X-Git-Url: https://git.saurik.com/wxWidgets.git/blobdiff_plain/e90c1d2a19361551eb07778280f22be3e759cf64..e58989615283771efcaa9e798568772ef5fcfb0c:/src/html/htmlfilt.cpp diff --git a/src/html/htmlfilt.cpp b/src/html/htmlfilt.cpp index c0a56d9891..7e1dc3c178 100644 --- a/src/html/htmlfilt.cpp +++ b/src/html/htmlfilt.cpp @@ -9,24 +9,56 @@ #ifdef __GNUG__ -#pragma implementation +#pragma implementation "htmlfilt.h" #endif #include "wx/wxprec.h" -#if wxUSE_HTML +#include "wx/defs.h" +#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/strconv.h" #include "wx/html/htmlfilt.h" #include "wx/html/htmlwin.h" +// utility function: read a wxString from a wxInputStream +static void ReadString(wxString& str, wxInputStream* s, wxMBConv& conv) +{ + size_t streamSize = s->GetSize(); + + if (streamSize == ~(size_t)0) + { + const size_t bufSize = 4095; + char buffer[bufSize+1]; + size_t lastRead; + + do + { + s->Read(buffer, bufSize); + lastRead = s->LastRead(); + buffer[lastRead] = 0; + str.Append(wxString(buffer, conv)); + } + while (lastRead == bufSize); + } + else + { + char* src = new char[streamSize+1]; + s->Read(src, streamSize); + src[streamSize] = 0; + str = wxString(src, conv); + delete[] src; + } +} /* @@ -53,19 +85,15 @@ bool wxHtmlFilterPlainText::CanRead(const wxFSFile& WXUNUSED(file)) const wxString wxHtmlFilterPlainText::ReadFile(const wxFSFile& file) const { wxInputStream *s = file.GetStream(); - char *src; wxString doc, doc2; if (s == NULL) return wxEmptyString; - src = new char[s -> GetSize()+1]; - src[s -> GetSize()] = 0; - s -> Read(src, s -> GetSize()); - doc = src; - delete [] src; - - doc.Replace("<", "<", TRUE); - doc.Replace(">", ">", TRUE); - doc2 = "
\n" + doc + "\n
"; + ReadString(doc, s, wxConvISO8859_1); + + doc.Replace(wxT("&"), wxT("&"), TRUE); + doc.Replace(wxT("<"), wxT("<"), TRUE); + doc.Replace(wxT(">"), wxT(">"), TRUE); + doc2 = wxT("
\n") + doc + wxT("\n
"); return doc2; } @@ -93,33 +121,25 @@ IMPLEMENT_DYNAMIC_CLASS(wxHtmlFilterImage, wxHtmlFilter) bool wxHtmlFilterImage::CanRead(const wxFSFile& file) const { - return (file.GetMimeType().Left(6) == "image/"); + return (file.GetMimeType().Left(6) == wxT("image/")); } wxString wxHtmlFilterImage::ReadFile(const wxFSFile& file) const { - return (""); + wxString res = wxT(""); + return res; } //-------------------------------------------------------------------------------- -// wxHtmlFilterPlainText -// filter for text/plain or uknown +// wxHtmlFilterHTML +// filter for text/html //-------------------------------------------------------------------------------- -class wxHtmlFilterHTML : public wxHtmlFilter -{ - DECLARE_DYNAMIC_CLASS(wxHtmlFilterHTML) - - public: - virtual bool CanRead(const wxFSFile& file) const; - virtual wxString ReadFile(const wxFSFile& file) const; -}; - IMPLEMENT_DYNAMIC_CLASS(wxHtmlFilterHTML, wxHtmlFilter) @@ -129,7 +149,7 @@ bool wxHtmlFilterHTML::CanRead(const wxFSFile& file) const // This is true in most case but some page can return: // "text/html; char-encoding=...." // So we use Find instead - return (file.GetMimeType().Find(T("text/html")) == 0); + return (file.GetMimeType().Find(wxT("text/html")) == 0); } @@ -137,15 +157,50 @@ bool wxHtmlFilterHTML::CanRead(const wxFSFile& file) const wxString wxHtmlFilterHTML::ReadFile(const wxFSFile& file) const { wxInputStream *s = file.GetStream(); - char *src; wxString doc; - if (s == NULL) return wxEmptyString; - src = new char[s -> GetSize() + 1]; - src[s -> GetSize()] = 0; - s -> Read(src, s -> GetSize()); - doc = src; - delete[] src; + if (s == NULL) + { + wxLogError(_("Cannot open HTML document: %s"), file.GetLocation().c_str()); + return wxEmptyString; + } + + // NB: We convert input file to wchar_t here in Unicode mode, based on + // either Content-Type header or tags. In ANSI mode, we don't + // do it as it is done by wxHtmlParser (for this reason, we add + // tag if we used Content-Type header). +#if wxUSE_UNICODE + int charsetPos; + if ((charsetPos = file.GetMimeType().Find(_T("; charset="))) != wxNOT_FOUND) + { + wxString charset = file.GetMimeType().Mid(charsetPos + 10); + wxCSConv conv(charset); + ReadString(doc, s, conv); + } + else + { + wxString tmpdoc; + ReadString(tmpdoc, s, wxConvISO8859_1); + wxString charset = wxHtmlParser::ExtractCharsetInformation(tmpdoc); + if (charset.empty()) + doc = tmpdoc; + else + { + wxCSConv conv(charset); + doc = wxString(tmpdoc.mb_str(wxConvISO8859_1), conv); + } + } +#else // !wxUSE_UNICODE + ReadString(doc, s, wxConvLibc); + // add meta tag if we obtained this through http: + if (!file.GetMimeType().empty()) + { + wxString hdr; + wxString mime = file.GetMimeType(); + hdr.Printf(_T(""), mime.c_str()); + return hdr+doc; + } +#endif return doc; }