]> code.citadel.org Git - citadel.git/blobdiff - citadel/mime_parser.c
* MSG4 (and CtdlOutputMsg() as well) now accepts an optional MIME part
[citadel.git] / citadel / mime_parser.c
index ff206dbcd8c9c2371c0b3b690cae938f9e401be6..edbdc52cb0bc0773d17ad246128d9e7ebac32ccf 100644 (file)
@@ -1,17 +1,13 @@
 /*
  * $Id$
  *
- * This is the MIME parser for Citadel.  Sometimes it actually works.
+ * This is the MIME parser for Citadel.
  *
- * Copyright (c) 1998-2002 by Art Cancro
+ * Copyright (c) 1998-2005 by Art Cancro
  * This code is distributed under the terms of the GNU General Public License.
  *
  */
 
-#ifdef DLL_EXPORT
-#define IN_LIBCIT
-#endif
-
 #include <stdlib.h>
 #include <unistd.h>
 #include <stdio.h>
@@ -71,7 +67,7 @@ int CtdlDecodeQuotedPrintable(char *decoded, char *encoded, int sourcelen) {
        char buf[SIZ];
        int buf_length = 0;
        int soft_line_break = 0;
-       int ch;
+       unsigned int ch;
        int decoded_length = 0;
        int i;
 
@@ -134,7 +130,7 @@ int CtdlDecodeQuotedPrintable(char *decoded, char *encoded, int sourcelen) {
  */
 void mime_decode(char *partnum,
                 char *part_start, size_t length,
-                char *content_type, char *encoding,
+                char *content_type, char *charset, char *encoding,
                 char *disposition,
                 char *name, char *filename,
                 void (*CallBack)
@@ -144,6 +140,7 @@ void mime_decode(char *partnum,
                   char *cbdisp,
                   void *cbcontent,
                   char *cbtype,
+                  char *cbcharset,
                   size_t cblength,
                   char *cbencoding,
                   void *cbuserdata),
@@ -154,6 +151,7 @@ void mime_decode(char *partnum,
                   char *cbdisp,
                   void *cbcontent,
                   char *cbtype,
+                  char *cbcharset,
                   size_t cblength,
                   char *cbencoding,
                   void *cbuserdata),
@@ -164,6 +162,7 @@ void mime_decode(char *partnum,
                   char *cbdisp,
                   void *cbcontent,
                   char *cbtype,
+                  char *cbcharset,
                   size_t cblength,
                   char *cbencoding,
                   void *cbuserdata),
@@ -188,7 +187,7 @@ void mime_decode(char *partnum,
                if (CallBack != NULL) {
                        CallBack(name, filename, fixed_partnum(partnum),
                                disposition, part_start,
-                               content_type, length, encoding, userdata);
+                               content_type, charset, length, encoding, userdata);
                        }
                return;
        }
@@ -203,7 +202,7 @@ void mime_decode(char *partnum,
         * will never be larger than the encoded data.  This is a safe
         * assumption with base64, uuencode, and quoted-printable.
         */
-       decoded = mallok(length+2048);
+       decoded = malloc(length+2048);
        if (decoded == NULL) {
                return;
        }
@@ -219,10 +218,10 @@ void mime_decode(char *partnum,
        if (bytes_decoded > 0) if (CallBack != NULL) {
                CallBack(name, filename, fixed_partnum(partnum),
                        disposition, decoded,
-                       content_type, bytes_decoded, "binary", userdata);
+                       content_type, charset, bytes_decoded, "binary", userdata);
        }
 
-       phree(decoded);
+       free(decoded);
 }
 
 /*
@@ -240,6 +239,7 @@ void the_mime_parser(char *partnum,
                       char *cbdisp,
                       void *cbcontent,
                       char *cbtype,
+                      char *cbcharset,
                       size_t cblength,
                       char *cbencoding,
                       void *cbuserdata),
@@ -250,6 +250,7 @@ void the_mime_parser(char *partnum,
                       char *cbdisp,
                       void *cbcontent,
                       char *cbtype,
+                      char *cbcharset,
                       size_t cblength,
                       char *cbencoding,
                       void *cbuserdata),
@@ -260,6 +261,7 @@ void the_mime_parser(char *partnum,
                       char *cbdisp,
                       void *cbcontent,
                       char *cbtype,
+                      char *cbcharset,
                       size_t cblength,
                       char *cbencoding,
                       void *cbuserdata),
@@ -269,13 +271,17 @@ void the_mime_parser(char *partnum,
 {
 
        char *ptr;
+       char *srch = NULL;
        char *part_start, *part_end = NULL;
        char buf[SIZ];
        char *header;
        char *boundary;
        char *startary;
+       size_t startary_len = 0;
        char *endary;
+       char *next_boundary;
        char *content_type;
+       char *charset;
        size_t content_length;
        char *encoding;
        char *disposition;
@@ -292,34 +298,37 @@ void the_mime_parser(char *partnum,
        ptr = content_start;
        content_length = 0;
 
