123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348349350351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381382383384385386387388389390391392393394395396397 |
- /***************************************************************************
- * _ _ ____ _
- * Project ___| | | | _ \| |
- * / __| | | | |_) | |
- * | (__| |_| | _ <| |___
- * \___|\___/|_| \_\_____|
- *
- * Copyright (C) 1998 - 2011, Daniel Stenberg, <daniel@haxx.se>, et al.
- *
- * This software is licensed as described in the file COPYING, which
- * you should have received as part of this distribution. The terms
- * are also available at http://curl.haxx.se/docs/copyright.html.
- *
- * You may opt to use, copy, modify, merge, publish, distribute and/or sell
- * copies of the Software, and permit persons to whom the Software is
- * furnished to do so, under the terms of the COPYING file.
- *
- * This software is distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY
- * KIND, either express or implied.
- *
- ***************************************************************************/
- #include "curl_setup.h"
- #ifndef CURL_DISABLE_HTTP
- #include "urldata.h" /* it includes http_chunks.h */
- #include "sendf.h" /* for the client write stuff */
- #include "content_encoding.h"
- #include "http.h"
- #include "curl_memory.h"
- #include "non-ascii.h" /* for Curl_convert_to_network prototype */
- #define _MPRINTF_REPLACE /* use our functions only */
- #include <curl/mprintf.h>
- /* The last #include file should be: */
- #include "memdebug.h"
- /*
- * Chunk format (simplified):
- *
- * <HEX SIZE>[ chunk extension ] CRLF
- * <DATA> CRLF
- *
- * Highlights from RFC2616 section 3.6 say:
- The chunked encoding modifies the body of a message in order to
- transfer it as a series of chunks, each with its own size indicator,
- followed by an OPTIONAL trailer containing entity-header fields. This
- allows dynamically produced content to be transferred along with the
- information necessary for the recipient to verify that it has
- received the full message.
- Chunked-Body = *chunk
- last-chunk
- trailer
- CRLF
- chunk = chunk-size [ chunk-extension ] CRLF
- chunk-data CRLF
- chunk-size = 1*HEX
- last-chunk = 1*("0") [ chunk-extension ] CRLF
- chunk-extension= *( ";" chunk-ext-name [ "=" chunk-ext-val ] )
- chunk-ext-name = token
- chunk-ext-val = token | quoted-string
- chunk-data = chunk-size(OCTET)
- trailer = *(entity-header CRLF)
- The chunk-size field is a string of hex digits indicating the size of
- the chunk. The chunked encoding is ended by any chunk whose size is
- zero, followed by the trailer, which is terminated by an empty line.
- */
- /* Check for an ASCII hex digit.
- We avoid the use of isxdigit to accommodate non-ASCII hosts. */
- static bool Curl_isxdigit(char digit)
- {
- return ( (digit >= 0x30 && digit <= 0x39) /* 0-9 */
- || (digit >= 0x41 && digit <= 0x46) /* A-F */
- || (digit >= 0x61 && digit <= 0x66) /* a-f */ ) ? TRUE : FALSE;
- }
- void Curl_httpchunk_init(struct connectdata *conn)
- {
- struct Curl_chunker *chunk = &conn->chunk;
- chunk->hexindex=0; /* start at 0 */
- chunk->dataleft=0; /* no data left yet! */
- chunk->state = CHUNK_HEX; /* we get hex first! */
- }
- /*
- * chunk_read() returns a OK for normal operations, or a positive return code
- * for errors. STOP means this sequence of chunks is complete. The 'wrote'
- * argument is set to tell the caller how many bytes we actually passed to the
- * client (for byte-counting and whatever).
- *
- * The states and the state-machine is further explained in the header file.
- *
- * This function always uses ASCII hex values to accommodate non-ASCII hosts.
- * For example, 0x0d and 0x0a are used instead of '\r' and '\n'.
