Skip to content
Commits on Jan 28, 2016
  1. @jeremyevans

    Work with ruby 2.3's --enable-frozen-string-literal

    jeremyevans committed
    These changes are the minimal ones necessary to allow Forme's specs
    to pass. There may well be other changes that are required.
Commits on Jan 25, 2016
  1. @manveru

    Merge pull request #1004 from kou/unuse-thread-exclusive

    manveru committed
    Use Mutex instead of Thread.exclusive for reloader
  2. @kou
Commits on Jan 23, 2016
  1. @raggi
Commits on Jan 16, 2016
  1. @eileencodes

    Move empty hash to it's own method for session

    eileencodes committed
    This sets the session's default to a new method called `default_session`
    which is set to an empty hash.
    
    Now we can depend on the `default_session` method rather than depending
    on the implementation of `Hash`.
Commits on Jan 10, 2016
  1. @zach-taylor
Commits on Jan 1, 2016
  1. @matthewd

    Merge pull request #981 from michaelsauter/fix/load-for-read

    matthewd committed
    Load session for read before accessing keys or values
Commits on Dec 23, 2015
  1. @davydovanton
Commits on Dec 22, 2015
  1. @ktheory

    Add support for HTTP 451 "Unavailable for Legal Reasons"

    ktheory committed
    Adds support for the newly approved HTTP 451 response code,
    “Unavailable for Legal Reasons”.
    
    IETF draft specification:
    https://tools.ietf.org/html/draft-tbray-http-legally-restricted-status-0
    5
    
    More info:
    http://www.451unavailable.org
Commits on Dec 10, 2015
  1. @michaelsauter
Commits on Dec 5, 2015
  1. @jeremy

    First-Party cookies, another line of CSRF defense

    jeremy committed
    Set `first_party: true` to set the First-Party attribute telling
    browsers to only send the cookie with legit first-party requests.
    
    * https://tools.ietf.org/html/draft-west-first-party-cookies-00
    * https://www.chromestatus.com/feature/4672634709082112
Commits on Nov 4, 2015
  1. @tenderlove

    add Rack::Events middleware

    tenderlove committed
    * Add `Rack::Events` middleware for adding event based middleware:
    middleware that does not care about the response body, but only cares
    about doing work at particular points in the request / response
    lifecycle.
Commits on Oct 11, 2015
  1. @frodsan

    Add support for custom hmac.

    frodsan committed
Commits on Oct 8, 2015
  1. @tenderlove

    add Rack::Request#authority

    tenderlove committed
    * Add `Rack::Request#authority` to calculate the authority under which
    the response is being made (this will be handy for h2 pushes).
Commits on Oct 6, 2015
  1. @tenderlove

    * Add `Rack::Response::Helpers#cache_control` and `cache_control=`.

    tenderlove committed
    Use this for setting cache control headers on your response objects.
  2. @tenderlove

    * Add `Rack::Response::Helpers#etag` and `etag=`. Use this for

    tenderlove committed
    setting etag values on the response.
Commits on Oct 5, 2015
  1. @tenderlove
Commits on Oct 4, 2015
  1. @jeremy

    `Response#add_header` to add to a value to a multivalued header

    jeremy committed
    * Introduce `Rack::Response::Helpers#add_header` to add a value to a
      multi-valued response header. Implemented in terms of other
      `Response#*_header` methods, so it's available to any response-like
      class that includes the `Helpers` module.
    * Add `Rack::Request#add_header` to match.
    * Rename `Response#have_header?` to `#has_header?` to match existing
      `Request#has_header?`
    * Add test coverage for this and other `*_header` methods.
Commits on Oct 1, 2015
  1. @jeremy
Commits on Sep 25, 2015
  1. @tenderlove

    ask the request for session options

    tenderlove committed
    The request is in charge.
  2. @tenderlove

    make `commit_session` public

    tenderlove committed
Commits on Sep 24, 2015
  1. @tenderlove

    use a response object when committing the session

    tenderlove committed
    Add an adapter object (Rack::Response::Raw) and use that internally in
    the session middleware.  After that we will make commit_session public
    so the session can be committed out of band of the middleware
  2. @tenderlove
  3. @tenderlove

    Merge pull request #943 from ggrossman/fix_name_token_at_eol

    tenderlove committed
    Fix bug in parsing of Content-Disposition header where an unquoted name at end-of-line sucked in the trailing newline
  4. @tenderlove

    Merge pull request #947 from davidrunger/doc-fix

    tenderlove committed
    fix rdoc formatting (remove accidental links)
  5. @tenderlove

    Merge pull request #950 from liamseanbrady/fix_minor_params_style_issue

    tenderlove committed
    Fix missing space in list of paramaters
Commits on Sep 23, 2015
  1. @tenderlove
Commits on Sep 21, 2015
  1. @tenderlove

    deflater: always finish zlib stream before closing

    Eric Wong committed with tenderlove
    This helps avoid Zlib::DataError when a client disconnects on
    the server while the server is writing the response.
    
    This fixes the following backtraces on my server:
    
      data error (Zlib::DataError)
      rack/deflater.rb:124:in `close'
      rack/deflater.rb:124:in `ensure in each'
      rack/deflater.rb:124:in `each'
      rack/chunked.rb:23:in `each'
      ...
  2. @liamseanbrady
Commits on Sep 13, 2015
  1. @davidrunger

    fix rdoc formatting (remove accidental links)

    davidrunger committed
    In several places the rdoc comments attempt to indicate hash access, for
    example `env['rack.input']`, but rdoc interprets this as a link with
    text "env" and href="%27rack.input%27/". Wrapping these snippets of hash
    access code in `<tt>` tags escapes the link syntax and also renders the
    code snippets in monospace.
Commits on Sep 8, 2015
  1. @ggrossman

    When parsing the name parameter of Content-Disposition, support quote…

    ggrossman committed
    …d chars in the quoted-string case.
Commits on Sep 6, 2015
  1. @ggrossman

    Fix bug in parsing of Content-Disposition header where an unquoted na…

    ggrossman committed
    …me at end-of-line sucked in the trailing newline.
Commits on Sep 5, 2015
  1. @tenderlove

    fixing the CI

    tenderlove committed
    I really don't understand what is wrong.  URI::Parser should exist.
  2. @tenderlove
  3. @tenderlove

    move Session::Pool to new superclass

    tenderlove committed
    ID is deprecated, and we only want to deal with request objects, so move
    to the new superclass.
Something went wrong with that request. Please try again.