corinthia-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Peter Kelly (JIRA)" <>
Subject [jira] [Commented] (COR-41) Use of /* ... */ preferable to //... in STDC Code
Date Fri, 13 Feb 2015 09:30:11 GMT


Peter Kelly commented on COR-41:

Indeed. Perhaps there is a feature test macro (#ifdef C99?) or similar that we could put in
one of the main header files to catch this case. I'll have a look into this. It is rather
important, as practically none of the code will compile with a non-C99 compiler.

Perhaps we could also maintain a list of officially supported compiler versions, to aid those
trying to build. Being outside of the windows world for much of the last decade I've always
taken C99 support for granted, but if we had something like "Requires Visual Studio 2013 or
later" on the build requirements page this could be helpful for some folks.

> Use of /* ... */ preferable to //... in STDC Code
> -------------------------------------------------
>                 Key: COR-41
>                 URL:
>             Project: Corinthia
>          Issue Type: Bug
>    Affects Versions: 0.5
>         Environment: All .c and .h files in the repository
>            Reporter: Dennis E. Hamilton
>            Priority: Trivial
>              Labels: Apache-License-Headers, C
> The ASF Notices on .c and .h files use // ... single-line comment triggers instead of
something like 
>    /* ...
>    *
>    ...
>    */
> the only form that some compilers recognize as STDC comments.  The use of -Wall and STDC
compile options will produce warnings in some products, such as most versions of Visual Studio,
clogging the error log with warning messages.
>  1. Agree that it is OK to change these to /* ... */ when the particular files are touched
for some purpose, such as clearing warnings out of a strict STDC compile.  Enough compilers
warn about this, apparently for down-level STDC compatibility.  
> (In the VC++ standard libraries, there is a file deep in the Windows SDK headers that
has a single //... that annoys me enough to actually fix it in the library.  I'd like to fix
this with some sort of platform quirk patch but I haven't dug enough into the include-path
rules to get that working.  Someday soon ... .) 
>  2. Fix these on an opportunistic basis.  There is no due date but it would be nice that
all new ones be in the universally-STDC form.  (Headers and code files that *require* C++
are another story and they will have different file extensions as well.)
>  3. Pray that rat, the IP checker, recognizes these notices either way.

This message was sent by Atlassian JIRA

View raw message