Manuel Jacob added the comment:
It’s possible to trigger the problem on Unix with much smaller sizes, e.g. by
interrupting the write() with a signal handler (even if the signal handler
doesn’t do anything). The following script starts a subprocess doing a 16MiB
write and sends a signal,
Manuel Jacob added the comment:
`io.TextIOWrapper.write()` returns the length of the passed string instead of
the actually written number of characters.
% python -u -c "import sys; print(sys.stdout.write('x'*4294967296),
file=sys.stderr)" | wc -c
4294967296
2147479552
So the possibility
Manuel Jacob added the comment:
2147479552 is the 0x7000 bytes limit documented for write() on Linux
(source: https://man7.org/linux/man-pages/man2/write.2.html). The limit could
be even smaller in other circumstances or other systems.
I’m adding Victor Stinner to the nosy list, as he
New submission from Manuel Jacob :
Without unbuffered mode, it works as expected:
% python -c "import sys; sys.stdout.write('x'*4294967296)" | wc -c
4294967296
% python -c "import sys; print('x'*4294967296)" | wc -c
4294967297
With unbuffered mode, writes get truncated to 2147479552
