Commit 40f1ab72 authored by jan.koester's avatar jan.koester
Browse files

html4 detection is working

parent e85ba366

src/html.cpp

+23 −9
Original line number Diff line number Diff line
@@ -1390,8 +1390,23 @@ bool libhtmlpp::HtmlPage::isHtml5(){ 


void libhtmlpp::HtmlPage::_CheckHeader(const HtmlString &page){

    const char type[] = { '!','D','O','C','T','Y','P','E' };



    int i = 0;



    bool start=false;



    do{

        ++i;

        if(page[i]== '<'){

            if(start==true){

                HTMLException excp;

                excp[HTMLException::Critical] << "Wrong Header arborting";

                throw excp;

            }

            start=true;

        }

    } while ( page[i]== '<' || page[i]== '!' || page[i] == ' ');



    if (page.size() < 8) {

        HTMLException excp;

        excp[HTMLException::Critical] << "No Doctype found arborting";
@@ -1411,7 +1426,7 @@ void libhtmlpp::HtmlPage::_CheckHeader(const HtmlString &page){ 
        ++i;

    }while (page[i] == ' ');



    const char doctype[] = { 'h','t','m','l' };

    const char doctype[] = { 'H','T','M','L' };

    size_t tpvl = 4;



    const char typevalue4[] = {'P','U','B','L','I','C'};
@@ -1419,32 +1434,31 @@ void libhtmlpp::HtmlPage::_CheckHeader(const HtmlString &page){ 


    if ((i + tpvl) > page.size()) {

        HTMLException excp;

        excp[HTMLException::Critical] << "Doctype header broken or wrong type";

        excp[HTMLException::Critical] << "Document to short broken !";

        throw excp;

    }



    int ii=0;

    size_t ii=0;



    while ( ii < tpvl) {

        if (tolower(page[i]) !=  tolower(doctype[ii])) {

        if (tolower(page[i++]) !=  tolower(doctype[ii++])) {

            HTMLException excp;

            excp[HTMLException::Critical] << "Doctype header broken or wrong type";

            throw excp;

        }

        ++i;

    }



    ii=0;



    do{

        ++i;

    } while (page[i] == ' ');



    ii=0;



    bool html4=true;



    if(i +tpvl4  <page.size()){

        while(ii < tpvl4){

            if (tolower(page[i]) !=  tolower(typevalue4[ii])) {

            if (tolower(page[i++]) !=  tolower(typevalue4[ii++])) {

                html4=false;

            }

        }

test/htmlfiles/html4.html

0 → 100644
+10 −0
Original line number Diff line number Diff line 
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">

<html>

<head>

    <title>My HTML4 Document</title>

</head>

<body>

    <h1>This is the main heading</h1>

    <p>This is a paragraph of text.</p>

</body>

</html>