X-Git-Url: http://git.marmaro.de/?p=mmh;a=blobdiff_plain;f=uip%2Fmhparse.c;h=7c1000060dd86ffb4dd11459a0bd382996edcf12;hp=1c46076d3d478aedd0f43f60034c125f9b6871ab;hb=d8916ff5d389de5ab225cd6f40aeda1b285d0f28;hpb=c085eca6dea41bf9c0ac2749215e256cc54ff628 diff --git a/uip/mhparse.c b/uip/mhparse.c index 1c46076..7c10000 100644 --- a/uip/mhparse.c +++ b/uip/mhparse.c @@ -3,6 +3,10 @@ * mhparse.c -- routines to parse the contents of MIME messages * * $Id$ + * + * This code is Copyright (c) 2002, by the authors of nmh. See the + * COPYRIGHT file in the root directory of the nmh distribution for + * complete copyright information. */ #include @@ -12,17 +16,17 @@ #include #include #include -#include -#include +#include +#include #include #include +#include #ifdef HAVE_SYS_WAIT_H # include #endif -extern int errno; extern int debugsw; extern int endian; /* mhmisc.c */ @@ -42,24 +46,16 @@ int checksw = 0; /* check Content-MD5 field */ char *tmp; /* - * Structure for mapping types to their internal flags - */ -struct k2v { - char *kv_key; - int kv_value; -}; - -/* * Structures for TEXT messages */ -static struct k2v SubText[] = { +struct k2v SubText[] = { { "plain", TEXT_PLAIN }, { "richtext", TEXT_RICHTEXT }, /* defined in RFC-1341 */ { "enriched", TEXT_ENRICHED }, /* defined in RFC-1896 */ { NULL, TEXT_UNKNOWN } /* this one must be last! */ }; -static struct k2v Charset[] = { +struct k2v Charset[] = { { "us-ascii", CHARSET_USASCII }, { "iso-8859-1", CHARSET_LATIN }, { NULL, CHARSET_UNKNOWN } /* this one must be last! */ @@ -68,7 +64,7 @@ static struct k2v Charset[] = { /* * Structures for MULTIPART messages */ -static struct k2v SubMultiPart[] = { +struct k2v SubMultiPart[] = { { "mixed", MULTI_MIXED }, { "alternative", MULTI_ALTERNATE }, { "digest", MULTI_DIGEST }, @@ -79,7 +75,7 @@ static struct k2v SubMultiPart[] = { /* * Structures for MESSAGE messages */ -static struct k2v SubMessage[] = { +struct k2v SubMessage[] = { { "rfc822", MESSAGE_RFC822 }, { "partial", MESSAGE_PARTIAL }, { "external-body", MESSAGE_EXTERNAL }, @@ -89,7 +85,7 @@ static struct k2v SubMessage[] = { /* * Structure for APPLICATION messages */ -static struct k2v SubApplication[] = { +struct k2v SubApplication[] = { { "octet-stream", APPLICATION_OCTETS }, { "postscript", APPLICATION_POSTSCRIPT }, { NULL, APPLICATION_UNKNOWN } /* this one must be last! */ @@ -113,34 +109,24 @@ void free_content (CT); void free_encoding (CT, int); /* - * prototypes - */ -int pidcheck (int); -CT parse_mime (char *); - -/* * static prototypes */ static CT get_content (FILE *, char *, int); -static int add_header (CT, char *, char *); -static int get_ctinfo (char *, CT); -static int get_comment (CT, char **, int); +static int get_comment (CT, unsigned char **, int); + static int InitGeneric (CT); static int InitText (CT); static int InitMultiPart (CT); static void reverse_parts (CT); static int InitMessage (CT); -static int params_external (CT, int); static int InitApplication (CT); static int init_encoding (CT, OpenCEFunc); -static void close_encoding (CT); static unsigned long size_encoding (CT); static int InitBase64 (CT); static int openBase64 (CT, char **); static int InitQuoted (CT); static int openQuoted (CT, char **); static int Init7Bit (CT); -static int open7Bit (CT, char **); static int openExternal (CT, CT, CE, char **, int *); static int