From james at colannino.org  Thu Aug  4 13:57:16 2005
From: james at colannino.org (James Colannino)
Date: Thu, 04 Aug 2005 13:57:16 -0700
Subject: [OCLUG-devel] Matching Strings
Message-ID: <42F2812C.2060100@colannino.org>

Hey everyone.  I'm writing a C program that will report either duplicate 
strings in a file or strings that are the same up to n characters.  For 
example, let's say that I have the following file named text.txt:

albert
albert
albany
bert
biscuit
true
trouble
lone

If I were to do the following:

match text.txt

It would search for complete matches and should output the following:

albert
albert

If I were to do the following:

match -n 3 text.txt

It should output the following:

albert
albert
albany

But instead, it outputs this:

albert
albert
albany
albert
albany

I know why it does this.  I look for matches in the following way:

read string
search rest of file below it for match
read next string
search rest of file below it for match

Since albert is there twice, it reads the first instance of albert, 
outputs all matches below it, reads the second albert, and outputs all 
the matches below it, and so on.

My question is, how can I write my program in such a way that these only 
show up once?  In essence, I'd like it to report:

albert
albert
albany

Instead of:

albert
albert
albany
albert
albany

Here's a link to the source (it's probably not very good, but I tried to 
structure and comment it in such a way that it's at least easy to read 
-- I hope...)

http://james.colannino.org/match.c

Any help would be greatly appreciated :)

James

From tnation at baipro.com  Thu Aug  4 14:01:18 2005
From: tnation at baipro.com (Tim Nation)
Date: Thu, 4 Aug 2005 14:01:18 -0700
Subject: [OCLUG-devel] Matching Strings
References: <42F2812C.2060100@colannino.org>
Message-ID: <000401c59937$aa5ba170$0317a8c0@win2000>

> Hey everyone.  I'm writing a C program that will report either duplicate
> strings in a file or strings that are the same up to n characters.  For
> example, let's say that I have the following file named text.txt:

<snip>

> But instead, it outputs this:
>
> albert
> albert
> albany
> albert
> albany
>
> I know why it does this.  I look for matches in the following way:
>
> read string
> search rest of file below it for match
> read next string
> search rest of file below it for match
>
> Since albert is there twice, it reads the first instance of albert,
> outputs all matches below it, reads the second albert, and outputs all
> the matches below it, and so on.

<snip>

How about removing the matching lines from the record in memory as they are
processed. That way they can't be reprocessed multiple times. As long as you
are only reading and not writing the text record, you should be free to
manipulate it in memory all you want.

Tim


From x at xman.org  Thu Aug  4 14:30:35 2005
From: x at xman.org (Christopher Smith)
Date: Thu, 04 Aug 2005 14:30:35 -0700
Subject: [OCLUG-devel] Matching Strings
In-Reply-To: <42F2812C.2060100@colannino.org>
References: <42F2812C.2060100@colannino.org>
Message-ID: <42F288FB.4000109@xman.org>

Hmm... I don't have time to look at your code right now, but the
algorithm is pretty straight forward for doing this right. In C++ it'd
be something like (filling out the necessary types and various bits of
prologue and epilogue code is left as an excercise to the reader):

//keep going until we've seen all the data
while (my_ifstream) {
        //read a new line
	getline(buffer, my_ifstream);
        //get whatever section is relevant as a key
	string key = buffer.substr(0, n);
        //check to see if we've seen this key before
	my_map_iterator = my_map.find(key);
	if (my_map_iterator != my_map.end()) {
                //check if we've printed out the previously
		//scanned line before
		if (my_map_iterator->second) {
			//print out the previous line
                	cout << *(my_map_iterator->second) << endl;
			//set last_time to null and wipe
			//out the string
			auto_ptr<string> temp(my_map_iterator->second);
			my_map_iterator->second = 0;
                }
		//print out this line
		cout << buffer << endl;
	} else {
		//add this new key to our map
		my_map_iterator.insert(make_pair(key, new string(buffer)));
        }
}

--Chris

From msimpson at braysimpson.com  Thu Aug  4 14:51:14 2005
From: msimpson at braysimpson.com (Morgan Simpson)
Date: Thu, 04 Aug 2005 14:51:14 -0700
Subject: [OCLUG-devel] Matching Strings
In-Reply-To: <42F2812C.2060100@colannino.org>
References: <42F2812C.2060100@colannino.org>
Message-ID: <42F28DD2.9010701@braysimpson.com>

James Colannino wrote:

> My question is, how can I write my program in such a way that these only 
> show up once?  In essence, I'd like it to report:
> 
> albert
> albert
> albany
> 
> Instead of:
> 
> albert
> albert
> albany
> albert
> albany

James, you may want to consider a dictionary construct where each unique 
word is a key and the argument (definition) is the number of times that 
word appears in the input. You traverse the file once to create the 
dictionary. Then you traverse the dictionary and report those words that 
occur more than once.

