URI functions in Windows Store Applications

Summary

The Modern SDK contains some URI related functionality as do libraries available in particular projection languages. Unfortunately, collectively these APIs do not cover all scenarios in all languages. Specifically, JavaScript and C++ have no URI building APIs, and C++ additionally has no percent-encoding/decoding APIs.

	WinRT (JS and C++)	JS Only	C++ Only	.NET Only
Parse	Windows.Foundation.Uri		WsDecodeUrl*	System.Uri
Build			WsEncodeUrl*	System.UriBuilder
Normalize	Windows.Foundation.Uri			System.Uri
Equality	Windows.Foundation.Uri. Equals			System.Uri.Compare
Relative resolution	Windows.Foundation.Uri. CombineUri			System.Uri
Encode data for including in URI property	Windows.Foundation.Uri. EscapeComponent	encodeURIComponent	WsEncodeUrl*	System.Uri. EscapeDataString
Decode data extracted from URI property	Windows.Foundation.Uri. UnescapeComponent	decodeURIComponent	WsDecodeUrl*	System.Uri. UnescapeDataString
Build Query
Parse Query	Windows.Foundation. Uri.QueryParsed / Windows.Foundation. WwwFormUrlDecoder*

The Windows.Foudnation.Uri type is not projected into .NET modern applications. Instead those applications use System.Uri and the platform ensures that it is correctly converted back and forth between Windows.Foundation.Uri as appropriate. Accordingly the column marked WinRT above is applicable to JS and C++ modern applications but not .NET modern applications. The only entries above applicable to .NET are the .NET Only column and the WwwFormUrlDecoder in the bottom left which is available to .NET.

Scenarios

Parse

This functionality is provided by the WinRT API Windows.Foundation.Uri in C++ and JS, and by System.Uri in .NET.

Parsing a URI pulls it apart into its basic components without decoding or otherwise modifying the contents.

var uri = new Windows.Foundation.Uri("http://example.com/path%20segment1/path%20segment2?key1=value1&key2=value2");

console.log(uri.path);// /path%20segment1/path%20segment2

WsDecodeUrl (C++)

WsDecodeUrl is not suitable for general purpose URI parsing.  Use Windows.Foundation.Uri instead.

Build (C#)

URI building is only available in C# via System.UriBuilder.

URI building is the inverse of URI parsing: URI building allows the developer to specify the value of basic components of a URI and the API assembles them into a URI. 

To work around the lack of a URI building API developers will likely concatenate strings to form their URIs.  This can lead to injection bugs if they don’t validate or encode their input properly, but if based on trusted or known input is unlikely to have issues.

            Uri originalUri = new Uri("http://example.com/path1/?query");

            UriBuilder uriBuilder = new UriBuilder(originalUri);

            uriBuilder.Path = "/path2/";

            Uri newUri = uriBuilder.Uri; // http://example.com/path2/?query

WsEncodeUrl (C++)

WsEncodeUrl, in addition to building a URI from components also does some encoding.  It encodes non-US-ASCII characters as UTF8, the percent, and a subset of gen-delims based on the URI property: all :/?#[]@ are percent-encoded except :/@ in the path and :/?@ in query and fragment.

Accordingly, WsEncodeUrl is not suitable for general purpose URI building.  It is acceptable to use in the following cases:

- You’re building a URI out of non-encoded URI properties and don’t care about the difference between encoded and decoded characters.  For instance you’re the only one consuming the URI and you uniformly decode URI properties when consuming – for instance using WsDecodeUrl to consume the URI.

- You’re building a URI with URI properties that don’t contain any of the characters that WsEncodeUrl encodes.

Normalize

This functionality is provided by the WinRT API Windows.Foundation.Uri in C++ and JS and by System.Uri in .NET.  Normalization is applied during construction of the Uri object.

URI normalization is the application of URI normalization rules (including DNS normalization, IDN normalization, percent-encoding normalization, etc.) to the input URI.

        var normalizedUri = new Windows.Foundation.Uri("HTTP://EXAMPLE.COM/p%61th foo/");

        console.log(normalizedUri.absoluteUri); // http://example.com/path%20foo/

This is modulo Win8 812823 in which the Windows.Foundation.Uri.AbsoluteUri property returns a normalized IRI not a normalized URI.  This bug does not affect System.Uri.AbsoluteUri which returns a normalized URI.

Equality

This functionality is provided by the WinRT API Windows.Foundation.Uri in C++ and JS and by System.Uri in .NET. 

