Skip to content

HTTPS clone URL

Subversion checkout URL

You can clone with HTTPS or Subversion.

Download ZIP
branch: master
Commits on May 24, 2013
  1. Improved Cache class documentation, added an overloaded alternative for

    authored
    Cache#set that accepts a scala.concurrent.duration.Duration instance.
  2. @nraychaudhuri

    Update select.scala.html

    nraychaudhuri authored
    Fixed the scaladoc
  3. @jroper

    Merge pull request #1126 from mandubian/perf/headers-cookies

    jroper authored
    optimized getHeaders/getCookies for perf
Commits on May 23, 2013
  1. @huntc
  2. @huntc

    Minimised the potential for a map conversion by retaining the data fi…

    huntc authored
    …eld and performing sequential searches on it. Given a typical number of headers (~10) this may be better than a hash map + the cost of coercing to a map in the first place. Its also roughly inline with how Netty deals headers.
    
    Performance measurements gathered show no significant performance benefit although there appears to be a slight improvement. In many cases the map should not have to be created and memory usage should also therefore be slightly better.
    
    New suite of unit tests for Http - mostly covering the internal changes made.
    
    Clarified the contract of the "data" member of Http.Headers such that multiple headers with the same name are not expected within the structure. This wasn't a problem previously as Fakes leverage the internal assumption that "data" would be transformed to a map and that duplicates would therefore be eliminated. The Fakes code has been updated to conform to Header's contract.
Commits on May 22, 2013
  1. @huntc
  2. @huntc

    Made the test more realistic by providing a handler that returns "hel…

    huntc authored
    …lo world!". This therefore exercises the router as well.
  3. @huntc
  4. @huntc

    Further improve the reliability of the benchmark results by increasin…

    huntc authored
    …g the warmup time to avoid GCs and other things.
  5. @huntc
Commits on May 21, 2013
  1. @huntc

    Merge pull request #1053 from jroper/results-refactoring

    huntc authored
    Results refactoring
  2. @huntc
  3. @huntc

    Correct usage of ramp up and duration instead of invocation and rampu…

    huntc authored
    …p. The idea is to get over 10K requests happening before we start sampling the tests. This allows GC to get out of the way.
  4. @huntc
  5. @huntc

    Improved the PT by making it loop 100 times per invocation. This yiel…

    huntc authored
    …ds a larger range of values to plot.
  6. @huntc
  7. @huntc
  8. @jroper

    Fixed merge conflict

    jroper authored
  9. @jroper

    Update from review

    jroper authored
  10. @jroper
  11. @jroper

    Merge pull request #1138 from huntc/request-perf-test

    jroper authored
    Performance test infrastructure
  12. @huntc

    Merge pull request #1087 from jroper/async-java-docs

    huntc authored
    Extracted java code samples for async
  13. @huntc
Commits on May 20, 2013
  1. @jroper
  2. @jroper
  3. @jroper
  4. @jroper
  5. @jroper
  6. @jroper

    Finished results refactoring

    jroper authored
    * Got all integration tests and sample apps running
    * Modified ActionBuilder to make building builders easier
    * JPA is now bound to HTTP context by default, so can be used
      asynchronously
  7. @jroper
  8. @jroper
  9. @jroper

    Formatted sources

    jroper authored
  10. @huntc

    Merge pull request #1078 from jroper/embed-routes

    huntc authored
    New integration test project
  11. @huntc

    Changes as recommended by @retronym - essentially the config did not …

    huntc authored
    …belong with the root project. In addition the pt filter should not overwrite all test options, but it does need to filter out the one it set.
  12. @huntc
Something went wrong with that request. Please try again.