X-Git-Url: http://git.madism.org/?p=apps%2Fmadmutt.git;a=blobdiff_plain;f=charset.c;h=11441e4a8f8918fdcd5c48c272f7fd858f33458e;hp=4f5700d6e065557727d9201f701fb09033813ff9;hb=7d29626ce4e1fa932c6349c7253e6f774df069fc;hpb=fa2fd1900f206da4bb975b5ece50ee29f5bd8a75 diff --git a/charset.c b/charset.c index 4f5700d..11441e4 100644 --- a/charset.c +++ b/charset.c @@ -1,3 +1,21 @@ +/* + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or (at + * your option) any later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, + * MA 02110-1301, USA. + * + * Copyright © 2006 Pierre Habouzit + */ /* * Copyright notice from original mutt: * Copyright (C) 1999-2000 Thomas Roessler @@ -7,630 +25,446 @@ * 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 +#ifdef HAVE_LANGINFO_CODESET +# include +#endif #include "mutt.h" #include "charset.h" -#include "ascii.h" - -#include "lib/mem.h" -#include "lib/intl.h" -#include "lib/str.h" #ifndef EILSEQ -# define EILSEQ EINVAL +# define EILSEQ EINVAL #endif -/* - * The following list has been created manually from the data under: - * http://www.isi.edu/in-notes/iana/assignments/character-sets - * Last update: 2000-09-07 - * - * Note that it includes only the subset of character sets for which - * a preferred MIME name is given. - */ - -static struct { - char *key; - char *pref; -} PreferredMIMENames[] = { - { - "ansi_x3.4-1968", "us-ascii"}, { - "iso-ir-6", "us-ascii"}, { - "iso_646.irv:1991", "us-ascii"}, { - "ascii", "us-ascii"}, { - "iso646-us", "us-ascii"}, { - "us", "us-ascii"}, { - "ibm367", "us-ascii"}, { - "cp367", "us-ascii"}, { - "csASCII", "us-ascii"}, { - "csISO2022KR", "iso-2022-kr"}, { - "csEUCKR", "euc-kr"}, { - "csISO2022JP", "iso-2022-jp"}, { - "csISO2022JP2", "iso-2022-jp-2"}, { - "ISO_8859-1:1987", "iso-8859-1"}, { - "iso-ir-100", "iso-8859-1"}, { - "iso_8859-1", "iso-8859-1"}, { - "latin1", "iso-8859-1"}, { - "l1", "iso-8859-1"}, { - "IBM819", "iso-8859-1"}, { - "CP819", "iso-8859-1"}, { - "csISOLatin1", "iso-8859-1"}, { - "ISO_8859-2:1987", "iso-8859-2"}, { - "iso-ir-101", "iso-8859-2"}, { - "iso_8859-2", "iso-8859-2"}, { - "latin2", "iso-8859-2"}, { - "l2", "iso-8859-2"}, { - "csISOLatin2", "iso-8859-2"}, { - "ISO_8859-3:1988", "iso-8859-3"}, { - "iso-ir-109", "iso-8859-3"}, { - "ISO_8859-3", "iso-8859-3"}, { - "latin3", "iso-8859-3"}, { - "l3", "iso-8859-3"}, { - "csISOLatin3", "iso-8859-3"}, { - "ISO_8859-4:1988", "iso-8859-4"}, { - "iso-ir-110", "iso-8859-4"}, { - "ISO_8859-4", "iso-8859-4"}, { - "latin4", "iso-8859-4"}, { - "l4", "iso-8859-4"}, { - "csISOLatin4", "iso-8859-4"}, { - "ISO_8859-6:1987", "iso-8859-6"}, { - "iso-ir-127", "iso-8859-6"}, { - "iso_8859-6", "iso-8859-6"}, { - "ECMA-114", "iso-8859-6"}, { - "ASMO-708", "iso-8859-6"}, { - "arabic", "iso-8859-6"}, { - "csISOLatinArabic", "iso-8859-6"}, { - "ISO_8859-7:1987", "iso-8859-7"}, { - "iso-ir-126", "iso-8859-7"}, { - "ISO_8859-7", "iso-8859-7"}, { - "ELOT_928", "iso-8859-7"}, { - "ECMA-118", "iso-8859-7"}, { - "greek