int
html2text(char *text, const char *content)
{
int ret;
xmlNodePtr root;
xmlErrorPtr err;
htmlParserCtxtPtr parser;
parser = htmlCreatePushParserCtxt(NULL, NULL, NULL, 0, NULL, 0);
/* htmlCtxtUseOptions(parser, HTML_PARSE_NOBLANKS | HTML_PARSE_NOERROR | HTML_PARSE_NOWARNING | HTML_PARSE_NONET); */
htmlCtxtUseOptions(parser, HTML_PARSE_RECOVER | HTML_PARSE_NOBLANKS | HTML_PARSE_NOERROR | HTML_PARSE_NOWARNING | HTML_PARSE_NONET);
ret = htmlParseChunk(parser, content, xmlStrlen(content), 0);
if (ret != 0) {
err = xmlCtxtGetLastError(parser);
fprintf(stderr, "htmlParseChunk failure: %d: %s\n", \
ret, err->message);
}
ret = htmlParseChunk(parser, NULL, 0, 1);
if (ret != 0) {
err = xmlCtxtGetLastError(parser);
fprintf(stderr, "htmlParseChunk failure 2: %d: %s\n", \
ret, err->message);
}
root = xmlDocGetRootElement(parser->myDoc);
walkTree(parser->myDoc, root, text);
return 0;
}
/*
* call-seq:
* initialize_native(xml_sax, filename)
*
* Initialize the push parser with +xml_sax+ using +filename+
*/
static VALUE initialize_native(VALUE self, VALUE _xml_sax, VALUE _filename,
VALUE encoding)
{
htmlSAXHandlerPtr sax;
const char * filename = NULL;
htmlParserCtxtPtr ctx;
xmlCharEncoding enc = XML_CHAR_ENCODING_NONE;
Data_Get_Struct(_xml_sax, xmlSAXHandler, sax);
if(_filename != Qnil) filename = StringValuePtr(_filename);
if (!NIL_P(encoding)) {
enc = xmlParseCharEncoding(StringValuePtr(encoding));
if (enc == XML_CHAR_ENCODING_ERROR)
rb_raise(rb_eArgError, "Unsupported Encoding");
}
ctx = htmlCreatePushParserCtxt(
sax,
NULL,
NULL,
0,
filename,
enc
);
if(ctx == NULL)
rb_raise(rb_eRuntimeError, "Could not create a parser context");
ctx->userData = NOKOGIRI_SAX_TUPLE_NEW(ctx, self);
ctx->sax2 = 1;
DATA_PTR(self) = ctx;
return self;
}
static void
_unhtmlize (gchar *string, ResultBuffer *buffer)
{
htmlParserCtxtPtr ctxt;
htmlSAXHandlerPtr sax_p;
sax_p = g_new0 (htmlSAXHandler, 1);
sax_p->characters = unhtmlizeHandleCharacters;
ctxt = htmlCreatePushParserCtxt (sax_p, buffer, string, strlen (string), "", XML_CHAR_ENCODING_UTF8);
htmlParseChunk (ctxt, string, 0, 1);
htmlFreeParserCtxt (ctxt);
g_free (sax_p);
}
static void parseHtml(const std::string &html,
std::string &title)
{
htmlParserCtxtPtr ctxt;
Context context;
ctxt = htmlCreatePushParserCtxt(&saxHandler, &context, "", 0, "",
XML_CHAR_ENCODING_NONE);
htmlParseChunk(ctxt, html.c_str(), html.size(), 0);
htmlParseChunk(ctxt, "", 0, 1);
htmlFreeParserCtxt(ctxt);
title = context.title;
}
void
sami_context_init (ParserState * state)
{
GstSamiContext *context;
g_assert (state->user_data == NULL);
state->user_data = (gpointer) g_new0 (GstSamiContext, 1);
context = (GstSamiContext *) state->user_data;
context->htmlctxt = htmlCreatePushParserCtxt (samiSAXHandler, context,
"", 0, NULL, XML_CHAR_ENCODING_UTF8);
context->buf = g_string_new ("");
context->rubybuf = g_string_new ("");
context->resultbuf = g_string_new ("");
context->state = g_string_new ("");
}
void parse(const std::string &page) {
htmlParserCtxtPtr ctxt;
htmlSAXHandler handler;
memset(&handler, 0, sizeof(handler));
handler.startElement = static_parser_start_element;
handler.endElement = static_parser_end_element;
handler.characters = static_parser_characters;
ctxt = htmlCreatePushParserCtxt(&handler, this, "", 0, "", XML_CHAR_ENCODING_NONE);
htmlParseChunk(ctxt, page.c_str(), page.size(), 0);
htmlParseChunk(ctxt, "", 0, 1);
htmlFreeParserCtxt(ctxt);
}
static int create_parser(saxctxt *ctxt, const char **bufp, apr_size_t *bytes)
{
xmlCharEncoding enc;
#ifndef USE_OLD_LIBXML2
int xmlopts = XML_PARSE_RECOVER | XML_PARSE_NONET |
XML_PARSE_NOBLANKS | XML_PARSE_NOERROR | XML_PARSE_NOWARNING;