InitFile (CT); static int openFile (CT, char **); @@ -150,17 +136,7 @@ static int InitMail (CT); static int openMail (CT, char **); static int readDigest (CT, char *); -/* - * Structures for mapping (content) types to - * the functions to handle them. - */ -struct str2init { - char *si_key; - int si_val; - InitFunc si_init; -}; - -static struct str2init str2cts[] = { +struct str2init str2cts[] = { { "application", CT_APPLICATION, InitApplication }, { "audio", CT_AUDIO, InitGeneric }, { "image", CT_IMAGE, InitGeneric }, @@ -172,12 +148,12 @@ static struct str2init str2cts[] = { { NULL, CT_UNKNOWN, NULL }, }; -static struct str2init str2ces[] = { +struct str2init str2ces[] = { { "base64", CE_BASE64, InitBase64 }, { "quoted-printable", CE_QUOTED, InitQuoted }, { "8bit", CE_8BIT, Init7Bit }, { "7bit", CE_7BIT, Init7Bit }, - { "binary", CE_BINARY, NULL }, + { "binary", CE_BINARY, Init7Bit }, { NULL, CE_EXTENSION, NULL }, /* these two must be last! */ { NULL, CE_UNKNOWN, NULL }, }; @@ -187,7 +163,7 @@ static struct str2init str2ces[] = { * * si_key is 1 if access method is anonymous. */ -static struct str2init str2methods[] = { +struct str2init str2methods[] = { { "afs", 1, InitFile }, { "anon-ftp", 1, InitFTP }, { "ftp", 0, InitFTP }, @@ -205,7 +181,8 @@ pidcheck (int status) fflush (stdout); fflush (stderr); - return done (1); + done (1); + return 1; } @@ -256,7 +233,6 @@ parse_mime (char *file) if (!(ct = get_content (fp, file, 1))) { if (is_stdin) unlink (file); - fclose (fp); advise (NULL, "unable to decode %s", file); return NULL; } @@ -290,6 +266,7 @@ parse_mime (char *file) * toplevel = 0 # we are inside message type or multipart type * # other than multipart/digest * toplevel = -1 # we are inside multipart/digest + * NB: on failure we will fclose(in)! */ static CT @@ -371,9 +348,10 @@ get_content (FILE *in, char *file, int toplevel) hp = ct->c_first_hf; /* start at first header field */ while (hp) { /* Get MIME-Version field */ - if (!strcasecmp (hp->name, VRSN_FIELD)) { + if (!mh_strcasecmp (hp->name, VRSN_FIELD)) { int ucmp; - char c, *cp, *dp; + char c; + unsigned char *cp, *dp; if (ct->c_vrsn) { advise (NULL, "message %s has multiple %s: fields", @@ -403,14 +381,14 @@ get_content (FILE *in, char *file, int toplevel) continue; c = *dp; *dp = '\0'; - ucmp = !strcasecmp (cp, VRSN_VALUE); + ucmp = !mh_strcasecmp (cp, VRSN_VALUE); *dp = c; if (!ucmp) { admonish (NULL, "message %s has unknown value for %s: field (%s)", ct->c_file, VRSN_FIELD, cp); } } - else if (!strcasecmp (hp->name, TYPE_FIELD)) { + else if (!mh_strcasecmp (hp->name, TYPE_FIELD)) { /* Get Content-Type field */ struct str2init *s2i; CI ci = &ct->c_ctinfo; @@ -423,7 +401,7 @@ get_content (FILE *in, char *file, int toplevel) } /* Parse the Content-Type field */ - if (get_ctinfo (hp->value, ct) == NOTOK) + if (get_ctinfo (hp->value, ct, 0) == NOTOK) goto out; /* @@ -431,16 +409,17 @@ get_content (FILE *in, char *file, int toplevel) * flag for this content type. */ for (s2i = str2cts; s2i->si_key; s2i++) - if (!strcasecmp (ci->ci_type, s2i->si_key)) + if (!mh_strcasecmp (ci->ci_type, s2i->si_key)) break; if (!s2i->si_key && !uprf (ci->ci_type, "X-")) s2i++; ct->c_type = s2i->si_val; ct->c_ctinitfnx = s2i->si_init; } - else if (!strcasecmp (hp->name, ENCODING_FIELD)) { + else if (!mh_strcasecmp (hp->name, ENCODING_FIELD)) { /* Get Content-Transfer-Encoding field */ - char c, *cp, *dp; + char c; + unsigned char *cp, *dp; struct str2init *s2i; /* @@ -468,7 +447,7 @@ get_content (FILE *in, char *file, int toplevel) * for this transfer encoding. */ for (s2i = str2ces; s2i->si_key; s2i++) - if (!strcasecmp (cp, s2i->si_key)) + if (!mh_strcasecmp (cp, s2i->si_key)) break; if (!s2i->si_key && !uprf (cp, "X-")) s2i++; @@ -479,9 +458,10 @@ get_content (FILE *in, char *file, int toplevel) if (s2i->si_init && (*s2i->si_init) (ct) == NOTOK) goto out; } - else if (!strcasecmp (hp->name, MD5_FIELD)) { + else if (!mh_strcasecmp (hp->name, MD5_FIELD)) { /* Get Content-MD5 field */ - char *cp, *dp, *ep; + unsigned char *cp, *dp; + char *ep; if (!checksw) goto next_header; @@ -518,14 +498,18 @@ get_content (FILE *in, char *file, int toplevel) free (ep); ct->c_digested++; } - else if (!strcasecmp (hp->name, ID_FIELD)) { + else if (!mh_strcasecmp (hp->name, ID_FIELD)) { /* Get Content-ID field */ ct->c_id = add (hp->value, ct->c_id); } - else if (!strcasecmp (hp->name, DESCR_FIELD)) { + else if (!mh_strcasecmp (hp->name, DESCR_FIELD)) { /* Get Content-Description field */ ct->c_descr = add (hp->value, ct->c_descr); } + else if (!mh_strcasecmp (hp->name, DISPO_FIELD)) { + /* Get Content-Disposition field */ + ct->c_dispo = add (hp->value, ct->c_dispo); + } next_header: hp = hp->next; /* next header field */ @@ -542,7 +526,7 @@ next_header: * so default type is message/rfc822 */ if (toplevel < 0) { - if (get_ctinfo ("message/rfc822", ct) == NOTOK) + if (get_ctinfo ("message/rfc822", ct, 0) == NOTOK) goto out; ct->c_type = CT_MESSAGE; ct->c_ctinitfnx = InitMessage; @@ -550,7 +534,7 @@ next_header: /* * Else default type is text/plain */ - if (get_ctinfo ("text/plain", ct) == NOTOK) + if (get_ctinfo ("text/plain", ct, 0) == NOTOK) goto out; ct->c_type = CT_TEXT; ct->c_ctinitfnx = InitText; @@ -575,14 +559,13 @@ out: * small routine to add header field to list */ -static int +int add_header (CT ct, char *name, char *value) { HF hp; /* allocate header field structure */ - if (!(hp = malloc (sizeof(*hp)))) - adios (NULL, "out of memory"); + hp = mh_xmalloc (sizeof(*hp)); /* link data into header structure */ hp->name = name; @@ -602,16 +585,99 @@ add_header (CT ct, char *name, char *value) } +/* Make sure that buf contains at least one appearance of name, + followed by =. If not, insert both name and value, just after + first semicolon, if any. Note that name should not contain a + trailing =. And quotes will be added around the value. Typical + usage: make sure that a Content-Disposition header contains + filename="foo". If it doesn't and value does, use value from + that. */ +static char * +incl_name_value (unsigned char *buf, char *name, char *value) { + char *newbuf = buf; + + /* Assume that name is non-null. */ + if (buf && value) { + char *name_plus_equal = concat (name, "=", NULL); + + if (! strstr (buf, name_plus_equal)) { + char *insertion; + unsigned char *cp; + char *prefix, *suffix; + + /* Trim trailing space, esp. newline. */ + for (cp = &buf[strlen (buf) - 1]; + cp >= buf && isspace (*cp); + --cp) { + *cp = '\0'; + } + + insertion = concat ("; ", name, "=", "\"", value, "\"", NULL); + + /* Insert at first semicolon, if any. If none, append to + end. */ + prefix = add (buf, NULL); + if ((cp = strchr (prefix, ';'))) { + suffix = concat (cp, NULL); + *cp = '\0'; + newbuf = concat (prefix, insertion, suffix, "\n", NULL); + free (suffix); + } else { + /* Append to end. */ + newbuf = concat (buf, insertion, "\n", NULL); + } + + free (prefix); + free (insertion); + free (buf); + } + + free (name_plus_equal); + } + + return newbuf; +} + +/* Extract just name_suffix="foo", if any, from value. If there isn't + one, return the entire value. Note that, for example, a name_suffix + of name will match filename="foo", and return foo. */ +static char * +extract_name_value (char *name_suffix, char *value) { + char *extracted_name_value = value; + char *name_suffix_plus_quote = concat (name_suffix, "=\"", NULL); + char *name_suffix_equals = strstr (value, name_suffix_plus_quote); + char *cp; + + free (name_suffix_plus_quote); + if (name_suffix_equals) { + char *name_suffix_begin; + + /* Find first \". */ + for (cp = name_suffix_equals; *cp != '"'; ++cp) /* empty */; + name_suffix_begin = ++cp; + /* Find second \". */ + for (; *cp != '"'; ++cp) /* empty */; + + extracted_name_value = mh_xmalloc (cp - name_suffix_begin + 1); + memcpy (extracted_name_value, + name_suffix_begin, + cp - name_suffix_begin); + extracted_name_value[cp - name_suffix_begin] = '\0'; + } + + return extracted_name_value; +} + /* - * Parse Content-Type line and fill in the - * information of the CTinfo structure. + * Parse Content-Type line and (if `magic' is non-zero) mhbuild composition + * directives. Fills in the information of the CTinfo structure. */ - -static int -get_ctinfo (char *cp, CT ct) +int +get_ctinfo (unsigned char *cp, CT ct, int magic) { int i; - char *dp, **ap, **ep; + unsigned char *dp; + char **ap, **ep; char c; CI ci; @@ -664,7 +730,8 @@ get_ctinfo (char *cp, CT ct) return NOTOK; if (*cp != '/') { - ci->ci_subtype = add ("", NULL); + if (!magic) + ci->ci_subtype = add ("", NULL); goto magic_skip; } @@ -705,7 +772,8 @@ magic_skip: */ ep = (ap = ci->ci_attrs) + NPARMS; while (*cp == ';') { - char *vp, *up; + char *vp; + unsigned char *up; if (ap >= ep) { advise (NULL, @@ -797,11 +865,108 @@ bad_quote: } /* + * Get any given in buffer + */ + if (magic && *cp == '<') { + if (ct->c_id) { + free (ct->c_id); + ct->c_id = NULL; + } + if (!(dp = strchr(ct->c_id = ++cp, '>'))) { + advise (NULL, "invalid ID in message %s", ct->c_file); + return NOTOK; + } + c = *dp; + *dp = '\0'; + if (*ct->c_id) + ct->c_id = concat ("<", ct->c_id, ">\n", NULL); + else + ct->c_id = NULL; + *dp++ = c; + cp = dp; + + while (isspace (*cp)) + cp++; + } + + /* + * Get any [Content-Description] given in buffer. + */ + if (magic && *cp == '[') { + ct->c_descr = ++cp; + for (dp = cp + strlen (cp) - 1; dp >= cp; dp--) + if (*dp == ']') + break; + if (dp < cp) { + advise (NULL, "invalid description in message %s", ct->c_file); + ct->c_descr = NULL; + return NOTOK; + } + + c = *dp; + *dp = '\0'; + if (*ct->c_descr) + ct->c_descr = concat (ct->c_descr, "\n", NULL); + else + ct->c_descr = NULL; + *dp++ = c; + cp = dp; + + while (isspace (*cp)) + cp++; + } + + /* + * Get any {Content-Disposition} given in buffer. + */ + if (magic && *cp == '{') { + ct->c_dispo = ++cp; + for (dp = cp + strlen (cp) - 1; dp >= cp; dp--) + if (*dp == '}') + break; + if (dp < cp) { + advise (NULL, "invalid disposition in message %s", ct->c_file); + ct->c_dispo = NULL; + return NOTOK; + } + + c = *dp; + *dp = '\0'; + if (*ct->c_dispo) + ct->c_dispo = concat (ct->c_dispo, "\n", NULL); + else + ct->c_dispo = NULL; + *dp++ = c; + cp = dp; + + while (isspace (*cp)) + cp++; + } + + /* * Check if anything is left over */ if (*cp) { - advise (NULL, "extraneous information in message %s's %s: field\n%*.