", "iso-8859-7"}, { - "greek8", "iso-8859-7"}, { - "csISOLatinGreek", "iso-8859-7"}, { - "ISO_8859-8:1988", "iso-8859-8"}, { - "iso-ir-138", "iso-8859-8"}, { - "ISO_8859-8", "iso-8859-8"}, { - "hebrew", "iso-8859-8"}, { - "csISOLatinHebrew", "iso-8859-8"}, { - "ISO_8859-5:1988", "iso-8859-5"}, { - "iso-ir-144", "iso-8859-5"}, { - "ISO_8859-5", "iso-8859-5"}, { - "cyrillic", "iso-8859-5"}, { - "csISOLatinCyrillic", "iso8859-5"}, { - "ISO_8859-9:1989", "iso-8859-9"}, { - "iso-ir-148", "iso-8859-9"}, { - "ISO_8859-9", "iso-8859-9"}, { - "latin5", "iso-8859-9"}, /* this is not a bug */ - { - "l5", "iso-8859-9"}, { - "csISOLatin5", "iso-8859-9"}, { - "ISO_8859-10:1992", "iso-8859-10"}, { - "iso-ir-157", "iso-8859-10"}, { - "latin6", "iso-8859-10"}, /* this is not a bug */ - { - "l6", "iso-8859-10"}, { - "csISOLatin6" "iso-8859-10"}, { - "csKOI8r", "koi8-r"}, { - "MS_Kanji", "Shift_JIS"}, /* Note the underscore! */ - { - "csShiftJis", "Shift_JIS"}, { - "Extended_UNIX_Code_Packed_Format_for_Japanese", "EUC-JP"}, { - "csEUCPkdFmtJapanese", "EUC-JP"}, { - "csGB2312", "gb2312"}, { - "csbig5", "big5"}, - /* - * End of official brain damage. What follows has been taken - * from glibc's localedata files. - */ - { - "iso_8859-13", "iso-8859-13"}, { - "iso-ir-179", "iso-8859-13"}, { - "latin7", "iso-8859-13"}, /* this is not a bug */ - { - "l7", "iso-8859-13"}, { - "iso_8859-14", "iso-8859-14"}, { - "latin8", "iso-8859-14"}, /* this is not a bug */ - { - "l8", "iso-8859-14"}, { - "iso_8859-15", "iso-8859-15"}, { - "latin9", "iso-8859-15"}, /* this is not a bug */ - /* Suggested by Ionel Mugurel Ciobica */ - { - "latin0", "iso-8859-15"}, /* this is not a bug */ - { - "iso_8859-16", "iso-8859-16"}, { - "latin10", "iso-8859-16"}, /* this is not a bug */ - /* - * David Champion has observed this with - * nl_langinfo under SunOS 5.8. - */ - { - "646", "us-ascii"}, - /* - * http://www.sun.com/software/white-papers/wp-unicode/ - */ - { - "eucJP", "euc-jp"}, { - "PCK", "Shift_JIS"}, { - "ko_KR-euc", "euc-kr"}, { - "zh_TW-big5", "big5"}, - /* seems to be common on some systems */ - { - "sjis", "Shift_JIS"}, { - "euc-jp-ms", "eucJP-ms"}, - /* - * If you happen to encounter system-specific brain-damage with - * respect to character set naming, please add it above this - * comment, and submit a patch to . - */ - /* End of aliases. Please keep this line last. */ - { - NULL, NULL} -}; +char *Charset; +int Charset_is_utf8 = 0; +wchar_t CharsetReplacement = '?'; -#ifdef HAVE_LANGINFO_CODESET -# include +/****************************************************************************/ +/* charset functions */ +/****************************************************************************/ -void mutt_set_langinfo_charset (void) +void charset_initialize(void) { - char buff[LONG_STRING]; - char buff2[LONG_STRING]; - - strfcpy (buff, nl_langinfo (CODESET), sizeof (buff)); - mutt_canonical_charset (buff2, sizeof (buff2), buff); - - /* finally, set $charset */ - if (!