| 1 |
#include "AsxReader.h" |
| 2 |
#include "avm_output.h" |
| 3 |
|
| 4 |
#include <ctype.h> |
| 5 |
#include <string.h> |
| 6 |
#include <stdio.h> |
| 7 |
|
| 8 |
AVM_BEGIN_NAMESPACE; |
| 9 |
|
| 10 |
enum asx_position |
| 11 |
{ |
| 12 |
BEGIN, |
| 13 |
OPEN_BRACKET, |
| 14 |
CLOSE_BRACKET, |
| 15 |
IN_COMMENT, |
| 16 |
IN_QUOTED_COMMENT, |
| 17 |
IN_TAGNAME, |
| 18 |
AFTER_TAGNAME, |
| 19 |
IN_ENDTAG, |
| 20 |
BEFORE_TAGCLOSE, |
| 21 |
IN_PROPNAME, |
| 22 |
AFTER_PROPNAME, |
| 23 |
PROP_EQUAL, |
| 24 |
AFTER_PROPEQUAL, |
| 25 |
BETWEEN_PROPS, |
| 26 |
IN_PROPCONTENTS, |
| 27 |
IN_QUOTED_PROPCONTENTS, |
| 28 |
IN_QUOTED_CONTENTS, |
| 29 |
IN_CONTENTS, |
| 30 |
}; |
| 31 |
|
| 32 |
bool ASXReader::addURL(const char* url) |
| 33 |
{ |
| 34 |
int i = 0; |
| 35 |
|
| 36 |
while (*url && (!isprint(*url) || (*url == '"'))) |
| 37 |
url++; |
| 38 |
|
| 39 |
while (isprint(url[i]) && (url[i] != '"')) |
| 40 |
i++; |
| 41 |
|
| 42 |
if (i > 0) |
| 43 |
{ |
| 44 |
avm::string vurl = avm::string(url, i); |
| 45 |
|
| 46 |
if (strncasecmp(vurl.c_str(), "mms://", 6) != 0 |
| 47 |
&& strncasecmp(vurl.c_str(), "http://", 7) != 0) |
| 48 |
{ |
| 49 |
char* h = new char[m_Server.size() * 2 + m_Filename.size()]; |
| 50 |
if (h) |
| 51 |
{ |
| 52 |
int p = sprintf(h, "http://%s", m_Server.c_str()); |
| 53 |
if (vurl[0] != '/') |
| 54 |
{ |
| 55 |
h[p++] = '/'; |
| 56 |
strcpy(h + p, m_Filename.c_str()); |
| 57 |
char* q = strchr(h + p, '?'); // strip all after '?' |
| 58 |
if (q) *q = 0; |
| 59 |
} |
| 60 |
//printf("_____addUrl>%s<>%s<____\n", h, vurl.c_str()); |
| 61 |
vurl.insert(0, h); |
| 62 |
} |
| 63 |
} |
| 64 |
|
| 65 |
//printf("URL:%s<\n", vurl.c_str()); |
| 66 |
m_Urls.push_back(vurl); |
| 67 |
return true; |
| 68 |
} |
| 69 |
return false; |
| 70 |
} |
| 71 |
|
| 72 |
bool ASXReader::create(const char* data, size_t size) |
| 73 |
{ |
| 74 |
int last_start = 0; |
| 75 |
avm::string last_tag; |
| 76 |
avm::string last_prop; |
| 77 |
avm::string last_prop_val; |
| 78 |
m_Urls.clear(); |
| 79 |
|
| 80 |
//printf("PARSE >%s< %d\n", data, size); |
| 81 |
asx_position state = BEGIN; |
| 82 |
for (unsigned i = 0; i < size; i++) |
| 83 |
{ |
| 84 |
unsigned char c = data[i]; |
| 85 |
if (c < 6) break; // stop processing |
| 86 |
//printf("i %d %c %d\n", i, c, state); |
| 87 |
int is_alpha = isalpha(c) || (c >= 0x80); |
| 88 |
int is_space = isspace(c); |
| 89 |
switch(state) |
| 90 |
{ |
| 91 |
case BEGIN: |
| 92 |
if (c == '<') |
| 93 |
state = OPEN_BRACKET; |
| 94 |
else if (is_alpha) |
| 95 |
{ |
| 96 |