URI equality determines if two URIs are equal or not necessarily equal.

            var uri1 = new Windows.Foundation.Uri("HTTP://EXAMPLE.COM/p%61th foo/"),

                uri2 = new Windows.Foundation.Uri("http://example.com/path%20foo/");

            console.log(uri1.equals(uri2)); // true

Relative resolution

This functionality is provided by the WinRT API Windows.Foundation.Uri in C++ and JS and by System.Uri in .NET 

Relative resolution is a function that given an absolute URI A and a relative URI B, produces a new absolute URI C.  C is the combination of A and B in which the basic components specified in B override or combine with those in A under rules specified in RFC 3986.

        var baseUri = new Windows.Foundation.Uri("http://example.com/index.html"),

            relativeUri = "/path?query#fragment",

            absoluteUri = baseUri.combineUri(relativeUri);

        console.log(baseUri.absoluteUri);       // http://example.com/index.html

        console.log(absoluteUri.absoluteUri);   // http://example.com/path?query#fragment

Encode data for including in URI property

This functionality is available in JavaScript via encodeURIComponent and in C# via System.Uri.EscapeDataString. Although the two methods mentioned above will suffice for this purpose, they do not perform exactly the same operation.

Additionally we now have Windows.Foundation.Uri.EscapeComponent in WinRT, which is available in JavaScript and C++ (not C# since it doesn’t have access to Windows.Foundation.Uri).  This is also slightly different from the previously mentioned mechanisms but works best for this purpose.

Encoding data for inclusion in a URI property is necessary when constructing a URI from data.  In all the above cases the developer is dealing with a URI or substrings of a URI and so the strings are all encoded as appropriate. For instance, in the parsing example the path contains “path%20segment1” and not “path segment1”.  To construct a URI one must first construct the basic components of the URI which involves encoding the data.  For example, if one wanted to include “path segment / example” in the path of a URI, one must percent-encode the ‘ ‘ since it is not allowed in a URI, as well as the ‘/’ since although it is allowed, it is a delimiter and won’t be interpreted as data unless encoded.

If a developer does not have this API provided they can write it themselves.  Percent-encoding methods appear simple to write, but the difficult part is getting the set of characters to encode correct, as well as handling non-US-ASCII characters.

        var uri = new Windows.Foundation.Uri("http://example.com" +

            "/" + Windows.Foundation.Uri.escapeComponent("path segment / example") +

            "?key=" + Windows.Foundation.Uri.escapeComponent("=&?#"));

        console.log(uri.absoluteUri); // http://example.com/path%20segment%20%2F%20example?key=%3D%26%3F%23

WsEncodeUrl (C++)

In addition to building a URI from components, WsEncodeUrl also percent-encodes some characters.  However the API is not recommend for this scenario given the particular set of characters that are encoded and the convoluted nature in which a developer would have to use this API in order to use it for this purpose.

There are no general purpose scenarios for which the characters WsEncodeUrl encodes make sense: encode the %, encode a subset of gen-delims but not also encode the sub-delims.  For instance this could not replace encodeURIComponent in a C++ version of the following code snippet since if ‘value’ contained ‘&’ or ‘=’ (both sub-delims) they wouldn’t be encoded and would be confused for delimiters in the name value pairs in the query:

"http://example.com/?key=" + Windows.Foundation.Uri.escapeComponent(value)

Since WsEncodeUrl produces a string URI, to obtain the property they want to encode they’d need to parse the resulting URI.  WsDecodeUrl won’t work because it decodes the property but Windows.Foundation.Uri doesn’t decode.  Accordingly the developer could run their string through WsEncodeUrl then Windows.Foundation.Uri to extract the property.

Decode data extracted from URI property

This functionality is available in JavaScript via decodeURIComponent and in C# via System.Uri.UnescapeDataString. Although the two methods mentioned above will suffice for this purpose, they do not perform exactly the same operation.

Additionally we now also have Windows.Foundation.Uri.UnescapeComponent in WinRT, which is available in JavaScript and C++ (not C# since it doesn’t have access to Windows.Foundation.Uri).  This is also slightly different from the previously mentioned mechanisms but works best for this purpose.

Decoding is necessary when extracting data from a parsed URI property.  For example, if a URI query contains a series of name and value pairs delimited by ‘=’ between names and values, and by ‘&’ between pairs, one must first parse the query into name and value entries and then decode the values.  It is necessary to make this an extra step separate from parsing the URI property so that sub-delimiters (in this case ‘&’ and ‘=’) that are encoded will be interpreted as data, and those that are decoded will be interpreted as delimiters.

