X-Git-Url: http://git.madism.org/?p=apps%2Fmadmutt.git;a=blobdiff_plain;f=lib-mime%2Frfc2047.c;h=5b88b4beee367013bfc5271461e7ffc9b71c306d;hp=442d2e4352f5c6ae78e1c45bbac868c6890b1986;hb=c5ca8b7eaa727392811813926cda758bceef80f1;hpb=df5b14084b39821395408679a2fb6f74605182ce diff --git a/lib-mime/rfc2047.c b/lib-mime/rfc2047.c index 442d2e4..5b88b4b 100644 --- a/lib-mime/rfc2047.c +++ b/lib-mime/rfc2047.c @@ -37,6 +37,7 @@ #include "charset.h" #include "thread.h" +#include #include #include #include @@ -49,7 +50,6 @@ #ifdef NDEBUG #define assert(x) #else -#include #endif #define ENCWORD_LEN_MAX 75 @@ -59,9 +59,6 @@ #define CONTINUATION_BYTE(c) (((c) & 0xc0) == 0x80) -typedef size_t (*encoder_t) (char *, const char *, size_t, - const char *); - /* converts f of len flen and charset from into *t of len *tlen and charset to @@ -69,14 +66,12 @@ typedef size_t (*encoder_t) (char *, const char *, size_t, returns number of converted chars from f, see iconv(3) */ static ssize_t -convert_string(const char *from, const char *f, size_t flen, - const char *to, char **t, size_t *tlen) +convert_string(const char *from, const char *f, ssize_t flen, + const char *to, char **t, ssize_t *tlen) { iconv_t cd; char *buf, *ob; - size_t obl; - ssize_t n; - int e; + ssize_t obl, n; cd = mutt_iconv_open(to, from, 0); @@ -88,109 +83,134 @@ convert_string(const char *from, const char *f, size_t flen, n = my_iconv(cd, &f, &flen, &ob, &obl); if (n < 0 || my_iconv(cd, 0, 0, &ob, &obl) < 0) { - e = errno; - p_delete(&buf); - iconv_close (cd); + int e = errno; + iconv_close(cd); errno = e; + p_delete(&buf); return -1; } + iconv_close(cd); *ob = '\0'; *tlen = ob - buf; - - p_realloc(&buf, ob - buf + 1); - *t = buf; - iconv_close (cd); + *t = buf; return n; } +/* choose the shortest encoding for u */ char *mutt_choose_charset(const char *fromcode, const char *charsets, - char *u, size_t ulen, char **d, size_t *dlen) + char *u, ssize_t ulen, char **dst, ssize_t *dlen) { - char canonical_buff[LONG_STRING]; - char *e = 0, *tocode = 0; - size_t elen = 0, bestn = 0; - const char *p, *q; + char *res = NULL; + ssize_t reslen = 0; - for (p = charsets; p; p = q ? q + 1 : 0) { - char *s, *t; - size_t slen, n; + char *tocode = NULL; + ssize_t bestn = 0; - q = strchr (p, ':'); + const char *p = charsets; - n = q ? q - p : m_strlen(p); + while (*p) { + char cset[SHORT_STRING]; + const char *q; + char *s; + ssize_t slen, n; - if (!n || + q = strchr(p, ':'); + if (q) { + n = m_strncpy(cset, sizeof(cset), p, q - p); + p = ++q; + } else { + n = m_strcpy(cset, sizeof(cset), p); + p += n; + } + + if (!n || n > (ENCWORD_LEN_MAX - ENCWORD_LEN_MIN + 2 - 12)) { /* Assume that we never need more than 12 characters of encoded-text to encode a single character. */ - n > (ENCWORD_LEN_MAX - ENCWORD_LEN_MIN + 2 - 12)) continue; + } - t = p_dupstr(p, n); - - n = convert_string(fromcode, u, ulen, t, &s, &slen); - if (n == (size_t) (-1)) + n = convert_string(fromcode, u, ulen, cset, &s, &slen); + if (n < 0) continue; if (!tocode || n < bestn) { + m_strreplace(&tocode, cset); bestn = n; - p_delete(&tocode); - tocode = t; - if (d) { - p_delete(&e); - e = s; - } else { - p_delete(&s); - } - elen = slen; + + p_delete(&res); + res = s; + reslen = slen; if (!bestn) break; } else { - p_delete(&t); p_delete(&s); } } if (tocode) { - if (d) - *d = e; - if (dlen) - *dlen = elen; + char buf[LONG_STRING]; + + if (dst && dlen) { + *dst = res; + *dlen = reslen; + } else { + p_delete(&res); + } - mutt_canonical_charset(canonical_buff, sizeof(canonical_buff), tocode); - m_strreplace(&tocode, canonical_buff); + mutt_canonical_charset(buf, sizeof(buf), tocode); + m_strreplace(&tocode, buf); } return tocode; } -static size_t b_encoder (char *s, const char *d, size_t dlen, - const char *tocode) + +/****************************************************************************/ +/* Encoding functions */ +/****************************************************************************/ + +static const char __qp_special[128] = { + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 0, 0, 1, 0, 0, 0, 0, 0, 1, 1, 0, 0, 1, 0, 1, 1, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, + 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 0, 1, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, +}; + +typedef size_t (encoder_t)(char *, const char *, ssize_t, const char *); + +static size_t +b_encoder(char *s, const char *d, ssize_t dlen, const char *tocode) { char *s0 = s; - memcpy (s, "=?", 2), s += 2; - memcpy (s, tocode, m_strlen(tocode)), s += m_strlen(tocode); - memcpy (s, "?B?", 3), s += 3; + s += sprintf(s, "=?%s?B?", tocode); + for (;;) { - if (!dlen) - break; - else if (dlen == 1) { + switch (dlen) { + case 0: + goto done; + + case 1: *s++ = __m_b64chars[(*d >> 2) & 0x3f]; *s++ = __m_b64chars[(*d & 0x03) << 4]; *s++ = '='; *s++ = '='; - break; - } - else if (dlen == 2) { + goto done; + + case 2: *s++ = __m_b64chars[(*d >> 2) & 0x3f]; *s++ = __m_b64chars[((*d & 0x03) << 4) | ((d[1] >> 4) & 0x0f)]; *s++ = __m_b64chars[(d[1] & 0x0f) << 2]; *s++ = '='; - break; - } - else { + goto done; + + default: *s++ = __m_b64chars[(*d >> 2) & 0x3f]; *s++ = __m_b64chars[((*d & 0x03) << 4) | ((d[1] >> 4) & 0x0f)]; *s++ = __m_b64chars[((d[1] & 0x0f) << 2) | ((d[2] >> 6) & 0x03)]; @@ -198,32 +218,36 @@ static size_t b_encoder (char *s, const char *d, size_t dlen, d += 3, dlen -= 3; } } - memcpy (s, "?=", 2), s += 2; + + done: + *s++ = '?'; + *s++ = '='; return s - s0; } -static size_t q_encoder (char *s, const char *d, size_t dlen, - const char *tocode) +static size_t +q_encoder(char *s, const char *d, ssize_t dlen, const char *tocode) { char *s0 = s; - memcpy (s, "=?", 2), s += 2; - memcpy (s, tocode, m_strlen(tocode)), s += m_strlen(tocode); - memcpy (s, "?Q?", 3), s += 3; + s += sprintf(s, "=?%s?Q?", tocode); while (dlen--) { unsigned char c = *d++; - if (c == ' ') + if (c == ' ') { *s++ = '_'; - else if (c >= 0x7f || c < 0x20 || c == '_' || strchr (MimeSpecials, c)) { + } else + if (c & 0x80 || __qp_special[c]) { *s++ = '='; *s++ = __m_b36chars_upper[c >> 4]; *s++ = __m_b36chars_upper[c & 0xf]; - } - else + } else { *s++ = c; + } } - memcpy (s, "?=", 2), s += 2; + + *s++ = '?'; + *s++ = '='; return s - s0; } @@ -236,83 +260,83 @@ static size_t q_encoder (char *s, const char *d, size_t dlen, * tocode, unless fromcode is 0, in which case the data is assumed to * be already in tocode, which should be 8-bit and stateless. */ -static size_t try_block (const char *d, size_t dlen, +static size_t try_block (const char *d, ssize_t dlen, const char *fromcode, const char *tocode, - encoder_t * encoder, size_t * wlen) + encoder_t **encoder, ssize_t *wlen) { char buf1[ENCWORD_LEN_MAX - ENCWORD_LEN_MIN + 1]; - iconv_t cd; - const char *ib; - char *ob, *p; - size_t ibl, obl; - int