Quick pull request question
Moderator: Forum Moderators
Quick pull request question
I have done a pull request (my first) for the project but the submitted code seems to have failed a build process (The Travis CI build failed)
I get:
Build Matrix
Job Duration Finished Compiler ENV
x 3917.1 23 min 36 sec 2 days ago gcc ALTERNATE_CONFIGURATION=false
x 3917.2 21 min 10 sec 2 days ago clang ALTERNATE_CONFIGURATION=false
v 3917.3 30 min 56 sec 2 days ago clang ALTERNATE_CONFIGURATION=true
I use the MSVC compiler (Windows) and the code builds correctly. Does this mean that with gcc and clang the code doesn't compile?
Thanks in advance,
Lovens
I get:
Build Matrix
Job Duration Finished Compiler ENV
x 3917.1 23 min 36 sec 2 days ago gcc ALTERNATE_CONFIGURATION=false
x 3917.2 21 min 10 sec 2 days ago clang ALTERNATE_CONFIGURATION=false
v 3917.3 30 min 56 sec 2 days ago clang ALTERNATE_CONFIGURATION=true
I use the MSVC compiler (Windows) and the code builds correctly. Does this mean that with gcc and clang the code doesn't compile?
Thanks in advance,
Lovens
Re: Quick pull request question
clang gives you a lo more warnings compared to MSVC, and the devs require the build to pass without any to succeed. So yes, you have to fix the issues reported by travis for clang. As I see, your second commit already fixed this.
However, the gcc build fails due to a standing issue with Wenoth master itself - your PR isn't responsible for that, just ignore it.
However, the gcc build fails due to a standing issue with Wenoth master itself - your PR isn't responsible for that, just ignore it.
-
- Inactive Developer
- Posts: 2461
- Joined: August 15th, 2008, 8:46 pm
- Location: Germany
Re: Quick pull request question
This happens very frequently.LovCAPONE wrote: I use the MSVC compiler (Windows) and the code builds correctly. Does this mean that with gcc and clang the code doesn't compile?
The general recommendation for someone working on a cross-platform project under windows (not just what I recommend) is to always compile with a parallel MinGw suite. Code and debug with MSVC, check with MinGw for any additional warnings/errors. That would be the CodeBlocks project files and dependency package.
Clang may be even a better option than MinGw if it works in a similar way for windows. The official releases are built with MinGw however, it's good to have a very similar testing suite.
projects (BfW 1.12):
A Simple Campaign: campaign draft for wml starters • Plan Your Advancements: mp mod
The Earth's Gut: sp campaign • Settlers of Wesnoth: mp scenario • Wesnoth Lua Pack: lua tags and utils
updated to 1.8 and handed over: A Gryphon's Tale: sp campaign
A Simple Campaign: campaign draft for wml starters • Plan Your Advancements: mp mod
The Earth's Gut: sp campaign • Settlers of Wesnoth: mp scenario • Wesnoth Lua Pack: lua tags and utils
updated to 1.8 and handed over: A Gryphon's Tale: sp campaign
Re: Quick pull request question
Note that we had some issues where gcc v 4.6 used by default on travis is segfaulting (internal compiler error) when trying to build master. This has been reproduced independently by 3 people on their systems as well, but at least two of them I know use ubuntu. Another person reported that gcc 4.6 worked fine, I believe they were on debian. So it might come down to subversions of 4.6, or even some kind of problem with the ubuntu package? (unlikely). We had a dev-list discussion recently about what if anything to do about this, but we don't have very much information.
When this started, I switched travis to use gcc 4.8 which made it work again, but I changed it back recently to the segfaulting version because someone said they wanted to try to fix it. That was a week ago now and it looks like they didn't manage to fix it, so I might change it back to gcc 4.8 again I guess.
Unless and until that happens, you should expect all travis gcc builds of master to fail. Sorry for the inconvenience of this...
When this started, I switched travis to use gcc 4.8 which made it work again, but I changed it back recently to the segfaulting version because someone said they wanted to try to fix it. That was a week ago now and it looks like they didn't manage to fix it, so I might change it back to gcc 4.8 again I guess.
Unless and until that happens, you should expect all travis gcc builds of master to fail. Sorry for the inconvenience of this...
Re: Quick pull request question
Ok thanks everyone for the replies, greatly appreciated.
Yesternight I setup a dev env with code::blocks as described in the wiki. Everything is good now.
Also as a side-note, I don't know if you guys heard of this but the gcc 4.9.* suite got alot of flames from the Linux dev team. Linus called it something that I won't repeat here. So before switching to it, it might be good to be sure that it's stable.
Lovens
Yesternight I setup a dev env with code::blocks as described in the wiki. Everything is good now.
Also as a side-note, I don't know if you guys heard of this but the gcc 4.9.* suite got alot of flames from the Linux dev team. Linus called it something that I won't repeat here. So before switching to it, it might be good to be sure that it's stable.
Lovens