dist is both more accurate (the 'build' directory is normally where you do the build,
and normally consists of intermediate build artifacts no one cares about). dist also
occurs less frequently in the code base: after this change \bbuild\b has 466 matches,
whereas \bdist\b has 101 matches. build has 1072 matches whereas dist has 362.
* Fix dangling webdriver
* Embed humor sans font
* Refactor matplotlib testing codes
* Trigger CI
* Trigger CI
* Catch error when import from js fails
* Update changelog
* Apply suggestions from the review
* benchmark html5 canvas renderer
* adhere to lint
* rectify condition
* increase timeout
* do suggested changes
* adhere to lint
* use correct backend for benchmarks
* reintroduce tests + fix futures callback
* adhere to lint
* increase timeout for tests
* use selenium_standalone for tests
* increase timeout for check_comparison
* add result() to Future
* add debugging print statement
* adhere to lint
* use draw() instead of draw_idle()
* add new reference image for math text
* return NaN for non-native
* replace math text image for chrome as well
* increase timeout
* selenium standalone for benchmarks
* only matplotlib benchmarks
* re-enable benchmarks for numpy
* adhere to lint
* increase timeout
* Don't automatically copy python objects into javascript, add new PyProxy.shallowCopyToJavascript and deepCopyToJavascript apis
* Lint
* Adjust conftest to automatically deep_copy results of run / run_async
* Update conftest again
* Fix deep/shallowCopyToJavascript
* Fix shallowCopyToJavascript and deepCopyToJavascript to return js object not hiwire id
* Fix the remaining tests (hopefully)
* Lint
* Fix more tests
* Fix some numpy tests
* Lint
* Fix more tests
* Lint
* Temporarily dummy out setLineDash since it breaks test
* Fix test_console
* Update changelog
* Don't automatically copy python objects into javascript, add new PyProxy.shallowCopyToJavascript and deepCopyToJavascript apis
* Lint
* Adjust conftest to automatically deep_copy results of run / run_async
* Update conftest again
* Fix deep/shallowCopyToJavascript
* Fix shallowCopyToJavascript and deepCopyToJavascript to return js object not hiwire id
* Fix the remaining tests (hopefully)
* Lint
* Fix more tests
* Fix some numpy tests
* Lint
* Fix more tests
* Lint
* Temporarily dummy out setLineDash since it breaks test
The library that is supplied by USE_PNG is png, but matplotlib seeks
png16. With the fastcomp backend, we end up ignoring it anyway, but with
the upstream backend this will be the correct course of action
Gyeongjae Choi