Skip to content

Added utility functions to print the dom as html #18

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 1 commit into
base: master
Choose a base branch
from
Open

Added utility functions to print the dom as html #18

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
68 changes: 66 additions & 2 deletions lib/htmlparser.js
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -686,8 +686,11 @@ function DefaultHandler (callback, options) {
var pos = this._tagStack.length - 1;
while (pos > -1 && this._tagStack[pos--].name != baseName) { }
if (pos > -1 || this._tagStack[0].name == baseName)
while (pos < this._tagStack.length - 1)
this._tagStack.pop();
var elem;
while (pos < this._tagStack.length - 1) {
elem = this._tagStack.pop();
}
elem.closed = true;
}
}
else { //This is not a closing tag
Expand Down Expand Up @@ -799,6 +802,67 @@ function DefaultHandler (callback, options) {
, getElementsByTagType: function DomUtils$getElementsByTagType (type, currentElement, recurse, limit) {
return(DomUtils.getElements({ tag_type: type }, currentElement, recurse, limit));
}

, printHtml: function DomUtils$print(nodes, output, encoding) {
for (var idx in nodes) {
var node = nodes[idx];
var text = node.data;
if (text === null || text === undefined) {
// need to reconstruct from name and attributes
text = node.name;
for (var attrName in node.attribs) {
var attrVal = node.attribs[attrName];
text += ' ' + attrName;
if (attrVal !== null && attrVal !== undefined) {
if (attrVal.match(/[^\\]"/)) {
text += "='" + attrVal + "'";
}
else {
text += '="' + attrVal + '"';
}
}
}
}
if (node.type == 'directive') {
output.write("<" + text + ">", encoding);
}
else if (node.type == 'comment') {
output.write("<!--" + text + "-->", encoding);
}
else if (node.type == 'tag' || node.type == 'script' || node.type == 'style') {
output.write("<" + text, encoding);
}
else {
output.write(text, encoding);
}

if (node.type == 'tag' || node.type == 'script' || node.type == 'style') {
output.write(">", encoding);
}
if (node.children) {
DomUtils.printHtml(node.children, output);
if ((node.closed && node.type == 'tag') || node.type == 'script' || node.type == 'style') {
output.write("</" + node.name + ">", encoding);
}
}
else {
if (((node.closed && node.type == 'tag') || node.type == 'script' || node.type == 'style') && !text.match(/\/\s*$/)) {
output.write("</" + node.name + ">", encoding);
}
}
}
}

, toHtml: function DomUtils$toString(nodes) {
var result = "";
DomUtils.printHtml(nodes, {
write: function(text, encoding) {
result += text;
}},
"utf-8"
);
return result;
}
}

function inherits (ctor, superCtor) {
Expand Down
21 changes: 19 additions & 2 deletions runtests.js
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -22,6 +22,7 @@ IN THE SOFTWARE.
var sys = require("sys");
var fs = require("fs");
var htmlparser = require("./lib/htmlparser");
var domutils = require("./lib/htmlparser").DomUtils;

var testFolder = "./tests";
var chunkSize = 5;
Expand Down Expand Up @@ -58,17 +59,33 @@ for (var i in testFiles) {
var testResult =
sys.inspect(resultComplete, false, null) === sys.inspect(test.expected, false, null)
&&
sys.inspect(resultChunk, false, null) === sys.inspect(test.expected, false, null)
;
sys.inspect(resultChunk, false, null) === sys.inspect(test.expected, false, null);
if (test.type != 'rss') {
testResult = testResult
&&
domutils.toHtml(resultComplete) == (test.expectedHtml ? test.expectedHtml : test.html)
&&
domutils.toHtml(resultChunk) == (test.expectedHtml ? test.expectedHtml : test.html)
;
}
sys.puts("[" + test.name + "\]: " + (testResult ? "passed" : "FAILED"));
if (!testResult) {
failedCount++;
sys.puts("== Complete ==");
sys.puts(sys.inspect(resultComplete, false, null));
if (test.type != 'rss') {
sys.puts(domutils.toHtml(resultComplete));
}
sys.puts("== Chunked ==");
sys.puts(sys.inspect(resultChunk, false, null));
if (test.type != 'rss') {
sys.puts(domutils.toHtml(resultChunk));
}
sys.puts("== Expected ==");
sys.puts(sys.inspect(test.expected, false, null));
if (test.type != 'rss') {
sys.puts(test.expectedHtml ? test.expectedHtml : test.html);
}
}
}
sys.puts("Total tests: " + testCount);
Expand Down
3 changes: 3 additions & 0 deletions tests/01-basic.js
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -42,6 +42,7 @@ exports.expected =
, type: 'tag'
, name: 'title'
, children: [ { raw: 'The Title', data: 'The Title', type: 'text' } ]
, closed: true
}
, { raw: 'body'
, data: 'body'
Expand All @@ -53,8 +54,10 @@ exports.expected =
, type: 'text'
}
]
, closed: true
}
]
, closed: true
}
];