*s(%s)", - ct->c_file, TYPE_FIELD, i, i, "", cp); + if (magic) { + ci->ci_magic = add (cp, NULL); + + /* If there is a Content-Disposition header and it doesn't + have a *filename=, extract it from the magic contents. + The r1bindex call skips any leading directory + components. */ + if (ct->c_dispo) + ct->c_dispo = + incl_name_value (ct->c_dispo, + "filename", + r1bindex (extract_name_value ("name", + ci-> + ci_magic), + '/')); + } + else + advise (NULL, + "extraneous information in message %s's %s: field\n%*.*s(%s)", + ct->c_file, TYPE_FIELD, i, i, "", cp); } return OK; @@ -809,10 +974,11 @@ bad_quote: static int -get_comment (CT ct, char **ap, int istype) +get_comment (CT ct, unsigned char **ap, int istype) { int i; - char *bp, *cp; + char *bp; + unsigned char *cp; char c, buffer[BUFSIZ], *dp; CI ci; @@ -892,7 +1058,7 @@ static int InitText (CT ct) { char buffer[BUFSIZ]; - char *chset; + char *chset = NULL; char **ap, **ep, *cp; struct k2v *kv; struct text *t; @@ -904,37 +1070,42 @@ InitText (CT ct) /* match subtype */ for (kv = SubText; kv->kv_key; kv++) - if (!strcasecmp (ci->ci_subtype, kv->kv_key)) + if (!mh_strcasecmp (ci->ci_subtype, kv->kv_key)) break; ct->c_subtype = kv->kv_value; - /* allocate text structure */ + /* allocate text character set structure */ if ((t = (struct text *) calloc (1, sizeof(*t))) == NULL) adios (NULL, "out of memory"); ct->c_ctparams = (void *) t; /* scan for charset parameter */ for (ap = ci->ci_attrs, ep = ci->ci_values; *ap; ap++, ep++) - if (!strcasecmp (*ap, "charset")) + if (!mh_strcasecmp (*ap, "charset")) break; - if (*ap) - chset = *ep; - else - chset = "US-ASCII"; /* default for text */ - - /* match character set, or set to unknown */ - for (kv = Charset; kv->kv_key; kv++) - if (!strcasecmp (chset, kv->kv_key)) - break; - t->tx_charset = kv->kv_value; + /* check if content specified a character set */ + if (*ap) { + /* match character set or set to CHARSET_UNKNOWN */ + for (kv = Charset; kv->kv_key; kv++) { + if (!mh_strcasecmp (*ep, kv->kv_key)) { + chset = *ep; + break; + } + } + t->tx_charset = kv->kv_value; + } else { + t->tx_charset = CHARSET_UNSPECIFIED; + } /* * If we can not handle character set natively, * then check profile for string to modify the * terminal or display method. + * + * termproc is for mhshow, though mhlist -debug prints it, too. */ - if (!check_charset (chset, strlen (chset))) { + if (chset != NULL && !check_charset (chset, strlen (chset))) { snprintf (buffer, sizeof(buffer), "%s-charset-%s", invo_name, chset); if ((cp = context_find (buffer))) ct->c_termproc = getcpy (cp); @@ -953,7 +1124,8 @@ InitMultiPart (CT ct) { int inout; long last, pos; - char *cp, *dp, **ap, **ep; + unsigned char *cp, *dp; + char **ap, **ep; char *bp, buffer[BUFSIZ]; struct multipart *m; struct k2v *kv; @@ -976,7 +1148,7 @@ InitMultiPart (CT ct) /* match subtype */ for (kv = SubMultiPart; kv->kv_key; kv++) - if (!strcasecmp (ci->ci_subtype, kv->kv_key)) + if (!mh_strcasecmp (ci->ci_subtype, kv->kv_key)) break; ct->c_subtype = kv->kv_value; @@ -984,8 +1156,9 @@ InitMultiPart (CT ct) * Check for "boundary" parameter, which is * required for multipart messages. */ + bp = 0; for (ap = ci->ci_attrs, ep = ci->ci_values; *ap; ap++, ep++) { - if (!strcasecmp (*ap, "boundary")) { + if (!mh_strcasecmp (*ap, "boundary")) { bp = *ep; break; } @@ -1052,7 +1225,6 @@ next_part: if (!