parse.c

mutt-1.5.12 源代码。linux 下邮件接受的工具。

/*
 * Copyright (C) 1996-2000 Michael R. Elkins <[e-mail]>
 * 
 *     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.
 */ 

#if HAVE_CONFIG_H
# include "config.h"
#endif

#include "mutt.h"
#include "mutt_regex.h"
#include "mailbox.h"
#include "mime.h"
#include "rfc2047.h"
#include "rfc2231.h"
#include "mutt_crypt.h"
#include "url.h"

#include <string.h>
#include <ctype.h>
#include <sys/stat.h>
#include <stdlib.h>

/* Reads an arbitrarily long header field, and looks ahead for continuation
 * lines.  ``line'' must point to a dynamically allocated string; it is
 * increased if more space is required to fit the whole line.
 */
char *mutt_read_rfc822_line (FILE *f, char *line, size_t *linelen)
{
  char *buf = line;
  char ch;
  size_t offset = 0;

  FOREVER
  {
    if (fgets (buf, *linelen - offset, f) == NULL ||	/* end of file or */
	(ISSPACE (*line) && !offset))			/* end of headers */ 
    {
      *line = 0;
      return (line);
    }

    buf += strlen (buf) - 1;
    if (*buf == '\n')
    {
      /* we did get a full line. remove trailing space */
      while (ISSPACE (*buf))
	*buf-- = 0;	/* we cannot come beyond line's beginning because
			 * it begins with a non-space */

      /* check to see if the next line is a continuation line */
      if ((ch = fgetc (f)) != ' ' && ch != '\t')
      {
	ungetc (ch, f);
	return (line); /* next line is a separate header field or EOH */
      }

      /* eat tabs and spaces from the beginning of the continuation line */
      while ((ch = fgetc (f)) == ' ' || ch == '\t')
	;
      ungetc (ch, f);
      *++buf = ' '; /* string is still terminated because we removed
		       at least one whitespace char above */
    }

    buf++;
    offset = buf - line;
    if (*linelen < offset + STRING)
    {
      /* grow the buffer */
      *linelen += STRING;
      safe_realloc (&line, *linelen);
      buf = line + offset;
    }
  }
  /* not reached */
}

static LIST *mutt_parse_references (char *s, int in_reply_to)
{
  LIST *t, *lst = NULL;
  int m, n = 0;
  char *o = NULL, *new, *at;

  while ((s = strtok (s, " \t;")) != NULL)
  {
    /*
     * some mail clients add other garbage besides message-ids, so do a quick
     * check to make sure this looks like a valid message-id
     * some idiotic clients also break their message-ids between lines, deal
     * with that too (give up if it's more than two lines, though)
     */
    t = NULL;
    new = NULL;

    if (*s == '<')
    {
      n = strlen (s);
      if (s[n-1] != '>')
      {
	o = s;
	s = NULL;
	continue;
      }

      new = safe_strdup (s);
    }
    else if (o)
    {
      m = strlen (s);
      if (s[m - 1] == '>')
      {
	new = safe_malloc (sizeof (char) * (n + m + 1));
	strcpy (new, o);	/* __STRCPY_CHECKED__ */
	strcpy (new + n, s);	/* __STRCPY_CHECKED__ */
      }
    }
    if (new)
    {
      /* make sure that this really does look like a message-id.
       * it should have exactly one @, and if we're looking at
       * an in-reply-to header, make sure that the part before
       * the @ has more than eight characters or it's probably
       * an email address
       */
      if (!(at = strchr (new, '@')) || strchr (at + 1, '@')
	  || (in_reply_to && at - new <= 8))
	FREE (&new);
      else
      {
	t = (LIST *) safe_malloc (sizeof (LIST));
	t->data = new;
	t->next = lst;
	lst = t;
      }
    }
    o = NULL;
    s = NULL;
  }

