X-Git-Url: http://git.madism.org/?p=apps%2Fmadmutt.git;a=blobdiff_plain;f=lib-mime%2Frfc822parse.c;h=121efb18b0d1641e09ed4ff8499c566ccd816841;hp=9643aeef9539f8cc0320affbc1e116bab8e0adab;hb=refs%2Fheads%2Ftry-some-d;hpb=108f3c7ab59844591f7540347914ea57be5245e2 diff --git a/lib-mime/rfc822parse.c b/lib-mime/rfc822parse.c index 9643aee..121efb1 100644 --- a/lib-mime/rfc822parse.c +++ b/lib-mime/rfc822parse.c @@ -26,23 +26,10 @@ * please see the file GPL in the top level source directory. */ -#if HAVE_CONFIG_H -# include "config.h" -#endif - -#include - -#include -#include -#include -#include -#include -#include -#include -#include +#include #include "recvattach.h" - +#include "charset.h" #include "mime.h" /* Reads an arbitrarily long header field, and looks ahead for continuation @@ -129,8 +116,8 @@ string_list_t *mutt_parse_references(char *s, int in_reply_to) o = NULL; } else { new = p_new(char, n + m + 1); - strcpy(new, o); - strcpy(new + n, s); + m_strcpy(new, n + m + 1, o); + m_strcpy(new + n, m + 1, s); } } @@ -201,14 +188,13 @@ int mutt_check_mime_type(const char *s) } } -static PARAMETER *parse_parameters(const char *s) +static parameters_t *parse_parameters(const char *s) { - PARAMETER *res = NULL; - PARAMETER **list = &res; + parameters_t *res = parameter_new(); while (*s) { + char *attr, *val; const char *p; - PARAMETER *new; int i; s = skipspaces(s); @@ -226,11 +212,10 @@ static PARAMETER *parse_parameters(const char *s) } i = p - s; - new = parameter_new(); - new->attribute = p_dupstr(s, i); + attr = p_dupstr(s, i); - while (--i >= 0 && ISSPACE(new->attribute[i])) { - new->attribute[i] = '\0'; + while (--i >= 0 && ISSPACE(attr[i])) { + attr[i] = '\0'; } s = skipspaces(p + 1); /* skip over the = */ @@ -240,12 +225,10 @@ static PARAMETER *parse_parameters(const char *s) s++; for (i = 0; *s && i < ssizeof(buffer) - 1; i++, s++) { - if (!option(OPTSTRICTMIME)) { - /* As iso-2022-* has a characer of '"' with non-ascii state, - * ignore it. */ - if (*s == 0x1b && i < ssizeof(buffer) - 2) { - state_ascii = s[1] == '(' && (s[2] == 'B' || s[2] == 'J'); - } + /* As iso-2022-* has a characer of '"' with non-ascii state, + * ignore it. */ + if (*s == 0x1b && i < ssizeof(buffer) - 2) { + state_ascii = s[1] == '(' && (s[2] == 'B' || s[2] == 'J'); } if (state_ascii && *s == '"') break; @@ -257,21 +240,21 @@ static PARAMETER *parse_parameters(const char *s) } } - new->value = p_dupstr(buffer, i); + val = p_dupstr(buffer, i); } else { for (p = s; *p && *p != ' ' && *p != ';'; p++); - new->value = p_dupstr(s, p - s); + val = p_dupstr(s, p - s); } - *list = new; - list = &new->next; - + parameter_setval(res, attr, val); + p_delete(&attr); + p_delete(&val); s = strchr(s, ';'); /* Find the next parameter */ if (!s) break; /* no more parameters */ } - rfc2231_decode_parameters(&res); + rfc2231_decode_parameters(res); return res; } @@ -281,7 +264,7 @@ void mutt_parse_content_type(char *s, BODY *ct) char *subtype; p_delete(&ct->subtype); - parameter_list_wipe(&ct->parameter); + ct->parameter = NULL; /* First extract any existing parameters */ if ((pc = strchr(s, ';')) != NULL) { @@ -291,7 +274,7 @@ void mutt_parse_content_type(char *s, BODY *ct) /* Some pre-RFC1521 gateways still use the "name=filename" convention, * but if a filename has already been set in the content-disposition, * let that take precedence, and don't set