(p = get_content (fp, ct->c_file, ct->c_subtype == MULTI_DIGEST ? -1 : 0))) { - fclose (ct->c_fp); ct->c_fp = NULL; return NOTOK; } @@ -1109,7 +1281,7 @@ last_part: char partnam[BUFSIZ]; if (ct->c_partno) { - snprintf (partnam, sizeof(partnum), "%s.", ct->c_partno); + snprintf (partnam, sizeof(partnam), "%s.", ct->c_partno); pp = partnam + strlen (partnam); } else { pp = partnam; @@ -1206,7 +1378,7 @@ InitMessage (CT ct) /* match subtype */ for (kv = SubMessage; kv->kv_key; kv++) - if (!strcasecmp (ci->ci_subtype, kv->kv_key)) + if (!mh_strcasecmp (ci->ci_subtype, kv->kv_key)) break; ct->c_subtype = kv->kv_value; @@ -1225,11 +1397,11 @@ InitMessage (CT ct) /* scan for parameters "id", "number", and "total" */ for (ap = ci->ci_attrs, ep = ci->ci_values; *ap; ap++, ep++) { - if (!strcasecmp (*ap, "id")) { + if (!mh_strcasecmp (*ap, "id")) { p->pm_partid = add (*ep, NULL); continue; } - if (!strcasecmp (*ap, "number")) { + if (!mh_strcasecmp (*ap, "number")) { if (sscanf (*ep, "%d", &p->pm_partno) != 1 || p->pm_partno < 1) { invalid_param: @@ -1241,7 +1413,7 @@ invalid_param: } continue; } - if (!strcasecmp (*ap, "total")) { + if (!mh_strcasecmp (*ap, "total")) { if (sscanf (*ep, "%d", &p->pm_maxno) != 1 || p->pm_maxno < 1) goto invalid_param; @@ -1281,7 +1453,6 @@ invalid_param: fseek (fp = ct->c_fp, ct->c_begin, SEEK_SET); if (!(p = get_content (fp, ct->c_file, 0))) { - fclose (ct->c_fp); ct->c_fp = NULL; return NOTOK; } @@ -1301,8 +1472,7 @@ invalid_param: goto no_body; } - if ((e->eb_body = bp = malloc ((unsigned) size)) == NULL) - adios (NULL, "out of memory"); + e->eb_body = bp = mh_xmalloc ((unsigned) size); fseek (p->c_fp, p->c_begin, SEEK_SET); while (size > 0) switch (cc = fread (bp, sizeof(*bp), size, p->c_fp)) { @@ -1355,7 +1525,7 @@ no_body: } -static int +int params_external (CT ct, int composing) { char **ap, **ep; @@ -1363,12 +1533,12 @@ params_external (CT ct, int composing) CI ci = &ct->c_ctinfo; for (ap = ci->ci_attrs, ep = ci->ci_values; *ap; ap++, ep++) { - if (!strcasecmp (*ap, "access-type")) { + if (!mh_strcasecmp (*ap, "access-type")) { struct str2init *s2i; CT p = e->eb_content; for (s2i = str2methods; s2i->si_key; s2i++) - if (!strcasecmp (*ep, s2i->si_key)) + if (!mh_strcasecmp (*ep, s2i->si_key)) break; if (!s2i->si_key) { e->eb_access = *ep; @@ -1385,39 +1555,39 @@ params_external (CT ct, int composing) return NOTOK; continue; } - if (!strcasecmp (*ap, "name")) { + if (!mh_strcasecmp (*ap, "name")) { e->eb_name = *ep; continue; } - if (!strcasecmp (*ap, "permission")) { + if (!mh_strcasecmp (*ap, "permission")) { e->eb_permission = *ep; continue; } - if (!strcasecmp (*ap, "site")) { + if (!mh_strcasecmp (*ap, "site")) { e->eb_site = *ep; continue; } - if (!strcasecmp (*ap, "directory")) { + if (!mh_strcasecmp (*ap, "directory")) { e->eb_dir = *ep; continue; } - if (!strcasecmp (*ap, "mode")) { + if (!mh_strcasecmp (*ap, "mode")) { e->eb_mode = *ep; continue; } - if (!strcasecmp (*ap, "size")) { + if (!mh_strcasecmp (*ap, "size")) { sscanf (*ep, "%lu", &e->eb_size); continue; } - if (!strcasecmp (*ap, "server")) { + if (!mh_strcasecmp (*ap, "server")) { e->eb_server = *ep; continue; } - if (!strcasecmp (*ap, "subject")) { + if (!mh_strcasecmp (*ap, "subject")) { e->eb_subject = *ep; continue; } - if (composing && !strcasecmp (*ap, "body")) { + if (composing && !mh_strcasecmp (*ap, "body")) { e->eb_body = getcpy (*ep); continue; } @@ -1446,7 +1616,7 @@ InitApplication (CT ct) /* match subtype */ for (kv = SubApplication; kv->kv_key; kv++) - if (!strcasecmp (ci->ci_subtype, kv->kv_key)) + if (!mh_strcasecmp (ci->ci_subtype, kv->kv_key)) break; ct->c_subtype = kv->kv_value; @@ -1475,7 +1645,7 @@ init_encoding (CT ct, OpenCEFunc openfnx) } -static void +void close_encoding (CT ct) { CE ce; @@ -1567,8 +1737,9 @@ openBase64 (CT ct, char **file) int fd, len, skip; unsigned long bits; unsigned char value, *b, *b1, *b2, *b3; - char *cp, *ep, buffer[BUFSIZ]; - /* sbeck -- handle prefixes */ + unsigned char *cp, *ep; + char buffer[BUFSIZ]; + /* sbeck -- handle suffixes */ CI ci; CE ce; MD5_CTX mdContext; @@ -1778,11 +1949,11 @@ static int openQuoted (CT ct, char **file) { int cc, digested, len, quoted; - char *cp, *ep; + unsigned char *cp, *ep; char buffer[BUFSIZ]; unsigned char mask; CE ce; - /* sbeck -- handle prefixes */ + /* sbeck -- handle suffixes */ CI ci; MD5_CTX mdContext; @@ -1849,8 +2020,6 @@ openQuoted (CT ct, char **file) fseek (ct->c_fp, ct->c_begin, SEEK_SET); while (len > 0) { - char *dp; - if (fgets (buffer, sizeof(buffer) - 1, ct->c_fp) == NULL) { content_error (NULL, ct, "premature eof"); goto clean_up; @@ -1866,79 +2035,67 @@ openQuoted (CT ct, char **file) *++ep = '\n', ep++; for (; cp < ep; cp++) { - if (quoted) { - if (quoted > 1) { - if (!isxdigit (*cp)) { -invalid_hex: - dp = "expecting hexidecimal-digit"; - goto invalid_encoding; - } + if (quoted > 0) { + /* in an escape sequence */ + if (quoted == 1) { + /* at byte 1 of an escape sequence */ + mask = hex2nib[*cp & 0x7f]; + /* next is byte 2 */ + quoted = 2; + } else { + /* at byte 2 of an escape sequence */ mask <<= 4; mask |= hex2nib[*cp & 0x7f]; putc (mask, ce->ce_fp); if (digested) MD5Update (&mdContext, &mask, 1); - } else { - switch (*cp) { - case ':': - putc (*cp, ce->ce_fp); - if (digested) - MD5Update (&mdContext, (unsigned char *) ":", 1); - break; - - default: - if (!isxdigit (*cp)) - goto invalid_hex; - mask = hex2nib[*cp & 0x7f]; - quoted = 2; - continue; + if (ferror (ce->ce_fp)) { + content_error (ce->ce_file, ct, "error writing to"); + goto clean_up; } + /* finished escape sequence; next may be literal or a new + * escape sequence */ + quoted = 0; } - - if (ferror (ce->ce_fp)) { - content_error (ce->ce_file, ct, "error writing to"); - goto clean_up; - } - quoted = 0; + /* on to next byte */ continue; } - switch (*cp) { - default: - if (*cp < '!' || *cp > '~') { - int i; - dp = "expecting character in range [!..~]"; - -invalid_encoding: - i = strlen (invo_name) + 2; - content_error (NULL, ct, - "invalid QUOTED-PRINTABLE encoding -- %s,\n%*.*sbut got char 0x%x", - dp, i, i, "", *cp); - goto clean_up; - } - /* and fall...*/ - case ' ': - case '\t': - case '\n': - putc (*cp, ce->ce_fp); - if (digested) { - if (*cp == '\n') - MD5Update (&mdContext, (unsigned char *) "\r\n",2); - else - MD5Update (&mdContext, (unsigned char *) cp, 1); + /* not in an escape sequence */ + if (*cp == '=') { + /* starting an escape sequence, or invalid '='? */ + if (cp + 1 < ep && cp[1] == '\n') { + /* "=\n" soft line break, eat the \n */ + cp++; + continue; } - if (ferror (ce->ce_fp)) { - content_error (ce->ce_file, ct, "error writing to"); - goto clean_up; + if (cp + 1 >= ep || cp + 2 >= ep) { + /* We don't have 2 bytes left, so this is an invalid + * escape sequence; just show the raw bytes (below). */ + } else if (isxdigit (cp[1]) && isxdigit (cp[2])) { + /* Next 2 bytes are hex digits, making this a valid escape + * sequence; let's decode it (above). */ + quoted = 1; + continue; + } else { + /* One or both of the next 2 is out of range, making this + * an invalid escape sequence; just show the raw bytes + * (below). */ } - break; + } - case '=': - if (*++cp != '\n') { - quoted = 1; - cp--; + /* Just show the raw byte. */ + putc (*cp, ce->ce_fp); + if (digested) { + if (*cp == '\n') { + MD5Update (&mdContext, (unsigned char *) "\r\n",2); + } else { + MD5Update (&mdContext, (unsigned char *) cp, 1); } - break; + } + if (ferror (ce->ce_fp)) { + content_error (ce->ce_file, ct, "error writing to"); + goto clean_up; } } } @@ -1995,12 +2152,12 @@ Init7Bit (CT ct) } -static int +int open7Bit (CT ct, char **file) { int cc, fd, len; char buffer[BUFSIZ]; - /* sbeck -- handle prefixes */ + /* sbeck -- handle suffixes */ char *cp; CI ci; CE ce; @@ -2087,6 +2244,8 @@ open7Bit (CT ct, char **file) fprintf (ce->ce_fp, "%s:%s", ID_FIELD, ct->c_id); if (ct->c_descr) fprintf (ce->ce_fp, "%s:%s", DESCR_FIELD, ct->c_descr); + if (ct->c_dispo) + fprintf (ce->ce_fp, "%s:%s", DISPO_FIELD, ct->c_dispo); fprintf (ce->ce_fp, "\n"); } @@ -2224,7 +2383,7 @@ openFile (CT ct, char **file) return NOTOK; } - if ((!e->eb_permission || strcasecmp (e->eb_permission, "read-write")) + if ((!e->eb_permission || mh_strcasecmp (e->eb_permission, "read-write")) && find_cache (NULL, wcachesw, &cachetype, e->eb_content->c_id, cachefile, sizeof(cachefile)) != NOTOK) { int mask; @@ -2372,7 +2531,7 @@ openFTP (CT ct, char **file) ce->ce_unlink = (*file == NULL); caching = 0; cachefile[0] = '\0'; - if ((!e->eb_permission || strcasecmp (e->eb_permission, "read-write")) + if ((!e->eb_permission || mh_strcasecmp (e->eb_permission, "read-write")) && find_cache (NULL, wcachesw, &cachetype, e->eb_content->c_id, cachefile, sizeof(cachefile)) != NOTOK) { if (*file == NULL) { @@ -2408,7 +2567,7 @@ openFTP (CT ct, char **file) vec[vecp++] = e->eb_dir; vec[vecp++] = e->eb_name; vec[vecp++] = ce->ce_file, - vec[vecp++] = e->eb_mode && !strcasecmp (e->eb_mode, "ascii") + vec[vecp++] = e->eb_mode && !mh_strcasecmp (e->eb_mode, "ascii") ? "ascii" : "binary"; vec[vecp] = NULL; @@ -2445,7 +2604,7 @@ losing_ftp: else if (ftp_get (e->eb_site, user, pass, e->eb_dir, e->eb_name, ce->ce_file, - e->eb_mode && !strcasecmp (e->eb_mode, "ascii"), 0) + e->eb_mode && !mh_strcasecmp (e->eb_mode, "ascii"), 0) == NOTOK) goto losing_ftp; #endif @@ -2600,6 +2759,8 @@ openMail (CT ct, char **file) return NOTOK; } + /* showproc is for mhshow and mhstore, though mhlist -debug + * prints it, too. */ if (ct->c_showproc) free (ct->c_showproc); ct->c_showproc = add ("true", NULL); @@ -2674,7 +2835,7 @@ invalid_digest: while (*cp) cp++; fprintf (stderr, "invalid MD5 digest (got %d octets)\n", - cp - bp); + (int)(cp - bp)); } return NOTOK;