potok/tinyxml2
1
0
Fork 0
Code Pull Requests Packages Projects Releases Wiki Activity

Merge branch 'pedantic-whitespace' of https://github.com/kcsaul/tinyxml2 into kcsaul-pedantic-whitespace

Browse Source
This commit is contained in:
Lee Thomason
2023-11-21 11:40:52 -08:00
parent 4c27b08f62 98303bbda3
commit 9d026527a4
3 changed files with 192 additions and 7 deletions
Download Patch File Download Diff File Expand all files Collapse all files

18
tinyxml2.cpp
View File

@@ -715,7 +715,7 @@ bool XMLUtil::ToUnsigned64(const char* str, uint64_t* value) {
} }
char* XMLDocument::Identify( char* p, XMLNode** node ) char* XMLDocument::Identify( char* p, XMLNode** node, bool first )
{ {
TIXMLASSERT( node ); TIXMLASSERT( node );
TIXMLASSERT( p ); TIXMLASSERT( p );
@@ -767,10 +767,20 @@ char* XMLDocument::Identify( char* p, XMLNode** node )
p += dtdHeaderLen; p += dtdHeaderLen;
} }
else if ( XMLUtil::StringEqual( p, elementHeader, elementHeaderLen ) ) { else if ( XMLUtil::StringEqual( p, elementHeader, elementHeaderLen ) ) {
returnNode = CreateUnlinkedNode<XMLElement>( _elementPool );
// Preserve whitespace pedantically before closing tag, when it's immediately after opening tag
if (WhitespaceMode() == PEDANTIC_WHITESPACE && first && p != start && *(p + elementHeaderLen) == '/') {
returnNode = CreateUnlinkedNode<XMLText>(_textPool);
returnNode->_parseLineNum = startLine;
p = start; // Back it up, all the text counts.
_parseCurLineNum = startLine;
}
else {
returnNode = CreateUnlinkedNode<XMLElement>(_elementPool);
returnNode->_parseLineNum = _parseCurLineNum; returnNode->_parseLineNum = _parseCurLineNum;
p += elementHeaderLen; p += elementHeaderLen;
} }
}
else { else {
returnNode = CreateUnlinkedNode<XMLText>( _textPool ); returnNode = CreateUnlinkedNode<XMLText>( _textPool );
returnNode->_parseLineNum = _parseCurLineNum; // Report line of first non-whitespace character returnNode->_parseLineNum = _parseCurLineNum; // Report line of first non-whitespace character
@@ -1098,14 +1108,16 @@ char* XMLNode::ParseDeep( char* p, StrPair* parentEndTag, int* curLineNumPtr )
if (_document->Error()) if (_document->Error())
return 0; return 0;
bool first = true;
while( p && *p ) { while( p && *p ) {
XMLNode* node = 0; XMLNode* node = 0;
p = _document->Identify( p, &node ); p = _document->Identify( p, &node, first );
TIXMLASSERT( p ); TIXMLASSERT( p );
if ( node == 0 ) { if ( node == 0 ) {
break; break;
} }
first = false;
const int initialLineNum = node->_parseLineNum; const int initialLineNum = node->_parseLineNum;

5
tinyxml2.h
View File

@@ -1710,7 +1710,8 @@ private:
enum Whitespace { enum Whitespace {
PRESERVE_WHITESPACE, PRESERVE_WHITESPACE,
COLLAPSE_WHITESPACE COLLAPSE_WHITESPACE,
PEDANTIC_WHITESPACE
}; };
@@ -1921,7 +1922,7 @@ public:
void DeepCopy(XMLDocument* target) const; void DeepCopy(XMLDocument* target) const;
// internal // internal
char* Identify( char* p, XMLNode** node ); char* Identify( char* p, XMLNode** node, bool first );
// internal // internal
void MarkInUse(const XMLNode* const); void MarkInUse(const XMLNode* const);

