http.c

harvest是一个下载html网页得机器人

        if(readthisamountnow > BUFSIZE)
          readthisamountnow = BUFSIZE;

        actuallyread =
          data->set.fread(data->state.buffer, 1, readthisamountnow,
                          data->set.in);

        passed += actuallyread;
        if(actuallyread != readthisamountnow) {
          failf(data, "Could only read %d bytes from the input",
                passed);
          return CURLE_READ_ERROR;
        }
      } while(passed != conn->resume_from); /* loop until done */

      /* now, decrease the size of the read */
      if(data->set.infilesize>0) {
        data->set.infilesize -= conn->resume_from;

        if(data->set.infilesize <= 0) {
          failf(data, "File already completely uploaded");
          return CURLE_PARTIAL_FILE;
        }
      }
      /* we've passed, proceed as normal */
    }
  }
  if(conn->bits.use_range) {
    /*
     * A range is selected. We use different headers whether we're downloading
     * or uploading and we always let customized headers override our internal
     * ones if any such are specified.
     */
    if((data->set.httpreq == HTTPREQ_GET) &&
       !checkheaders(data, "Range:")) {
      /* if a line like this was already allocated, free the previous one */
      if(conn->allocptr.rangeline)
        free(conn->allocptr.rangeline);
      conn->allocptr.rangeline = aprintf("Range: bytes=%s\r\n", conn->range);
    }
    else if((data->set.httpreq != HTTPREQ_GET) &&
            !checkheaders(data, "Content-Range:")) {

      if(conn->resume_from) {
        /* This is because "resume" was selected */
        long total_expected_size= conn->resume_from + data->set.infilesize;
        conn->allocptr.rangeline = aprintf("Content-Range: bytes %s%ld/%ld\r\n",
                                      conn->range, total_expected_size-1,
                                      total_expected_size);
      }
      else {
        /* Range was selected and then we just pass the incoming range and 
           append total size */
        conn->allocptr.rangeline = aprintf("Content-Range: bytes %s/%d\r\n",
                                      conn->range, data->set.infilesize);
      }
    }
  }

  {
    /* Use 1.1 unless the use specificly asked for 1.0 */
    const char *httpstring=
      data->set.httpversion==CURL_HTTP_VERSION_1_0?"1.0":"1.1";

    send_buffer *req_buffer;
    struct curl_slist *headers=data->set.headers;
    size_t postsize;

    /* initialize a dynamic send-buffer */
    req_buffer = add_buffer_init();

    if(!req_buffer)
      return CURLE_OUT_OF_MEMORY;

    /* add the main request stuff */
    result =
      add_bufferf(req_buffer,
                  "%s " /* GET/HEAD/POST/PUT */
                  "%s HTTP/%s\r\n" /* path + HTTP version */
                  "%s" /* proxyuserpwd */
                  "%s" /* userpwd */
                  "%s" /* range */
                  "%s" /* user agent */
                  "%s" /* cookie */
                  "%s" /* host */
                  "%s" /* pragma */
                  "%s" /* accept */
                  "%s" /* accept-encoding */
                  "%s" /* referer */
                  "%s",/* transfer-encoding */

                request,
                ppath,
                httpstring,
                (conn->bits.httpproxy && conn->allocptr.proxyuserpwd)?
                conn->allocptr.proxyuserpwd:"",
                conn->allocptr.userpwd?conn->allocptr.userpwd:"",
                (conn->bits.use_range && conn->allocptr.rangeline)?
                conn->allocptr.rangeline:"",
                (data->set.useragent && *data->set.useragent && conn->allocptr.uagent)?
                conn->allocptr.uagent:"",
                (conn->allocptr.cookie?conn->allocptr.cookie:""), /* Cookie: <data> */
                (conn->allocptr.host?conn->allocptr.host:""), /* Host: host */
                http->p_pragma?http->p_pragma:"",
                http->p_accept?http->p_accept:"",
                (data->set.encoding && *data->set.encoding && conn->allocptr.accept_encoding)?
                conn->allocptr.accept_encoding:"", /* 08/28/02 jhrg */
                (data->change.referer && conn->allocptr.ref)?conn->allocptr.ref:"" /* Referer: <data> <CRLF> */,
                te
                );

    if(result)
      return result;

    if(co) {
      int count=0;
      struct Cookie *store=co;
      /* now loop through all cookies that matched */
      while(co) {
        if(co->value) {
          if(0 == count) {
            add_bufferf(req_buffer, "Cookie: ");
          }
          add_bufferf(req_buffer,
                      "%s%s=%s", count?"; ":"", co->name, co->value);
          count++;
        }
        co = co->next; /* next cookie please */
      }
      if(count) {
        add_buffer(req_buffer, "\r\n", 2);
      }
      Curl_cookie_freelist(store); /* free the cookie list */
      co=NULL;
    }

    if(data->set.timecondition) {
      struct tm *thistime;

