SURVEY: what extensions are used on C++ source code

0 views
Skip to first unread message

Steinar Bang

unread,
Oct 3, 1996, 3:00:00 AM10/3/96
to

Please reply to me, do not followup to the list. I'll count the
results and summarize the conclusions. To ease the creation of
statistics, it would help if you followed the format hinted at, at the
end of this message (I'll write a perl script to run through the
responses and extract the statistics).

The reason I'm asking this, is that we're having a discussion right
now, about what extension to use on C++ source files. Our sourcefile
will be used both on different flavours of UNIX, and on Win32
(Visual/C++).

Our contenders, are cpp (from the MSWin-world. I personally find it
unaestetic, and also confusing wrt. to the C pre-prosessor), and cc.

One reason this becomes a problem at all, is because we've already
checked in sources with different extensions in our CVS repository.
We would like to coordinate the source code, to use only one type
extension, and we want one that gives the least set of problems.

Note! I'm interested in what extensions you have on the source code
*you* write. Not on source code you have on 2nd party tools that are
a part of your product (eg. libraries).

Format of response:
- Just do a "reply" to this article. A procmail rule will send
articles with this subject (and replies) directly to a folder
- The response should have the following format:
HEADEREXT: here comes the extension you use for header files (eg. "h", "hh")
SRCEXT: here comes the extension you use for source files (eg. "cc", "cpp")
PLATFORM: development environment and platform, separated by a comma. ex:
PLATFORM: Visual/C++, Win32
PLATFORM: SPARCWorks 4.1, Solaris 2.5
PLATFORM: there can be several "PLATFORM" lines
WHY: the reasons for your choice, in free form text. This has to be
the last part of the message, because I'll just chop out everything
from the "WHY:" and to the end, and use that as the reason.

I'll let the survey run for about a week.


- - Steinar

[ Send an empty e-mail to c++-...@netlab.cs.rpi.edu for info ]
[ about comp.lang.c++.moderated. First time posters: do this! ]

Steinar Bang

unread,
Oct 10, 1996, 3:00:00 AM10/10/96
to

>>>>> Steinar Bang <s...@metis.no>:

I asked about what file name extensions people were using on their
C++ source code.

> Please reply to me, do not followup to the list. I'll count the
> results and summarize the conclusions. To ease the creation of
> statistics, it would help if you followed the format hinted at, at the
> end of this message (I'll write a perl script to run through the
> responses and extract the statistics).

Here are the results in short. If I can find the time over the
weekend (ie. if it rains), I will add code to the perl script that
will make an HTML table of the results, showing the platforms the
respondents use, and the reasons behind their choices.

Interesting reading.

Header extensions:
header extension "(missing response)": 1 (5%)
header extension "h": 14 (74%)
header extension "hh": 1 (5%)
header extension "hpp": 1 (5%)
header extension "hxx": 1 (5%)
header extension "none": 1 (5%)

Source extensions:
source extension "(missing response)": 2 (11%)
source extension "C": 2 (11%)
source extension "cc": 3 (16%)
source extension "cpp": 11 (58%)
source extension "cxx": 1 (5%)
Total # of responses: 19 (as of 961010, 18:50 MET DST)

19 probably isn't a significant number, so if you want to get counted,
the procmail setup is still open, and counting. See
<URL:news:53064p$a...@netlab.cs.rpi.edu>
for how to respond.

Steinar Bang

unread,
Oct 14, 1996, 3:00:00 AM10/14/96
to

>>>>> Steinar Bang <s...@metis.no>:

> I asked about what file name extensions people were using on their
> C++ source code.

>> Please reply to me, do not followup to the list. I'll count the
>> results and summarize the conclusions. To ease the creation of
>> statistics, it would help if you followed the format hinted at, at the
>> end of this message (I'll write a perl script to run through the
>> responses and extract the statistics).

> Here are the results in short. If I can find the time over the
> weekend (ie. if it rains), I will add code to the perl script that
> will make an HTML table of the results, showing the platforms the
> respondents use, and the reasons behind their choices.

This table is now available at
<URL:http://www.metis.no/private/sb/cppsurvey.html>
It may be updated occasionally, if responses to the survey keeps
ticking in.

The most interesting part, is actually the "WHY:" free text field of
the responses.

A handful of new responses *have* ticked in, since the previous
message was posted (raising the number of responses from 19, to a
fabulous 23...), so I'm enclosing a new summary of the results:

Header extensions:
header extension "(missing response)": 1 (4%)
header extension "h": 16 (70%)
header extension "hh": 2 (9%)
header extension "hpp": 2 (9%)
header extension "hxx": 1 (4%)
header extension "none": 1 (4%)

Source extensions:
source extension "(missing response)": 2 (9%)
source extension "C": 2 (9%)
source extension "cc": 5 (22%)
source extension "cpp": 13 (57%)
source extension "cxx": 1 (4%)
Total # of responses: 23

Reply all
Reply to author
Forward
0 new messages