// sometime only single one line redirection appears |
| 97 |
if (!strncasecmp(data + i, "mms://", 6) |
| 98 |
|| !strncasecmp(data + i, "http://", 7)) |
| 99 |
return addURL(data + i); |
| 100 |
} |
| 101 |
continue; |
| 102 |
case OPEN_BRACKET: |
| 103 |
if (c == '!') |
| 104 |
state = IN_COMMENT; |
| 105 |
else if (c == '/') |
| 106 |
{ |
| 107 |
state = IN_ENDTAG; |
| 108 |
last_start = i + 1; |
| 109 |
} |
| 110 |
else if (is_alpha) |
| 111 |
{ |
| 112 |
state = IN_TAGNAME; |
| 113 |
last_start = i; |
| 114 |
} |
| 115 |
else if (!is_space) |
| 116 |
return false; |
| 117 |
continue; |
| 118 |
case IN_COMMENT: |
| 119 |
if (c == '"') |
| 120 |
state = IN_QUOTED_COMMENT; |
| 121 |
else if (c == '>') |
| 122 |
state = CLOSE_BRACKET; |
| 123 |
continue; |
| 124 |
case IN_QUOTED_COMMENT: |
| 125 |
if (c == '"') |
| 126 |
state = IN_COMMENT; |
| 127 |
continue; |
| 128 |
case CLOSE_BRACKET: |
| 129 |
if (c == '<') |
| 130 |
state = OPEN_BRACKET; |
| 131 |
else |
| 132 |
state = IN_CONTENTS; |
| 133 |
continue; |
| 134 |
case IN_TAGNAME: |
| 135 |
if (is_space) |
| 136 |
{ |
| 137 |
last_tag = avm::string(&data[last_start], i-last_start); |
| 138 |
//printf("TAG1: %s\n", last_tag.c_str()); |
| 139 |
state = AFTER_TAGNAME; |
| 140 |
} |
| 141 |
else if ((c == '/') || (c == '>')) |
| 142 |
{ |
| 143 |
last_tag = avm::string(&data[last_start], i-last_start); |
| 144 |
//printf("TAG2: %s\n", last_tag.c_str()); |
| 145 |
if (c == '/') |
| 146 |
state = BEFORE_TAGCLOSE; |
| 147 |
else |
| 148 |
state = CLOSE_BRACKET; |
| 149 |
} |
| 150 |
continue; |
| 151 |
case AFTER_TAGNAME: |
| 152 |
if (c == '/') |
| 153 |
state = BEFORE_TAGCLOSE; |
| 154 |
else if (!is_space) |
| 155 |
{ |
| 156 |
last_start = i; |
| 157 |
state = IN_PROPNAME; |
| 158 |
} |
| 159 |
continue; |
| 160 |
case IN_ENDTAG: |
| 161 |
if (c == '>') |
| 162 |
{ |
| 163 |
last_tag = avm::string(&data[last_start], i-last_start); |
| 164 |
state = CLOSE_BRACKET; |
| 165 |
AVM_WRITE("ASX reader", 1, " TAG: %s\n", last_tag.c_str()); |
| 166 |
if (last_tag == "asx") |
| 167 |
return true; |
| 168 |
} |
| 169 |
continue; |
| 170 |
case BEFORE_TAGCLOSE: |
| 171 |
if (c == '>') |
| 172 |
state = CLOSE_BRACKET; |
| 173 |
else if (!is_space) |
| 174 |
return false; |
| 175 |
continue; |
| 176 |
case IN_PROPNAME: |
| 177 |
if (c == '=') |
| 178 |
{ |
| 179 |
last_prop = avm::string(&data[last_start], i-last_start); |
| 180 |
state = PROP_EQUAL; |
| 181 |
} |
| 182 |
else if (is_space) |
| 183 |
{ |
| 184 |
last_prop = avm::string(&data[last_start], i-last_start); |