it here */ - pc = mutt_get_parameter("name", ct->parameter); + pc = parameter_getval(ct->parameter, "name"); if (pc && !ct->filename) ct->filename = m_strdup(pc); } @@ -315,7 +298,7 @@ void mutt_parse_content_type(char *s, BODY *ct) * field, so we can attempt to convert the type to BODY here. */ switch (ct->type) { - char buffer[SHORT_STRING]; + char buffer[STRING]; case TYPETEXT: ct->subtype = m_strdup("plain"); @@ -343,17 +326,15 @@ void mutt_parse_content_type(char *s, BODY *ct) /* Default character set for text types. */ if (ct->type == TYPETEXT) { - pc = mutt_get_parameter("charset", ct->parameter); + pc = parameter_getval(ct->parameter, "charset"); if (!pc) { - mutt_set_parameter("charset", - option(OPTSTRICTMIME) ? "us-ascii" : - mutt_get_first_charset(AssumedCharset), - &ct->parameter); + parameter_setval(ct->parameter, "charset", + charset_getfirst(mod_cset.assumed_charset)); } } } -static void parse_content_disposition(char *s, BODY *ct) +static void parse_content_disposition(const char *s, BODY *ct) { if (!ascii_strncasecmp(s, "inline", 6)) { ct->disposition = DISPINLINE; @@ -365,14 +346,12 @@ static void parse_content_disposition(char *s, BODY *ct) /* Check to see if a default filename was given */ if ((s = strchr (s, ';'))) { - PARAMETER *parms = parse_parameters(vskipspaces(s)); + parameters_t *parms = parse_parameters(vskipspaces(s)); - if ((s = mutt_get_parameter("filename", parms))) + if ((s = parameter_getval(parms, "filename"))) m_strreplace(&ct->filename, s); - if ((s = mutt_get_parameter ("name", parms))) + if ((s = parameter_getval(parms, "name"))) ct->form_name = m_strdup(s); - - parameter_list_wipe(&parms); } } @@ -384,7 +363,7 @@ static void parse_content_disposition(char *s, BODY *ct) */ BODY *mutt_read_mime_header(FILE *fp, int digest) { - BODY *body = mutt_new_body (); + BODY *body = body_new(); char *line = p_new(char, LONG_STRING); ssize_t linelen = LONG_STRING; char *p; @@ -402,7 +381,6 @@ BODY *mutt_read_mime_header(FILE *fp, int digest) if (!*p) continue; } else { - debug_print (1, ("bogus MIME header: %s\n", line)); break; } @@ -437,7 +415,7 @@ BODY *mutt_read_mime_header(FILE *fp, int digest) } p_delete(&line); - return (body); + return body; } void mutt_parse_part(FILE *fp, BODY *b) @@ -446,7 +424,7 @@ void mutt_parse_part(FILE *fp, BODY *b) switch (b->type) { case TYPEMULTIPART: - bound = mutt_get_parameter("boundary", b->parameter); + bound = parameter_getval(b->parameter, "boundary"); fseeko(fp, b->offset, SEEK_SET); b->parts = mutt_parse_multipart(fp, bound, b->offset + b->length, mime_which_token(b->subtype, -1) == MIME_DIGEST); @@ -456,7 +434,7 @@ void mutt_parse_part(FILE *fp, BODY *b) if (b->subtype) { fseeko(fp, b->offset, SEEK_SET); - if (mutt_is_message_type(b->type, b->subtype)) { + if (mutt_is_message_type(b)) { b->parts = mutt_parse_messageRFC822(fp, b); } else if (mime_which_token(b->subtype, -1) == MIME_EXTERNAL_BODY) { @@ -574,7 +552,7 @@ mutt_parse_multipart(FILE *fp, const char *bound, off_t end_off, int digest) */ if (new->offset > end_off) { - mutt_free_body(&new); + body_list_wipe(&new); break; } @@ -596,82 +574,9 @@ mutt_parse_multipart(FILE *fp, const char *bound, off_t end_off, int digest) mutt_parse_part(fp, b); } - return (head); -} - -static const char * -uncomment_timezone(char *buf, size_t buflen, const char *tz) -{ - char *p; - - if (*tz != '(') - return