- */
- CHUNKcode Curl_httpchunk_read(struct connectdata *conn,
- char *datap,
- ssize_t datalen,
- ssize_t *wrotep)
- {
- CURLcode result=CURLE_OK;
- struct SessionHandle *data = conn->data;
- struct Curl_chunker *ch = &conn->chunk;
- struct SingleRequest *k = &data->req;
- size_t piece;
- size_t length = (size_t)datalen;
- size_t *wrote = (size_t *)wrotep;
- *wrote = 0; /* nothing's written yet */
- /* the original data is written to the client, but we go on with the
- chunk read process, to properly calculate the content length*/
- if(data->set.http_te_skip && !k->ignorebody) {
- result = Curl_client_write(conn, CLIENTWRITE_BODY, datap, datalen);
- if(result)
- return CHUNKE_WRITE_ERROR;
- }
- while(length) {
- switch(ch->state) {
- case CHUNK_HEX:
- if(Curl_isxdigit(*datap)) {
- if(ch->hexindex < MAXNUM_SIZE) {
- ch->hexbuffer[ch->hexindex] = *datap;
- datap++;
- length--;
- ch->hexindex++;
- }
- else {
- return CHUNKE_TOO_LONG_HEX; /* longer hex than we support */
- }
- }
- else {
- if(0 == ch->hexindex) {
- /* This is illegal data, we received junk where we expected
- a hexadecimal digit. */
- return CHUNKE_ILLEGAL_HEX;
- }
- /* length and datap are unmodified */
- ch->hexbuffer[ch->hexindex]=0;
- /* convert to host encoding before calling strtoul */
- result = Curl_convert_from_network(conn->data, ch->hexbuffer,
- ch->hexindex);
- if(result) {
- /* Curl_convert_from_network calls failf if unsuccessful */
- /* Treat it as a bad hex character */
- return(CHUNKE_ILLEGAL_HEX);
- }
- ch->datasize=strtoul(ch->hexbuffer, NULL, 16);
- ch->state = CHUNK_POSTHEX;
- }
- break;
- case CHUNK_POSTHEX:
- /* In this state, we're waiting for CRLF to arrive. We support
- this to allow so called chunk-extensions to show up here
- before the CRLF comes. */
- if(*datap == 0x0d)
- ch->state = CHUNK_CR;
- length--;
- datap++;
- break;
- case CHUNK_CR:
- /* waiting for the LF */
- if(*datap == 0x0a) {
- /* we're now expecting data to come, unless size was zero! */
- if(0 == ch->datasize) {
- ch->state = CHUNK_TRAILER; /* now check for trailers */
- conn->trlPos=0;
- }
- else {
- ch->state = CHUNK_DATA;
- }
- }
- else
- /* previously we got a fake CR, go back to CR waiting! */
- ch->state = CHUNK_CR;
- datap++;
- length--;
- break;
- case CHUNK_DATA:
- /* we get pure and fine data
- We expect another 'datasize' of data. We have 'length' right now,
- it can be more or less than 'datasize'. Get the smallest piece.
- */
- piece = (ch->datasize >= length)?length:ch->datasize;
- /* Write the data portion available */
- #ifdef HAVE_LIBZ
- switch (conn->data->set.http_ce_skip?
- IDENTITY : data->req.auto_decoding) {
- case IDENTITY:
- #endif
- if(!k->ignorebody) {
- if(!data->set.http_te_skip)
- result = Curl_client_write(conn, CLIENTWRITE_BODY, datap,
- piece);
- else
- result = CURLE_OK;
- }
- #ifdef HAVE_LIBZ
- break;
- case DEFLATE:
- /* update data->req.keep.str to point to the chunk data. */
- data->req.str = datap;
- result = Curl_unencode_deflate_write(conn, &data->req,
- (ssize_t)piece);
- break;
- case GZIP:
- /* update data->req.keep.str to point to the chunk data. */
- data->req.str = datap;
- result = Curl_unencode_gzip_write(conn, &data->req,
- (ssize_t)piece);
- break;
- case COMPRESS:
- default:
- failf (conn->data,
- "Unrecognized content encoding type. "
- "libcurl understands `identity', `deflate' and `gzip' "
- "content encodings.");
- return CHUNKE_BAD_ENCODING;
- }
- #endif
- if(result)
- return CHUNKE_WRITE_ERROR;
- *wrote += piece;
- ch->datasize -= piece; /* decrease amount left to expect */
- datap += piece; /* move read pointer forward */
- length -= piece; /* decrease space left in this round */
- if(0 == ch->datasize)
- /* end of data this round, we now expect a trailing CRLF */
- ch->state = CHUNK_POSTCR;
- break;
- case CHUNK_POSTCR:
- if(*datap == 0x0d) {
- ch->state = CHUNK_POSTLF;
- datap++;
- length--;
- }
- else
- return CHUNKE_BAD_CHUNK;
- break;
- case CHUNK_POSTLF:
- if(*datap == 0x0a) {
- /*
- * The last one before we go back to hex state and start all
- * over.
- */
- Curl_httpchunk_init(conn);
- datap++;
- length--;
- }
- else
- return CHUNKE_BAD_CHUNK;
- break;
- case CHUNK_TRAILER:
- if(*datap == 0x0d) {
- /* this is the end of a trailer, but if the trailer was zero bytes
- there was no trailer and we move on */
- if(conn->trlPos) {
- /* we allocate trailer with 3 bytes extra room to fit this */
- conn->trailer[conn->trlPos++]=0x0d;
- conn->trailer[conn->trlPos++]=0x0a;
- conn->trailer[conn->trlPos]=0;
- /* Convert to host encoding before calling Curl_client_write */
- result = Curl_convert_from_network(conn->data, conn->trailer,
- conn->trlPos);
- if(result)
- /* Curl_convert_from_network calls failf if unsuccessful */
- /* Treat it as a bad chunk */
- return CHUNKE_BAD_CHUNK;
- if(!data->set.http_te_skip) {
- result = Curl_client_write(conn, CLIENTWRITE_HEADER,
- conn->trailer, conn->trlPos);
- if(result)
- return CHUNKE_WRITE_ERROR;
- }
- conn->trlPos=0;
- ch->state = CHUNK_TRAILER_CR;
- }
- else {
- /* no trailer, we're on the final CRLF pair */
- ch->state = CHUNK_TRAILER_POSTCR;
- break; /* don't advance the pointer */
- }
- }
- else {
- /* conn->trailer is assumed to be freed in url.c on a
- connection basis */
- if(conn->trlPos >= conn->trlMax) {
- /* we always allocate three extra bytes, just because when the full
- header has been received we append CRLF\0 */
- char *ptr;
- if(conn->trlMax) {
- conn->trlMax *= 2;
- ptr = realloc(conn->trailer, conn->trlMax + 3);
- }
- else {
- conn->trlMax=128;
- ptr = malloc(conn->trlMax + 3);
- }
- if(!ptr)
- return CHUNKE_OUT_OF_MEMORY;
- conn->trailer = ptr;
- }
- conn->trailer[conn->trlPos++]=*datap;
- }
- datap++;
- length--;
- break;
- case CHUNK_TRAILER_CR:
- if(*datap == 0x0a) {
- ch->state = CHUNK_TRAILER_POSTCR;
- datap++;
- length--;
- }
- else
- return CHUNKE_BAD_CHUNK;
- break;
- case CHUNK_TRAILER_POSTCR:
- /* We enter this state when a CR should arrive so we expect to
- have to first pass a CR before we wait for LF */
- if(*datap != 0x0d) {
- /* not a CR then it must be another header in the trailer */
- ch->state = CHUNK_TRAILER;
- break;
- }
- datap++;
- length--;
- /* now wait for the final LF */
- ch->state = CHUNK_STOP;
- break;
- case CHUNK_STOPCR:
- /* Read the final CRLF that ends all chunk bodies */
- if(*datap == 0x0d) {
- ch->state = CHUNK_STOP;
- datap++;
- length--;
- }
- else
- return CHUNKE_BAD_CHUNK;
- break;
- case CHUNK_STOP:
- if(*datap == 0x0a) {
- length--;
- /* Record the length of any data left in the end of the buffer
- even if there's no more chunks to read */
- ch->dataleft = length;
- return CHUNKE_STOP; /* return stop */
- }
- else
- return CHUNKE_BAD_CHUNK;
- default:
- return CHUNKE_STATE_ERROR;
- }
- }
- return CHUNKE_OK;
- }
- #endif /* CURL_DISABLE_HTTP */
|