  return (lst);
}

int mutt_check_encoding (const char *c)
{
  if (ascii_strncasecmp ("7bit", c, sizeof ("7bit")-1) == 0)
    return (ENC7BIT);
  else if (ascii_strncasecmp ("8bit", c, sizeof ("8bit")-1) == 0)
    return (ENC8BIT);
  else if (ascii_strncasecmp ("binary", c, sizeof ("binary")-1) == 0)
    return (ENCBINARY);
  else if (ascii_strncasecmp ("quoted-printable", c, sizeof ("quoted-printable")-1) == 0)
    return (ENCQUOTEDPRINTABLE);
  else if (ascii_strncasecmp ("base64", c, sizeof("base64")-1) == 0)
    return (ENCBASE64);
  else if (ascii_strncasecmp ("x-uuencode", c, sizeof("x-uuencode")-1) == 0)
    return (ENCUUENCODED);
#ifdef SUN_ATTACHMENT
  else if (ascii_strncasecmp ("uuencode", c, sizeof("uuencode")-1) == 0)
    return (ENCUUENCODED);
#endif
  else
    return (ENCOTHER);
}

static PARAMETER *parse_parameters (const char *s)
{
  PARAMETER *head = 0, *cur = 0, *new;
  char buffer[LONG_STRING];
  const char *p;
  size_t i;

  dprint (2, (debugfile, "parse_parameters: `%s'\n", s));
  
  while (*s)
  {
    if ((p = strpbrk (s, "=;")) == NULL)
    {
      dprint(1, (debugfile, "parse_parameters: malformed parameter: %s\n", s));
      goto bail;
    }

    /* if we hit a ; now the parameter has no value, just skip it */
    if (*p != ';')
    {
      i = p - s;

      new = mutt_new_parameter ();

      new->attribute = safe_malloc (i + 1);
      memcpy (new->attribute, s, i);
      new->attribute[i] = 0;

      /* remove whitespace from the end of the attribute name */
      while (ISSPACE (new->attribute[--i]))
	new->attribute[i] = 0;

      s = p + 1; /* skip over the = */
      SKIPWS (s);

      if (*s == '"')
      {
	s++;
	for (i=0; *s && *s != '"' && i < sizeof (buffer) - 1; i++, s++)
	{
	  if (*s == '\\')
	  {
	    /* Quote the next character */
	    buffer[i] = s[1];
	    if (!*++s)
	      break;
	  }
	  else
	    buffer[i] = *s;
	}
	buffer[i] = 0;
	if (*s)
	  s++; /* skip over the " */
      }
      else
      {
	for (i=0; *s && *s != ' ' && *s != ';' && i < sizeof (buffer) - 1; i++, s++)
	  buffer[i] = *s;
	buffer[i] = 0;
      }

      new->value = safe_strdup (buffer);

      dprint (2, (debugfile, "parse_parameter: `%s' = `%s'\n",
		  new->attribute ? new->attribute : "",
		  new->value ? new->value : ""));
      
      /* Add this parameter to the list */
      if (head)
      {
	cur->next = new;
	cur = cur->next;
      }
      else
	head = cur = new;
    }
    else
    {
      dprint (1, (debugfile, "parse_parameters(): parameter with no value: %s\n", s));
      s = p;
    }

    /* Find the next parameter */
    if (*s != ';' && (s = strchr (s, ';')) == NULL)
	break; /* no more parameters */

    do
    {
      s++;

      /* Move past any leading whitespace */
      SKIPWS (s);
    }
    while (*s == ';'); /* skip empty parameters */
  }    

  bail:

  rfc2231_decode_parameters (&head);
  return (head);
}

int mutt_check_mime_type (const char *s)
{
  if (ascii_strcasecmp ("text", s) == 0)
    return TYPETEXT;
  else if (ascii_strcasecmp ("multipart", s) == 0)
    return TYPEMULTIPART;
#ifdef SUN_ATTACHMENT 
  else if (ascii_strcasecmp ("x-sun-attachment", s) == 0)
    return TYPEMULTIPART;
#endif
  else if (ascii_strcasecmp ("application", s) == 0)
    return TYPEAPPLICATION;
  else if (ascii_strcasecmp ("message", s) == 0)
    return TYPEMESSAGE;
  else if (ascii_strcasecmp ("image", s) == 0)
    return TYPEIMAGE;
  else if (ascii_strcasecmp ("audio", s) == 0)
    return TYPEAUDIO;
  else if (ascii_strcasecmp ("video", s) == 0)
    return TYPEVIDEO;
  else if (ascii_strcasecmp ("model", s) == 0)
    return TYPEMODEL;
  else if (ascii_strcasecmp ("*", s) == 0)
    return TYPEANY;
  else if (ascii_strcasecmp (".*", s) == 0)
    return TYPEANY;
  else
    return TYPEOTHER;
}

void mutt_parse_content_type (char *s, BODY *ct)
{
  char *pc;
  char *subtype;

  FREE (&ct->subtype);
  mutt_free_parameter(&ct->parameter);

  /* First extract any existing parameters */
  if ((pc = strchr(s, ';')) != NULL)
  {
    *pc++ = 0;
    while (*pc && ISSPACE (*pc))
      pc++;
    ct->parameter = parse_parameters(pc);

    /* 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 */
    if ((pc = mutt_get_parameter( "name", ct->parameter)) != 0 && !ct->filename)
      ct->filename = safe_strdup(pc);
    
#ifdef SUN_ATTACHMENT
    /* this is deep and utter perversion */
    if ((pc = mutt_get_parameter ("conversions", ct->parameter)) != 0)
      ct->encoding = mutt_check_encoding (pc);
#endif
    
  }
  
  /* Now get the subtype */
  if ((subtype = strchr(s, '/')))
  {
    *subtype++ = '\0';
    for(pc = subtype; *pc && !ISSPACE(*pc) && *pc != ';'; pc++)
      ;
    *pc = '\0';
    ct->subtype = safe_strdup (subtype);
  }

  /* Finally, get the major type */
  ct->type = mutt_check_mime_type (s);

#ifdef SUN_ATTACHMENT
  if (ascii_strcasecmp ("x-sun-attachment", s) == 0)
      ct->subtype = safe_strdup ("x-sun-attachment");
#endif

  if (ct->type == TYPEOTHER)
  {
    ct->xtype = safe_strdup (s);
  }

  if (ct->subtype == NULL)
  {
    /* Some older non-MIME mailers (i.e., mailtool, elm) have a content-type
     * field, so we can attempt to convert the type to BODY here.
     */
    if (ct->type == TYPETEXT)
      ct->subtype = safe_strdup ("plain");
    else if (ct->type == TYPEAUDIO)
      ct->subtype = safe_strdup ("basic");
    else if (ct->type == TYPEMESSAGE)
      ct->subtype = safe_strdup ("rfc822");
    else if (ct->type == TYPEOTHER)
    {
      char buffer[SHORT_STRING];

      ct->type = TYPEAPPLICATION;
      snprintf (buffer, sizeof (buffer), "x-%s", s);
      ct->subtype = safe_strdup (buffer);
    }
    else
      ct->subtype = safe_strdup ("x-unknown");
  }

  /* Default character set for text types. */
  if (ct->type == TYPETEXT)
  {
    if (!(pc = mutt_get_parameter ("charset", ct->parameter)))
      mutt_set_parameter ("charset", "us-ascii", &ct->parameter);
  }

}

static void parse_content_disposition (char *s, BODY *ct)
{
  PARAMETER *parms;

  if (!ascii_strncasecmp ("inline", s, 6))
    ct->disposition = DISPINLINE;
  else if (!ascii_strncasecmp ("form-data", s, 9))
    ct->disposition = DISPFORMDATA;
  else
    ct->disposition = DISPATTACH;

  /* Check to see if a default filename was given */
  if ((s = strchr (s, ';')) != NULL)
  {
    s++;
    SKIPWS (s);
    if ((s = mutt_get_parameter ("filename", (parms = parse_parameters (s)))) != 0)
      mutt_str_replace (&ct->filename, s);
    if ((s = mutt_get_parameter ("name", parms)) != 0)
      ct->form_name = safe_strdup (s);
    mutt_free_parameter (&parms);
  }
}