      /* Phil Karn (Fri, 13 Apr 2001) pointed out that the If-Modified-Since
       * header family should have their times set in GMT as RFC2616 defines:
       * "All HTTP date/time stamps MUST be represented in Greenwich Mean Time
       * (GMT), without exception. For the purposes of HTTP, GMT is exactly
       * equal to UTC (Coordinated Universal Time)." (see page 20 of RFC2616).
       */

#ifdef HAVE_GMTIME_R
      /* thread-safe version */
      struct tm keeptime;
      thistime = (struct tm *)gmtime_r(&data->set.timevalue, &keeptime);
#else
      thistime = gmtime(&data->set.timevalue);
#endif

#ifdef HAVE_STRFTIME
      /* format: "Tue, 15 Nov 1994 12:45:26 GMT" */
      strftime(buf, BUFSIZE-1, "%a, %d %b %Y %H:%M:%S GMT", thistime);
#else
      /* TODO: Right, we *could* write a replacement here */
      strcpy(buf, "no strftime() support");
#endif
      switch(data->set.timecondition) {
      case CURL_TIMECOND_IFMODSINCE:
      default:
        add_bufferf(req_buffer,
                    "If-Modified-Since: %s\r\n", buf);
        break;
      case CURL_TIMECOND_IFUNMODSINCE:
        add_bufferf(req_buffer,
                    "If-Unmodified-Since: %s\r\n", buf);
        break;
      case CURL_TIMECOND_LASTMOD:
        add_bufferf(req_buffer,
                    "Last-Modified: %s\r\n", buf);
        break;
      }
    }

    while(headers) {
      ptr = strchr(headers->data, ':');
      if(ptr) {
        /* we require a colon for this to be a true header */

        ptr++; /* pass the colon */
        while(*ptr && isspace((int)*ptr))
          ptr++;

        if(*ptr) {
          /* only send this if the contents was non-blank */

          add_bufferf(req_buffer, "%s\r\n", headers->data);
        }
      }
      headers = headers->next;
    }

    http->postdata = NULL;  /* nothing to post at this point */
    Curl_pgrsSetUploadSize(data, 0); /* upload size is 0 atm */

    if(!authdone)
      /* until the auth is done, pretend we only do GET */
      httpreq = HTTPREQ_GET;
    else
      httpreq = data->set.httpreq;

    switch(httpreq) {

    case HTTPREQ_POST_FORM:
      if(Curl_FormInit(&http->form, http->sendit)) {
        failf(data, "Internal HTTP POST error!");
        return CURLE_HTTP_POST_ERROR;
      }

      /* set the read function to read from the generated form data */
      conn->fread = (curl_read_callback)Curl_FormReader;
      conn->fread_in = &http->form;

      http->sending = HTTPSEND_BODY;

      if(!conn->bits.upload_chunky)
        /* only add Content-Length if not uploading chunked */
        add_bufferf(req_buffer,
                    "Content-Length: %d\r\n", http->postsize);

      if(!checkheaders(data, "Expect:")) {
        /* if not disabled explicitly we add a Expect: 100-continue
           to the headers which actually speeds up post operations (as
           there is one packet coming back from the web server) */
        add_bufferf(req_buffer,
                    "Expect: 100-continue\r\n");
        data->set.expect100header = TRUE;
      }

      if(!checkheaders(data, "Content-Type:")) {
        /* Get Content-Type: line from Curl_FormReadOneLine, which happens
           to always be the first line. We can know this for sure since
           we always build the formpost linked list the same way!

           The Content-Type header line also contains the MIME boundary
           string etc why disabling this header is likely to not make things
           work, but we support it anyway.
        */
        char contentType[256];
        int linelength=0;
        linelength = Curl_FormReadOneLine (contentType,
                                           sizeof(contentType),
                                           1,
                                           (FILE *)&http->form);
        if(linelength == -1) {
          failf(data, "Could not get Content-Type header line!");
          return CURLE_HTTP_POST_ERROR;
        }
        add_buffer(req_buffer, contentType, linelength);
      }

      /* make the request end in a true CRLF */
      add_buffer(req_buffer, "\r\n", 2);

      /* set upload size to the progress meter */
      Curl_pgrsSetUploadSize(data, http->postsize);