Best regards,

Morgan Simpson
Bray, Simpson & Associates, Inc.

+1 714 390 5040
+1 714 549 3064

From strombrg at dcs.nac.uci.edu  Thu Aug  4 15:15:30 2005
From: strombrg at dcs.nac.uci.edu (Dan Stromberg)
Date: Thu, 04 Aug 2005 15:15:30 -0700
Subject: [OCLUG-devel] Matching Strings
In-Reply-To: <42F2812C.2060100@colannino.org>
References: <42F2812C.2060100@colannino.org>
Message-ID: <1123193731.14126.194.camel@seki.nac.uci.edu>


There's a variety of ways of doing this sort of thing, but some of the
easier ones that should still have a good running time, in pseudo code
look like:

A) Not absolutely blazing, but not slow, and very simple coding:
	1) go through file, stuffing all matches in an array
	2) pass the resulting array to qsort with an equivalence
          function that only compares -all- characters in the
          array
	3) go through and print out all matches that aren't
	   adjacent and identical

Or perhaps a little bit better:

	1) Initialize a suitably-sized hash table
	2) Go through the file, stuffing all matches in the hash
	   table.  The table is indexed by the matching prefix,
          but contains an associated piece of data for every
          matching line
	3) Just run through the hash table, outputting all data
          at each key

...or you could do much the same thing with a binary tree or something,
but a suitably-sized hash table will probably outperform the binary
tree, -and- be less likely to exhibit pathological behavior.

-Or-, if you have a truly huge list, then you might use a
database-backed "hash table", rather than keeping everything in memory.

On Thu, 2005-08-04 at 13:57 -0700, James Colannino wrote:
> Hey everyone.  I'm writing a C program that will report either duplicate 
> strings in a file or strings that are the same up to n characters.  For 
> example, let's say that I have the following file named text.txt:
> 
> albert
> albert
> albany
> bert
> biscuit
> true
> trouble
> lone
> 
> If I were to do the following:
> 
> match text.txt
> 
> It would search for complete matches and should output the following:
> 
> albert
> albert
> 
> If I were to do the following:
> 
> match -n 3 text.txt
> 
> It should output the following:
> 
> albert
> albert
> albany
> 
> But instead, it outputs this:
> 
> albert
> albert
> albany
> albert
> albany
> 
> I know why it does this.  I look for matches in the following way:
> 
> read string
> search rest of file below it for match
> read next string
> search rest of file below it for match
> 
> Since albert is there twice, it reads the first instance of albert, 
> outputs all matches below it, reads the second albert, and outputs all 
> the matches below it, and so on.
> 
> My question is, how can I write my program in such a way that these only 
> show up once?  In essence, I'd like it to report:
> 
> albert
> albert
> albany
> 
> Instead of:
> 
> albert
> albert
> albany
> albert
> albany
> 
> Here's a link to the source (it's probably not very good, but I tried to 
> structure and comment it in such a way that it's at least easy to read 
> -- I hope...)
> 
> http://james.colannino.org/match.c
> 
> Any help would be greatly appreciated :)
> 
> James
> _______________________________________________
> OCLUG-devel mailing list -- OCLUG-devel at oclug.org
> http://mailman.oclug.org/mailman/listinfo/oclug-devel
> 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part
Url : http://localhost.localdomain/pipermail/oclug-devel/attachments/20050804/40b565cf/attachment.bin 

From james at colannino.org  Thu Aug  4 16:25:59 2005
From: james at colannino.org (James Colannino)
Date: Thu, 04 Aug 2005 16:25:59 -0700
Subject: [OCLUG-devel] Matching Strings
In-Reply-To: <42F2812C.2060100@colannino.org>
References: <42F2812C.2060100@colannino.org>
Message-ID: <42F2A407.9020708@colannino.org>

Thanks everyone for all your answers :)

