Skip to content

Escaping and quoting issues #34

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 2 commits into
base: master
Choose a base branch
from
Open

Escaping and quoting issues #34

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
a063b87
Always quote attribute values
SirNickolas Nov 12, 2018
e549264
Escape entities that must be escaped
SirNickolas Nov 12, 2018
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
53 changes: 30 additions & 23 deletions src/html/dom.d
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -601,14 +601,9 @@ class Node {
app.put(attr);

if (value.length) {
if (value.requiresQuotes) {
app.put("=\"");
app.writeQuotesEscaped(value);
app.put("\"");
} else {
app.put('=');
app.put(value);
}
app.put("=\"");
app.writeHTMLEscaped!(Yes.escapeQuotes)(value);
app.put("\"");
}
}

Expand All @@ -633,7 +628,7 @@ class Node {
}
break;
case Text:
app.put(tag_);
app.writeHTMLEscaped!(No.escapeQuotes)(tag_);
break;
case Comment:
app.put("<!--");
Expand Down Expand Up @@ -669,14 +664,9 @@ class Node {
app.put(attr);

if (value.length) {
if (value.requiresQuotes) {
app.put("=\"");
app.writeQuotesEscaped(value);
app.put("\"");
} else {
app.put('=');
app.put(value);
}
app.put("=\"");
app.writeHTMLEscaped!(Yes.escapeQuotes)(value);
app.put("\"");
}
}

Expand Down Expand Up @@ -768,7 +758,7 @@ class Node {
} else {
space = false;
}
app.put(ch);
app.writeHTMLEscaped!(No.escapeQuotes)(ch);
}
}
break;
Expand Down Expand Up @@ -1105,13 +1095,24 @@ unittest {
auto doc = createDocument(`<html><body>&nbsp;</body></html>`);
assert(doc.root.outerHTML == "<root><html><body>\&nbsp;</body></html></root>");
doc = createDocument!(DOMCreateOptions.None)(`<html><body>&nbsp;</body></html>`);
assert(doc.root.outerHTML == `<root><html><body>&nbsp;</body></html></root>`);
assert(doc.root.outerHTML == `<root><html><body>&amp;nbsp;</body></html></root>`);
doc = createDocument(`<script>&nbsp;</script>`);
assert(doc.root.outerHTML == `<root><script>&nbsp;</script></root>`, doc.root.outerHTML);
doc = createDocument(`<style>&nbsp;</style>`);
assert(doc.root.outerHTML == `<root><style>&nbsp;</style></root>`, doc.root.outerHTML);
}

unittest {
string src = `&lt;script&gt;alert("test&amp;");&lt;/script&gt;`;
auto doc = createDocument(src);
assert(doc.root.html == src, doc.root.html);
assert(doc.root.compactHTML == src, doc.root.compactHTML);
src = `<div title="&gt;&lt;script&gt;alert('test&amp;');&lt;/script&gt;"></div>`;
doc = createDocument(src);
assert(doc.root.html == src, doc.root.html);
assert(doc.root.compactHTML == src, doc.root.compactHTML);
}

unittest {
const doc = createDocument(`<html><body><div title='"Тест&apos;"'>"К"ириллица</div></body></html>`);
assert(doc.root.html == `<html><body><div title="&#34;Тест'&#34;">"К"ириллица</div></body></html>`);
Expand All @@ -1127,6 +1128,12 @@ unittest {
assert(doc.root.outerHTML == `<root><br><div></div></root>`, doc.root.outerHTML);
}

// quotes should be preserved as they are mandatory in SVG
unittest {
auto doc = createDocument(`<svg><text x="7" /></svg>`);
assert(doc.root.outerHTML == `<root><svg><text x="7"></text></svg></root>`, doc.root.outerHTML);
}

// toString prints elements with content as <tag attr="value"/>
unittest {
// self-closed element w/o content
Expand Down Expand Up @@ -1346,8 +1353,8 @@ private:


unittest {
const(char)[] src = `<parent attr=value><child></child>text</parent>`;
auto doc = createDocument(src);
const(char)[] src = `<parent attr="value"><child></child>text</parent>`;
auto doc = createDocument(`<parent attr=value><child></child>text</parent>`);
assert(doc.root.html == src, doc.root.html);

const(char)[] srcq = `<parent attr="v a l u e"><child></child>text</parent>`;
Expand All @@ -1366,11 +1373,11 @@ unittest {
auto child = cloned.find("child").front.clone;
child.attr("attr", "test");
cloned.find("parent").front.appendChild(child);
assert(cloned.html == `<parent attr=value><child></child>text<child attr=test></child></parent>`, cloned.html);
assert(cloned.html == `<parent attr="value"><child></child>text<child attr="test"></child></parent>`, cloned.html);
assert(doc.root.html == src, doc.root.html);

child.text = "text";
assert(cloned.html == `<parent attr=value><child></child>text<child attr=test>text</child></parent>`, cloned.html);
assert(cloned.html == `<parent attr="value"><child></child>text<child attr="test">text</child></parent>`, cloned.html);
assert(doc.root.html == src, doc.root.html);

// document cloning
Expand Down
64 changes: 16 additions & 48 deletions src/html/utils.d
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -17,28 +17,6 @@ bool isAllWhite(Char)(Char[] value) {
}


bool requiresQuotes(Char)(Char[] value) {
auto ptr = value.ptr;
const end = ptr + value.length;

while (ptr != end) {
switch (*ptr++) {
case 'a': .. case 'z':
case 'A': .. case 'Z':
case '0': .. case '9':
case '-':
case '_':
case '.':
case ':':
continue;
default:
return true;
}
}
return false;
}


bool equalsCI(CharA, CharB)(const(CharA)[] a, const(CharB)[] b) {
if (a.length == b.length) {
for (size_t i; i < a.length; ++i) {
Expand Down Expand Up @@ -79,42 +57,32 @@ hash_t tagHashOf(const(char)[] x) {
}


void writeQuotesEscaped(Appender)(ref Appender app, const(char)[] x) {
foreach (dchar ch; x) {
switch (ch) {
void writeHTMLEscaped(Flag!q{escapeQuotes} escapeQuotes, Appender)(ref Appender app, dchar ch) {
switch (ch) {
static if (escapeQuotes) {
case '"':
app.put("&#34;"); // shorter than &quot;
static assert('"' == 34);
break;
default:
app.put(ch);
break;
}
case '<':
app.put("&lt;");
break;
case '>':
app.put("&gt;");
break;
case '&':
app.put("&amp;");
break;
default:
app.put(ch);
break;
}
}


void writeHTMLEscaped(Flag!q{escapeQuotes} escapeQuotes, Appender)(ref Appender app, const(char)[] x) {
foreach (dchar ch; x) {
switch (ch) {
static if (escapeQuotes) {
case '"':
app.put("&#34;"); // shorter than &quot;
static assert('"' == 34);
break;
}
case '<':
app.put("&lt;");
break;
case '>':
app.put("&gt;");
break;
case '&':
app.put("&amp;");
break;
default:
app.put(ch);
break;
}
writeHTMLEscaped!escapeQuotes(app, ch);
}
}