Gets a value that indicates if HttpListener uses the raw unescaped URI instead of the converted URI.
Assembly: System (in System.dll)
The property indicates if HttpListener uses the raw unescaped URI instead of the converted URI where any percent-encoded values are converted and other normalization steps are taken.
The http.sys service exposes two request URI strings:
The raw URI is the System.Uri provided in the request line of a HTTP request:
The raw URI provided by http.sys for the request mentioned above, is "/path/". This represents the string following the HTTP verb as it was sent over the network.
The http.sys service creates a converted URI from the information provided in the request by using the URI provided in the HTTP request line and the Host header to determine the origin server the request should be forwarded to. This is done by comparing the information from the request with a set of registered URI prefixes. The HTTP Server SDK documentation refers to this converted URI as the HTTP_COOKED_URL structure.
In order to be able to compare the request with registered URI prefixes, some normalization to the request needs to be done. For the sample above the converted URI would be as follows:
Un-escapes all percent encoded values.
Converts percent-encoded non-ASCII characters into a UTF-16 character representation. Note that UTF-8 and ANSI/DBCS characters are supported as well as Unicode characters (Unicode encoding using the %uXXXX format).
Executes other normalization steps, like path compression.
Since the request doesn't contain any information about the encoding used for percent-encoded values, it may not be possible to determine the correct encoding just by parsing the percent-encoded values.
Therefore http.sys provides two registry keys for modifying the process:
If zero, http.sys accepts only UTF-8-encoded URLs.
If non-zero, http.sys also accepts ANSI-encoded or DBCS-encoded URLs in requests.
If non-zero, http.sys always tries to decode a URL as UTF-8 first; if that conversion fails and EnableNonUTF8 is non-zero, Http.sys then tries to decode it as ANSI or DBCS.
If zero (and EnableNonUTF8 is non-zero), http.sys tries to decode it as ANSI or DBCS; if that is not successful, it tries a UTF-8 conversion.
There is a need for supporting characters besides characters and numbers in URIs. An example is the following URI, which is used to retrieve customer information for customer number "1/3812":
Note the percent-encoded slash in the Uri (%2F). This is necessary, since in this case the slash character represents data and not a path delimiter.
Passing the string to Uri constructor will lead to the following URI:
Splitting the path into its segments would result in the following elements:
This is not the intent of the sender of the request.