James

From hugo at teknofx.com  Tue Aug  9 00:27:49 2005
From: hugo at teknofx.com (Hugo Samayoa)
Date: Tue, 09 Aug 2005 00:27:49 -0700
Subject: [OCLUG-devel] mod_dav customization
Message-ID: <42F85AF5.3050806@teknofx.com>

*I was wondering if anyone here has done any webdav customization work. 
I'm in the process of re-writing some functions in mod_dav so that when 
a user uploads/puts an image to the folder that it automatically creates 
a thumbnail for the image. Just wondering if anyone has messed around 
with either dav_method_copymove, dav_method_delete, dav_method_put or 
any other method in mod_dav for apache. I just dont want to duplicate 
efforts. Thanks in advance.


-Hugo*

From msimpson at braysimpson.com  Tue Aug  9 10:46:10 2005
From: msimpson at braysimpson.com (Morgan Simpson)
Date: Tue, 09 Aug 2005 10:46:10 -0700
Subject: [OCLUG-devel] mod_dav customization
In-Reply-To: <42F85AF5.3050806@teknofx.com>
References: <42F85AF5.3050806@teknofx.com>
Message-ID: <42F8EBE2.90805@braysimpson.com>

Hugo Samayoa wrote:
> *I was wondering if anyone here has done any webdav customization work. 
> I'm in the process of re-writing some functions in mod_dav so that when 
> a user uploads/puts an image to the folder that it automatically creates 
> a thumbnail for the image. Just wondering if anyone has messed around 
> with either dav_method_copymove, dav_method_delete, dav_method_put or 
> any other method in mod_dav for apache. I just dont want to duplicate 
> efforts. Thanks in advance.

We are just starting a DAV implementation as a replacement for Samba. 
We'd be interested in following your progress. Good luck!

-- 

Best regards,

Morgan Simpson
Bray, Simpson & Associates, Inc.

+1 714 390 5040
+1 714 549 3064

From hugo at teknofx.com  Tue Aug  9 14:08:10 2005
From: hugo at teknofx.com (Hugo Samayoa)
Date: Tue, 09 Aug 2005 14:08:10 -0700
Subject: [OCLUG-devel] mod_dav customization
In-Reply-To: <42F8EBE2.90805@braysimpson.com>
References: <42F85AF5.3050806@teknofx.com> <42F8EBE2.90805@braysimpson.com>
Message-ID: <42F91B3A.6060508@teknofx.com>

Morgan,
    I have a DAV implementation already in production with backend 
authentication to mysql. We have Windows, Linux, Mac OS 9 and OS X 
clients using it. Right now I'm just customizing the modules to do image 
capture on movies uploaded and thumbnail creation for those images and 
other images file uploaded. Our webapp will take of that as well but 
since we allow webdav connections to the app i need to customize this 
work. All of it is backed by a GFS filesystem. So yeah if you have any 
question of the implementation send me an email off list.


-Hugo

Morgan Simpson wrote:

> Hugo Samayoa wrote:
>
>> *I was wondering if anyone here has done any webdav customization 
>> work. I'm in the process of re-writing some functions in mod_dav so 
>> that when a user uploads/puts an image to the folder that it 
>> automatically creates a thumbnail for the image. Just wondering if 
>> anyone has messed around with either dav_method_copymove, 
>> dav_method_delete, dav_method_put or any other method in mod_dav for 
>> apache. I just dont want to duplicate efforts. Thanks in advance.
>
>
> We are just starting a DAV implementation as a replacement for Samba. 
> We'd be interested in following your progress. Good luck!
>


From ddjolley at gmail.com  Fri Aug 12 16:15:05 2005
From: ddjolley at gmail.com (Doug Jolley)
Date: Fri, 12 Aug 2005 19:15:05 -0400
Subject: [OCLUG-devel] Conditionally Included Sections in HTML Doc
Message-ID: <b31d342a05081216157c469d5c@mail.gmail.com>

I have a rather large HTML document which which I'm going to call the
master document and which includes a variety of sections.  Actually,
this large document represents a composite of several smaller separate
documents each of which are a derivitive of the master document
drawing certain sections alternatively and choosing whether to include
other sections conditionally.  The point is that all of the smaller
documents can be built from the larger document by selecting and/or
conditionally including sections from the mater document.  The reason
I maintain the master document is that it is much easier to maintain
the one master document and then develop the derivitive documents on
the fly (i.e., programmatically) as needed.  I am trying to figure out
how to best develop the derivitive documents programmatically from the
master document.  PGP is an obvious solution.  Using the here document
feature of bash script seems appealing.  Using server-side includes
seems a bit clumsey; but, could also be considered a possibility. 
That's about the only 3 possibilities I recognize as being practical. 
I'm wondering what I might be missing.  Suggestions?  Thanks for any
input.