172
xmltest.cpp
View File

@@ -1869,6 +1869,178 @@ int main( int argc, const char ** argv )
XMLTest( "Whitespace all space", true, 0 == doc.FirstChildElement()->FirstChild() ); XMLTest( "Whitespace all space", true, 0 == doc.FirstChildElement()->FirstChild() );
} }
// ----------- Preserve Whitespace ------------
{
const char* xml = "<element>This is &apos; \n\n text &apos;</element>";
XMLDocument doc(true, PRESERVE_WHITESPACE);
doc.Parse(xml);
XMLTest("Parse with whitespace preserved", false, doc.Error());
XMLTest("Whitespace preserved", "This is ' \n\n text '", doc.FirstChildElement()->GetText());
}
{
const char* xml = "<element> This \nis &apos; text &apos; </element>";
XMLDocument doc(true, PRESERVE_WHITESPACE);
doc.Parse(xml);
XMLTest("Parse with whitespace preserved", false, doc.Error());
XMLTest("Whitespace preserved", " This \nis ' text ' ", doc.FirstChildElement()->GetText());
}
{
const char* xml = "<element> \n This is &apos; text &apos; \n</element>";
XMLDocument doc(true, PRESERVE_WHITESPACE);
doc.Parse(xml);
XMLTest("Parse with whitespace preserved", false, doc.Error());
XMLTest("Whitespace preserved", " \n This is ' text ' \n", doc.FirstChildElement()->GetText());
}
// Following cases are for text that is all whitespace which are not preserved intentionally
{
const char* xml = "<element> </element>";
XMLDocument doc(true, PRESERVE_WHITESPACE);
doc.Parse(xml);
XMLTest("Parse with whitespace preserved", false, doc.Error());
XMLTest("Whitespace preserved", true, 0 == doc.FirstChildElement()->GetText());
}
{
const char* xml = "<element> </element>";
XMLDocument doc(true, PRESERVE_WHITESPACE);
doc.Parse(xml);
XMLTest("Parse with whitespace preserved", false, doc.Error());
XMLTest("Whitespace preserved", true, 0 == doc.FirstChildElement()->GetText());
}
{
const char* xml = "<element>\n\n</element>";
XMLDocument doc(true, PRESERVE_WHITESPACE);
doc.Parse(xml);
XMLTest("Parse with whitespace preserved", false, doc.Error());
XMLTest("Whitespace preserved", true, 0 == doc.FirstChildElement()->GetText());
}
{
const char* xml = "<element> \n</element>";
XMLDocument doc(true, PRESERVE_WHITESPACE);
doc.Parse(xml);
XMLTest("Parse with whitespace preserved", false, doc.Error());
XMLTest("Whitespace preserved", true, 0 == doc.FirstChildElement()->GetText());
}
{
const char* xml = "<element> \n \n </element>";
XMLDocument doc(true, PRESERVE_WHITESPACE);
doc.Parse(xml);
XMLTest("Parse with whitespace preserved", false, doc.Error());
XMLTest("Whitespace preserved", true, 0 == doc.FirstChildElement()->GetText());
}
// ----------- Pedantic Whitespace ------------
{
const char* xml = "<element>This is &apos; \n\n text &apos;</element>";
XMLDocument doc(true, PEDANTIC_WHITESPACE);
doc.Parse(xml);
XMLTest("Parse with pedantic whitespace", false, doc.Error());
XMLTest("Pedantic whitespace", "This is ' \n\n text '", doc.FirstChildElement()->GetText());
}
{
const char* xml = "<element> This \nis &apos; text &apos; </element>";
XMLDocument doc(true, PEDANTIC_WHITESPACE);
doc.Parse(xml);
XMLTest("Parse with pedantic whitespace", false, doc.Error());
XMLTest("Pedantic whitespace", " This \nis ' text ' ", doc.FirstChildElement()->GetText());
}
{
const char* xml = "<element> \n This is &apos; text &apos; \n</element>";
XMLDocument doc(true, PEDANTIC_WHITESPACE);
doc.Parse(xml);
XMLTest("Parse with pedantic whitespace", false, doc.Error());
XMLTest("Pedantic whitespace", " \n This is ' text ' \n", doc.FirstChildElement()->GetText());
}
// Following cases are for text that is all whitespace which is preserved with pedantic mode
{
const char* xml = "<element> </element>";
XMLDocument doc(true, PEDANTIC_WHITESPACE);
doc.Parse(xml);
XMLTest("Parse with pedantic whitespace", false, doc.Error());
XMLTest("Pedantic whitespace", " ", doc.FirstChildElement()->GetText());
}
{
const char* xml = "<element> </element>";
XMLDocument doc(true, PEDANTIC_WHITESPACE);
doc.Parse(xml);
XMLTest("Parse with pedantic whitespace", false, doc.Error());
XMLTest("Pedantic whitespace", " ", doc.FirstChildElement()->GetText());
}
{
const char* xml = "<element>\n\n</element>\n";
XMLDocument doc(true, PEDANTIC_WHITESPACE);
doc.Parse(xml);
XMLTest("Parse with pedantic whitespace", false, doc.Error());
XMLTest("Pedantic whitespace", "\n\n", doc.FirstChildElement()->GetText());
}
{
const char* xml = "<element> \n</element> \n ";
XMLDocument doc(true, PEDANTIC_WHITESPACE);
doc.Parse(xml);
XMLTest("Parse with pedantic whitespace", false, doc.Error());
XMLTest("Pedantic whitespace", " \n", doc.FirstChildElement()->GetText());
}
{
const char* xml = "<element> \n \n </element> ";
XMLDocument doc(true, PEDANTIC_WHITESPACE);
doc.Parse(xml);
XMLTest("Parse with pedantic whitespace", false, doc.Error());
XMLTest("Pedantic whitespace", " \n \n ", doc.FirstChildElement()->GetText());
}
// Following cases are for checking nested elements are still parsed with pedantic whitespace
{
const char* xml = "<element>\n\t<a> This is nested text </a>\n</element> ";
XMLDocument doc(true, PEDANTIC_WHITESPACE);
doc.Parse(xml);
XMLTest("Parse nested elements with pedantic whitespace", false, doc.Error());
XMLTest("Pedantic whitespace", " This is nested text ", doc.RootElement()->FirstChildElement()->GetText());
}
{
const char* xml = "<element> <b> </b> </element>\n";
XMLDocument doc(true, PEDANTIC_WHITESPACE);
doc.Parse(xml);
XMLTest("Parse nested elements with pedantic whitespace", false, doc.Error());
XMLTest("Pedantic whitespace", " ", doc.RootElement()->FirstChildElement()->GetText());
}
{
const char* xml = "<element> <c attribute=\"test\"/> </element>\n ";
XMLDocument doc(true, PEDANTIC_WHITESPACE);
doc.Parse(xml);
XMLTest("Parse nested elements with pedantic whitespace", false, doc.Error());
XMLTest("Pedantic whitespace", true, 0 == doc.RootElement()->FirstChildElement()->GetText());
}
// Check sample xml can be parsed with pedantic mode
{
XMLDocument doc(true, PEDANTIC_WHITESPACE);
doc.LoadFile("resources/dream.xml");
XMLTest("Load dream.xml with pedantic whitespace mode", false, doc.Error());
XMLTest("Dream", "xml version=\"1.0\"",
doc.FirstChild()->ToDeclaration()->Value());
XMLTest("Dream", true, doc.FirstChild()->NextSibling()->ToUnknown() != 0);
XMLTest("Dream", "DOCTYPE PLAY SYSTEM \"play.dtd\"",
doc.FirstChild()->NextSibling()->ToUnknown()->Value());
XMLTest("Dream", "And Robin shall restore amends.",
doc.LastChild()->LastChild()->LastChild()->LastChild()->LastChildElement()->GetText());
}
{ {
// An assert should not fire. // An assert should not fire.
const char* xml = "<element/>"; const char* xml = "<element/>";