uuscan.c

来自「UUDeview是一个编码解码器」· C语言 代码 · 共 2,492 行 · 第 1/5 页

      }
    }
    variable = NULL;
  }
  else {
    /*
     * nothing interesting
     */
    return theheaders;
  }

  /*
   * okay, so extract the actual data
   */
  if (variable) {
    length = 0;
    ptr = uuscan_phtext;

    while (isspace (*value))
      value++;
    while (*value && (delimit==0 || *value!=delimit) &&
	   *value != '\012' && *value != '\015' && length < 255) {
      *ptr++ = *value++;
      length++;
    }
    while (length && isspace(*(ptr-1))) {
      ptr--; length--;
    }
    *ptr = '\0';

    if ((*variable = _FP_strdup (uuscan_phtext)) == NULL)
      return NULL;
  }

  return theheaders;
}

/*
 * is this a header line we know about?
 */

static int
IsKnownHeader (char *line)
{
  char **iter = knownmsgheaders;

  while (iter && *iter) {
    if (_FP_strnicmp (line, *iter, strlen (*iter)) == 0)
      return 1;
    iter++;
  }

  iter = knownmimeheaders;

  while (iter && *iter) {
    if (_FP_strnicmp (line, *iter, strlen (*iter)) == 0)
      return 2;
    iter++;
  }

  return 0;
}

/*
 * Scan a header
 */

int
UUScanHeader (FILE *datei, headers *envelope)
{
  char *ptr;

  while (!feof (datei)) {
    if ((ptr = ScanHeaderLine (datei, NULL)) == NULL)
      break;
    if (*ptr == '\0' || *ptr == '\012' || *ptr == '\015')
      break;
    ParseHeader (envelope, ptr);
  }
  return 0;
}

/*
 * Scan something for encoded data and fill the fileread* structure.
 * If boundary is non-NULL, we stop after having read it. If Check-
 * Headers != 0, we also stop after we've found uu_headercount recog-
 * nized header lines.
 * If we have a boundary, then we also don't accept Base64; MIME mails
 * really should handle this properly.
 * We return -1 if something went wrong, 0 if everything is normal,
 * 1 if we found a new header and 2 if we found a boundary.
 * In MIME message bodies (not multiparts), we also disable our reduced
 * MIME handling.
 */

static int
ScanData (FILE *datei, char *fname, int *errcode,
	  char *boundary, int ismime, int checkheaders,
	  fileread *result)
{
  char *line=uuscan_sdline, *bhds1=uuscan_sdbhds1, *bhds2=uuscan_sdbhds2;
  static char *ptr, *p2, *p3=NULL, *bhdsp, bhl;
  int islen[10], isb64[10], isuue[10], isxxe[10], isbhx[10], iscnt;
  int cbb64, cbuue, cbxxe, cbbhx;
  int bhflag=0, vflag, haddh=0, hadct=0;
  int bhrpc=0, bhnf=0, c, hcount, lcount, blen=0;
  int encoding=0, dflag=0, ctline=42;
  int dontcare=0, hadnl=0;
  long preheaders=0, oldposition;
  long yefilesize=0, yepartends=0;
  size_t dcc, bhopc;

  *errcode = UURET_OK;
  (void) UUDecodeLine (NULL, NULL, 0);          /* init */
  bhdsp = bhds2;

  if (datei == NULL || feof (datei))
    return -1;

  result->startpos = ftell (datei);
  hcount = lcount  = 0;

  for (iscnt=0; iscnt<10; iscnt++) {
    isb64[iscnt] = isuue[iscnt] = isxxe[iscnt] = isbhx[iscnt] = 0;
    islen[iscnt] = -1;
  }

  iscnt = 0;

  if (boundary)
    blen = strlen (boundary);

  while (!feof (datei)) {
    oldposition = ftell (datei);
    if (_FP_fgets (line, 255, datei) == NULL)
      break;
    if (ferror (datei))
      break;

    line[255] = '\0'; /* For Safety of string functions */

    /*
     * Make Busy Polls
     */

    if (UUBUSYPOLL(ftell(datei),progress.fsize)) {
      UUMessage (uuscan_id, __LINE__, UUMSG_NOTE,
		 uustring (S_SCAN_CANCEL));
      *errcode = UURET_CANCEL;
      break;
    }

    if (IsLineEmpty (line)) { /* line empty? */
      hcount = 0;
      hadnl  = 1;
      continue;               /* then ignore */
    }

    if (checkheaders) {
      if (IsKnownHeader (line)) {
	(void) ScanHeaderLine (datei, line);

	if (hcount == 0) {
	  preheaders = oldposition;
	  lcount     = 0;
	}
	hcount++;
	lcount++;