      ... doug

From x at xman.org  Fri Aug 12 16:31:05 2005
From: x at xman.org (Christopher Smith)
Date: Fri, 12 Aug 2005 16:31:05 -0700
Subject: [OCLUG-devel] Conditionally Included Sections in HTML Doc
In-Reply-To: <b31d342a05081216157c469d5c@mail.gmail.com>
References: <b31d342a05081216157c469d5c@mail.gmail.com>
Message-ID: <42FD3139.4090802@xman.org>

Doug Jolley wrote:
> I have a rather large HTML document which which I'm going to call the
> master document and which includes a variety of sections.  Actually,
> this large document represents a composite of several smaller separate
> documents each of which are a derivitive of the master document
> drawing certain sections alternatively and choosing whether to include
> other sections conditionally.  The point is that all of the smaller
> documents can be built from the larger document by selecting and/or
> conditionally including sections from the mater document.  The reason
> I maintain the master document is that it is much easier to maintain
> the one master document and then develop the derivitive documents on
> the fly (i.e., programmatically) as needed.  I am trying to figure out
> how to best develop the derivitive documents programmatically from the
> master document.  PGP is an obvious solution.

I'm guessing you mean PHP. ;-)

> Using the here document
> feature of bash script seems appealing.  Using server-side includes
> seems a bit clumsey; but, could also be considered a possibility. 
> That's about the only 3 possibilities I recognize as being practical. 
> I'm wondering what I might be missing.  Suggestions?  Thanks for any
> input.

I should think the proper thing would be to have files for each of the
component parts, which are then assembled as externally parsed entities.
There's a good document on this stuff (and other stuff):

http://www.xml.org/xml/xslt_efficient_programming_techniques.pdf

--Chris

From james at colannino.org  Mon Aug 15 12:07:45 2005
From: james at colannino.org (James Colannino)
Date: Mon, 15 Aug 2005 12:07:45 -0700
Subject: [OCLUG-devel] Perl and Recursion
Message-ID: <4300E801.6060600@colannino.org>

Hey everyone.  Here's a Perl question for you guys.  Here's some pseudo 
code to show what I've been working on (it's going backward recursively 
through a list of directories looking for the previous version of a 
specified package - note that the function below, except for the first 
two variables, is private to another function which is why I wrote it in 
this way):

    #These have scope outside of the private function below
    my @dirs2check;
    my @dirs2check_elements;

    $look_for_rpm = sub {

        #If it wasn't found and we've reached the end of directories we 
can search,
        #we must return 0 to indicate our failure to find the package.

        if ($dirs2check_elements <= -1) {return 0;}

        while (files exist in $dirs2check[$dirs2check_elements]) {

            examine file;

            if match found, {
                install older version of package;
                return 1;
            }

            else {
                --$dirs2check_elements;
                &$look_for_rpm();
            }
        }
    };

My problem here is that if a package is found, 1 is only returned to the 
previous instance of &$look_for_rpm() and is not returned to where I 
want it to go (past all the recursion and to the function that 
&$look_for_rpm() is private to.)  If there is an error, than 0 is also 
not returned to where I want it to go.  I want to find a way for either 
1 or 0 to be returned outside of the recursion that's taking place so I 
can tell whether or not a file was found.  I should probably be able to 
figure this out, and I'm sure the answer is a stupid one, but I can't 
for the life of me figure this out.  Thanks in advance.

James

From ddjolley at gmail.com  Mon Aug 15 12:48:01 2005
From: ddjolley at gmail.com (Doug Jolley)
Date: Mon, 15 Aug 2005 15:48:01 -0400
Subject: [OCLUG-devel] Conditionally Included Sections in HTML Doc
In-Reply-To: <42FD3139.4090802@xman.org>
References: <b31d342a05081216157c469d5c@mail.gmail.com>
	 <42FD3139.4090802@xman.org>
Message-ID: <b31d342a05081512481bef20d1@mail.gmail.com>

> I'm guessing you mean PHP. ;-)

Yes.  I seem to have a propensity for making that mistake.

