+3
−1
Loading
Gitlab 现已全面支持 git over ssh 与 git over https。通过 HTTPS 访问请配置带有 read_repository / write_repository 权限的 Personal access token。通过 SSH 端口访问请使用 22 端口或 13389 端口。如果使用CAS注册了账户但不知道密码,可以自行至设置中更改;如有其他问题,请发邮件至 service@cra.moe 寻求协助。
I just wasted dozens of failed build attempts and more than 5 minutes
staring at this cryptic build error:
```
Traceback (most recent call last):
File "scripts/build/parse_syscalls.py", line 216, in <module>
main()
File "scripts/build/parse_syscalls.py", line 194, in main
syscalls, tagged = analyze_headers(args.include, args.scan,
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "scripts/build/parse_syscalls.py", line 118, in analyze_headers
with open(one_file, "r", encoding="utf-8") as fp:
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
FileNotFoundError: [Errno 2] No such file or directory:
'zephyr/include/zephyr/pm/.#device.h'
```
Because this file didn't show in `git status`, I kept believing that
building from scratch would make it go away.
`git status` should never hide source files that the build system uses.
Maybe `parse_syscalls.py` should be "Emacs-aware"? Or maybe not. As long
as it's not, `git status` must show every spurious file that might
affect the build and break it.
This is the first time that I see an error like this; so Emacs does not
leave files like this behind on a regular basis. Only when there
is a some sort of glitch; and then we definitely want to know about it.
More generally speaking, the choice of an editor is a personal
preference. So patterns matching Emacs files belong to personal
$HOME/.gitignore files; they shouldn't be mass-duplicated into every
single git repo instead. Stuffing every project with editor-specific
configuration of every editor obviously does not scale. But that's
probably for another day.
This is a partial revert of commit 3f4159c5 ("gitignore: Ignore
emacs temporary edit files")
Signed-off-by:
Marc Herbert <marc.herbert@intel.com>
CRA Git | Maintained and supported by SUSTech CRA and CCSE