Trying to build an ASP.Net application using UrlRewriter.Net for some of the pages, as sketched by this blog entry of ScottGu.
The rewriting generally works based on these rules:
<rewriter>
<rewrite url="^~/text/(.+)$" to="~/text.aspx?title=${encode($1)}" />
<rewrite url="^~/text/$" to="~/default.aspx" />
<rewrite url="^~/text$" to="~/default.aspx" />
....
</rewriter>
but there are some issues, which are probably caused by IIS (IIS 7 on Windows 7 in this case) rather than the rewriting module.
The configuration above redirects any request to the /text/ directory to the test.aspx Web Page.
First lesson learned: if you expect “funny” characters in the URL parameter (i.e. everything that needs to get URL-encoded), use the encode transformation.
Second, ASP.Net somehow loses track of what the “real” current URL directory is, so the “~/” (application root path) may not evaluate as expected. Use ResolveUrl() in pages that are affected by a rewrite module.
Two mysteries remain, though:
- a final dot in the URL causes a 404
- a sequence of two dots in the URL causes a 400
If you follow the MS links on a related Stack Overflow question, you will find that they are aware of similar problems for about 2 years, with the issues still unresolved.
The best strategy seems to be to avoid generating these kind of URLs in the first place.