Anyway, thanks for the input.  I finally went with a JS solution.  I
know, that left me spending a couple of hours typing in
'document.write' and
adding backslashes in front of quotes.  But, it kept me from having to
install an interpreter which probably wouldn't be needed for anything
else.  It's all a trade off.  I just wanted to make sure I wasn't
missing something.  Thanks again for the input.

      ... doug

From x at xman.org  Mon Aug 15 16:24:42 2005
From: x at xman.org (Christopher Smith)
Date: Mon, 15 Aug 2005 16:24:42 -0700
Subject: [OCLUG-devel] Perl and Recursion
In-Reply-To: <4300E801.6060600@colannino.org>
References: <4300E801.6060600@colannino.org>
Message-ID: <4301243A.101@xman.org>

James Colannino wrote:
>    $look_for_rpm = sub {
> 
>        #If it wasn't found and we've reached the end of directories we
> can search,
>        #we must return 0 to indicate our failure to find the package.
> 
>        if ($dirs2check_elements <= -1) {return 0;}
> 
>        while (files exist in $dirs2check[$dirs2check_elements]) {
> 
>            examine file;
> 
>            if match found, {
>                install older version of package;
>                return 1;
>            }
> 
>            else {
>                --$dirs2check_elements;
>                &$look_for_rpm();
>            }
>        }
>    };

Okay, the biggest problem I see with this code is that it isn't using a
return statement for all of it's exit points. Technically that can work
with Perl, but probably not the way you want it to.

So, just for clarity I'd do "return &$look_for_rpm();", but you still
have the case where you fall out of the while loop. I'm guessing you
want to return "0" for that case.

Generally with recursion you want to structure things as such:

sub recursiveFunction(arguments) {
    if (atEndOfRecursion) {
        return someAppropriateValue;
    } else {
        return recursiveFunction(someChange(arguments));
    }
}

The way to think of it is you can prove correctness of the code via
induction. For some end case you can prove the function returns the
correct value. Then for all other cases it returns the correct value as
long as some reduced form (i.e. n-1) of the recursive function returns
the correct value.

--Chris

From james at colannino.org  Wed Aug 24 13:53:57 2005
From: james at colannino.org (James Colannino)
Date: Wed, 24 Aug 2005 13:53:57 -0700
Subject: [OCLUG-devel] C and external variables
Message-ID: <430CDE65.50108@colannino.org>

I have a quick question about C and external variables; let's say that 
in one file I have the following:

typedef struct thing_ {
    variables;
} thing;

Now let's say that I want to declare a variable in another file with 
that external type.  How would I declare it?  Would I have to do another 
typedef in the other file, such as:

typedef extern thing thing;

I'm just guessing here, so I know I'm most likely wrong.  Thanks in advance.

James

From x at xman.org  Wed Aug 24 16:36:30 2005
From: x at xman.org (Christopher Smith)
Date: Wed, 24 Aug 2005 16:36:30 -0700
Subject: [OCLUG-devel] C and external variables
In-Reply-To: <430CDE65.50108@colannino.org>
References: <430CDE65.50108@colannino.org>
Message-ID: <430D047E.7070902@xman.org>

James Colannino wrote:
> I have a quick question about C and external variables; let's say that
> in one file I have the following:
> 
> typedef struct thing_ {
>    variables;
> } thing;
> 
> Now let's say that I want to declare a variable in another file with
> that external type.  How would I declare it?  Would I have to do another
> typedef in the other file, such as:
> 
> typedef extern thing thing;
> 
> I'm just guessing here, so I know I'm most likely wrong.  Thanks in
> advance.

extern is a modifier on the variable, not the type. Normally what you do
is the type definition is a header that is included by the "other file".

extern just tells you that the storage space for that variable is being
allocated in some other object file (and the linker is going to get
right mad at you if one of the object file's it links doesn't claim
ownership).

So, if I have type defined in "type_defined.h" that I want to have
shared between multiple object files, then any file that uses that type
should '#include "type_defined.h"'. No extern necessary.

If I have a *variable* that is used by multiple objects files, then to
avoid a conflict I need to assign ownership to one of the object files
and have the rest have the variable as an extern. So that probably means
"variable_owned.c" has the variable declared as a global (as opposed to
static), and "other_file.c" has the variable declared as extern (both
will have '#include "type_defined.h"' to get the type definition. The
linker will end up linking together "variable_owned.o" and
"other_file.o" and marry up the storage in variable_owned.o with the
references to the externed variable in other_file.o.

--Chris

From james at colannino.org  Wed Aug 24 22:29:54 2005
From: james at colannino.org (James Colannino)
Date: Wed, 24 Aug 2005 22:29:54 -0700
Subject: [OCLUG-devel] C and external variables
In-Reply-To: <430D047E.7070902@xman.org>
References: <430CDE65.50108@colannino.org> <430D047E.7070902@xman.org>
Message-ID: <430D5752.9030603@colannino.org>

Christopher Smith wrote:

>James Colannino wrote:
>  
>
>>I have a quick question about C and external variables; let's say that
>>in one file I have the following:
>>
>>typedef struct thing_ {
>>   variables;
>>} thing;
>>
>>Now let's say that I want to declare a variable in another file with
>>that external type.  How would I declare it?  Would I have to do another
>>typedef in the other file, such as:
>>
>>typedef extern thing thing;
>>
>>I'm just guessing here, so I know I'm most likely wrong.  Thanks in
>>advance.
>>    
>>
>
>extern is a modifier on the variable, not the type. Normally what you do
>is the type definition is a header that is included by the "other file".
>  
>

Now I understand.

>So, if I have type defined in "type_defined.h" that I want to have
>shared between multiple object files, then any file that uses that type
>should '#include "type_defined.h"'. No extern necessary.
>  
>

That's what I'll be doing then.  Thanks for the help.

James


From ddjolley at gmail.com  Sat Aug 27 11:52:43 2005
From: ddjolley at gmail.com (Doug Jolley)
Date: Sat, 27 Aug 2005 14:52:43 -0400
Subject: [OCLUG-devel] 
	warning: incompatible implicit declaration of built-in function blah
Message-ID: <b31d342a0508271152624b550@mail.gmail.com>

I am persistently trying to learn C. It is a struggle for me.

I had a program written on a FC2 machine and it compiled just fine and 
seemed to work OK as well. For reasons beyond my control, I had to move it 
to a FC4 machine to continue my development. Now
when I try to compile the program I get 3 warnings of the type. 'warning: 
incompatible implicit declaration of built-in function whatever'. In my case 
the functions are strlen, strcpy, and strcat. Here are the lines involved:

char *path2CountFile=(char*)malloc(strlen(cgi_val(entries,"__dataDir"))+12);
strcpy(path2CountFile,cgi_val(entries,"__dataDir"));
strcat(path2CountFile,"/count.dat");

I did some googling. One suggestion that I came across was to include 
stdlib.h. That didn't work; but, even if it had, I don't see why it should 
be suddenly required just because I switched machines. I did encounter one 
comment from an individual stating that he was having this problem on FC4; 
but, from the context it sounded to me like his problems were originating 
elsewhere.

Anyway, I'd really like to understand why I'm suddenly encountering this 
problem. I'd certainly like to know how to fix it. I'm beginning to wonder 
if I need to move back to an FC2 machine for some reason. If I do have to 
move back, I'd sure like to know why I'm making the move.

Thanks for any input.

... doug
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://localhost.localdomain/pipermail/oclug-devel/attachments/20050827/9e3f8ed4/attachment.html 

From ddjolley at gmail.com  Mon Aug 29 11:17:43 2005
From: ddjolley at gmail.com (Doug Jolley)
Date: Mon, 29 Aug 2005 14:17:43 -0400
Subject: [OCLUG-devel] Single Element Structures
Message-ID: <b31d342a0508291117377ce733@mail.gmail.com>

Hi --

I am a real newbie to C and in way over my head on a project that I'd really 
like to get to work. I'm thinking that the thing to do may be to try to get 
some help at the next OCLUG meeting; but, in the meantime I'm going to try 
to muddle along as best I can.

To continue my muddling, I have encountered the following type definition:

typedef struct {
node* head;
} llist;

My question is: Why would anyone go to the trouble of defining a structure 
with only one element in it? I thought the essence of structures was that 
they contained a collection of dissimilar data elements. I'm sure the guy 
that did this had a very good reason and I realize that I'm asking a 
question without giving a lot of context; so, from what I've given there may 
be no answer. Anyway, if an answer may be deduced from what I've given, I'd 
love to hear it. Thanks for any input.

... doug
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://localhost.localdomain/pipermail/oclug-devel/attachments/20050829/dd047c35/attachment.html