1 /***************************************************************************
2 * _ _ ____ _
3 * Project ___| | | | _ \| |
4 * / __| | | | |_) | |
5 * | (__| |_| | _ <| |___
6 * \___|\___/|_| \_\_____|
7 *
8 * Copyright (C) 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.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 * SPDX-License-Identifier: curl
22 *
23 ***************************************************************************/
24 /* <DESC>
25 * Stream-parse a document using the streaming Expat parser.
26 * </DESC>
27 */
28 /* Written by David Strauss
29 *
30 * Expat => https://libexpat.github.io/
31 *
32 * gcc -Wall -I/usr/local/include xmlstream.c -lcurl -lexpat -o xmlstream
33 *
34 */
35
36 #include <stdio.h>
37 #include <stdlib.h>
38 #include <string.h>
39
40 #include <expat.h>
41 #include <curl/curl.h>
42
43 struct MemoryStruct {
44 char *memory;
45 size_t size;
46 };
47
48 struct ParserStruct {
49 int ok;
50 size_t tags;
51 size_t depth;
52 struct MemoryStruct characters;
53 };
54
startElement(void * userData,const XML_Char * name,const XML_Char ** atts)55 static void startElement(void *userData, const XML_Char *name,
56 const XML_Char **atts)
57 {
58 struct ParserStruct *state = (struct ParserStruct *) userData;
59 state->tags++;
60 state->depth++;
61
62 (void)name;
63 (void)atts;
64
65 /* Get a clean slate for reading in character data. */
66 free(state->characters.memory);
67 state->characters.memory = NULL;
68 state->characters.size = 0;
69 }
70
characterDataHandler(void * userData,const XML_Char * s,int len)71 static void characterDataHandler(void *userData, const XML_Char *s, int len)
72 {
73 struct ParserStruct *state = (struct ParserStruct *) userData;
74 struct MemoryStruct *mem = &state->characters;
75
76 char *ptr = realloc(mem->memory, mem->size + (unsigned long)len + 1);
77 if(!ptr) {
78 /* Out of memory. */
79 fprintf(stderr, "Not enough memory (realloc returned NULL).\n");
80 state->ok = 0;
81 return;
82 }
83
84 mem->memory = ptr;
85 memcpy(&(mem->memory[mem->size]), s, len);
86 mem->size += (unsigned long)len;
87 mem->memory[mem->size] = 0;
88 }
89
endElement(void * userData,const XML_Char * name)90 static void endElement(void *userData, const XML_Char *name)
91 {
92 struct ParserStruct *state = (struct ParserStruct *) userData;
93 state->depth--;
94
95 printf("%5lu %10lu %s\n", state->depth, state->characters.size, name);
96 }
97
parseStreamCallback(void * contents,size_t length,size_t nmemb,void * userp)98 static size_t parseStreamCallback(void *contents, size_t length, size_t nmemb,
99 void *userp)
100 {
101 XML_Parser parser = (XML_Parser) userp;
102 size_t real_size = length * nmemb;
103 struct ParserStruct *state = (struct ParserStruct *) XML_GetUserData(parser);
104
105 /* Only parse if we are not already in a failure state. */
106 if(state->ok && XML_Parse(parser, contents, (int)real_size, 0) == 0) {
107 enum XML_Error error_code = XML_GetErrorCode(parser);
108 fprintf(stderr, "Parsing response buffer of length %lu failed"
109 " with error code %d (%s).\n",
110 real_size, error_code, XML_ErrorString(error_code));
111 state->ok = 0;
112 }
113
114 return real_size;
115 }
116
main(void)117 int main(void)
118 {
119 CURL *curl_handle;
120 CURLcode res;
121 XML_Parser parser;
122 struct ParserStruct state;
123
124 /* Initialize the state structure for parsing. */
125 memset(&state, 0, sizeof(struct ParserStruct));
126 state.ok = 1;
127
128 /* Initialize a namespace-aware parser. */
129 parser = XML_ParserCreateNS(NULL, '\0');
130 XML_SetUserData(parser, &state);
131 XML_SetElementHandler(parser, startElement, endElement);
132 XML_SetCharacterDataHandler(parser, characterDataHandler);
133
134 /* Initialize a libcurl handle. */
135 curl_global_init(CURL_GLOBAL_DEFAULT);
136 curl_handle = curl_easy_init();
137 curl_easy_setopt(curl_handle, CURLOPT_URL,
138 "https://www.w3schools.com/xml/simple.xml");
139 curl_easy_setopt(curl_handle, CURLOPT_WRITEFUNCTION, parseStreamCallback);
140 curl_easy_setopt(curl_handle, CURLOPT_WRITEDATA, (void *)parser);
141
142 printf("Depth Characters Closing Tag\n");
143
144 /* Perform the request and any follow-up parsing. */
145 res = curl_easy_perform(curl_handle);
146 if(res != CURLE_OK) {
147 fprintf(stderr, "curl_easy_perform() failed: %s\n",
148 curl_easy_strerror(res));
149 }
150 else if(state.ok) {
151 /* Expat requires one final call to finalize parsing. */
152 if(XML_Parse(parser, NULL, 0, 1) == 0) {
153 enum XML_Error error_code = XML_GetErrorCode(parser);
154 fprintf(stderr, "Finalizing parsing failed with error code %d (%s).\n",
155 error_code, XML_ErrorString(error_code));
156 }
157 else {
158 printf(" --------------\n");
159 printf(" %lu tags total\n", state.tags);
160 }
161 }
162
163 /* Clean up. */
164 free(state.characters.memory);
165 XML_ParserFree(parser);
166 curl_easy_cleanup(curl_handle);
167 curl_global_cleanup();
168
169 return 0;
170 }
171
