Need recommendations for code modularization

[ern]

I inherited a huge C application and most of it is in one file. I'd
like to modularize it into several different files. Is there a good
resource for learning the best way to link C files together... and
maybe some good conventions to use?

Thanks,

[Karl Malbrain]

ern wrote:

I inherited a huge C application and most of it is in one file. I'd
like to modularize it into several different files. Is there a good
resource for learning the best way to link C files together... and
maybe some good conventions to use?

Places to start:

1) global variables: divide the functions by the variables they access.
2) structure definitions: divide the functions by the structures they
use.

Hope this helps. karl m

[Ian]

ern wrote:

I inherited a huge C application and most of it is in one file. I'd
like to modularize it into several different files. Is there a good
resource for learning the best way to link C files together... and
maybe some good conventions to use?

Your compiler manuals?

Look for logical groupings of functions and variables and keep them
together in the new files.

The biggest problem you will find will be converting between static
variables (or functions) in the one file to global/shared across many
files. You will either have to make things global, or create some form
of encapsulation, depending or you style preference. In my experience,
the pain will lead to a better design, but you had better have a good
set of tests before you start the process.

Ian

[matt]

I am currently going through the same thing. I have been trying to
make accessors and mutator functions that act on the global variables.
Also, I have been using a program called Understand for C++ by
Scientific Toolworks Inc (www.scitools.com). It generates a lot of
information about your code. For example if you click on a variable it
tells you were it is defined, used and modified. It tells you what
functions call other functions. It has saved me a great deal of time,
although the cost was a little step.

[Chuck F.]

matt wrote:

I am currently going through the same thing. I have been trying
Same thing as what? Without context you make no sense. See my sig
below. Please be sure to read the URL before posting again.
to make accessors and mutator functions that act on the global
variables. Also, I have been using a program called Understand
There are no such things in C.
for C++ by Scientific Toolworks Inc (www.scitools.com). It
C++ is another language, found down the hall to the right.
generates a lot of information about your code. For example if
you click on a variable it tells you were it is defined, used
and modified. It tells you what functions call other functions.
It has saved me a great deal of time, although the cost was a
little step.

It is remotely possible that with context your comments could have
been topical. Without it, they are just off topic nonsense here.

--
"If you want to post a followup via groups.google.c om, don't use
the broken "Reply" link at the bottom of the article. Click on
"show options" at the top of the article, then click on the
"Reply" at the bottom of the article headers." - Keith Thompson
More details at: <http://cfaj.freeshell. org/google/>

[Keith Thompson]

"Chuck F. " <cb********@yah oo.com> writes:

matt wrote:

I am currently going through the same thing. I have been trying
Same thing as what? Without context you make no sense. See my sig
below. Please be sure to read the URL before posting again.

[snip]
--
"If you want to post a followup via groups.google.c om, don't use
the broken "Reply" link at the bottom of the article. Click on
"show options" at the top of the article, then click on the
"Reply" at the bottom of the article headers." - Keith Thompson
More details at: <http://cfaj.freeshell. org/google/>

I suggest updating the boilerplate slightly. It's been brought to my
attention that there's another workaround for Google's interface
problems, one that might be more convenient for some users. Rather
than going through the "show options" link, you can:

Click the "Reply" link at the bottom of the article.
Click the "Preview" button.
Click the "Edit Message" button.

Chris F.A. Johnson has updated his web page to reflect this.

Since Google users are going to be viewing this from their web
browsers anyway, it's probably sufficient just to point to Chris's web
page rather than repeating the instructions.

--
Keith Thompson (The_Other_Keit h) ks***@mib.org <http://www.ghoti.net/~kst>
San Diego Supercomputer Center <*> <http://users.sdsc.edu/~kst>
We must do something. This is something. Therefore, we must do this.

[Chuck F.]

Keith Thompson wrote:

[snip]

I suggest updating the boilerplate slightly. It's been brought to my
attention that there's another workaround for Google's interface
problems, one that might be more convenient for some users. Rather
than going through the "show options" link, you can:

Click the "Reply" link at the bottom of the article.
Click the "Preview" button.
Click the "Edit Message" button.

