v2.0.1 v2.0.1.rc1 v2.0.0 v2.0.0.rc6 v2.0.0.rc5 v2.0.0.rc4 v2.0.0.rc3 v2.0.0.rc2 v2.0.0.rc1 v2.0.0.beta2 v1.4.8 v1.4.7 v1.4.6 v1.4.5 v1.4.4 v1.4.3 v1.4.2 v1.4.1 v1.4.0 v1.4.0.d v1.4.0.c v1.4.0.b v1.4.0.a v1.3.6 v1.3.5 v1.3.4 v1.3.3 v1.3.2 v1.3.1 v1.3.0 v1.3.0.g v1.3.0.f v1.3.0.e v1.3.0.d v1.3.0.c v1.3.0.b v1.3.0.a v1.2.9 v1.2.8 v1.2.7 v1.2.6 v1.2.5 v1.2.4 v1.2.3 v1.2.2 v1.2.1 v1.2.0 v1.1.4 v1.1.3 v1.1.2 v1.1.1 v1.1.0 v1.0.0 semver filter_pattern 2.0.0.rc5 2.0.0.rc4 2.0.0.rc3 1.4.3 1.4.2 1.4.1 1.4.0 1.4.0.d 1.4.0.c 1.4.0.b 1.4.0.a 1.3.6 1.3.5 1.3.4 1.3.3 1.3.2 1.3.1 1.3.0 1.3.0.g 1.3.0.f 1.3.0.e 1.3.0.d 1.3.0.c 1.3.0.b 1.3.0.a 1.2.9 1.2.8 1.2.7 1.2.6 1.2.5 1.2.4 1.2.3 1.2.2 1.2.1 1.2.0 1.2.0.d 1.2.0.c 1.2.0.b 1.2.0.a 1.1.4 1.1.3 1.1.2 1.1.1 1.1.0 1.1.b
Nothing to show
Commits on Mar 3, 2009
Commits on Mar 2, 2009
Commits on Feb 27, 2009
This changes the parent commit's :session option semantics just slightly. The :session option must be passed as part of the options Hash -- the body arg is assumed to be params or a POST/PUT body. Also, the mapping from :session to HTTP_COOKIE has been switched to 'rack.session' - mapping it to HTTP_COOKIE doesn't make any sense. While here, refactor the make_request method to make it more obvious that we're really just building up the options Hash for MockRequest.
The app's middleware pipeline was ignored when the app itself was run as middleware. This was due to the separate call paths for middleware vs. endpoint apps. This change makes it so that both endpoint and middleware apps are invoked via the same instance level #call method. One potentially confusing aspect of this change is that Base.new now returns the head of the app's middleware pipeline. If no middleware is used by the app, this will be an instance of the Base class; however, if middleware is used, Base.new will return the head of the middleware chain leading to the Base instance.
Commits on Feb 25, 2009
Feb 25, 2009
Responses to HEAD requests _should_ have a Content-Length header that's identical to GET requests. When a body is provided by the application, set the Content-Length header to the size of the body and replace the body with an empty Array; when no body is provided by the application in response to a HEAD request, and the Content-Length is calculated to be "0", remove the Content-Length header entirely to avoid mis-matched values. The idea here is that it's better to omit the header when we believe the Content-Length is not indicative of the same in response to GET than to send the Content-Length with a mismatched value. Logic taken from Apache and is generally in line with RFC 2616.
Commits on Feb 24, 2009
Feb 24, 2009
sr commented this out during some refactoring but I'd like to leave it in for now.
The "registered" message is sent to extension modules immediately after the module is mixed into a Sinatra::Base class. This can be used to initialize the class with options, add templates, define error handlers, etc.
Commits on Feb 23, 2009
Commits on Feb 22, 2009
Commits on Feb 20, 2009
Commits on Feb 18, 2009