If a developer does not have this API provided they can write it themselves.  Percent-decoding methods appear simple to write, but have some tricky parts including correctly handling non-US-ASCII, and remembering not to decode .

In the following example, note that if unescapeComponent were called first, the encoded ‘&’ and ‘=’ would be decoded and interfere with the parsing of the name value pairs in the query.

            var uri = new Windows.Foundation.Uri("http://example.com/?foo=bar&array=%5B%27%E3%84%93%27%2C%27%26%27%2C%27%3D%27%2C%27%23%27%5D");

            uri.query.substr(1).split("&").forEach(

                function (keyValueString) {

                    var keyValue = keyValueString.split("=");

                    console.log(Windows.Foundation.Uri.unescapeComponent(keyValue[0]) + ": " + Windows.Foundation.Uri.unescapeComponent(keyValue[1]));

                    // foo: bar

                    // array: ['ㄓ','&','=','#']

                });

WsDecodeUrl (C++)

Since WsDecodeUrl decodes all percent-encoded octets it could be used for general purpose percent-decoding but it takes a URI so would require the dev to construct a stub URI around the string they want to decode.  For example they could prefix “http:///#” to their string, run it through WsDecodeUrl and then extract the fragment property.  It is convoluted but will work correctly.

Parse Query

The query of a URI is often encoded as application/x-www-form-urlencoded which is percent-encoded name value pairs delimited by ‘&’ between pairs and ‘=’ between corresponding names and values.

In WinRT we have a class to parse this form of encoding using Windows.Foundation.WwwFormUrlDecoder.  The queryParsed property on the Windows.Foundation.Uri class is of this type and created with the query of its Uri:

    var uri = Windows.Foundation.Uri("http://example.com/?foo=bar&array=%5B%27%E3%84%93%27%2C%27%26%27%2C%27%3D%27%2C%27%23%27%5D");

    uri.queryParsed.forEach(

        function (pair) {

            console.log("name: " + pair.name + ", value: " + pair.value);

            // name: foo, value: bar

            // name: array, value: ['ㄓ','&','=','#']

        });

    console.log(uri.queryParsed.getFirstValueByName("array")); // ['ㄓ','&','=','#']

The QueryParsed property is only on Windows.Foundation.Uri and not System.Uri and accordingly is not available in .NET.  However the Windows.Foundation.WwwFormUrlDecoder class is available in C# and can be used manually:

            Uri uri = new Uri("http://example.com/?foo=bar&array=%5B%27%E3%84%93%27%2C%27%26%27%2C%27%3D%27%2C%27%23%27%5D");

            WwwFormUrlDecoder decoder = new WwwFormUrlDecoder(uri.Query);

            foreach (IWwwFormUrlDecoderEntry entry in decoder)

            {

                System.Diagnostics.Debug.WriteLine("name: " + entry.Name + ", value: " + entry.Value);

                // name: foo, value: bar

                // name: array, value: ['ㄓ','&','=','#']

            }

 

Build Query

To build a query of name value pairs encoded as application/x-www-form-urlencoded there is no WinRT API to do this directly.  Instead a developer must do this manually making use of the code described in “Encode data for including in URI property”.

In terms of public releases, this property is only in the RC and later builds.

For example in JavaScript a developer may write:

            var uri = new Windows.Foundation.Uri("http://example.com/"),

                query = "?" + Windows.Foundation.Uri.escapeComponent("array") + "=" + Windows.Foundation.Uri.escapeComponent("['ㄓ','&','=','#']");

 

            console.log(uri.combine(new Windows.Foundation.Uri(query)).absoluteUri); // http://example.com/?array=%5B'%E3%84%93'%2C'%26'%2C'%3D'%2C'%23'%5D

 

C++ constructor member initializers run in member declaration order

TL;DR: Keep your C++ class member declaration order the same as your constructor member initializers order.

C++ guarantees that the member initializers in a constructor are called in order. However the order in which they are called is the order in which the associated members are declared in the class, not the order in which they appear in the member initializer list. For instance, take the following code. I would have thought it would print "three, one, two", but in fact it prints, "one, two, three".

   
      #include "stdafx.h"
#include <iostream>

class PrintSomething {
public:
 PrintSomething(const wchar_t *name) { std::wcout << name << std::endl; }
};

class NoteOrder {
public:
        // This order doesn't matter.
 NoteOrder() : three(L"three"), one(L"one"), two(L"two") { }

 PrintSomething one;
 PrintSomething two;
 PrintSomething three;
};

int wmain(const int argc, const wchar_t* argv[])
{
 NoteOrder note; // Prints one, two, three, not three, one, two!
 return 0;
}

Subtleties of postMessage

In IE10 and other new browsers one may create MessageChannel objects that have two MessagePorts each connected (w3c spec calls it entangled) to one another such that postMessage on one port results in the message event firing on the other. You can pass an array of ports as the last parameter to postMessage and they show up in the ports property of the message event arg.

