[Gluster-infra] [Bug 1564149] Agree upon a coding standard, and automate check for this in smoke
bugzilla at redhat.com
bugzilla at redhat.com
Wed Apr 18 15:06:36 UTC 2018
--- Comment #16 from Jeff Darcy <jeff at pl.atyp.us> ---
FWIW, since this came up in the maintainers' meeting...
The inconsistency of style has had a measurable negative impact on ramp-up time
for new developers at Facebook. I'm sure the same has been true of new
developers at Red Hat and elsewhere. This impact manifests in several ways:
(1) Every developer has to spend time adjusting editor settings to have code
not seem randomly indented in one set of files, then again for another set of
files which violate our existing standard in a different way. Readability
matters. Also, when they're *writing* code, each developer has to tweak
settings some more because our standards are so unlike any other.
(2) Many aspects of our old standard (to which most code still conforms) are
simply alien to someone who learned C in the last ten years. Again, readability
matters. It might not seem like a big deal to read code that's formatted
"strangely" and indeed it's a good skill for all developers to have, but on top
of everything else it slows down new contributors.
(3) Time is wasted addressing these issues in code reviews, especially when a
patch is submitted to code that was already in violation of all standards and
Lastly, I'll add that the combination of eight-column indents, 80-column limit,
and gigantic functions nested ten levels deep has led to a lot of code wedged
up against the right margin. Relaxing the 80-column limit would be ill advised
in light of real human-factors studies showing that anything more than 80-100
columns (results vary) makes it harder to scan from one line to the next. Of
the other two factors, the indent width can be changed with a reformat. The
gigantic functions and deep nesting are terrible habits, but those habits
aren't likely to change any time soon. Even if the only change was to reduce
indentation and spurious line breaks, I think a reformat would be justified.
You are receiving this mail because:
You are on the CC list for the bug.
Unsubscribe from this bug https://bugzilla.redhat.com/token.cgi?t=eLTCNJSFpj&a=cc_unsubscribe
More information about the Gluster-infra