Remove noProxy option from -sPROXY_TO_WORKER
#23297
Conversation
The `noProxy` options was added back in 2014 in c2295c7, but doesn't have any test coverage. This means there is is good chance its already broken. What is more it a fair mount of complexity since it adds an additional way in which certain JS files get pre-processed and included. I propose that we simply remove it, but and alternative would be to fine a way to test it. I'm not quite sure who the end user of this feature might be though.
sbc100
changed the title
noProxy options from -sPROXY_TO_WORKERnoProxy option from -sPROXY_TO_WORKER
Was it not tested here? c2295c7#diff-644556461d8f2f2bdd70b1f0d5455342db92e9535b5bca4ca5223224bd1a21f3R751 |
Ah, you are correct. I was kind of hoping it wasn't tested and that it had been broken for a while :) Do you want to keep it around? |
|
But how did the browser tests pass then ... ? |
|
This option is probably not that important, given it is proxy-to-worker and proxy-to-pthread is much more useful. And IIRC one can just build without the option to run on the main thread, which is all the option did. So I think I'm ok with removing it. Maybe worth asking on the mailing list first? |
|
Oh I see the test was removed in #13244. I can't really see why .. maybe because it was marked as |
|
Hopefully we can remove it. lgtm if no one complains on the mailing list. |
The
noProxyoptions was added back in 2014 in c2295c7, but doesn't have any test coverage. This means there is is good chance its already broken. What is more it a fair mount of complexity since it adds an additional way in which certain JS files get pre-processed and included.I propose that we simply remove it, but and alternative would be to fine a way to test it. I'm not quite sure who the end user of this feature might be though.