tz; /* no need to do anything */ - - tz = vskipspaces(tz + 1); - p = strpbrk(tz, " )"); - if (!p) - return tz; - - m_strncpy(buf, buflen, tz, p - tz); - return buf; + return head; } -static struct tz_t { - char tzname[5]; - unsigned char zhours; - unsigned char zminutes; - unsigned char zoccident; /* west of UTC? */ -} TimeZones[] = { - {"aat", 1, 0, 1}, /* Atlantic Africa Time */ - {"adt", 4, 0, 0}, /* Arabia DST */ - {"ast", 3, 0, 0}, /* Arabia */ - /*{ "ast", 4, 0, 1 }, *//* Atlantic */ - {"bst", 1, 0, 0}, /* British DST */ - {"cat", 1, 0, 0}, /* Central Africa */ - {"cdt", 5, 0, 1}, - {"cest", 2, 0, 0}, /* Central Europe DST */ - {"cet", 1, 0, 0}, /* Central Europe */ - {"cst", 6, 0, 1}, - /*{ "cst", 8, 0, 0 }, *//* China */ - /*{ "cst", 9, 30, 0 }, *//* Australian Central Standard Time */ - {"eat", 3, 0, 0}, /* East Africa */ - {"edt", 4, 0, 1}, - {"eest", 3, 0, 0}, /* Eastern Europe DST */ - {"eet", 2, 0, 0}, /* Eastern Europe */ - {"egst", 0, 0, 0}, /* Eastern Greenland DST */ - {"egt", 1, 0, 1}, /* Eastern Greenland */ - {"est", 5, 0, 1}, - {"gmt", 0, 0, 0}, - {"gst", 4, 0, 0}, /* Presian Gulf */ - {"hkt", 8, 0, 0}, /* Hong Kong */ - {"ict", 7, 0, 0}, /* Indochina */ - {"idt", 3, 0, 0}, /* Israel DST */ - {"ist", 2, 0, 0}, /* Israel */ - /*{ "ist", 5, 30, 0 }, *//* India */ - {"jst", 9, 0, 0}, /* Japan */ - {"kst", 9, 0, 0}, /* Korea */ - {"mdt", 6, 0, 1}, - {"met", 1, 0, 0}, /* this is now officially CET */ - {"msd", 4, 0, 0}, /* Moscow DST */ - {"msk", 3, 0, 0}, /* Moscow */ - {"mst", 7, 0, 1}, - {"nzdt", 13, 0, 0}, /* New Zealand DST */ - {"nzst", 12, 0, 0}, /* New Zealand */ - {"pdt", 7, 0, 1}, - {"pst", 8, 0, 1}, - {"sat", 2, 0, 0}, /* South Africa */ - {"smt", 4, 0, 0}, /* Seychelles */ - {"sst", 11, 0, 1}, /* Samoa */ - /*{ "sst", 8, 0, 0 }, *//* Singapore */ - {"utc", 0, 0, 0}, - {"wat", 0, 0, 0}, /* West Africa */ - {"west", 1, 0, 0}, /* Western Europe DST */ - {"wet", 0, 0, 0}, /* Western Europe */ - {"wgst", 2, 0, 1}, /* Western Greenland DST */ - {"wgt", 3, 0, 1}, /* Western Greenland */ - {"wst", 8, 0, 0}, /* Western Australia */ -}; - /* parses a date string in RFC822 format: * * Date: [ weekday , ] day-of-month month year hour:minute:second timezone @@ -681,117 +586,24 @@ static struct tz_t { */ time_t mutt_parse_date(const char *s, HEADER *h) { - int zhours = 0, zminutes = 0, zoccident = 0; - char scratch[SHORT_STRING]; struct tm tm; - int count = 0; - char *p; - - /* Don't modify our argument. Fixed-size buffer is ok here since - the date format imposes a natural limit. */ - - m_strcpy(scratch, sizeof(scratch), s); - - /* kill the day of the week, if it exists. */ - p = strchr(scratch, ','); - p = vskipspaces(p ? p + 1 : scratch); + const char *loc; + time_t tz; + loc = setlocale(LC_ALL, "C"); p_clear(&tm, 1); + if (strptime(s, "%a, %d %b %Y %H:%M:%S %z", &tm)) + goto ok; + p_clear(&tm, 1); + if (strptime(s, "%a, %d %b %Y %H:%M %z", &tm)) + goto ok; + setlocale(LC_ALL, ""); + return 0; - while ((p = strtok (p, " \t")) != NULL) { - char tzstr[SHORT_STRING]; - const char *ptz; - - switch (count) { - case 0: /* day of the month */ - if (!isdigit((unsigned char)*p)) - return -1; - tm.tm_mday = atoi(p); - if (tm.tm_mday > 31) - return -1; - break; - - case 1: /* month of the year */ - tm.tm_mon = mutt_check_month(p); - if (tm.tm_mon < 0) - return -1; - break; - - case 2: /* year */ - tm.tm_year = atoi(p); - if (tm.tm_year < 50) - tm.tm_year += 100; - else if (tm.tm_year >= 1900) - tm.tm_year -= 1900; - break; - - case 3: /* time of day */ - tm.tm_hour = strtol(p, &p, 10); - if (*p++ != ':') - return -1; - tm.tm_min = strtol(p, &p, 10); - if (*p++ == ':') { - tm.tm_sec = strtol(p, &p, 10); - } else { - tm.tm_sec = 0; - } - break; - - case 4: /* timezone */ - /* sometimes we see things like (MST) or (-0700) so attempt to - * compensate by uncommenting the string if non-RFC822 compliant - */ - ptz = uncomment_timezone(tzstr, sizeof(tzstr), p); - - if (*ptz == '+' || *ptz == '-') { - if (isdigit((unsigned char)ptz[1]) - && isdigit((unsigned char)ptz[2]) - && isdigit((unsigned char)ptz[3]) - && isdigit((unsigned char)ptz[4])) - { - zoccident = ptz[0] == '-'; - zhours = (ptz[1] - '0') * 10 + (ptz[2] - '0'); - zminutes = (ptz[3] - '0') * 10 + (ptz[4] - '0'); - } - } else { - struct tz_t *tz; - - /* This is safe to do: A pointer to a struct equals a pointer to its - * first element*/ - tz = bsearch(ptz, TimeZones, countof(TimeZones), sizeof(TimeZones[0]), - (int (*)(const void *, const void *))ascii_strcasecmp); - - if (tz) { - zhours = tz->zhours; - zminutes = tz->zminutes; - zoccident = tz->zoccident; - } - - /* ad hoc support for the European MET (now officially CET) TZ */ - if (ascii_strcasecmp(p, "MET") == 0) { - if ((p = strtok (NULL, " \t")) && !ascii_strcasecmp(p, "DST")) { - zhours++; - } - } - } - break; - } - count++; - p = NULL; - } - - if (count < 4) { /* don't check for missing timezone */ - debug_print (1, ("error parsing date format, using received time\n")); - return -1; - } - - if (h) { - h->zhours = zhours; - h->zminutes = zminutes; - h->zoccident = zoccident; - } - - return mutt_mktime(&tm, 0) + (zoccident ? 1 : -1) * (zhours * 3600 + zminutes * 60); + ok: + setlocale(LC_ALL, ""); + tz = tm.tm_gmtoff; + return timegm(&tm) - tz; } string_list_t **mutt_parse_rfc822_line(ENVELOPE *e, HEADER *hdr, char *line, char *p, @@ -854,15 +666,6 @@ string_list_t **mutt_parse_rfc822_line(ENVELOPE *e, HEADER *hdr, char *line, cha hdr->expired = 1; break; -#ifdef USE_NNTP - case MIME_FOLLOWUP_TO: - if (!e->followup_to) { - m_strrtrim(p); - e->followup_to = m_strdup(skipspaces(p)); - } - break; -#endif - case MIME_FROM: e->from = rfc822_parse_adrlist(e->from, p); /* don't leave from info NULL if there's an invalid address (or @@ -890,7 +693,7 @@ string_list_t **mutt_parse_rfc822_line(ENVELOPE *e, HEADER *hdr, char *line, cha case MIME_LIST_POST: /* RFC 2369. FIXME: We should ignore whitespace, but don't. */ - if (strncmp(p, "NO", 2)) { + if (m_strncmp(p, "NO", 2)) { char *beg, *end; for (beg = strchr (p, '<'); beg; beg = strchr (end, ',')) { @@ -934,14 +737,6 @@ string_list_t **mutt_parse_rfc822_line(ENVELOPE *e, HEADER *hdr, char *line, cha hdr->mime = 1; break; -#ifdef USE_NNTP - case MIME_NEWSGROUPS: - p_delete(&e->newsgroups); - m_strrtrim(p); - e->newsgroups = m_strdup(skipspaces(p)); - break; -#endif - case MIME_ORGANIZATION: if (!e->organization && mime_which_token(p, -1) == MIME_UNKNOWN) e->organization = m_strdup(p); @@ -1006,24 +801,10 @@ string_list_t **mutt_parse_rfc822_line(ENVELOPE *e, HEADER *hdr, char *line, cha e->to = rfc822_parse_adrlist(e->to, p); break; -#ifdef USE_NNTP - case MIME_X_COMMENT_TO: - if (!e->x_comment_to) - e->x_comment_to = m_strdup(p); - break; -#endif - case