count, len, len_b, len_q; + ssize_t obl = sizeof(buf1) - m_strlen(tocode); + char *ob; if (fromcode) { - cd = mutt_iconv_open (tocode, fromcode, 0); - assert (cd != (iconv_t) (-1)); - ib = d, ibl = dlen, ob = buf1, obl = sizeof (buf1) - m_strlen(tocode); + const char *ib = d; + ssize_t ibl = dlen; + iconv_t cd = mutt_iconv_open(tocode, fromcode, 0); + + assert (cd != (iconv_t)(-1)); + + ob = buf1; + if (my_iconv(cd, &ib, &ibl, &ob, &obl) < 0 || my_iconv(cd, 0, 0, &ob, &obl) < 0) { - assert (errno == E2BIG); - iconv_close (cd); - assert (ib > d); + assert (errno == E2BIG && ib > d); + iconv_close(cd); return (ib - d == dlen) ? dlen : ib - d + 1; } iconv_close (cd); - } - else { - if (dlen > sizeof (buf1) - m_strlen(tocode)) - return sizeof (buf1) - m_strlen(tocode) + 1; - memcpy (buf1, d, dlen); + } else { + if (dlen > obl) + return obl + 1; + memcpy(buf1, d, dlen); ob = buf1 + dlen; } - count = 0; - for (p = buf1; p < ob; p++) { - unsigned char c = *p; + { + const char *p; + int count, len, len_b, len_q; - assert (strchr (MimeSpecials, '?')); - if (c >= 0x7f || c < 0x20 || *p == '_' || - (c != ' ' && strchr (MimeSpecials, *p))) - ++count; - } - - len = ENCWORD_LEN_MIN - 2 + m_strlen(tocode); - len_b = len + (((ob - buf1) + 2) / 3) * 4; - len_q = len + (ob - buf1) + 2 * count; - - /* Apparently RFC 1468 says to use B encoding for iso-2022-jp. */ - if (!ascii_strcasecmp (tocode, "ISO-2022-JP")) - len_q = ENCWORD_LEN_MAX + 1; + count = 0; + for (p = buf1; p < ob; p++) { + count += (*p & 0x80 || __qp_special[(int)*p]); + } - if (len_b < len_q && len_b <= ENCWORD_LEN_MAX) { - *encoder = b_encoder; - *wlen = len_b; - return 0; - } - else if (len_q <= ENCWORD_LEN_MAX) { - *encoder = q_encoder; - *wlen = len_q; - return 0; + len = ENCWORD_LEN_MIN - 2 + m_strlen(tocode); + len_b = len + (((ob - buf1) + 2) / 3) * 4; + len_q = len + (ob - buf1) + 2 * count; + + /* Apparently RFC 1468 says to use B encoding for iso-2022-jp. */ + if (!ascii_strcasecmp(tocode, "ISO-2022-JP")) + len_q = ENCWORD_LEN_MAX + 1; + + if (len_b < len_q && len_b <= ENCWORD_LEN_MAX) { + *encoder = b_encoder; + *wlen = len_b; + return 0; + } else + if (len_q <= ENCWORD_LEN_MAX) { + *encoder = q_encoder; + *wlen = len_q; + return 0; + } else { + return dlen; + } } - else - return dlen; } /* * Encode the data (d, dlen) into s using the encoder. * Return the length of the encoded word. */ -static size_t encode_block (char *s, char *d, size_t dlen, +static size_t encode_block (char *s, char *d, ssize_t dlen, const char *fromcode, const char *tocode, - encoder_t encoder) + encoder_t *encoder) { char buf1[ENCWORD_LEN_MAX - ENCWORD_LEN_MIN + 1]; + ssize_t ibl, obl, n1, n2; iconv_t cd; const char *ib; char *ob; - size_t ibl, obl, n1, n2; if (fromcode) { cd = mutt_iconv_open (tocode, fromcode, 0); @@ -320,12 +344,12 @@ static size_t encode_block (char *s, char *d, size_t dlen, ib = d, ibl = dlen, ob = buf1, obl = sizeof (buf1) - m_strlen(tocode); n1 = my_iconv(cd, &ib, &ibl, &ob, &obl); n2 = my_iconv(cd, 0, 0, &ob, &obl); - assert (n1 != (size_t) (-1) && n2 != (size_t) (-1)); + assert (n1 >= 0 && n2 >= 0); iconv_close (cd); return (*encoder) (s, buf1, ob - buf1, tocode); - } - else + } else { return (*encoder) (s, d, dlen, tocode); + } } /* @@ -334,9 +358,9 @@ static size_t encode_block (char *s, char *d, size_t dlen, * and set the length *wlen of the encoded word and *encoder. * We