#endif
enc = sniff_encoding(ctxt, *bufp, *bytes);
ctxt->parser = htmlCreatePushParserCtxt(&sax, ctxt, *bufp, 0, 0, enc);
if (ctxt->parser == NULL)
return 0;
apr_pool_cleanup_register
(ctxt->f->r->pool, ctxt->parser, (int(*)(void*))htmlFreeParserCtxt,
apr_pool_cleanup_null);
#ifndef USE_OLD_LIBXML2
if (xmlopts = xmlCtxtUseOptions(ctxt->parser, xmlopts), xmlopts)
ap_log_error(APLOG_MARK, APLOG_WARNING, APR_SUCCESS, ctxt->f->r,
"create_parser: unsupported parser opts %x", xmlopts);
#endif
return 1;
}
void se_parser::parse_output(char *output,
std::vector<search_snippet*> *snippets,
const int &count_offset) throw (sp_exception)
{
_count = count_offset;
htmlParserCtxtPtr ctxt = NULL;
parser_context pc;
pc._parser = this;
pc._snippets = snippets;
pc._current_snippet = NULL;
htmlSAXHandler saxHandler =
{
NULL,
NULL,
NULL,
NULL,
NULL,
NULL,
NULL,
NULL,
NULL,
NULL,
NULL,
NULL,
NULL,
NULL,
start_element_wrapper,
end_element_wrapper,
NULL,
characters_wrapper,
NULL,
NULL,
NULL,
NULL,
NULL,
NULL,
NULL,
cdata_wrapper,
NULL,
NULL,
NULL,
NULL,
NULL,
NULL
};
//mutex_lock(&se_parser::_se_parser_mutex);
int status = 0;
try
{
ctxt = htmlCreatePushParserCtxt(&saxHandler, &pc, "", 0, "",
XML_CHAR_ENCODING_UTF8); // encoding here.
htmlCtxtUseOptions(ctxt,HTML_PARSE_NOERROR);
status = htmlParseChunk(ctxt,output,strlen(output),0);
}
catch (std::exception e)
{
errlog::log_error(LOG_LEVEL_PARSER,"Error %s in xml/html parsing of search results.",
e.what());
throw sp_exception(WB_ERR_PARSE,e.what());
}
catch (...) // catch everything else to avoid crashes.
{
std::string msg = "Unknown error in xml/html parsing of search results";
errlog::log_error(LOG_LEVEL_PARSER,msg.c_str());
throw sp_exception(WB_ERR_PARSE,msg);
}
if (status == 0)
{
if (ctxt)
xmlFreeParserCtxt(ctxt);
//mutex_unlock(&se_parser::_se_parser_mutex);
}
else // an error occurred.
{
xmlErrorPtr xep = xmlCtxtGetLastError(ctxt);
if (xep)
{
std::string err_msg = std::string(xep->message);
miscutil::replace_in_string(err_msg,"\n","");
errlog::log_error(LOG_LEVEL_PARSER, "html level parsing error (libxml2): %s",
err_msg.c_str());
// check on error level.
if (xep->level == 3) // fatal or recoverable error.
{
std::string msg = "libxml2 fatal error";
errlog::log_error(LOG_LEVEL_PARSER,msg.c_str());
if (ctxt)
xmlFreeParserCtxt(ctxt);
//mutex_unlock(&se_parser::_se_parser_mutex);
throw sp_exception(WB_ERR_PARSE,msg);
}
// XXX: too verbose, and confusing to users.
else if (xep->level == 2)
{
std::string msg = "libxml2 recoverable error";
errlog::log_error(LOG_LEVEL_DEBUG,msg.c_str());
if (ctxt)
xmlFreeParserCtxt(ctxt);
//mutex_unlock(&se_parser::_se_parser_mutex);
//throw sp_exception(WB_ERR_PARSE,msg);
}
}
}
}
void
html_file_open (G_GNUC_UNUSED GOFileOpener const *fo, GOIOContext *io_context,
WorkbookView *wb_view, GsfInput *input)
{
guint8 const *buf;
gsf_off_t size;
int len, bomlen;
htmlParserCtxtPtr ctxt;
htmlDocPtr doc = NULL;
xmlCharEncoding enc;
GnmHtmlTableCtxt tc;
g_return_if_fail (input != NULL);
if (gsf_input_seek (input, 0, G_SEEK_SET))
return;
size = gsf_input_size (input);
if (size >= 4) {
size -= 4;
buf = gsf_input_read (input, 4, NULL);
if (buf != NULL) {
enc = xmlDetectCharEncoding(buf, 4);
switch (enc) { /* Skip byte order mark */
case XML_CHAR_ENCODING_UCS4BE:
case XML_CHAR_ENCODING_UCS4LE:
case XML_CHAR_ENCODING_UCS4_2143:
case XML_CHAR_ENCODING_UCS4_3412:
case XML_CHAR_ENCODING_EBCDIC:
bomlen = 4;
break;
case XML_CHAR_ENCODING_UTF16BE:
case XML_CHAR_ENCODING_UTF16LE:
bomlen = 2;
break;
case XML_CHAR_ENCODING_UTF8:
if (buf[0] == 0xef)
bomlen = 3;
else if (buf[0] == 0x3c)