-       boundary = mallok(SIZ);
+       boundary = malloc(SIZ);
        memset(boundary, 0, SIZ);
 
-       startary = mallok(SIZ);
+       startary = malloc(SIZ);
        memset(startary, 0, SIZ);
 
-       endary = mallok(SIZ);
+       endary = malloc(SIZ);
        memset(endary, 0, SIZ);
 
-       header = mallok(SIZ);
+       header = malloc(SIZ);
        memset(header, 0, SIZ);
 
-       content_type = mallok(SIZ);
+       content_type = malloc(SIZ);
        memset(content_type, 0, SIZ);
 
-       encoding = mallok(SIZ);
+       charset = malloc(SIZ);
+       memset(charset, 0, SIZ);
+
+       encoding = malloc(SIZ);
        memset(encoding, 0, SIZ);
 
-       content_type_name = mallok(SIZ);
+       content_type_name = malloc(SIZ);
        memset(content_type_name, 0, SIZ);
 
-       content_disposition_name = mallok(SIZ);
+       content_disposition_name = malloc(SIZ);
        memset(content_disposition_name, 0, SIZ);
 
-       filename = mallok(SIZ);
+       filename = malloc(SIZ);
        memset(filename, 0, SIZ);
 
-       disposition = mallok(SIZ);
+       disposition = malloc(SIZ);
        memset(disposition, 0, SIZ);
 
        /* If the caller didn't supply an endpointer, generate one by measure */
@@ -335,13 +344,17 @@ void the_mime_parser(char *partnum,
                        goto end_parser;
                }
 
-               for (i = 0; i < strlen(buf); ++i)
-                       if (isspace(buf[i]))
+               for (i = 0; i < strlen(buf); ++i) {
+                       if (isspace(buf[i])) {
                                buf[i] = ' ';
+                       }
+               }
+
                if (!isspace(buf[0])) {
                        if (!strncasecmp(header, "Content-type: ", 14)) {
                                strcpy(content_type, &header[14]);
                                extract_key(content_type_name, content_type, "name");
+                               extract_key(charset, content_type, "charset");
                                /* Deal with weird headers */
                                if (strchr(content_type, ' '))
                                        *(strchr(content_type, ' ')) = '\0';
@@ -387,56 +400,71 @@ void the_mime_parser(char *partnum,
                /* Tell the client about this message's multipartedness */
                if (PreMultiPartCallBack != NULL) {
                        PreMultiPartCallBack("", "", partnum, "",
-                               NULL, content_type,
+                               NULL, content_type, charset,
                                0, encoding, userdata);
                }
 