(Charset = str_dup (buff2))) - Charset = str_dup ("iso-8859-1"); -} - -#else +#ifdef HAVE_LANGINFO_CODESET + char buff[STRING]; + char buff2[STRING]; -void mutt_set_langinfo_charset (void) -{ - Charset = str_dup ("iso-8859-1"); -} + m_strcpy(buff, sizeof(buff), nl_langinfo(CODESET)); + charset_canonicalize(buff2, sizeof(buff2), buff); + /* finally, set $charset */ + if (!m_strisempty(buff2)) { + m_strreplace(&Charset, buff2); + } else #endif - -void mutt_canonical_charset (char *dest, size_t dlen, const char *name) -{ - size_t i; - char *p; - char scratch[LONG_STRING]; - - /* catch some common iso-8859-something misspellings */ - if (!ascii_strncasecmp (name, "8859", 4) && name[4] != '-') - snprintf (scratch, sizeof (scratch), "iso-8859-%s", name + 4); - else if (!ascii_strncasecmp (name, "8859-", 5)) - snprintf (scratch, sizeof (scratch), "iso-8859-%s", name + 5); - else if (!ascii_strncasecmp (name, "iso8859", 7) && name[7] != '-') - snprintf (scratch, sizeof (scratch), "iso_8859-%s", name + 7); - else if (!ascii_strncasecmp (name, "iso8859-", 8)) - snprintf (scratch, sizeof (scratch), "iso_8859-%s", name + 8); - else - strfcpy (scratch, NONULL (name), sizeof (scratch)); - - for (i = 0; PreferredMIMENames[i].key; i++) - if (!ascii_strcasecmp (scratch, PreferredMIMENames[i].key) || - !str_casecmp (scratch, PreferredMIMENames[i].key)) { - strfcpy (dest, PreferredMIMENames[i].pref, dlen); - return; + { + m_strreplace(&Charset, "iso-8859-1"); } - strfcpy (dest, scratch, dlen); + Charset_is_utf8 = !m_strcmp(Charset, "utf-8"); + CharsetReplacement = Charset_is_utf8 ? 0xfffd : '?'; - /* for cosmetics' sake, transform to lowercase. */ - for (p = dest; *p; p++) - *p = ascii_tolower (*p); +#ifdef HAVE_BIND_TEXTDOMAIN_CODESET + bind_textdomain_codeset(PACKAGE, Charset); +#endif } -int mutt_chscmp (const char *s, const char *chs) +#include "charset.gperf" +void charset_canonicalize(char *dest, ssize_t dlen, const char *name) { - char buffer[STRING]; - - if (!s) - return 0; + const struct cset_pair *cp; + char scratch[STRING]; + const char *p; + int i = 0; + + if (!name) { + m_strcpy(dest, dlen, "us-ascii"); + return; + } - mutt_canonical_charset (buffer, sizeof (buffer), s); - return !ascii_strcasecmp (buffer, chs); + // canonize name: only keep a-z0-9 and dots, put into lowercase + for (p = name; *p && *p != ':' && i < ssizeof(scratch) - 1; p++) { + if (isalnum(*p) || *p== '.') { + scratch[i++] = tolower((unsigned char)*p); + } + } + scratch[i] = '\0'; + + cp = charset_canonicalize_aux(scratch, strlen(scratch)); + if (cp) { + m_strcpy(dest, dlen, cp->pref); + } else { + m_strcpy(dest, dlen, name); + m_strtolower(dest); + } } +/* XXX: MC: UGLY return of local static */ +const char *charset_getfirst(const char *charset) +{ + static char fcharset[STRING]; + const char *p; -#ifndef HAVE_ICONV + if (m_strisempty(charset)) + return "us-ascii"; -iconv_t iconv_open (const char *tocode, const char *fromcode) -{ - return (iconv_t) (-1); + p = m_strchrnul(charset, ':'); + m_strncpy(fcharset, sizeof(fcharset), charset, p - charset); + return fcharset; } -size_t iconv (iconv_t cd, ICONV_CONST char **inbuf, size_t * inbytesleft, - char **outbuf, size_t * outbytesleft) +int charset_is_utf8(const char *s) { - return 0; + char buf[STRING]; + charset_canonicalize(buf, sizeof(buf), s); + return !m_strcmp(buf, "utf-8"); } -int iconv_close (iconv_t cd) +int charset_is_us_ascii(const char *s) { - return 0; + char buf[STRING]; + charset_canonicalize(buf, sizeof(buf), s); + return !m_strcmp(buf, "us-ascii"); } -#endif /* !HAVE_ICONV */ +/****************************************************************************/ +/* iconv-line functions */ +/****************************************************************************/ -/* - * Like iconv_open, but canonicalises the charsets - */ - -iconv_t mutt_iconv_open (const char *tocode, const char *fromcode, int flags) +/* Like iconv_open, but canonicalises the charsets */ +iconv_t mutt_iconv_open(const char *tocode, const char *fromcode, int flags) { - char tocode1[SHORT_STRING]; - char fromcode1[SHORT_STRING]; - char *tocode2, *fromcode2; - char *tmp; + char tocode1[STRING]; + char fromcode1[STRING]; + const char *tmp; - iconv_t cd; + iconv_t cd; - mutt_canonical_charset (tocode1, sizeof (tocode1), tocode); + if ((flags & M_ICONV_HOOK_TO) && (tmp = mutt_charset_hook(tocode1))) { + charset_canonicalize(tocode1, sizeof(tocode1), tmp); + } else { + charset_canonicalize(tocode1, sizeof(tocode1), tocode); + } -#ifdef M_ICONV_HOOK_TO - /* Not used. */ - if ((flags & M_ICONV_HOOK_TO) && (tmp = mutt_charset_hook (tocode1))) - mutt_canonical_charset (tocode1, sizeof (tocode1), tmp); -#endif + if ((flags & M_ICONV_HOOK_FROM) && (tmp = mutt_charset_hook(fromcode1))) { + charset_canonicalize(fromcode1, sizeof(fromcode1), tmp); + } else { + charset_canonicalize(fromcode1, sizeof(fromcode1), fromcode); + } - mutt_canonical_charset (fromcode1, sizeof (fromcode1), fromcode); - if ((flags & M_ICONV_HOOK_FROM) && (tmp = mutt_charset_hook (fromcode1))) - mutt_canonical_charset (fromcode1, sizeof (fromcode1), tmp); + cd = iconv_open(tocode1, fromcode1); + if (cd != MUTT_ICONV_ERROR) + return cd; - if ((cd = iconv_open (tocode1, fromcode1)) != (iconv_t) - 1) - return cd; - if ((tocode2 = mutt_iconv_hook (tocode1)) - && (fromcode2 = mutt_iconv_hook (fromcode1))) - return iconv_open (tocode2, fromcode2); + { + const char *to = mutt_iconv_hook(tocode1); + const char *from = mutt_iconv_hook(fromcode1); - return (iconv_t) - 1; + return to && from ? iconv_open(to, from) : MUTT_ICONV_ERROR; + } } -/* - * Like iconv, but keeps going even when the input is invalid - * If you're supplying inrepls, the source charset should be stateless; - * if you're supplying an outrepl, the target charset should be. - */ - -size_t mutt_iconv (iconv_t cd, ICONV_CONST char **inbuf, size_t * inbytesleft, - char **outbuf, size_t * outbytesleft, - ICONV_CONST char **inrepls, const char *outrepl) +/* Like