MIME_X_LABEL: e->x_label = m_strdup(p); break; -#ifdef USE_NNTP - case MIME_XREF: - if (!e->xref) - e->xref = m_strdup(p); - break; -#endif - case MIME_X_STATUS: if (hdr) { while (*p) { @@ -1052,8 +833,8 @@ string_list_t **mutt_parse_rfc822_line(ENVELOPE *e, HEADER *hdr, char *line, cha /* restore the original line */ line[m_strlen(line)] = ':'; - if (weed && option(OPTWEED) && mutt_matches_ignore(line, Ignore) - && !mutt_matches_ignore(line, UnIgnore)) { + if (weed && string_list_contains(Ignore, line, "*") + && !string_list_contains(UnIgnore, line, "*")) { break; } @@ -1096,7 +877,7 @@ mutt_read_rfc822_header(FILE *f, HEADER *hdr, short user_hdrs, short weed) off_t loc; if (hdr && !hdr->content) { - hdr->content = mutt_new_body (); + hdr->content = body_new(); /* set the defaults from RFC1521 */ hdr->content->type = TYPETEXT; @@ -1116,46 +897,29 @@ mutt_read_rfc822_header(FILE *f, HEADER *hdr, short user_hdrs, short weed) p = strpbrk(line, ": \t"); if (!p || *p != ':') { - char return_path[LONG_STRING]; - time_t t; - /* some bogus MTAs will quote the original "From " line */ - if (!m_strncmp(">From ", line, 6)) + if (!m_strncmp(">From ", line, 6) || !m_strncmp("From ", line, 5)) continue; /* just ignore */ - if (is_from(line, return_path, sizeof(return_path), &t)) { - /* MH somtimes has the From_ line in the middle of the header! */ - if (hdr && !hdr->received) - hdr->received = t - mutt_local_tz(t); - continue; - } - fseeko(f, loc, 0); break; /* end of header */ } - if (mutt_match_spam_list(line, SpamList, buf, sizeof(buf))) { - if (!rx_list_match(NoSpamList, line)) { - /* if spam tag already exists, figure out how to amend it */ - if (e->spam && *buf) { - if (SpamSep) { - /* If SpamSep defined, append with separator */ - mutt_buffer_addstr(e->spam, SpamSep); - mutt_buffer_addstr(e->spam, buf); - } else { - /* else overwrite */ - mutt_buffer_reset(e->spam); - mutt_buffer_addstr(e->spam, buf); - } - } - else if (!e->spam && *buf) { - /* spam tag is new, and match expr is non-empty; copy */ - e->spam = mutt_buffer_from(NULL, buf); - } - else if (!e->spam) { - /* match expr is empty; plug in null string if no existing tag */ - e->spam = mutt_buffer_from(NULL, ""); + if (rx_list_match2(SpamList, line, buf, sizeof(buf)) + && !rx_list_match(NoSpamList, line)) + { + /* if spam tag already exists, figure out how to amend it */ + if (e->spam && *buf) { + if (mod_mime.spam_separator) { + mutt_buffer_addstr(e->spam, mod_mime.spam_separator); + } else { + mutt_buffer_reset(e->spam); } + mutt_buffer_addstr(e->spam, buf); + } + + if (!e->spam) { + e->spam = mutt_buffer_from(NULL, buf); } } @@ -1175,8 +939,6 @@ mutt_read_rfc822_header(FILE *f, HEADER *hdr, short user_hdrs, short weed) rfc2047_decode_envelope(e); /* check for missing or invalid date */ if (hdr->date_sent <= 0) { - debug_print(1, ("no date found, using received " - "time from msg separator\n")); hdr->date_sent = hdr->received; } } @@ -1258,9 +1020,6 @@ static int count_body_parts (BODY *body, int flags) int mutt_count_body_parts(HEADER *hdr, int flags) { - if (!option(OPTCOUNTATTACH)) - return 0; - if (hdr->attach_valid && !(flags & M_PARTS_RECOUNT)) return hdr->attach_total; @@ -1273,3 +1032,72 @@ int mutt_count_body_parts(HEADER *hdr, int flags) hdr->attach_valid = 1; return hdr->attach_total; } + +/* + * A valid message separator looks like: + * + * From [ ]