bomlen = 4;
else
bomlen = 0;
break;
case XML_CHAR_ENCODING_NONE:
bomlen = 0;
/* Try to detect unmarked UTF16LE
(Firefox Windows clipboard, drag data all platforms) */
if ((buf[0] >= 0x20 || g_ascii_isspace(buf[0])) &&
buf[1] == 0 &&
(buf[2] >= 0x20 || g_ascii_isspace(buf[2])) &&
buf[3] == 0)
enc = XML_CHAR_ENCODING_UTF16LE;
break;
default:
bomlen = 0;
}
ctxt = htmlCreatePushParserCtxt (
NULL, NULL, (char const *)(buf + bomlen),
4 - bomlen, gsf_input_name (input), enc);
for (; size > 0 ; size -= len) {
len = MIN (4096, size);
buf = gsf_input_read (input, len, NULL);
if (buf == NULL)
break;
htmlParseChunk (
ctxt, (char const *)buf, len, 0);
}
htmlParseChunk (ctxt, (char const *)buf, 0, 1);
doc = ctxt->myDoc;
htmlFreeParserCtxt (ctxt);
}
}
if (doc != NULL) {
xmlNodePtr ptr;
tc.sheet = NULL;
tc.row = -1;
tc.wb_view = wb_view;
for (ptr = doc->children; ptr != NULL ; ptr = ptr->next)
html_search_for_tables (ptr, doc, wb_view, &tc);
xmlFreeDoc (doc);
} else
go_io_error_info_set (io_context,
go_error_info_new_str (_("Unable to parse the html.")));
}
void HttpPullData::parseHtml(const std::string &html,int pos,PARSE_ACTION action){
htmlParserCtxtPtr ctxt;
Context context;
ctxt = htmlCreatePushParserCtxt(&saxHandler, &context, "", 0, "",XML_CHAR_ENCODING_NONE);
htmlParseChunk(ctxt,html.c_str()+pos,max((unsigned long)0,html.size()-pos), 0);
htmlParseChunk(ctxt, "", 0, 1);
htmlFreeParserCtxt(ctxt);
switch(action){
case GET_OPTIONS:
options=context.options;
break;
case GET_OZONE:
ozone=-1;
for(unsigned int i=0;i<context.tableText.size();i++){
if(context.tableText[i].find(" DU",0)>=0){
ozone=atof(StringTools::stripString(context.tableText[i]," DU").c_str());
break;
}
}
break;
case GET_LOCATION:
for(int i=0;i<context.boldText.size();i++){
if(context.boldText[i]=="Ecoregion "){
info.ecoregion=context.boldText[i+1];
} else if(context.boldText[i]=="Latitude"){
info.latitude=atof(StringTools::stripString(context.boldText[i+1]," N").c_str());
} else if(context.boldText[i]=="Longitude"){
if((int)context.boldText[i+1].find(" W")>=0){
info.longitude=atof(StringTools::stripString(context.boldText[i+1]," W").c_str());
} else if((int)context.boldText[i+1].find(" E")>=0){
info.longitude=-atof(StringTools::stripString(context.boldText[i+1]," E").c_str());
}
} else if(context.boldText[i]=="Elevation"){
info.elevation=atof(context.boldText[i+1].c_str());
}
}
case GET_AEROSOL_VISIBLE:
for(int i=0;i<context.tableText.size();i++){
if(context.tableText[i]==" Morning "){
aerosols.depthsAM[2]=StringTools::stof(context.tableText[i+1]);
aerosols.depthsAM[3]=StringTools::stof(context.tableText[i+2]);
aerosols.depthsAM[4]=StringTools::stof(context.tableText[i+3]);
aerosols.depthsAM[5]=StringTools::stof(context.tableText[i+4]);
aerosols.depthsAM[6]=StringTools::stof(context.tableText[i+5]);
}
if(context.tableText[i]==" Afternoon "){
aerosols.depthsPM[2]=StringTools::stof(context.tableText[i+1]);
aerosols.depthsPM[3]=StringTools::stof(context.tableText[i+2]);
aerosols.depthsPM[4]=StringTools::stof(context.tableText[i+3]);
aerosols.depthsPM[5]=StringTools::stof(context.tableText[i+4]);
aerosols.depthsPM[6]=StringTools::stof(context.tableText[i+5]);
}
}
break;
case GET_AEROSOL_UV:
for(int i=0;i<context.tableText.size();i++){
if(context.tableText[i]==" Morning "){
aerosols.depthsAM[0]=StringTools::stof(context.tableText[i+1]);
aerosols.depthsAM[1]=StringTools::stof(context.tableText[i+2]);
}
if(context.tableText[i]==" Afternoon "){
aerosols.depthsPM[0]=StringTools::stof(context.tableText[i+1]);
aerosols.depthsPM[1]=StringTools::stof(context.tableText[i+2]);
}
}
break;
default:
break;
}
}