iconv, but keeps going even when the input is invalid + If you're supplying inrepls, the source charset should be stateless; + if you're supplying an outrepl, the target charset should be. */ +/* XXX: MC: I do not understand what it does yet */ +ssize_t mutt_iconv(iconv_t cd, + const char **inbuf, ssize_t *inbytesleft, + char **outbuf, ssize_t *outbytesleft, + const char **inrepls, const char *outrepl) { - size_t ret = 0, ret1; - ICONV_CONST char *ib = *inbuf; - size_t ibl = *inbytesleft; - char *ob = *outbuf; - size_t obl = *outbytesleft; - - for (;;) { - ret1 = iconv (cd, &ib, &ibl, &ob, &obl); - if (ret1 != (size_t) - 1) - ret += ret1; - if (ibl && obl && errno == EILSEQ) { - if (inrepls) { - /* Try replacing the input */ - ICONV_CONST char **t; - - for (t = inrepls; *t; t++) { - ICONV_CONST char *ib1 = *t; - size_t ibl1 = str_len (*t); - char *ob1 = ob; - size_t obl1 = obl; - - iconv (cd, &ib1, &ibl1, &ob1, &obl1); - if (!ibl1) { - ++ib, --ibl; - ob = ob1, obl = obl1; - ++ret; - break; - } - } - if (*t) - continue; - } - /* Replace the output */ - if (!outrepl) - outrepl = "?"; - iconv (cd, 0, 0, &ob, &obl); - if (obl) { - int n = str_len (outrepl); - - if (n > obl) { - outrepl = "?"; - n = 1; + ssize_t ret = 0, ret1; + const char *ib = *inbuf; + ssize_t ibl = *inbytesleft; + char *ob = *outbuf; + ssize_t obl = *outbytesleft; + + for (;;) { + ret1 = my_iconv(cd, &ib, &ibl, &ob, &obl); + if (ret1 != -1) + ret += ret1; + + if (ibl && obl && errno == EILSEQ) { + if (inrepls) { + /* Try replacing the input */ + const char **t; + + for (t = inrepls; *t; t++) { + const char *ib1 = *t; + ssize_t ibl1 = m_strlen(*t); + char *ob1 = ob; + ssize_t obl1 = obl; + + my_iconv(cd, &ib1, &ibl1, &ob1, &obl1); + if (!ibl1) { + ++ib, --ibl; + ob = ob1, obl = obl1; + ++ret; + break; + } + } + if (*t) + continue; + } + /* Replace the output */ + if (!outrepl) + outrepl = "?"; + my_iconv(cd, 0, 0, &ob, &obl); + if (obl) { + ssize_t n = m_strlen(outrepl); + + if (n > obl) { + outrepl = "?"; + n = 1; + } + memcpy(ob, outrepl, n); + ++ib, --ibl; + ob += n, obl -= n; + ++ret; + my_iconv(cd, 0, 0, 0, 0); /* for good measure */ + continue; + } } - memcpy (ob, outrepl, n); - ++ib, --ibl; - ob += n, obl -= n; - ++ret; - iconv (cd, 0, 0, 0, 0); /* for good measure */ - continue; - } + *inbuf = ib, *inbytesleft = ibl; + *outbuf = ob, *outbytesleft = obl; + return ret; } - *inbuf = ib, *inbytesleft = ibl; - *outbuf = ob, *outbytesleft = obl; - return ret; - } } - -/* - * Convert a string - * Used in rfc2047.c and rfc2231.c - */ - -int mutt_convert_string (char **ps, const char *from, const char *to, - int flags) +/* Convert a string */ +int +mutt_convert_string(char **ps, const char *from, const char *to, int flags) { - iconv_t cd; - ICONV_CONST char *repls[] = { "\357\277\275", "?", 0 }; - char *s = *ps; + iconv_t cd; + const char *repls[] = { "\357\277\275", "?", 0 }; + + if (m_strisempty(*ps)) + return 0; + + cd = mutt_iconv_open(to, from, flags); + if (cd != MUTT_ICONV_ERROR) { + const char **inrepls = NULL; + const char *outrepl = NULL; + const char *ib; + char *buf, *ob; + ssize_t ibl, obl; + + if (charset_is_utf8(to)) + outrepl = "\357\277\275"; + else + if (charset_is_utf8(from)) + inrepls = repls; + else + outrepl = "?"; + + ibl = m_strlen(*ps) + 1; + ib = *ps; + + obl = MB_LEN_MAX * ibl; + ob = buf = p_new(char, obl + 1); + + mutt_iconv(cd, &ib, &ibl, &ob, &obl, inrepls, outrepl); + iconv_close(cd); + + *ob = '\0'; + + p_delete(ps); + *ps = buf; + return 0; + } - if (!s || !*s) - return 0; + return -1; +} - if (to && from && (cd = mutt_iconv_open (to, from, flags)) != (iconv_t) - 1) { - int len; - ICONV_CONST char *ib; +static ssize_t convert_string(const char *f, ssize_t flen, + const char *from, const char *to, + char **t, ssize_t * tlen) +{ + iconv_t cd; char *buf, *ob; - size_t ibl, obl; - ICONV_CONST char **inrepls = 0; - char *outrepl = 0; - - if (mutt_is_utf8 (to)) - outrepl = "\357\277\275"; - else if (mutt_is_utf8 (from)) - inrepls = repls; - else - outrepl = "?"; - - len = str_len (s); - ib = s, ibl = len + 1; - obl = MB_LEN_MAX * ibl; - ob = buf = mem_malloc (obl + 1); - - mutt_iconv (cd, &ib, &ibl, &ob, &obl, inrepls, outrepl); - iconv_close (cd); - - *ob = '\0'; - - mem_free (ps); - *ps = buf; - - str_adjust (ps); - return 0; - } - else - return -1; + ssize_t obl; + ssize_t n; + int e; + + if ((cd = mutt_iconv_open(to, from, 0)) == MUTT_ICONV_ERROR) + return -1; + + obl = 4 * flen + 1; + ob = buf = p_new(char, obl); + 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); + errno = e; + return -1; + } + + *ob = '\0'; + *tlen = ob - buf; + *t = buf; + iconv_close(cd); + return n; } +int mutt_convert_nonmime_string(char **ps) +{ + const char *p = AssumedCharset; + ssize_t ulen = m_strlen(*ps); + char *u = *ps; -/* - * FGETCONV stuff for converting a file while reading it - * Used in sendlib.c for converting from mutt's Charset - */ + while (*p) { + const char *q; + char fromcode[LONG_STRING], *s = NULL; + ssize_t slen; -struct fgetconv_s { - FILE *file; - iconv_t cd; - char bufi[512]; - char bufo[512]; - char *p; - char *ob; - char *ib; - size_t ibl; - ICONV_CONST char **inrepls; -}; + if (!ulen) + return 0; + + while (*p == ':') + *p++; + + q = m_strchrnul(p, ':'); + m_strncpy(fromcode, sizeof(fromcode), p, q - p); + p = q; -struct fgetconv_not { - FILE *file; - iconv_t cd; + if (convert_string(u, ulen, fromcode, Charset, &s, &slen) >= 0) { + p_delete(ps); + *ps = s; + return 0; + } + } + + return -1; +} + +/****************************************************************************/ +/* fgetconv functions */ +/****************************************************************************/ + +/* fgetconv_t stuff for converting a file while reading it + Used in sendlib.c for converting from mutt's Charset */ + +struct fgetconv_t { + FILE *file; + iconv_t cd; + char bufi[BUFSIZ]; + char bufo[BUFSIZ]; + char *p; + char *ob; + char *ib; + ssize_t ibl; + const char **inrepls; }; -FGETCONV *fgetconv_open (FILE * file, const char *from, const char *to, - int flags) +fgetconv_t * +fgetconv_open(FILE *file, const char *from, const char *to, int flags) { - struct fgetconv_s *fc; - iconv_t cd = (iconv_t) - 1; - static ICONV_CONST char *repls[] = { "\357\277\275", "?", 0 }; + static const char *repls[] = { "\357\277\275", "?", 0 }; - if (from && to) - cd = mutt_iconv_open (to, from, flags); + struct fgetconv_t *fc = p_new(struct fgetconv_t, 1); - if (cd != (iconv_t) - 1) { - fc = mem_malloc (sizeof (struct fgetconv_s)); - fc->p = fc->ob = fc->bufo; - fc->ib = fc->bufi; - fc->ibl = 0; - fc->inrepls = mutt_is_utf8 (to) ? repls : repls + 1; - } - else - fc = mem_malloc (sizeof (struct fgetconv_not)); - fc->file = file; - fc->cd = cd; - return (FGETCONV *) fc; + fc->file = file; + fc->cd = MUTT_ICONV_ERROR; + if (from && to) + fc->cd = mutt_iconv_open(to, from, flags); + + if (fc->cd != MUTT_ICONV_ERROR) { + fc->p = fc->ob = fc->bufo; + fc->ib = fc->bufi; + fc->ibl = 0; + fc->inrepls = repls + charset_is_utf8(to); + } + return fc; } -char *fgetconvs (char *buf, size_t l, FGETCONV * _fc) +void fgetconv_close(fgetconv_t **fcp) { - int c; - size_t r; - - for (r = 0; r + 1 < l;) { - if ((c = fgetconv (_fc)) == EOF) - break; - buf[r++] = (char) c; - if (c == '\n') - break; - } - buf[r] = '\0'; - - if (r) - return buf; - else - return NULL; + struct fgetconv_t *fc = *fcp; + + if (fc->cd != MUTT_ICONV_ERROR) + iconv_close (fc->cd); + p_delete(fcp); } -int fgetconv (FGETCONV * _fc) -{ - struct fgetconv_s *fc = (struct fgetconv_s *) _fc; - if (!fc) - return EOF; - if (fc->cd == (iconv_t) - 1) - return fgetc (fc->file); - if (!fc->p) - return EOF; - if (fc->p < fc->ob) - return (unsigned char) *(fc->p)++; +int fgetconv(fgetconv_t *fc) +{ + if (!fc) + return EOF; - /* Try to convert some more */ - fc->p = fc->ob = fc->bufo; - if (fc->ibl) { - size_t obl = sizeof (fc->bufo); + if (fc->cd == MUTT_ICONV_ERROR) + return fgetc(fc->file); - iconv (fc->cd, (ICONV_CONST char **) &fc->ib, &fc->ibl, &fc->ob, &obl); + if (!fc->p) + return EOF; if (fc->p < fc->ob) - return (unsigned char) *(fc->p)++; - } + return (unsigned char)*(fc->p)++; - /* If we trusted iconv a bit more, we would at this point - * ask why it had stopped converting ... */ + /* Try to convert some more */ + fc->p = fc->ob = fc->bufo; + if (fc->ibl) { + ssize_t obl = ssizeof(fc->bufo); - /* Try to read some more */ - if (fc->ibl == sizeof (fc->bufi) || - (fc->ibl && fc->ib + fc->ibl < fc->bufi + sizeof (fc->bufi))) { - fc->p = 0; - return EOF; - } - if (fc->ibl) - memcpy (fc->bufi, fc->ib, fc->ibl); - fc->ib = fc->bufi; - fc->ibl += - fread (fc->ib + fc->ibl, 1, sizeof (fc->bufi) - fc->ibl, fc->file); - - /* Try harder this time to convert some */ - if (fc->ibl) { - size_t obl = sizeof (fc->bufo); - - mutt_iconv (fc->cd, (ICONV_CONST char **) &fc->ib, &fc->ibl, &fc->ob, - &obl, fc->inrepls, 0); - if (fc->p < fc->ob) - return (unsigned