Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.lang.c > #3087 > unrolled thread

malloc

Back to article view | Back to comp.lang.c

Contents

  malloc "Bill Cunningham" <nospam@nspam.invalid> - 2011-05-01 01:01 -0400
    Re: malloc "Bill Cunningham" <nospam@nspam.invalid> - 2011-05-01 01:03 -0400
    Re: malloc luser- -droog <mijoryx@yahoo.com> - 2011-04-30 22:31 -0700
      Re: malloc "Bill Cunningham" <nospam@nspam.invalid> - 2011-05-01 02:15 -0400
      Re: malloc "BartC" <bc@freeuk.com> - 2011-05-01 13:52 +0100
        Re: malloc luser- -droog <mijoryx@yahoo.com> - 2011-05-02 00:05 -0700
          Re: malloc crisgoogle <crisd@telus.net> - 2011-05-02 17:43 -0700
    Re: malloc Ian Collins <ian-news@hotmail.com> - 2011-05-01 17:35 +1200
    Re: malloc "Bill Cunningham" <nospam@nspam.invalid> - 2011-05-01 02:14 -0400
      Re: malloc Mickey Mouse <fake@email.com> - 2011-05-01 00:18 -0700
        Re: malloc "Bill Cunningham" <nospam@nspam.invalid> - 2011-05-01 04:07 -0400
          Re: malloc Barry Schwarz <schwarzb@dqel.com> - 2011-05-01 10:31 -0700
            Re: malloc "Bill Cunningham" <nospam@nspam.invalid> - 2011-05-02 16:55 -0400
            Re: malloc "Bill Cunningham" <nospam@nspam.invalid> - 2011-05-02 23:51 -0400
        Re: malloc "Bill Cunningham" <nospam@nspam.invalid> - 2011-05-02 16:58 -0400
          Re: malloc Mickey Mouse <fake@email.com> - 2011-05-02 14:34 -0700
            Re: malloc Ian Collins <ian-news@hotmail.com> - 2011-05-03 09:55 +1200
              Re: malloc "Bill Cunningham" <nospam@nspam.invalid> - 2011-05-03 10:20 -0400
                Re: malloc Angel <angel+news@spamcop.net> - 2011-05-03 14:55 +0000
                  Re: malloc "Bill Cunningham" <nospam@nspam.invalid> - 2011-05-03 15:34 -0400
                    Re: malloc Angel <angel+news@spamcop.net> - 2011-05-03 20:56 +0000
              Re: malloc "Bill Cunningham" <nospam@nspam.invalid> - 2011-05-03 11:29 -0400
            Re: malloc Seebs <usenet-nospam@seebs.net> - 2011-05-02 22:18 +0000
            Re: malloc "Default User" <defaultuserbr@yahoo.com> - 2011-05-04 12:20 -0500
              Re: malloc Keith Thompson <kst-u@mib.org> - 2011-05-04 10:52 -0700
                Re: malloc Joe Wright <joewwright@comcast.net> - 2011-05-04 14:57 -0400
                Re: malloc "Bill Cunningham" <nospam@nspam.invalid> - 2011-05-04 18:53 -0400
                  Re: malloc Angel <angel+news@spamcop.net> - 2011-05-05 08:12 +0000
      Re: malloc John Gordon <gordon@panix.com> - 2011-05-02 15:31 +0000
    Re: malloc "osmium" <r124c4u102@comcast.net> - 2011-05-01 09:56 -0500
      Re: malloc "Bill Cunningham" <nospam@nspam.invalid> - 2011-05-01 18:20 -0400
      Re: malloc "Bill Cunningham" <nospam@nspam.invalid> - 2011-05-02 21:38 -0400
        Re: malloc "osmium" <r124c4u102@comcast.net> - 2011-05-03 09:02 -0500
          Re: malloc "Bill Cunningham" <nospam@nspam.invalid> - 2011-05-03 15:21 -0400
            Re: malloc "osmium" <r124c4u102@comcast.net> - 2011-05-03 14:35 -0500
              Re: malloc "Bill Cunningham" <nospam@nspam.invalid> - 2011-05-03 18:43 -0400
    Re: malloc Tim Rentsch <txr@alumni.caltech.edu> - 2011-05-01 12:51 -0700
      Re: malloc "Bill Cunningham" <nospam@nspam.invalid> - 2011-05-02 23:49 -0400
        Re: malloc Ian Collins <ian-news@hotmail.com> - 2011-05-03 13:35 +1200
          Re: malloc "Bill Cunningham" <nospam@nspam.invalid> - 2011-05-03 11:27 -0400
            Re: malloc Ian Collins <ian-news@hotmail.com> - 2011-05-04 07:54 +1200
            Re: malloc Ben Bacarisse <ben.usenet@bsb.me.uk> - 2011-05-03 21:50 +0100

