X-Git-Url: https://git.saurik.com/wxWidgets.git/blobdiff_plain/01dba85a7fc9a1103f866c7c3a5cc72f21c0784e..cbcf084b0501737b4450d015853282d6193eb6f1:/src/html/htmlpars.cpp diff --git a/src/html/htmlpars.cpp b/src/html/htmlpars.cpp index 5c07054f58..ae752eafba 100644 --- a/src/html/htmlpars.cpp +++ b/src/html/htmlpars.cpp @@ -8,30 +8,62 @@ ///////////////////////////////////////////////////////////////////////////// -#ifdef __GNUG__ -#pragma implementation +#if defined(__GNUG__) && !defined(NO_GCC_PRAGMA) +#pragma implementation "htmlpars.h" #endif #include "wx/wxprec.h" #include "wx/defs.h" -#if wxUSE_HTML +#if wxUSE_HTML && wxUSE_STREAMS -#ifdef __BORDLANDC__ +#ifdef __BORLANDC__ #pragma hdrstop #endif #ifndef WXPRECOMP -#include "wx/wx.h" + #include "wx/log.h" + #include "wx/intl.h" #endif #include "wx/tokenzr.h" #include "wx/wfstream.h" #include "wx/url.h" +#include "wx/fontmap.h" #include "wx/html/htmldefs.h" #include "wx/html/htmlpars.h" +#include "wx/dynarray.h" +#include "wx/arrimpl.cpp" +// DLL options compatibility check: +#include "wx/app.h" +WX_CHECK_BUILD_OPTIONS("wxHTML") + +//----------------------------------------------------------------------------- +// wxHtmlParser helpers +//----------------------------------------------------------------------------- + +class wxHtmlTextPiece +{ +public: + wxHtmlTextPiece(int pos, int lng) : m_pos(pos), m_lng(lng) {} + int m_pos, m_lng; +}; + +WX_DECLARE_OBJARRAY(wxHtmlTextPiece, wxHtmlTextPieces); +WX_DEFINE_OBJARRAY(wxHtmlTextPieces); + +class wxHtmlParserState +{ +public: + wxHtmlTag *m_curTag; + wxHtmlTag *m_tags; + wxHtmlTextPieces *m_textPieces; + int m_curTextPiece; + wxString m_source; + wxHtmlParserState *m_nextState; +}; //----------------------------------------------------------------------------- // wxHtmlParser @@ -39,82 +71,256 @@ IMPLEMENT_ABSTRACT_CLASS(wxHtmlParser,wxObject) +wxHtmlParser::wxHtmlParser() + : wxObject(), m_HandlersHash(wxKEY_STRING), + m_FS(NULL), m_HandlersStack(NULL) +{ + m_entitiesParser = new wxHtmlEntitiesParser; + m_Tags = NULL; + m_CurTag = NULL; + m_TextPieces = NULL; + m_CurTextPiece = 0; + m_SavedStates = NULL; +} + +wxHtmlParser::~wxHtmlParser() +{ + while (RestoreState()) {} + DestroyDOMTree(); + + if (m_HandlersStack) + { + wxList& tmp = *m_HandlersStack; + wxList::iterator it, en; + for( it = tmp.begin(), en = tmp.end(); it != en; ++it ) + delete (wxHashTable*)*it; + tmp.clear(); + } + delete m_HandlersStack; + m_HandlersHash.Clear(); + WX_CLEAR_LIST(wxList, m_HandlersList); + delete m_entitiesParser; +} wxObject* wxHtmlParser::Parse(const wxString& source) { - wxObject *result; - InitParser(source); DoParsing(); - result = GetProduct(); + wxObject *result = GetProduct(); DoneParser(); return result; } - - void wxHtmlParser::InitParser(const wxString& source) { - m_Source = source; - m_Cache = new wxHtmlTagsCache(m_Source); + SetSource(source); + m_stopParsing = FALSE; } - - - + void wxHtmlParser::DoneParser() { - delete m_Cache; - m_Cache = NULL; + DestroyDOMTree(); } +void wxHtmlParser::SetSource(const wxString& src) +{ + DestroyDOMTree(); + m_Source = src; + CreateDOMTree(); + m_CurTag = NULL; + m_CurTextPiece = 0; +} - -#define wxHTML_MAX_BUFLEN 1024 - -void wxHtmlParser::DoParsing(int begin_pos, int end_pos) +void wxHtmlParser::CreateDOMTree() { - char temp[wxHTML_BUFLEN], c; - int i; - int templen; + wxHtmlTagsCache cache(m_Source); + m_TextPieces = new wxHtmlTextPieces; + CreateDOMSubTree(NULL, 0, m_Source.Length(), &cache); + m_CurTextPiece = 0; +} - templen = 0; - i = begin_pos; +extern bool wxIsCDATAElement(const wxChar *tag); - while (i < end_pos) { - c = m_Source[(unsigned int) i]; +void wxHtmlParser::CreateDOMSubTree(wxHtmlTag *cur, + int begin_pos, int end_pos, + wxHtmlTagsCache *cache) +{ + if (end_pos <= begin_pos) return; + + wxChar c; + int i = begin_pos; + int textBeginning = begin_pos; + + // If the tag contains CDATA text, we include the text between beginning + // and ending tag verbosely. Setting i=end_pos will skip to the very + // end of this function where text piece is added, bypassing any child + // tags parsing (CDATA element can't have child elements by definition): + if (cur != NULL && wxIsCDATAElement(cur->GetName().c_str())) + { + i = end_pos; + } - // continue building word: - if (c != '<') { - temp[templen++] = c; - if (templen == wxHTML_BUFLEN-1) { - temp[templen] = 0; - AddText(temp); - templen = 0; + while (i < end_pos) + { + c = m_Source.GetChar(i); + + if (c == wxT('<')) + { + // add text to m_TextPieces: + if (i - textBeginning > 0) + m_TextPieces->Add( + wxHtmlTextPiece(textBeginning, i - textBeginning)); + + // if it is a comment, skip it: + if (i < end_pos-6 && m_Source.GetChar(i+1) == wxT('!') && + m_Source.GetChar(i+2) == wxT('-') && + m_Source.GetChar(i+3) == wxT('-')) + { + // Comments begin with "