char) *(fc->p)++; - } + my_iconv(fc->cd, (const char **)&fc->ib, &fc->ibl, &fc->ob, &obl); + if (fc->p < fc->ob) + return (unsigned char)*(fc->p)++; + } - /* Either the file has finished or one of the buffers is too small */ - fc->p = 0; - return EOF; -} + /* If we trusted iconv a bit more, we would at this point + * ask why it had stopped converting ... */ -void fgetconv_close (FGETCONV ** _fc) -{ - struct fgetconv_s *fc = (struct fgetconv_s *) *_fc; + /* Try to read some more */ + if (fc->ibl == sizeof(fc->bufi) + || (fc->ibl && fc->ib + fc->ibl < fc->bufi + sizeof(fc->bufi))) { + fc->p = NULL; + return EOF; + } - if (fc->cd != (iconv_t) - 1) - iconv_close (fc->cd); - mem_free (_fc); -} + if (fc->ibl) { + memcpy(fc->bufi, fc->ib, fc->ibl); + } + fc->ib = fc->bufi; + fc->ibl += fread(fc->ib + fc->ibl, 1, sizeof(fc->bufi) - fc->ibl, + fc->file); -char *mutt_get_first_charset (const char *charset) -{ - static char fcharset[SHORT_STRING]; - const char *c, *c1; - - c = charset; - if (!str_len (c)) - return "us-ascii"; - if (!(c1 = strchr (c, ':'))) - return ((char*) charset); - strfcpy (fcharset, c, c1 - c + 1); - return fcharset; -} + /* Try harder this time to convert some */ + if (fc->ibl) { + ssize_t obl = ssizeof(fc->bufo); -static size_t convert_string (ICONV_CONST char *f, size_t flen, - const char *from, const char *to, - char **t, size_t * tlen) -{ - iconv_t cd; - char *buf, *ob; - size_t obl, n; - int e; - - cd = mutt_iconv_open (to, from, 0); - if (cd == (iconv_t) (-1)) - return (size_t) (-1); - obl = 4 * flen + 1; - ob = buf = mem_malloc (obl); - n = iconv (cd, &f, &flen, &ob, &obl); - if (n == (size_t) (-1) || iconv (cd, 0, 0, &ob, &obl) == (size_t) (-1)) { - e = errno; - mem_free (&buf); - iconv_close (cd); - errno = e; - return (size_t) (-1); - } - *ob = '\0'; - - *tlen = ob - buf; - - mem_realloc (&buf, ob - buf + 1); - *t = buf; - iconv_close (cd); - - return n; + mutt_iconv(fc->cd, (const char **)&fc->ib, &fc->ibl, &fc->ob, &obl, + fc->inrepls, 0); + if (fc->p < fc->ob) { + return (unsigned char)*(fc->p)++; + } + } + + /* Either the file has finished or one of the buffers is too small */ + fc->p = NULL; + return EOF; } -int mutt_convert_nonmime_string (char **ps) +char *fgetconvs(char *buf, ssize_t len, fgetconv_t *fc) { - const char *c, *c1; + ssize_t pos = 0; - for (c = AssumedCharset; c; c = c1 ? c1 + 1 : 0) { - char *u = *ps; - char *s = NULL; - char *fromcode; - size_t m, n; - size_t ulen = str_len (*ps); - size_t slen; - - if (!u || !*u) - return 0; - - c1 = strchr (c, ':'); - n = c1 ? c1 - c : str_len (c); - if (!n) - continue; - fromcode = mem_malloc (n + 1); - strfcpy (fromcode, c, n + 1); - m = convert_string (u, ulen, fromcode, Charset, &s, &slen); - mem_free (&fromcode); - if (m != (size_t) (-1)) { - mem_free (ps); - *ps = s; - return 0; + while (pos < len - 1) { + int c = fgetconv(fc); + if (c == EOF) + break; + + buf[pos++] = c; + if (c == '\n') + break; } - } - return -1; + buf[pos] = '\0'; + + return pos ? buf : NULL; }