Page 2 of 3 — ← Prev page 1 [2] 3  Next page →

#3200

On 2011-05-03, Bill Cunningham <nospam@nspam.invalid> wrote:
> Angel wrote:
>
> [snip]
>
>> realloc() could, for example, be used to implement a dynamic "array"
>> that grows in size as elements are added and shrinks as elements are
>> removed.
>
>     Exactly what I was looking for. Using the example of streams and the 
> FILE* struct. I don't think fread should even be used for this come to think 
> of it. fgetc might be a better way. But I have no idea where to even begin 
> with designing a function to set aside storage and allow it to grow and 
> shrink dynamically.

Here is a very quick, very dirty (no error handling or boundary
checking) implementation of a variable-length first-in, first-out
character buffer. It's only as large as it needs to be,
and the maximum size is only limited by available memory.

(Yes, I know there are far better ways to do this. It's just an example
of using realloc(), nothing more.)


#include <stdlib.h>
#include <string.h>

static char *fifo = NULL;
static size_t fifo_count = 0;

size_t fifo_write(const char *data, const size_t len)
{
  fifo = realloc(fifo, fifo_count + len);
  memcpy(fifo + fifo_count, data, len);
  fifo_count += len;
  return len;
}

size_t fifo_read(char *data, const size_t len)
{
  memcpy(data, fifo, len);
  memmove(fifo, fifo + len, fifo_count - len);
  fifo_count -= len;
  fifo = realloc(fifo, fifo_count);
  return len;
}


-- 
The perfected state of a spam server is a smoking crater.
- The Crater Corollary to Rule #4

[toc] | [prev] | [next] | [standalone]

#3183

Ian Collins wrote:

> "Bill" first asked about malloc in 2003 (as seen from google).
>
> Go figure.

    Fill me in.

Bill

[toc] | [prev] | [next] | [standalone]

#3141

On 2011-05-02, Mickey Mouse <fake@email.com> wrote:
> You might be better posting questions in 'comp.lang.c.moderated', it
> is a mostly quiet group but one where questions get answered.

But it's also a group where the moderator might not approve posts which
were blatantly trolling.  You never know.

I honestly don't know what I'd do if one of Bill's surrealist pieces showed
up.  I have spent a fair amount of my recreational time over the last twenty
years marvelling at the many and diverse ways in which humans can be
mind-numbingly stupid, and I still can't figure out for sure whether his
posts are *possibly* the result of sincere stupidity.  I tend to suspect
that he's an extremely talented and persistent troll.  If not, I imagine
someone could get a doctorate by mapping the boundaries of his cognitive
dysfunction.

-s
-- 
Copyright 2011, all wrongs reversed.  Peter Seebach / usenet-nospam@seebs.net
http://www.seebs.net/log/ <-- lawsuits, religion, and funny pictures
http://en.wikipedia.org/wiki/Fair_Game_(Scientology) <-- get educated!
I am not speaking for my employer, although they do rent some of my opinions.

[toc] | [prev] | [next] | [standalone]

#3234

"Mickey Mouse" <fake@email.com> wrote in message 
news:kh8ur6hs3huci2rqq66aunn0f7rgd4kv4h@4ax.com...

