1 /***************************************************************************
2 * _ _ ____ _
3 * Project ___| | | | _ \| |
4 * / __| | | | |_) | |
5 * | (__| |_| | _ <| |___
6 * \___|\___/|_| \_\_____|
7 *
8 * Copyright (C) 1998 - 2018, Daniel Stenberg, <daniel@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 => https://libexpat.github.io/
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
startElement(void * userData,const XML_Char * name,const XML_Char ** atts)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
characterDataHandler(void * userData,const XML_Char * s,int len)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 char *ptr = realloc(mem->memory, mem->size + len + 1);
73 if(!ptr) {
74 /* Out of memory. */
75 fprintf(stderr, "Not enough memory (realloc returned NULL).\n");
76 state->ok = 0;
77 return;
78 }
79
80 mem->memory = ptr;
81 memcpy(&(mem->memory[mem->size]), s, len);
82 mem->size += len;
83 mem->memory[mem->size] = 0;
84 }
85
endElement(void * userData,const XML_Char * name)86 static void endElement(void *userData, const XML_Char *name)
87 {
88 struct ParserStruct *state = (struct ParserStruct *) userData;
89 state->depth--;
90
91 printf("%5lu %10lu %s\n", state->depth, state->characters.size, name);
92 }
93
parseStreamCallback(void * contents,size_t length,size_t nmemb,void * userp)94 static size_t parseStreamCallback(void *contents, size_t length, size_t nmemb,
95 void *userp)
96 {
97 XML_Parser parser = (XML_Parser) userp;
98 size_t real_size = length * nmemb;
99 struct ParserStruct *state = (struct ParserStruct *) XML_GetUserData(parser);
100
101 /* Only parse if we're not already in a failure state. */
102 if(state->ok && XML_Parse(parser, contents, real_size, 0) == 0) {
103 int error_code = XML_GetErrorCode(parser);
104 fprintf(stderr, "Parsing response buffer of length %lu failed"
105 " with error code %d (%s).\n",
106 real_size, error_code, XML_ErrorString(error_code));
107 state->ok = 0;
108 }
109
110 return real_size;
111 }
112
main(void)113 int main(void)
114 {
115 CURL *curl_handle;
116 CURLcode res;
117 XML_Parser parser;
118 struct ParserStruct state;
119
120 /* Initialize the state structure for parsing. */
121 memset(&state, 0, sizeof(struct ParserStruct));
122 state.ok = 1;
123
124 /* Initialize a namespace-aware parser. */
125 parser = XML_ParserCreateNS(NULL, '\0');
126 XML_SetUserData(parser, &state);
127 XML_SetElementHandler(parser, startElement, endElement);
128 XML_SetCharacterDataHandler(parser, characterDataHandler);
129
130 /* Initialize a libcurl handle. */
131 curl_global_init(CURL_GLOBAL_DEFAULT);
132 curl_handle = curl_easy_init();
133 curl_easy_setopt(curl_handle, CURLOPT_URL,
134 "https://www.w3schools.com/xml/simple.xml");
135 curl_easy_setopt(curl_handle, CURLOPT_WRITEFUNCTION, parseStreamCallback);
136 curl_easy_setopt(curl_handle, CURLOPT_WRITEDATA, (void *)parser);
137
138 printf("Depth Characters Closing Tag\n");
139
140 /* Perform the request and any follow-up parsing. */
141 res = curl_easy_perform(curl_handle);
142 if(res != CURLE_OK) {
143 fprintf(stderr, "curl_easy_perform() failed: %s\n",
144 curl_easy_strerror(res));
145 }
146 else if(state.ok) {
147 /* Expat requires one final call to finalize parsing. */
148 if(XML_Parse(parser, NULL, 0, 1) == 0) {
149 int error_code = XML_GetErrorCode(parser);
150 fprintf(stderr, "Finalizing parsing failed with error code %d (%s).\n",
151 error_code, XML_ErrorString(error_code));
152 }
153 else {
154 printf(" --------------\n");
155 printf(" %lu tags total\n", state.tags);
156 }
157 }
158
159 /* Clean up. */
160 free(state.characters.memory);
161 XML_ParserFree(parser);
162 curl_easy_cleanup(curl_handle);
163 curl_global_cleanup();
164
165 return 0;
166 }
167