      /* fire away the whole request to the server */
      result = add_buffer_send(req_buffer, conn->firstsocket, conn, 
                               &data->info.request_size);
      if(result)
        failf(data, "Failed sending POST request");
      else
        /* setup variables for the upcoming transfer */
        result = Curl_Transfer(conn, conn->firstsocket, -1, TRUE,
                               &http->readbytecount,
                               conn->firstsocket,
                               &http->writebytecount);
      if(result) {
        Curl_formclean(http->sendit); /* free that whole lot */
        return result;
      }
      break;

    case HTTPREQ_PUT: /* Let's PUT the data to the server! */

      if((data->set.infilesize>0) && !conn->bits.upload_chunky)
        /* only add Content-Length if not uploading chunked */
        add_bufferf(req_buffer,
                    "Content-Length: %d\r\n", /* file size */
                    data->set.infilesize );

      add_bufferf(req_buffer, "\r\n");

      /* set the upload size to the progress meter */
      Curl_pgrsSetUploadSize(data, data->set.infilesize);

      /* this sends the buffer and frees all the buffer resources */
      result = add_buffer_send(req_buffer, conn->firstsocket, conn,
                               &data->info.request_size);
      if(result)
        failf(data, "Failed sending POST request");
      else
        /* prepare for transfer */
        result = Curl_Transfer(conn, conn->firstsocket, -1, TRUE,
                               &http->readbytecount,
                               conn->firstsocket,
                               &http->writebytecount);
      if(result)
        return result;
      break;

    case HTTPREQ_POST:
      /* this is the simple POST, using x-www-form-urlencoded style */

      /* store the size of the postfields */
      postsize = data->set.postfieldsize?
        data->set.postfieldsize:
        (data->set.postfields?strlen(data->set.postfields):0);
      
      if(!conn->bits.upload_chunky) {
        /* We only set Content-Length and allow a custom Content-Length if
           we don't upload data chunked, as RFC2616 forbids us to set both
           kinds of headers (Transfer-Encoding: chunked and Content-Length) */

        if(!checkheaders(data, "Content-Length:"))
          /* we allow replacing this header, although it isn't very wise to
             actually set your own */
          add_bufferf(req_buffer, "Content-Length: %d\r\n", postsize);
      }

      if(!checkheaders(data, "Content-Type:"))
        add_bufferf(req_buffer,
                    "Content-Type: application/x-www-form-urlencoded\r\n");

      add_buffer(req_buffer, "\r\n", 2);

      if(data->set.postfields) {

        if(postsize < (100*1024)) {
          /* The post data is less than 100K, then append it to the header.
             This limit is no magic limit but only set to prevent really huge
             POSTs to get the data duplicated with malloc() and family. */

          if(!conn->bits.upload_chunky)
            /* We're not sending it 'chunked', append it to the request
               already now to reduce the number if send() calls */
            add_buffer(req_buffer, data->set.postfields, postsize);
          else {
            /* Append the POST data chunky-style */
            add_bufferf(req_buffer, "%x\r\n", postsize);
            add_buffer(req_buffer, data->set.postfields, postsize);
            add_buffer(req_buffer, "\r\n0\r\n", 5); /* end of a chunked
                                                       transfer stream */
          }
        }
        else {
          /* A huge POST coming up, do data separate from the request */
          http->postsize = postsize;
          http->postdata = data->set.postfields;

          http->sending = HTTPSEND_BODY;

          conn->fread = (curl_read_callback)readmoredata;
          conn->fread_in = (void *)conn;

          /* set the upload size to the progress meter */
          Curl_pgrsSetUploadSize(data, http->postsize);
        }
      }
      else {
        /* set the upload size to the progress meter */
        Curl_pgrsSetUploadSize(data, data->set.infilesize);

        /* set the pointer to mark that we will send the post body using
           the read callback */
        http->postdata = (char *)&http->postdata;
      }
      /* issue the request */
      result = add_buffer_send(req_buffer, conn->firstsocket, conn,
                               &data->info.request_size);

      if(result)
        failf(data, "Failed sending HTTP POST request");
      else
        result =
          Curl_Transfer(conn, conn->firstsocket, -1, TRUE,
                        &http->readbytecount,
                        http->postdata?conn->firstsocket:-1,
                        http->postdata?&http->writebytecount:NULL);
      break;

    default:
      add_buffer(req_buffer, "\r\n", 2);
      
      /* issue the request */
      result = add_buffer_send(req_buffer, conn->firstsocket, conn,
                               &data->info.request_size);

      if(result)
        failf(data, "Failed sending HTTP request");
      else
        /* HTTP GET/HEAD download: */
        result = Curl_Transfer(conn, conn->firstsocket, -1, TRUE,
                               &http->readbytecount,
                               http->postdata?conn->firstsocket:-1,
                               http->postdata?&http->writebytecount:NULL);
    }
    if(result)
      return result;
  }

  return CURLE_OK;
}
#endif