> You might be better posting questions in 'comp.lang.c.moderated', it
> is a mostly quiet group but one where questions get answered.

Bill has supposedly been learning C for the better part of a decade. Over 
the course of numerous threads during that time, his knowledge of the 
language and specific parts of the language have waxed and waned. He will 
use some function correctly in one program, then a month later use it with 
the wrong inputs, or even claim to be unaware of it.

Many of us have come to the conclusion that it's a long-con troll. If it's 
not, then his capacity for learning the language simply doesn't exist. He 
demonstrates less capability than a first-sememster student at mid-terms. 
Either way, a waste of time to respond.



Brian

[toc] | [prev] | [next] | [standalone]

#3235

"Default User" <defaultuserbr@yahoo.com> writes:
> "Mickey Mouse" <fake@email.com> wrote in message 
> news:kh8ur6hs3huci2rqq66aunn0f7rgd4kv4h@4ax.com...
>> You might be better posting questions in 'comp.lang.c.moderated', it
>> is a mostly quiet group but one where questions get answered.
>
> Bill has supposedly been learning C for the better part of a decade. Over 
> the course of numerous threads during that time, his knowledge of the 
> language and specific parts of the language have waxed and waned. He will 
> use some function correctly in one program, then a month later use it with 
> the wrong inputs, or even claim to be unaware of it.
>
> Many of us have come to the conclusion that it's a long-con troll. If it's 
> not, then his capacity for learning the language simply doesn't exist. He 
> demonstrates less capability than a first-sememster student at mid-terms. 
> Either way, a waste of time to respond.

He's said in the past that it's due to side effects of some
medication he's on.  Either it's true, and he's (sadly) wasting his
time, or it's not, and he's (annoyingly) wasting ours.  (I'm aware
that some people here dismiss the former possibility; I don't care
to debate the point.)

If Bill really wants to learn programming, I suggest that C may be one
of the worst possible languages for him to use.  Of the languages I'm
familiar with, Python might be a much better choice.

-- 
Keith Thompson (The_Other_Keith) kst-u@mib.org  <http://www.ghoti.net/~kst>
Nokia
"We must do something.  This is something.  Therefore, we must do this."
    -- Antony Jay and Jonathan Lynn, "Yes Minister"

[toc] | [prev] | [next] | [standalone]

#3236

On 5/4/2011 13:52, Keith Thompson wrote:
> "Default User"<defaultuserbr@yahoo.com>  writes:
>> "Mickey Mouse"<fake@email.com>  wrote in message
>> news:kh8ur6hs3huci2rqq66aunn0f7rgd4kv4h@4ax.com...
>>> You might be better posting questions in 'comp.lang.c.moderated', it
>>> is a mostly quiet group but one where questions get answered.
>>
>> Bill has supposedly been learning C for the better part of a decade. Over
>> the course of numerous threads during that time, his knowledge of the
>> language and specific parts of the language have waxed and waned. He will
>> use some function correctly in one program, then a month later use it with
>> the wrong inputs, or even claim to be unaware of it.
>>
>> Many of us have come to the conclusion that it's a long-con troll. If it's
>> not, then his capacity for learning the language simply doesn't exist. He
>> demonstrates less capability than a first-sememster student at mid-terms.
>> Either way, a waste of time to respond.
>
> He's said in the past that it's due to side effects of some
> medication he's on.  Either it's true, and he's (sadly) wasting his
> time, or it's not, and he's (annoyingly) wasting ours.  (I'm aware
> that some people here dismiss the former possibility; I don't care
> to debate the point.)
>
> If Bill really wants to learn programming, I suggest that C may be one
> of the worst possible languages for him to use.  Of the languages I'm
> familiar with, Python might be a much better choice.
>
I got involved with Bill several years ago here in clc. I am not a doctor. 
Bill's condition plus his medication causes severe memory loss. He couldn't 
tell you what he had for lunch yesterday. Bill is not a troll. My take on 
him is that he is 'trying to get better' and using usenet and comp.lang.c 
to do it. I can't fault him for that.

Replying to his quests here is kind if not helpful to him. Your reply to 
Bill might be helpful to other myriad readers here. Don't stop.

-- 
Joe Wright
"If you rob Peter to pay Paul you can depend on the support of Paul."

[toc] | [prev] | [next] | [standalone]

#3237

Keith Thompson wrote:

[...]

> If Bill really wants to learn programming, I suggest that C may be one
> of the worst possible languages for him to use.  Of the languages I'm
> familiar with, Python might be a much better choice.

    I have looked at python and it totally blows my mind. Just looking at 
the code mind you and not trying to learn. I have read so many times the C 
is easy to learn. I've looked at C++ and it is alittle more understnadable 
to me. clc is the only place I know to go if I have any C questions. I know 
nothing about algorithms and computer science. I have no college training. 
I just want to learn a language. I claim nothing more. I know how to use 
functions half wy decent in C but coding for results is beyond me. Right now 
anyway. I hope that changes some day.

Bill

[toc] | [prev] | [next] | [standalone]

#3252

On 2011-05-04, Bill Cunningham <nospam@nspam.invalid> wrote:
> Keith Thompson wrote:
>
> [...]
>
>> If Bill really wants to learn programming, I suggest that C may be one
>> of the worst possible languages for him to use.  Of the languages I'm
>> familiar with, Python might be a much better choice.
>
>     I have looked at python and it totally blows my mind. Just looking at 
> the code mind you and not trying to learn. I have read so many times the C 
> is easy to learn. I've looked at C++ and it is alittle more understnadable 
> to me. clc is the only place I know to go if I have any C questions. I know 
> nothing about algorithms and computer science. I have no college training. 
> I just want to learn a language. I claim nothing more. I know how to use 
> functions half wy decent in C but coding for results is beyond me. Right now 
> anyway. I hope that changes some day.

While the syntax of C is fairly simple, the language allows way too much
freedom that can result in bad code or subtle bugs, so IMHO it is not
the best language for someone completely new to programming. For
learning purposes, a strongly typed and rigidly structured language
might be better.

I started with Pascal myself back in high school (okay, I did C64 BASIC
before that), but nowadays Java seems to be the educational language of
choice. If you've seen C++, Java will look very familiar.

If you still want to learn C, well the FAQ for this newsgroup holds a
lot of answers to common questions, take a look there. There are also
free online courses available, just Google a bit for them.

Of course, like with so many computer things, the best way to learn it
is to play with it. Start with small things, like for example a program
that works like the Unix command "cat", and work your way up from there.


-- 
The perfected state of a spam server is a smoking crater.
- The Crater Corollary to Rule #4

[toc] | [prev] | [next] | [standalone]

#3128

In <4dbcfa24$0$8384$bbae4d71@news.suddenlink.net> "Bill Cunningham" <nospam@nspam.invalid> writes:


>     Let me see if I can make this easier to understand.

> fread(p,sizeof(int),20,fp);

> Ok nothing will go above 20 size_t's but if only 15 size_t's are filled I 
> want to return 5 to memory. Is there a way to do that?

Allocate a suitably large buffer with malloc().
Call fread() to put stuff in the buffer.
If the buffer ends up being too large, resize it with realloc().

I noticed a problem with your original code:

> char p[20];
> t = fread(p, sizeof(int), 20, fp);

p is declared as an array of 20 characters, however you're trying to stuff
20 ints into it, which is too much.  If you know you will put ints into p,
declare it that way.

-- 
John Gordon                   A is for Amy, who fell down the stairs
gordon@panix.com              B is for Basil, assaulted by bears
                                -- Edward Gorey, "The Gashlycrumb Tinies"

[toc] | [prev] | [next] | [standalone]

#3105

"Bill Cunningham" wrote:

>    I have this unfinished untested code that I would like to enquire to 
> someone that has used malloc() before if I can use it here.
>
> #include <stdio.h>
> #include <stdlib.h>
>
> int main(int argc, char **argv)
> {
>    if (argc > 10) {
>        printf(" too many characters in name\n try 10 or less\n");
>        exit(1);
>    }
>    char p[20];
>    FILE *fp;
>    size_t t;
>    if ((fopen("argv[1]", "r")) == EOF) {
>        fprintf(stderr, "fopen error\n");
>        exit(1);
>    }
>    t = fread(p, sizeof(int), 20, fp);
>    if (t = ferror(fp)) {
>        fprintf(stderr, " file error indicated\n");
>        exit(1);
>    } else if (t = feof(fp)) {
>        fprintf(stderr, " EOF indicated\n");
>        exit(1);
>    }
>
>    In the code above the 3rd parameter of malloc() is taking 20 character 
> members. That's not enough. But I don't want buffer overflow either. I 
> know this is kind of reinventing the wheel so far in what is above but 
> there is a reason for this. I am trying to write a text editor that takes 
> code that only ends with "+" instead of C's normal ";". Can malloc be put 
> where the 20 is and just take a malloc? malloc returns void * and this 3rd 
> parameter a size_t so maybe a struct or union would be easier. Such as
>
> struct values{
>    size_t t;
>    void * val;
> };
>
>    Would that eliminate the need for casting?

For the umpteenth time.

I suggest you try something simpler that you might actually get to work. 
Write a program that tries to read 20 char from a file that may or may not 
exist.  The file's name is specified by a command line argument.

You can create a file to test your program by using a text editor.

[toc] | [prev] | [next] | [standalone]

#3112

osmium wrote:

> For the umpteenth time.
>
> I suggest you try something simpler that you might actually get to
> work. Write a program that tries to read 20 char from a file that may
> or may not exist.  The file's name is specified by a command line
> argument.
> You can create a file to test your program by using a text editor.

    Looking over that code I have noticed several bugs. the argv[1] 
shouldn't be in quotes. The tests for t isn't an assignment like written but 
should be t== not t=. If I can't write something like this I'm a very sad 
story. I've had success with fgetc more anyway.

Bill

[toc] | [prev] | [next] | [standalone]

#3154

osmium wrote:
> For the umpteenth time.
>
> I suggest you try something simpler that you might actually get to
> work. Write a program that tries to read 20 char from a file that may
> or may not exist.

    Why?

The file's name is specified by a command line
> argument.

So?

>
> You can create a file to test your program by using a text editor.

    What's this post all about? What's it's title? I could've picked picked 
a better title true. Which do you think would've made a better title?

1) stream question or
2) memory allocation question ?

Bill

[toc] | [prev] | [next] | [standalone]

#3176

"Bill Cunningham" wrote:

> osmium wrote:
>> For the umpteenth time.
>>
>> I suggest you try something simpler that you might actually get to
>> work. Write a program that tries to read 20 char from a file that may
>> or may not exist.
>
>    Why?
>
> The file's name is specified by a command line
>> argument.
>
> So?
>
>>
>> You can create a file to test your program by using a text editor.
>
>    What's this post all about? What's it's title? I could've picked picked 
> a better title true. Which do you think would've made a better title?
>
> 1) stream question or
> 2) memory allocation question ?

Your post made clear that you were trying to skip the fundamentals and go 
right to the meat of the thing; for this week your target was "malloc".  My 
suggestion was for a more ordered approach in which you might actually learn 
something and get a program you had written to work.  But clearly you don't 
like that approach.

Your approach reminds me of someone in a foreign country opening an English 
dictionary at random, and asking questions about a few words every week. 
What is a pharmacist?  What is a dormer?  What is rubidium?  What is a 
carnival?

Since you are unlikely to live for much more than 100 years, I think that 
method is doomed to failure. 

[toc] | [prev] | [next] | [standalone]

#3189

osmium wrote:

> Your post made clear that you were trying to skip the fundamentals
> and go right to the meat of the thing; for this week your target was
> "malloc".  My suggestion was for a more ordered approach in which you
> might actually learn something and get a program you had written to
> work.  But clearly you don't like that approach.
[...]

    How much more to "the meat of the thing" is malloc? I'm not sure that 
I'm understanding you. What do you mean by a "more ordered approach" ? The 
program that was written did not included malloc. And I noticed later it was 
full of bugs as posted. I must admit I need to start testing code and 
examining it more closely before posting. fread as was in the program is 
designed to take a set number of data.
    Hence my point. I am in no position to write code that expands and 
ontracts memory storage on the fly. That was what my inquiry was concerning. 
Using the FILE* and streams as an example.

Bill

[toc] | [prev] | [next] | [standalone]

#3192

"Bill Cunningham" wrote:

> osmium wrote:
>
>> Your post made clear that you were trying to skip the fundamentals
>> and go right to the meat of the thing; for this week your target was
>> "malloc".  My suggestion was for a more ordered approach in which you
>> might actually learn something and get a program you had written to
>> work.  But clearly you don't like that approach.
> [...]
>
>    How much more to "the meat of the thing" is malloc? I'm not sure that 
> I'm understanding you. What do you mean by a "more ordered approach" ? The 
> program that was written did not included malloc. And I noticed later it 
> was full of bugs as posted. I must admit I need to start testing code and 
> examining it more closely before posting. fread as was in the program is 
> designed to take a set number of data.
>    Hence my point. I am in no position to write code that expands and 
> ontracts memory storage on the fly. That was what my inquiry was 
> concerning. Using the FILE* and streams as an example.

Memory that expands or contracts is more complicated than a fixed target of 
known maximum size.Your post tells me you can't even handle that.  Thus the 
simpler assignment I suggested.  Your knowledge of argc, for example is 
woefully wrong..  If you understand the basics you should be able to do what 
I wrote in five minutes or less. 

[toc] | [prev] | [next] | [standalone]

#3204

osmium wrote:

> Memory that expands or contracts is more complicated than a fixed
> target of known maximum size.Your post tells me you can't even handle
> that.  Thus the simpler assignment I suggested.  Your knowledge of
> argc, for example is woefully wrong..  If you understand the basics
> you should be able to do what I wrote in five minutes or less.

    You're obviously more experienced than I. And have a better 
understanding of C and programming in general.

Bill

[toc] | [prev] | [next] | [standalone]

#3110

"Bill Cunningham" <nospam@nspam.invalid> writes:

>     I have this unfinished untested code that I would like to enquire to 
> someone that has used malloc() before if I can use it here.
>
> [snip]

You might want to try posting this again 11 months from now.

[toc] | [prev] | [next] | [standalone]

#3149

"Tim Rentsch" <txr@alumni.caltech.edu> wrote in

> You might want to try posting this again 11 months from now.

    I don't believe I've ever posted any question like this before 
concerning memory reallocation.

Bill

[toc] | [prev] | [next] | [standalone]

#3153

On 05/ 3/11 03:49 PM, Bill Cunningham wrote:
> "Tim Rentsch"<txr@alumni.caltech.edu>  wrote in
>
>> You might want to try posting this again 11 months from now.
>
>      I don't believe I've ever posted any question like this before
> concerning memory reallocation.

http://groups.google.com/group/comp.lang.c/browse_thread/thread/8505637713d6e1eb/885a55feeb4d25d0

-- 
Ian Collins

[toc] | [prev] | [next] | [standalone]

#3182

Ian Collins wrote:

 http://groups.google.com/group/comp.lang.c/browse_thread/thread/8505637713d6e1eb/885a55feeb4d25d0    Nope after viewing the link above, I think I can still say I don'tbelieve I've ever posted anything about memory allocation. Encrypting andrandom strings maybe. I've never used malloc before in code but it is verypopular.B

[toc] | [prev] | [next] | [standalone]

Page 2 of 3 — ← Prev page 1 [2] 3  Next page →

Back to top | Article view | comp.lang.c

csiph-web