Origin

The postMessage here is like the worker postMessage and unlike the window and iframe postMessage in that it applies no origin checking:

1. No origin postMessage in workers and MessagePorts: postMessage(messageData, ports)
2. Origin postMessage in windows and iframes: postMessage(messageData, targetOrigin, ports)

Unfortunately the origin isn't an optional parameter at the end to make the two postMessages have the same signature.

On the event handler side, the event arg always has an origin property. But in the no origin case it is always the empty string.

Source

There is also a source property on the message event arg which if set is an object that has a postMessage property allowing you to post back to your caller. It is set for the origin case, however, in the no origin case this property is null. This is somewhat reasonable because in the case of MessagePort and Workers there are only two endpoints so you always know the source of a message implicitly. Unlike the origin case in which any iframe or window can be calling postMessage on any other iframe or window and the caller is unknown. So not unreasonable but it would be nice if the source property was always set for consistency.

MessageChannel start

When a MessageChannel is created it has two MessagePorts, but until those ports are started they will queue up any messages they receive. Once started they will dispatch all queued messages. Ports don't have to be started to send messages.

A port may be started in two ways, either by explicitly calling the start method on the port, or by setting the onmessage callback property on the port. However, adding an event listener via addEventListener("message", does not start the port. It works this way in IE and Chrome and the spec states this as well.

The justification is that since you can have only one callback via onmessage that once set you must implicitly be ready to receive messages and its fine to start the port. As opposed to the addEventListener in which case the user agent cannot start implicitly because it doesn't know how many event listeners will be added.  I found Hixie stating this justification in geoloc meeting notes.

Links

W3C Spec

Opera introduction

Percent Clcok Windows Store App Development Notes

My third completed Windows Store app is Percent Clock which displays portions of a time span like the time of the day or time until your next birthday, as a percentage. This was a small project I had previously started as a webpage and converted and finished as an HTML JavaScript Windows Store app.

The only somewhat interesting aspect of this app is that its the first app for which I tried charging. I picked the minimum amount for price 1.49 USD as it is a simple app and unsurprisingly it has sold very poorly. I'm considering releasing new instances of the app for specific scenarios:

• Death Clock: viewing your current age with respect to your life expectancy as a percentage.
• New Year Countdown: percentage of the year until New Years.

WinRT PropertySet Changed Event Danger

The Windows Runtime API Windows.Foundation.Collections.PropertySet class​ is a nice string name to object value map that has a changed event that fires when the contents of the map is modified. Be careful with this event because it fires synchronously from the thread on which the PropertySet was modified. If modified from the UI thread, the UI thread will then wait as it synchronously dispatches the changed event to all listeners which could lead to performance issues or especially from the UI thread deadlock. For instance, deadlock if you have two threads both trying to tell each other about changed events for different PropertySets.

Words with Hints Windows 8 App Development Notes

My second completed app for the Windows Store was Words with Hints a companion to Words with Friends or other Scrabble like games that gives you *ahem* hints. You provide your tiles and optionally letters placed in a line on the board and Words with Hints gives you word options.

I wrote this the first time by building a regular expression to check against my dictionary of words which made for a slow app on the Surface. In subsequent release of the app I now spawn four web workers (one for each of the Surface's cores) each with its own fourth of my dictionary. Each fourth of the dictionary is a trie which makes it easy for me to discard whole chunks of possible combinations of Scrabble letters as I walk the tree of possibilities.

The dictionaries are large and takes a noticeable amount of time to load on the Surface. The best performing mechanism I found to load them is as JavaScript source files that simply define their portion of the dictionary on the global object and synchronously (only on the worker so not blocking the UI thread). Putting them into .js files means they take advantage of bytecode caching making them load faster. However because the data is mostly strings and not code there is a dramatic size increase when the app is installed. The total size of the four dictionary .js files is about 44Mb. The bytecode cache for the dictionary files is about double that 88Mb meaning the dictionary plus the bytecode cache is 132Mb.

To handle the bother of postMessage communication and web workers this was the first app in which I used my promise MessagePort project which I'll discuss more in the future.

This is the first app in which I used the Microsoft Ad SDK. It was difficult to find the install for the SDK and difficult to use their website, but once setup, the Ad SDK was easy to import into VS and easy to use in my app.