Chris F.A. Johnson has updated his web page to reflect this.

Since Google users are going to be viewing this from their web
browsers anyway, it's probably sufficient just to point to Chris's web
page rather than repeating the instructions.

I'm not sure I agree. I suspect the googler doesn't want to break
off whatever thread he is reading to check another webpage,
although they probably can without harm on most browsers. Thus I
think having a suitable mechanism right under their noses will help.

I gather that the new method is simply another alternative, and
requires conscious effort, rather than being the default. This is
stupid, because nobody wants to wait for more intermediate pages to
arrive.
--
"If you want to post a followup via groups.google.c om, don't use
the broken "Reply" link at the bottom of the article. Click on
"show options" at the top of the article, then click on the
"Reply" at the bottom of the article headers." - Keith Thompson
More details at: <http://cfaj.freeshell. org/google/>

[Randy Howard]

Chuck F. wrote
(in article <gd************ *************** ***@maineline.n et>):

Keith Thompson wrote:

[snip]

I suggest updating the boilerplate slightly. It's been brought to my
attention that there's another workaround for Google's interface
problems, one that might be more convenient for some users. Rather
than going through the "show options" link, you can:

Click the "Reply" link at the bottom of the article.
Click the "Preview" button.
Click the "Edit Message" button.

Chris F.A. Johnson has updated his web page to reflect this.

Since Google users are going to be viewing this from their web
browsers anyway, it's probably sufficient just to point to Chris's web
page rather than repeating the instructions.

I'm not sure I agree. I suspect the googler doesn't want to break
off whatever thread he is reading to check another webpage,
although they probably can without harm on most browsers. Thus I
think having a suitable mechanism right under their noses will help.

BTW, I seem to remember that Google snips sigs ENTIRELY when
displayed in their web interface. As such, when you post text
in your sig on how to work around google interface issues, they
may never see it at all.
--
Randy Howard (2reply remove FOOBAR)
"The power of accurate observation is called cynicism by those
who have not got it." - George Bernard Shaw
How 'bout them Horns?

[mensanator]

Randy Howard wrote:

Chuck F. wrote
(in article <gd************ *************** ***@maineline.n et>):

Keith Thompson wrote:

[snip]

I suggest updating the boilerplate slightly. It's been brought to my
attention that there's another workaround for Google's interface
problems, one that might be more convenient for some users. Rather
than going through the "show options" link, you can:

Click the "Reply" link at the bottom of the article.
Click the "Preview" button.
Click the "Edit Message" button.

Chris F.A. Johnson has updated his web page to reflect this.

Since Google users are going to be viewing this from their web
browsers anyway, it's probably sufficient just to point to Chris's web
page rather than repeating the instructions.
I'm not sure I agree. I suspect the googler doesn't want to break
off whatever thread he is reading to check another webpage,
although they probably can without harm on most browsers. Thus I
think having a suitable mechanism right under their noses will help.

BTW, I seem to remember that Google snips sigs ENTIRELY when
displayed in their web interface.

No, it doesn't.
As such, when you post text
in your sig on how to work around google interface issues, they
may never see it at all.
In fact, I get yelled at because not only are they displayed,
but they are included in the quoting.


--
Randy Howard (2reply remove FOOBAR)
"The power of accurate observation is called cynicism by those
who have not got it." - George Bernard Shaw
How 'bout them Horns?

[Richard Heathfield]

Chuck F. said:

Thus I
think having a suitable mechanism right under their noses will help.

(Subject line: "Housebreak ing googlers")

Did you mean "Housetraining" ? In the UK at least, "housebreak ing" has a
completely different meaning.

I can't help thinking of a guy in a black beret, a mask over the eyes, a
hooped yellow-and-black jersey, a coarse sack with "SWAG" written on it in
large black letters, a jemmy or crowbar, and a laptop. Presumably he only
has dial-up at home, is disgusted with his phone bill, and would therefore
rather use someone else's landline for his surfing.

--
Richard Heathfield
"Usenet is a strange place" - dmr 29/7/1999
http://www.cpax.org.uk
email: rjh at above domain (but drop the www, obviously)