https://bugs.winehq.org/show_bug.cgi?id=54379
--- Comment #19 from Eugene Savelov savelov@gmail.com --- (In reply to Piotr Caban from comment #15)
(In reply to Eugene Savelov from comment #13)
is it just a typo in this patch?
https://gitlab.winehq.org/wine/wine/-/commit/ 1cf1412a12c6ee6191d387162fecb456f0df2d89#2ca91ec591036cfc37280681ab0602cab666 64e1_254_273
No, this patch is correct in this regard (port is NULL when output is NULL).
There was an regression caused by 1cf1412a12c6ee6191d387162fecb456f0df2d89 and fixed by b6204639ba864231bd14164363613963a9696278 (gdi32: NULL terminate output string in CreateDCW).
It might be useful to do regression test starting from b6204639ba864231bd14164363613963a9696278 commit if printing works with it.
I was doing git bisect regression between 7.21 and 7.22, and printing did not work after the 1cf1412a12c6ee6191d387162fecb456f0df2d89, but worked before it.