Expand Down
2 changes: 1 addition & 1 deletion tests/02-single_tag_1.js
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -35,5 +35,5 @@ exports.expected =
[ { raw: 'br', data: 'br', type: 'tag', name: 'br' }
, { raw: 'text', data: 'text', type: 'text' }
];

exports.expectedHtml = "<br>text";
})();
2 changes: 2 additions & 0 deletions tests/04-unescaped_in_script.js
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -48,8 +48,10 @@ exports.expected =
, type: 'text'
}
]
, closed: true
}
]
, closed: true
}
];

Expand Down
1 change: 1 addition & 0 deletions tests/05-tags_in_comment.js
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -42,6 +42,7 @@ exports.expected =
, type: 'comment'
}
]
, closed: true
}
];

Expand Down
1 change: 1 addition & 0 deletions tests/06-comment_in_script.js
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -42,6 +42,7 @@ exports.expected =
, type: 'comment'
}
]
, closed: true
}
];

Expand Down
1 change: 1 addition & 0 deletions tests/07-unescaped_in_style.js
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -43,6 +43,7 @@ exports.expected =
, type: 'text'
}
]
, closed: true
}
];

Expand Down
2 changes: 2 additions & 0 deletions tests/08-extra_spaces_in_tag.js
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -43,7 +43,9 @@ exports.expected =
, type: 'text'
}
]
, closed: true
}
];
exports.expectedHtml = "<font \n size='14'>the text</font>";

})();
1 change: 1 addition & 0 deletions tests/09-unquoted_attrib.js
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -43,6 +43,7 @@ exports.expected =
, type: 'text'
}
]
, closed: true
}
];

Expand Down
2 changes: 1 addition & 1 deletion tests/14-comment_in_text_in_script.js
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -49,8 +49,8 @@ exports.expected =
, data: ' the text'
, type: 'text'
}

]
, closed: true
}
];

Expand Down
2 changes: 2 additions & 0 deletions tests/15-non-verbose.js
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -40,7 +40,9 @@ exports.expected =
, type: 'text'
}
]
, closed: true
}
];
exports.expectedHtml = '<font size="14">the text</font>';

})();
2 changes: 2 additions & 0 deletions tests/16-ignore_whitespace.js
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -65,7 +65,9 @@ exports.expected =
, type: 'text'
}
]
, closed: true
}
];
exports.expectedHtml = "Line one\n<br><br>\nline two<font><br> x </font>";

})();
2 changes: 1 addition & 1 deletion tests/17-xml_namespace.js
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -32,7 +32,7 @@ exports.options = {
};
exports.html = "<ns:tag>text</ns:tag>";
exports.expected =
[ { raw: 'ns:tag', data: 'ns:tag', type: 'tag', name: 'ns:tag', children: [ { raw: 'text', data: 'text', type: 'text' } ] }
[ { raw: 'ns:tag', data: 'ns:tag', type: 'tag', name: 'ns:tag', children: [ { raw: 'text', data: 'text', type: 'text' } ], closed: true }
];

})();
1 change: 1 addition & 0 deletions tests/18-enforce_empty_tags.js
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -36,5 +36,6 @@ exports.expected =
{ raw: 'link', data: 'link', type: 'tag', name: 'link' }
, { raw: 'text', data: 'text', type: 'text' }
];
exports.expectedHtml = "<link>text";

})();
2 changes: 1 addition & 1 deletion tests/19-ignore_empty_tags.js
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -35,7 +35,7 @@ exports.expected =
[
{ raw: 'link', data: 'link', type: 'tag', name: 'link', children: [
{ raw: 'text', data: 'text', type: 'text' }
] }
], closed: true }
];

})();
9 changes: 6 additions & 3 deletions tests/22-position_data.js
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -66,7 +66,8 @@ exports.expected = [
line: 3,
col: 9
}
}]
}],
closed: true
}, {
raw: 'body',
data: 'body',
Expand All @@ -84,7 +85,8 @@ exports.expected = [
line: 3,
col: 32
}
}]
}],
closed: true
}, {
raw: '\n\n',
data: '\n\n',
Expand All @@ -93,7 +95,8 @@ exports.expected = [
line: 6,
col: 8
}
}]
}],
closed: true
}
];

Expand Down