	/*
	 * check for the various restart counters
	 */

	if ((hcount >= hlcount.restart) ||
	    (hcount >= hlcount.afterdata && ismime == 0) ||
	    (hcount >= hlcount.afterdata && result->uudet) ||
	    (hcount >= hlcount.afternl   && result->uudet && hadnl)) {
	  /*
	   * Hey, a new header starts here
	   */
	  fseek (datei, preheaders, SEEK_SET);
	  break;
	}
	/* continue; */
      }
      else if (lcount > WAITHEADER) {
	hcount = 0;
	lcount = 0;
	dontcare=0;
      }
      else if (hcount) {
	lcount++;
	dontcare=1;
      }
      else {
	dontcare=0;
      }
    }
    else {
      dontcare=0;
    }

    if (boundary != NULL && 
	line[0] == '-' && line[1] == '-' &&
	strncmp (line+2, boundary, blen) == 0) {
      fseek (datei, oldposition, SEEK_SET);
      break;
    }
    if (boundary != NULL && line[0] == 'C' && line[1] == 'o' &&
	_FP_strnicmp (line, "Content-Type:", 13) == 0) {
      ptr = ScanHeaderLine (datei, line);
      p2  = (ptr)?_FP_stristr(ptr,"boundary"):NULL;
      p3  = (p2)?ParseValue(p2):NULL;

      if (p3 && strcmp (p3, boundary) == 0) {
	fseek (datei, oldposition, SEEK_SET);
	break;
      }
      else {
	p3 = NULL;
      }
    }

    if (strncmp      (line, "begin ",       6) == 0 ||
	_FP_strnicmp (line, "<pre>begin ", 11) == 0) {
      if ((result->begin || result->end ||
	   result->uudet == B64ENCODED ||
	   result->uudet == BH_ENCODED) && !uu_more_mime) {
	fseek (datei, oldposition, SEEK_SET);
	break;
      }
      
      if (*line == '<')
	ptr = line + 10;
      else
	ptr = line + 5;

      while (*ptr == ' ') ptr++;
      while (isdigit (*ptr)) 
	result->mode = result->mode * 8 + *ptr++ - '0';
      while (*ptr == ' ') ptr++;

      /*
       * We may have picked up a filename from a uuenview-style header
       */
      _FP_free (result->filename);
      result->filename = _FP_strdup (ptr);
      result->begin    = 1;

      while (isspace (result->filename[strlen(result->filename)-1]))
	result->filename[strlen(result->filename)-1] = '\0';

      continue;
    }

    if ((strncmp (line, "end", 3) == 0) &&
	result->uudet != BH_ENCODED &&
	result->uudet != YENC_ENCODED) {
      if (result->uudet == B64ENCODED && result->begin)
	result->uudet = XX_ENCODED;

      if (result->uudet != B64ENCODED) {
	result->end = 1;
	if (dflag && encoding)
	  result->uudet = encoding;
	continue;
      }
    }

    hadnl = 0;

    /*
     * Detect a UUDeview-Style header
     */

    if (_FP_strnicmp (line, "_=_ Part ", 9) == 0 &&
	result->uudet != YENC_ENCODED) {
      if (result->uudet) {
	fseek (datei, oldposition, SEEK_SET);
	break;
      }
      result->partno = atoi (line + 8);
      if ((ptr = _FP_stristr (line, "of file ")) != NULL) {
	ptr += 8;
	while (isspace (*ptr)) ptr++;
	p2 = ptr;
	while (isalnum(*p2) || 
	       *p2 == '.' || *p2=='_' || *p2 == '-' ||
	       *p2 == '!' || *p2=='@' || *p2 == '$')
	  p2++;
	c = *p2; *p2 = '\0';
	if (p2 != ptr && result->filename == NULL)
	  result->filename = _FP_strdup (ptr);
	else if (p2 - ptr > 5 && strchr (ptr, '.') != NULL) {
	  /*
	   * This file name looks good, too. Let's use it
	   */
	  _FP_free (result->filename);
	  result->filename = _FP_strdup (ptr);
	}
	*p2 = c;
      }
    }