                /* Figure out where the boundaries are */
                snprintf(startary, SIZ, "--%s", boundary);
                snprintf(endary, SIZ, "--%s--", boundary);
+               startary_len = strlen(startary);
+
+               part_start = NULL;
                do {
-                       if ( (!strncasecmp(ptr, startary, strlen(startary)))
-                          || (!strncasecmp(ptr, endary, strlen(endary))) ) {
-                               if (part_start != NULL) {
-                                       if (strlen(partnum) > 0) {
-                                               snprintf(nested_partnum,
-                                                        sizeof nested_partnum,
-                                                        "%s.%d", partnum,
-                                                        ++part_seq);
-                                       }
-                                       else {
-                                               snprintf(nested_partnum,
-                                                        sizeof nested_partnum,
-                                                        "%d", ++part_seq);
-                                       }
-                                       the_mime_parser(nested_partnum,
-                                                   part_start, part_end,
-                                                       CallBack,
-                                                       PreMultiPartCallBack,
-                                                       PostMultiPartCallBack,
-                                                       userdata,
-                                                       dont_decode);
+                       next_boundary = NULL;
+                       for (srch=ptr; srch<content_end; ++srch) {
+                               if (!memcmp(srch, startary, startary_len)) {
+                                       next_boundary = srch;
+                                       srch = content_end;
                                }
-                               ptr = memreadline(ptr, buf, SIZ);
-                               part_start = ptr;
                        }
-                       else {
-                               part_end = ptr;
-                               ++ptr;
+
+                       if ( (part_start != NULL) && (next_boundary != NULL) ) {
+                               part_end = next_boundary;
+                               --part_end;
+
+                               if (strlen(partnum) > 0) {
+                                       snprintf(nested_partnum,
+                                                sizeof nested_partnum,
+                                                "%s.%d", partnum,
+                                                ++part_seq);
+                               }
+                               else {
+                                       snprintf(nested_partnum,
+                                                sizeof nested_partnum,
+                                                "%d", ++part_seq);
+                               }
+                               the_mime_parser(nested_partnum,
+                                           part_start, part_end,
+                                               CallBack,
+                                               PreMultiPartCallBack,
+                                               PostMultiPartCallBack,
+                                               userdata,
+                                               dont_decode);
                        }
-                       /* If we pass out of scope in the MIME multipart (by
-                        * hitting the end boundary), force the pointer out
-                        * of scope so this loop ends.
-                        */
-                       if (ptr < content_end) {
-                               if (!strcasecmp(ptr, endary)) {
-                                       ptr = content_end++;
+
+                       if (next_boundary != NULL) {
+                               /* If we pass out of scope, don't attempt to
+                                * read past the end boundary. */
+                               if (!strcmp(next_boundary, endary)) {
+                                       ptr = content_end;
+                               }
+                               else {
+                                       /* Set up for the next part. */
+                                       part_start = strstr(next_boundary, "\n");
+                                       ++part_start;
+                                       ptr = part_start;
                                }
                        }
-               } while (ptr <= content_end);
+                       else {
+                               /* Invalid end of multipart.  Bail out! */
+                               ptr = content_end;
+                       }
+               } while ( (ptr < content_end) && (next_boundary != NULL) );
+
                if (PostMultiPartCallBack != NULL) {
                        PostMultiPartCallBack("", "", partnum, "", NULL,
-                               content_type, 0, encoding, userdata);
+                               content_type, charset, 0, encoding, userdata);
                }
                goto end_parser;
        }
@@ -450,9 +478,9 @@ void the_mime_parser(char *partnum,
                        ++length;
                }
                part_end = content_end;
-                /* fix an off-by-one error */
-                --part_end;
-                --length;
+               /* fix an off-by-one error */
+               --part_end;
+               --length;
                
                /* Truncate if the header told us to */
                if ( (content_length > 0) && (length > content_length) ) {
@@ -469,26 +497,33 @@ void the_mime_parser(char *partnum,
                else {
                        name = content_type_name;
                }
-               
+       
+               /* lprintf(CTDL_DEBUG, "mime_decode part=%s, len=%d, type=%s, charset=%s, encoding=%s\n",
+                       partnum, length, content_type, charset, encoding); */
+
+               /* Ok, we've got a non-multipart part here, so do something with it.
+                */
                mime_decode(partnum,
-                           part_start, length,
-                           content_type, encoding, disposition,
-                           name, filename,
-                           CallBack, NULL, NULL,
-                           userdata, dont_decode);
+                       part_start, length,
+                       content_type, charset, encoding, disposition,
+                       name, filename,
+                       CallBack, NULL, NULL,
+                       userdata, dont_decode
+               );
        }
 
 end_parser:    /* free the buffers!  end the oppression!! */
-       phree(boundary);
-       phree(startary);
-       phree(endary);  
-       phree(header);
-       phree(content_type);
-       phree(encoding);
-       phree(content_type_name);
-       phree(content_disposition_name);
-       phree(filename);
-       phree(disposition);
+       free(boundary);
+       free(startary);
+       free(endary);   
+       free(header);
+       free(content_type);
+       free(charset);
+       free(encoding);
+       free(content_type_name);
+       free(content_disposition_name);
+       free(filename);
+       free(disposition);
 }
 
 
@@ -509,6 +544,7 @@ void mime_parser(char *content_start,
                   char *cbdisp,
                   void *cbcontent,
                   char *cbtype,
+                  char *cbcharset,
                   size_t cblength,
                   char *cbencoding,
                   void *cbuserdata),
@@ -520,6 +556,7 @@ void mime_parser(char *content_start,
                   char *cbdisp,
                   void *cbcontent,
                   char *cbtype,
+                  char *cbcharset,
                   size_t cblength,
                   char *cbencoding,
                   void *cbuserdata),
@@ -531,6 +568,7 @@ void mime_parser(char *content_start,
                   char *cbdisp,
                   void *cbcontent,
                   char *cbtype,
+                  char *cbcharset,
                   size_t cblength,
                   char *cbencoding,
                   void *cbuserdata),