content(CT ct)
                len = strlen(prefix);
 
                while (fgets(buffer, sizeof(buffer) - 1, in)) {
-                       /*
-                       ** Check for 8bit data.
-                       */
                        if (check8bit) {
                                for (cp = buffer; *cp; cp++) {
                                        if (!isascii(*cp)) {
@@ -1347,17 +1334,11 @@ scan_content(CT ct)
                                }
                        }
 
-                       /*
-                       ** Check line length.
-                       */
                        if (checklinelen && (strlen(buffer) > CPERLIN + 1)) {
                                linelen = 1;
                                checklinelen = 0;  /* no need to keep checking */
                        }
 
-                       /*
-                       ** Check if line ends with a space.
-                       */
                        if (checklinespace &&
                                        (cp = buffer + strlen(buffer) - 2) >
                                        buffer && isspace(*cp)) {
@@ -1426,24 +1407,19 @@ scan_content(CT ct)
                        ct->c_encoding = CE_7BIT;
                break;
 
-       case CT_APPLICATION:
-               /* For application type, use base64, except when postscript */
-               if (contains8bit || linelen || linespace)
-                       ct->c_encoding = (ct->c_subtype ==
-                                       APPLICATION_POSTSCRIPT) ?
-                                       CE_QUOTED : CE_BASE64;
-               else
-                       ct->c_encoding = CE_7BIT;
-               break;
-
        case CT_MESSAGE:
                ct->c_encoding = CE_7BIT;
                break;
 
+       case CT_APPLICATION:
        case CT_AUDIO:
        case CT_IMAGE:
        case CT_VIDEO:
-               /* For audio, image, and video contents, just use base64 */
+               /*
+               ** Forcing use of base64, because these types likely
+               ** contain binary data and NUL bytes. Don't care about
+               ** files that would be clean.
+               */
                ct->c_encoding = CE_BASE64;
                break;
        }