| 185 |
state = AFTER_PROPNAME; |
| 186 |
} |
| 187 |
continue; |
| 188 |
case AFTER_PROPNAME: |
| 189 |
if (c == '=') |
| 190 |
state = PROP_EQUAL; |
| 191 |
continue; |
| 192 |
case PROP_EQUAL: |
| 193 |
if (is_space) |
| 194 |
continue; |
| 195 |
if (c == '"') |
| 196 |
state = IN_QUOTED_PROPCONTENTS; |
| 197 |
else |
| 198 |
state = IN_PROPCONTENTS; |
| 199 |
last_start = i; |
| 200 |
continue; |
| 201 |
case IN_PROPCONTENTS: |
| 202 |
if (c == '"') |
| 203 |
state = IN_QUOTED_PROPCONTENTS; |
| 204 |
else if (is_space || (c == '/') || (c == '>')) |
| 205 |
{ |
| 206 |
last_prop_val = avm::string(&data[last_start], i-last_start); |
| 207 |
|
| 208 |
AVM_WRITE("ASX reader", 1, "VALUE: %s t:%s p:%s\n", last_prop_val.c_str(), last_tag.c_str(), last_prop.c_str()); |
| 209 |
if (last_tag.tolower() == "ref" |
| 210 |
|| last_tag == "entryref" |
| 211 |
|| last_tag == "a" |
| 212 |
|| last_tag == "location" |
| 213 |
|| last_tag == "embed") // part of some javascript |
| 214 |
if (last_prop.tolower() == "href" |
| 215 |
|| last_prop == "src") |
| 216 |
addURL(last_prop_val.c_str()); |
| 217 |
if (c == '/') |
| 218 |
state = BEFORE_TAGCLOSE; |
| 219 |
else if (c == '>') |
| 220 |
state = CLOSE_BRACKET; |
| 221 |
else |
| 222 |
state = BETWEEN_PROPS; |
| 223 |
} |
| 224 |
continue; |
| 225 |
case IN_QUOTED_PROPCONTENTS: |
| 226 |
if (c == '"') |
| 227 |
state = IN_PROPCONTENTS; |
| 228 |
continue; |
| 229 |
case BETWEEN_PROPS: |
| 230 |
if (c == '/') |
| 231 |
state = BEFORE_TAGCLOSE; |
| 232 |
else if (c == '>') |
| 233 |
state = CLOSE_BRACKET; |
| 234 |
else if (is_alpha) |
| 235 |
{ |
| 236 |
last_start = i; |
| 237 |
state = IN_PROPNAME; |
| 238 |
} |
| 239 |
else if (!is_space) |
| 240 |
return false; |
| 241 |
continue; |
| 242 |
case IN_CONTENTS: |
| 243 |
if (c == '"') |
| 244 |
state = IN_QUOTED_CONTENTS; |
| 245 |
else if (c == '<') |
| 246 |
state = OPEN_BRACKET; |
| 247 |
continue; |
| 248 |
case IN_QUOTED_CONTENTS: |
| 249 |
if (c == '"') |
| 250 |
state = IN_CONTENTS; |
| 251 |
continue; |
| 252 |
default: // after_propequal |
| 253 |
AVM_WRITE("ASX reader", "FIXME ERROR after_propequal not handled\n"); |
| 254 |
break; |
| 255 |
} |
| 256 |
} |
| 257 |
return (m_Urls.size() > 0); |
| 258 |
} |
| 259 |
|
| 260 |
bool ASXReader::getURLs(avm::vector<avm::string>& urls) const |
| 261 |
{ |
| 262 |
//for (unsigned i = 0; i < m_Urls.size(); i++) printf("URLs: %s\n", m_Urls[i].c_str()); |
| 263 |
urls = m_Urls; |
| 264 |
return true; |
| 265 |
} |
| 266 |
|
| 267 |
AVM_END_NAMESPACE; |