    /*
     * Some reduced MIME handling. Only use if boundary == NULL. Also
     * accept the "X-Orcl-Content-Type" used by some braindead program.
     */
    if (boundary == NULL && !ismime && !uu_more_mime &&
	result->uudet != YENC_ENCODED) {
      if (_FP_strnicmp (line, "Content-Type", 12) == 0 ||
	  _FP_strnicmp (line, "X-Orcl-Content-Type", 19) == 0) {
	/*
	 * We use Content-Type to mark a new attachment and split the file.
	 * However, we do not split if we haven't found anything encoded yet.
	 */
	if (result->uudet) {
	  fseek (datei, oldposition, SEEK_SET);
	  break;
	}
	if ((ptr = strchr (line, ':')) != NULL) {
	  ptr++;
	  while (isspace (*ptr)) ptr++; p2 = ptr;
	  while (!isspace (*p2) && *p2 != ';') p2++;
	  c = *p2; *p2 = '\0';
	  if (p2 != ptr) {
	    _FP_free (result->mimetype);
	    result->mimetype = _FP_strdup (ptr);
	  }
	  *p2 = c;
	}
	ctline=0;
	hadct=1;
      }
      if ((ptr = _FP_stristr (line, "number=")) && ctline<4) {
	ptr += 7; if (*ptr == '"') ptr++;
	result->partno = atoi (ptr);
      }
      if ((ptr = _FP_stristr (line, "total=")) && ctline<4) {
	ptr += 6; if (*ptr == '"') ptr++;
	result->maxpno = atoi (ptr);
      }
      if ((ptr = _FP_stristr (line, "name=")) && ctline<4) {
	ptr += 5;
	while (isspace (*ptr)) ptr++;
	if (*ptr == '"' && *(ptr+1) && (p2 = strchr (ptr+2, '"')) != NULL) {
	  c = *p2; *p2 = '\0';
	  _FP_free (result->filename);
	  result->filename = _FP_strdup (ptr+1);
	  *p2 = c;
	}
	else if (*ptr=='\''&&*(ptr+1)&&(p2 = strchr(ptr+2, '\'')) != NULL) {
	  c = *p2; *p2 = '\0';
	  _FP_free (result->filename);
	  result->filename = _FP_strdup (ptr+1);
	  *p2 = c;
	}
	else {
	  p2 = ptr;
	  while (isalnum(*p2) || 
		 *p2 == '.' || *p2=='_' || *p2 == '-' ||
		 *p2 == '!' || *p2=='@' || *p2 == '$')
	    p2++;
	  c = *p2; *p2 = '\0';
	  if (p2 != ptr && result->filename == NULL)
	    result->filename = _FP_strdup (ptr);
	  else if (p2 - ptr > 5 && strchr (ptr, '.') != NULL) {
	    /*
	     * This file name looks good, too. Let's use it
	     */
	    _FP_free (result->filename);
	    result->filename = _FP_strdup (ptr);
	  }
	  *p2 = c;
	}
      }
      if ((ptr = _FP_stristr (line, "id=")) && ctline<4) {
	p2 = ptr += 3;
	if (*p2 == '"') {
	  p2 = strchr (++ptr, '"');
	}
	else {
	  while (*p2 && isprint(*p2) && !isspace(*p2) && *p2 != ';')
	    p2++;
	}
	if (p2 && *p2 && p2!=ptr) {
	  c = *p2; *p2 = '\0';
	  if (result->mimeid)
	    _FP_free (result->mimeid);
	  result->mimeid = _FP_strdup (ptr);
	  *p2 = c;
	}
      }
      
      /* 
       * Handling for very short Base64 files.
       */
      if (uu_tinyb64 && !ismime && !uu_more_mime) {
	if (line[0] == '-' && line[1] == '-') {
	  if (dflag && (encoding==B64ENCODED || result->uudet==B64ENCODED)) {
	    if (encoding==B64ENCODED && result->uudet==0 && (haddh||hadct)) {
	      result->uudet = encoding;
	      encoding = dflag = 0;
	    }
	    haddh = 1;
	    continue;
	  }
	  hadct = 0;
	}
      }
    } /* end of reduced MIME handling */

    /*
     * If we're in "freestyle" mode, have not encountered anything
     * interesting yet, and stumble upon something that looks like
     * a boundary, followed by a Content-* line, try to use it.
     */

    if (boundary == NULL && !ismime && !uu_more_mime && dflag <= 1 &&
	line[0] == '-' && line[1] == '-' && strlen(line+2)>10 &&
	(((ptr = _FP_strrstr (line+2, "--")) == NULL) ||
	 (*(ptr+2) != '\012' && *(ptr+2) != '\015')) &&
	_FP_strstr (line+2, "_=_") != NULL) {
      if (_FP_fgets (line, 255, datei) == NULL) {
	break;
      }
      if (_FP_strnicmp (line, "Content-", 8) == 0) {
	/*
	 * Okay, let's do it. This breaks out of ScanData. ScanPart will
	 * recognize the boundary on the next call and use it.
	 */
	fseek (datei, oldposition, SEEK_SET);
	break;
      }
    }

    /*
     * Detection for yEnc encoding
     */

    if (strncmp (line, "=ybegin ", 8) == 0 &&
	_FP_strstr (line, " name=") != NULL) {
      if ((result->begin || result->end || result->uudet) && !uu_more_mime) {
	fseek (datei, oldposition, SEEK_SET);
	break;
      }

      /*
       * name continues to the end of the line
       */
      
      _FP_free (result->filename);
      ptr = _FP_strstr (line, " name=") + 6;
      result->filename = _FP_strdup (ptr);

      while (isspace (result->filename[strlen(result->filename)-1]))
	result->filename[strlen(result->filename)-1] = '\0';

      /*
       * Determine size
       */

      if ((ptr = _FP_strstr (line, " size=")) != NULL) {
	ptr += 6;
	yefilesize = atoi (ptr);
      }
      else {
	yefilesize = -1;
      }

      /*
       * check for multipart file and read =ypart line
       */

      if ((ptr = _FP_strstr (line, " part=")) != NULL) {
	result->partno = atoi (ptr + 6);

	if (result->partno == 1) {
	  result->begin = 1;
	}

	if (_FP_fgets (line, 255, datei) == NULL) {
	  break;
	}

	line[255] = '\0';

	if (strncmp (line, "=ypart ", 7) != 0) {
	  break;
	}

	if ((ptr = _FP_strstr (line, " end=")) == NULL) {
	  break;
	}
       
	yepartends = atoi (ptr + 5);
      }
      else {
	result->partno = 1;