start in column col, which limits the length of the word. */ -static size_t choose_block (char *d, size_t dlen, int col, - const char *fromcode, const char *tocode, - encoder_t * encoder, size_t * wlen) +static size_t choose_block(char *d, size_t dlen, int col, + const char *fromcode, const char *tocode, + encoder_t **encoder, ssize_t *wlen) { size_t n, nn; int utf8 = fromcode && !ascii_strcasecmp (fromcode, "UTF-8"); @@ -366,17 +390,17 @@ static size_t choose_block (char *d, size_t dlen, int col, * The input data is assumed to be a single line starting at column col; * if col is non-zero, the preceding character was a space. */ -static int rfc2047_encode (const char *d, size_t dlen, int col, +static int rfc2047_encode (const char *d, ssize_t dlen, int col, const char *fromcode, const char *charsets, - char **e, size_t * elen, char *specials) + char **e, ssize_t *elen, const char *specials) { int ret = 0; char *buf; - size_t bufpos, buflen; + ssize_t bufpos, buflen; char *u, *t0, *t1, *t; char *s0, *s1; - size_t ulen, r, n, wlen; - encoder_t encoder; + ssize_t ulen, r, n, wlen; + encoder_t *encoder; char *tocode1 = 0; const char *tocode; const char *icode = "UTF-8"; @@ -420,7 +444,8 @@ static int rfc2047_encode (const char *d, size_t dlen, int col, /* Choose target charset. */ tocode = fromcode; if (icode) { - if ((tocode1 = mutt_choose_charset (icode, charsets, u, ulen, 0, 0))) + if ((tocode1 = mutt_choose_charset(icode, charsets, u, ulen, + NULL, NULL))) tocode = tocode1; else ret = 2, icode = 0; @@ -539,7 +564,7 @@ static int rfc2047_encode (const char *d, size_t dlen, int col, void _rfc2047_encode_string (char **pd, int encode_specials, int col) { char *e; - size_t elen; + ssize_t elen; const char *charsets; if (!Charset || !*pd) @@ -748,7 +773,7 @@ void rfc2047_decode (char **pd) } if (ascii_strcasecmp (AssumedCharset, "us-ascii")) { char *t; - size_t tlen; + ssize_t tlen; t = p_dupstr(s, n); if (mutt_convert_nonmime_string (&t) == 0) { @@ -770,7 +795,7 @@ void rfc2047_decode (char **pd) } if (p != s) { - n = (size_t) (p - s); + n = (p - s); /* ignore spaces between encoded words * and linear white spaces between encoded word and *text */ if (!option (OPTSTRICTMIME)) { @@ -813,32 +838,31 @@ void rfc2047_decode (char **pd) str_adjust (pd); } -void rfc2047_decode_adrlist (address_t * a) +void rfc2047_decode_adrlist(address_t *a) { while (a) { if (a->personal) - rfc2047_decode (&a->personal); + rfc2047_decode(&a->personal); a = a->next; } } -void rfc2047_decode_envelope (ENVELOPE* e) { - - if (!e) - return; +void rfc2047_decode_envelope(ENVELOPE* e) +{ + assert (e); /* do RFC2047 decoding */ - rfc2047_decode_adrlist (e->from); - rfc2047_decode_adrlist (e->to); - rfc2047_decode_adrlist (e->cc); - rfc2047_decode_adrlist (e->bcc); - rfc2047_decode_adrlist (e->reply_to); - rfc2047_decode_adrlist (e->mail_followup_to); - rfc2047_decode_adrlist (e->return_path); - rfc2047_decode_adrlist (e->sender); + rfc2047_decode_adrlist(e->from); + rfc2047_decode_adrlist(e->to); + rfc2047_decode_adrlist(e->cc); + rfc2047_decode_adrlist(e->bcc); + rfc2047_decode_adrlist(e->reply_to); + rfc2047_decode_adrlist(e->mail_followup_to); + rfc2047_decode_adrlist(e->return_path); + rfc2047_decode_adrlist(e->sender); if (e->subject) { - rfc2047_decode (&e->subject); - mutt_adjust_subject (e); + rfc2047_decode(&e->subject); + mutt_adjust_subject(e); } }