You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Feb 22, 2018. It is now read-only.
At the moment, pub run is executed for every test suite. Maybe execution could be speed up by batching test suites together that are compatible (vm vs content_shell, vm arguments, ...).
There is also a TODO comment in the code for limiting the amount of parallel pub run executions. Implementing this would be quite useful, as my computer is very unresponsible running all my tests at the moment. I have many small test suites.
The text was updated successfully, but these errors were encountered:
Limiting the number of concurrently running Processes is definitely next on my TODO list. Especially since, for large projects, this causes failures on Mac (see #9), make the computer irresponsive for a while and slows things down because there is a lot of Process Context Switch being done.
Also currently we rely on dart2js to detect if a test is a browser test and dart2js is very slow at doing this (2 sec per file) and very CPU intensive. Soon we'll have an option in the analyzer to get this information and this should be much faster.
FYI I've implemented the max number of processes limit a few days ago. By default the max parallel processes is equal to the number of processors available on the system. This can be overridden using the option --max-processes
The parallel process limit makes the test runner much more useable then before.
But I think batching would still be usefull if you have many tests and using a more complicated pub setup with many files and transformers. In that case the startup time of pub run is quite long.
Sign up for freeto subscribe to this conversation on GitHub.
Already have an account?
Sign in.
At the moment, pub run is executed for every test suite. Maybe execution could be speed up by batching test suites together that are compatible (vm vs content_shell, vm arguments, ...).
There is also a TODO comment in the code for limiting the amount of parallel
pub run
executions. Implementing this would be quite useful, as my computer is very unresponsible running all my tests at the moment. I have many small test suites.The text was updated successfully, but these errors were encountered: