<mattip>
no problem. Was the goal to get the correct syntax error,
<mattip>
i.e. to get the indentation error before the generator error?
<mattip>
also: should py3.8-fix-lineno go into 7.3.11?
<cfbolz>
mattip: yes, I think the 3.9 changes should make sure that the "expected an indented block" message should appear
<cfbolz>
about the branch: I don't know :-(. It definitely is more correct, but it's a bit of a big change. we should run some tests to see that it doesn't regress anything
yizawa has joined #pypy
epony has quit [Remote host closed the connection]
yizawa has quit [Quit: Connection closed for inactivity]
epony has joined #pypy
jinsun has quit [Ping timeout: 256 seconds]
jinsun has joined #pypy
jinsun has quit [Read error: Connection reset by peer]
jinsun has joined #pypy
jinsun has quit [Read error: Connection reset by peer]
jinsun has joined #pypy
jinsun has quit [Read error: Connection reset by peer]
jinsun has joined #pypy
jinsun has quit [Read error: Connection reset by peer]
jinsun has joined #pypy
jinsun has quit [Read error: Connection reset by peer]
jinsun has joined #pypy
jcea has joined #pypy
jinsun has quit [Quit: No Ping reply in 180 seconds.]
jinsun has joined #pypy
jinsun has quit [Client Quit]
jinsun has joined #pypy
<mattip>
what project's tests could you suggest that check the ast/dis modules?
Dejan has quit [Read error: No route to host]
Atque has quit [Remote host closed the connection]
Atque_ has joined #pypy
<cfbolz>
mattip: if ned really takes a look, that would probably find a lot of possible problems
<mattip>
+1
<cfbolz>
mattip: should we point him at a 3.8 binary from the branch?
<mattip>
sure. I started going down the rabbit hole of why lib-python/3/tests/test_regrtest.py hangs
<mattip>
which I think is why the pypy-c-jit tests of the latest build on the branch is still running
jcea has quit [Remote host closed the connection]
jcea has joined #pypy
slav0nic has joined #pypy
epony has quit [Quit: QUIT]
<cfbolz>
mattip: ouch
<mattip>
seems to be some multiprocessing problem: they use subprocess.Popen to run tests in 4 threads
<cfbolz>
"great"
<cfbolz>
but why did that regress?!
<mattip>
dunno if it regresses or if it was always flaky
slav0nic has quit [Ping timeout: 264 seconds]
<mattip>
it seems that test failed forever, and recently started timing out too
<mattip>
if I put a print statement just before the call to subprocess.Popen, the test no longer hangs
<mattip>
maybe something to do with flushing stdout/stderr/stdin?
<mattip>
adding sys.stderr.flush() prevents the hang as far as I can tell
epony has joined #pypy
srs has joined #pypy
<tumbleweed>
mattip: if you want an insane example, ponyorm exercises ast/dis quite a lot (but with all sorts of cheating and inaccuracy)