Home | History | Annotate | Download | only in examples
      1 /***************************************************************************
      2  *                                  _   _ ____  _
      3  *  Project                     ___| | | |  _ \| |
      4  *                             / __| | | | |_) | |
      5  *                            | (__| |_| |  _ <| |___
      6  *                             \___|\___/|_| \_\_____|
      7  *
      8  * Copyright (C) 1998 - 2017, Daniel Stenberg, <daniel (at) haxx.se>, et al.
      9  *
     10  * This software is licensed as described in the file COPYING, which
     11  * you should have received as part of this distribution. The terms
     12  * are also available at https://curl.haxx.se/docs/copyright.html.
     13  *
     14  * You may opt to use, copy, modify, merge, publish, distribute and/or sell
     15  * copies of the Software, and permit persons to whom the Software is
     16  * furnished to do so, under the terms of the COPYING file.
     17  *
     18  * This software is distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY
     19  * KIND, either express or implied.
     20  *
     21  ***************************************************************************/
     22 /* <DESC>
     23  * Stream-parse a document using the streaming Expat parser.
     24  * </DESC>
     25  */
     26 /* Written by David Strauss
     27  *
     28  * Expat => http://www.libexpat.org/
     29  *
     30  * gcc -Wall -I/usr/local/include xmlstream.c -lcurl -lexpat -o xmlstream
     31  *
     32  */
     33 
     34 #include <stdio.h>
     35 #include <stdlib.h>
     36 #include <string.h>
     37 #include <assert.h>
     38 
     39 #include <expat.h>
     40 #include <curl/curl.h>
     41 
     42 struct MemoryStruct {
     43   char *memory;
     44   size_t size;
     45 };
     46 
     47 struct ParserStruct {
     48   int ok;
     49   size_t tags;
     50   size_t depth;
     51   struct MemoryStruct characters;
     52 };
     53 
     54 static void startElement(void *userData, const XML_Char *name,
     55                          const XML_Char **atts)
     56 {
     57   struct ParserStruct *state = (struct ParserStruct *) userData;
     58   state->tags++;
     59   state->depth++;
     60 
     61   /* Get a clean slate for reading in character data. */
     62   free(state->characters.memory);
     63   state->characters.memory = NULL;
     64   state->characters.size = 0;
     65 }
     66 
     67 static void characterDataHandler(void *userData, const XML_Char *s, int len)
     68 {
     69   struct ParserStruct *state = (struct ParserStruct *) userData;
     70   struct MemoryStruct *mem = &state->characters;
     71 
     72   mem->memory = realloc(mem->memory, mem->size + len + 1);
     73   if(mem->memory == NULL) {
     74     /* Out of memory. */
     75     fprintf(stderr, "Not enough memory (realloc returned NULL).\n");
     76     state->ok = 0;
     77     return;
     78   }
     79 
     80   memcpy(&(mem->memory[mem->size]), s, len);
     81   mem->size += len;
     82   mem->memory[mem->size] = 0;
     83 }
     84 
     85 static void endElement(void *userData, const XML_Char *name)
     86 {
     87   struct ParserStruct *state = (struct ParserStruct *) userData;
     88   state->depth--;
     89 
     90   printf("%5lu   %10lu   %s\n", state->depth, state->characters.size, name);
     91 }
     92 
     93 static size_t parseStreamCallback(void *contents, size_t length, size_t nmemb,
     94                                   void *userp)
     95 {
     96   XML_Parser parser = (XML_Parser) userp;
     97   size_t real_size = length * nmemb;
     98   struct ParserStruct *state = (struct ParserStruct *) XML_GetUserData(parser);
     99 
    100   /* Only parse if we're not already in a failure state. */
    101   if(state->ok && XML_Parse(parser, contents, real_size, 0) == 0) {
    102     int error_code = XML_GetErrorCode(parser);
    103     fprintf(stderr, "Parsing response buffer of length %lu failed"
    104             " with error code %d (%s).\n",
    105             real_size, error_code, XML_ErrorString(error_code));
    106     state->ok = 0;
    107   }
    108 
    109   return real_size;
    110 }
    111 
    112 int main(void)
    113 {
    114   CURL *curl_handle;
    115   CURLcode res;
    116   XML_Parser parser;
    117   struct ParserStruct state;
    118 
    119   /* Initialize the state structure for parsing. */
    120   memset(&state, 0, sizeof(struct ParserStruct));
    121   state.ok = 1;
    122 
    123   /* Initialize a namespace-aware parser. */
    124   parser = XML_ParserCreateNS(NULL, '\0');
    125   XML_SetUserData(parser, &state);
    126   XML_SetElementHandler(parser, startElement, endElement);
    127   XML_SetCharacterDataHandler(parser, characterDataHandler);
    128 
    129   /* Initialize a libcurl handle. */
    130   curl_global_init(CURL_GLOBAL_DEFAULT);
    131   curl_handle = curl_easy_init();
    132   curl_easy_setopt(curl_handle, CURLOPT_URL,
    133                    "http://www.w3schools.com/xml/simple.xml");
    134   curl_easy_setopt(curl_handle, CURLOPT_WRITEFUNCTION, parseStreamCallback);
    135   curl_easy_setopt(curl_handle, CURLOPT_WRITEDATA, (void *)parser);
    136 
    137   printf("Depth   Characters   Closing Tag\n");
    138 
    139   /* Perform the request and any follow-up parsing. */
    140   res = curl_easy_perform(curl_handle);
    141   if(res != CURLE_OK) {
    142     fprintf(stderr, "curl_easy_perform() failed: %s\n",
    143             curl_easy_strerror(res));
    144   }
    145   else if(state.ok) {
    146     /* Expat requires one final call to finalize parsing. */
    147     if(XML_Parse(parser, NULL, 0, 1) == 0) {
    148       int error_code = XML_GetErrorCode(parser);
    149       fprintf(stderr, "Finalizing parsing failed with error code %d (%s).\n",
    150               error_code, XML_ErrorString(error_code));
    151     }
    152     else {
    153       printf("                     --------------\n");
    154       printf("                     %lu tags total\n", state.tags);
    155     }
    156   }
    157 
    158   /* Clean up. */
    159   free(state.characters.memory);
    160   XML_ParserFree(parser);
    161   curl_easy_cleanup(curl_handle);
    162   curl_global_cleanup();
    163 
    164   return 0;
    165 }
    166