/* args:
 *	fp	stream to read from
 *
 *	digest	1 if reading subparts of a multipart/digest, 0
 *		otherwise
 */

BODY *mutt_read_mime_header (FILE *fp, int digest)
{
  BODY *p = mutt_new_body();
  char *c;
  char *line = safe_malloc (LONG_STRING);
  size_t linelen = LONG_STRING;
  
  p->hdr_offset  = ftello (fp);

  p->encoding    = ENC7BIT; /* default from RFC1521 */
  p->type        = digest ? TYPEMESSAGE : TYPETEXT;
  p->disposition = DISPINLINE;
  
  while (*(line = mutt_read_rfc822_line (fp, line, &linelen)) != 0)
  {
    /* Find the value of the current header */
    if ((c = strchr (line, ':')))
    {
      *c = 0;
      c++;
      SKIPWS (c);
      if (!*c)
      {
	dprint (1, (debugfile, "mutt_read_mime_header(): skipping empty header field: %s\n", line));
	continue;
      }
    }
    else
    {
      dprint (1, (debugfile, "read_mime_header: bogus MIME header: %s\n", line));
      break;
    }

    if (!ascii_strncasecmp ("content-", line, 8))
    {
      if (!ascii_strcasecmp ("type", line + 8))
	mutt_parse_content_type (c, p);
      else if (!ascii_strcasecmp ("transfer-encoding", line + 8))
	p->encoding = mutt_check_encoding (c);
      else if (!ascii_strcasecmp ("disposition", line + 8))
	parse_content_disposition (c, p);
      else if (!ascii_strcasecmp ("description", line + 8))
      {
	mutt_str_replace (&p->description, c);
	rfc2047_decode (&p->description);
      }
    } 
#ifdef SUN_ATTACHMENT
    else if (!ascii_strncasecmp ("x-sun-", line, 6))
    {
      if (!ascii_strcasecmp ("data-type", line + 6))
        mutt_parse_content_type (c, p);
      else if (!ascii_strcasecmp ("encoding-info", line + 6))
        p->encoding = mutt_check_encoding (c);
      else if (!ascii_strcasecmp ("content-lines", line + 6))
        mutt_set_parameter ("content-lines", c, &(p->parameter));
      else if (!ascii_strcasecmp ("data-description", line + 6))
      {
	mutt_str_replace (&p->description, c);
        rfc2047_decode (&p->description);
      }
    }
#endif
  }
  p->offset = ftello (fp); /* Mark the start of the real data */
  if (p->type == TYPETEXT && !p->subtype)
    p->subtype = safe_strdup ("plain");
  else if (p->type == TYPEMESSAGE && !p->subtype)
    p->subtype = safe_strdup ("rfc822");

  FREE (&line);

  return (p);
}

void mutt_parse_part (FILE *fp, BODY *b)
{
  char *bound = 0;

  switch (b->type)
  {
    case TYPEMULTIPART:
#ifdef SUN_ATTACHMENT
      if ( !ascii_strcasecmp (b->subtype, "x-sun-attachment") )
          bound = "--------";
      else
#endif
          bound = mutt_get_parameter ("boundary", b->parameter);

      fseeko (fp, b->offset, SEEK_SET);
      b->parts =  mutt_parse_multipart (fp, bound, 
					b->offset + b->length,
					ascii_strcasecmp ("digest", b->subtype) == 0);
      break;

    case TYPEMESSAGE:
      if (b->subtype)
      {
	fseeko (fp, b->offset, SEEK_SET);
	if (mutt_is_message_type(b->type, b->subtype))
	  b->parts = mutt_parse_messageRFC822 (fp, b);
	else if (ascii_strcasecmp (b->subtype, "external-body") == 0)
	  b->parts = mutt_read_mime_header (fp, 0);
	else
	  return;
      }
      break;

    default:
      return;
  }

  /* try to recover from parsing error */
  if (!b->parts)
  {
    b->type = TYPETEXT;