Changes between Initial Version and Version 1 of Ticket #40656, comment 20
- Timestamp:
- May 8, 2021, 3:48:27 PM (4 years ago)
Legend:
- Unmodified
- Added
- Removed
- Modified
-
Ticket #40656, comment 20
initial v1 1 1 Greetings. I'm speaking as a developer of GNU LilyPond investigating our [https://gitlab.com/lilypond/lilypond/-/issues/6135 issue 6135], which is an error while compiling the MacPorts port. 2 2 3 From reading the build log provided to me, I believe that the main issue is that MacPorts sets up the build to search /opt/local/include with -I rather than -isystem. (Note that I am not currently a MacPorts user and I have not tried to reproduce the issue myself.)3 From reading the build log provided to me, I believe that the main issue is that MacPorts sets up the build to search /opt/local/include with `-I` rather than `-isystem`. (Note that I am not currently a MacPorts user and I have not tried to reproduce the issue myself.) 4 4 5 Using -I makes headers in that directory subject to the same compiler warnings as project-specific headers, which is undesirable in my view. There are warnings (one of which is currently treated as an error) that we benefit from enabling for LilyPond headers, but which we do not want to to enable for external headers. In our own build scripts, we distinguish external headers with -isystem.5 Using `-I` makes headers in that directory subject to the same compiler warnings as project-specific headers, which is undesirable in my view. There are warnings (one of which is currently treated as an error) that we benefit from enabling for LilyPond headers, but which we do not want to to enable for external headers. In our own build scripts, we distinguish external headers with `-isystem`. 6 6 7 7 I would like to find out whether MacPorts developers are willing to address this issue in MacPorts the near future. Seeing that the previous comment is seven years old, I guess that this issue has either been forgotten or that there is no real interest in ever doing it. Which one is it?