|
Thread Rules
1. This is not a "do my homework for me" thread. If you have specific questions, ask, but don't post an assignment or homework problem and expect an exact solution.
2. No recruiting for your cockamamie projects (you won't replace facebook with 3 dudes you found on the internet and $20)
3. If you can't articulate why a language is bad, don't start slinging shit about it. Just remember that nothing is worse than making CSS IE6 compatible.
4. Use [code] tags to format code blocks. |
On January 27 2014 21:28 actionbastrd wrote:So i am trying to learn C programing, and am hoping to dive really deep into it. I have a task which i am having trouble with. I feel like accomplishing this task could be done way better, but none the less, Prompt the user and accept the following types of values from a SINGLE input line: char float int char. What i have is this(spoiler), which works when the question asked is "char int char float", but breaks, i am assuming because it cannot tell when the float number ends. I understand spacing is an issue to and i commented on that in the prompt,.. but i need a separator somehow, i am learning out of books and having come across anything. I personally would just have it ask in separate lines or something. D: This is actually my first language and am really stoked on it, but its mostly a grind with little to no in person help lol. If anyone has good C programming resources i could check out and read that would be awesome too, about anything! + Show Spoiler +
char f, g;
int h;
float i;
printf("Enter a char, float, int, char; with no spaces in between each input: ");
scanf("%c%f%d%c", &f, &i, &h, &g);
printf("You entered: %c, %.3f, %d, %c.\n", f, i, h, g);
I believe this would work:
char f, g;
int h;
float i;
printf("Enter a char, float, int, char; with no spaces in between each input: ");
scanf("%c %f %d %c", &f, &i, &h, &g);
printf("You entered: %c, %.3f, %d, %c.\n", f, i, h, g);
Only difference is the scanf() format string.
Just tested it, it works! The scanf() documentation probably explains it well, but it's very similar to a Regex.
See the following:
int i=0;
sscanf("Test s15tring", "Test s%dtring", &i);
// i==15
It'll try its hardest to match it.
|
Mavvie, I already explained earlier that you need sscanf for that, not scanf 
|
Ooh silly me! Missed that one.
scanf() is okay to get the job he's doing done; it's not like security is an issue. But yeah sscanf is the better choice. Note I used it in the second example!
I'm mostly a Rails guy, nobody here ever seems to have Rails questions 
|
Hey guys I need some with formatting a number in SQL, How can I add a comma to a number after every thousand? ex: 1234567 to turn into 1,234,567?
SELECT COUNT(*) as "Tweets in Database"
from tweets_sml;
is what I have to count the number of "tweets"
|
Hyrule19209 Posts
FORMAT(COUNT(*), 0)
also, COUNT(*) is generally considered bad
|
On January 30 2014 12:55 tofucake wrote:
FORMAT(COUNT(*), 0)
also, COUNT(*) is generally considered bad
Out of curiosity, why is count(*) bad?
|
Hyrule19209 Posts
first, because you have no where clause
second, because you are counting everything returned from the result
third, because you are doing both of those things at the same time
it's even worse if you're using innodb instead of myisam because of how inno works.
I don't know how your table is structured, but if tweets aren't ever deleted (unlikely) and you have a unique id column (which you should) that happens to be an auto_incrementing int you could do SELECT MAX(id) or SELECT id ORDER BY id DESC LIMIT 1
If (more likely) tweets are deleted, and you have a silly number of rows, you should use a set of triggers which update another table (preferably in a memory table for speed) with a count.
|
|
|
On January 30 2014 13:07 tofucake wrote:
first, because you have no where clause
second, because you are counting everything returned from the result
third, because you are doing both of those things at the same time
it's even worse if you're using innodb instead of myisam because of how inno works.
I don't know how your table is structured, but if tweets aren't ever deleted (unlikely) and you have a unique id column (which you should) that happens to be an auto_incrementing int you could do SELECT MAX(id) or SELECT id ORDER BY id DESC LIMIT 1
If (more likely) tweets are deleted, and you have a silly number of rows, you should use a set of triggers which update another table (preferably in a memory table for speed) with a count.
I can only strongly disagree with this advice.
If you need exact count, using MAX(id) to determine it is probably the worst idea. There are multiple ways where auto_incrementing columns can produce holes even without deletes.
Also do not worry about optimizations if there is no need to worry about optimizations. If your application works perfectly fine, it is actually bad idea to store the count separately as you introduce unnecessary complexity and possible point of inconsistency. In general never duplicate data if performance reasons do not force you to do it. Storing count is duplicating data. Of course in real-life systems you will be in the end forced to actually do that sometimes, but do it only as a response to the actual need.
Triggers are work of the devil  and should be avoided.
Deleting data seems to me nearly always bad idea. If you have performance issues, partition, if not what is the point. Again do optimizations only in response to actual need.
EDIT: In short SELECT COUNT(*) is perfectly fine.
|
Let's play a game...
What is bar equal to? [C]:
uint8_t foo = 0x70;
uint8_t bar = (~foo) >> 4;
|
On January 31 2014 03:15 RoyGBiv_13 wrote:Let's play a game... What is bar equal to? [C]:
uint8_t foo = 0x70;
uint8_t bar = (~foo) >> 4;
In base-2:
foo = 01110000
~foo = 10001111
(~foo) >> 4 = 00001000
So bar = 8
|
On January 31 2014 03:25 Rannasha wrote:Show nested quote +On January 31 2014 03:15 RoyGBiv_13 wrote:Let's play a game... What is bar equal to? [C]:
uint8_t foo = 0x70;
uint8_t bar = (~foo) >> 4;
In base-2: foo = 01110000 ~foo = 10001111 (~foo) >> 4 = 00001000 So bar = 8
That is what you would expect. But it's wrong.
+ Show Spoiler [Solution] +
foo = 01110000
foo >> 4 = 00000111
(~foo) >> 4 = 11111000
bar == 248
|
On January 31 2014 03:38 Shenghi wrote:Show nested quote +On January 31 2014 03:25 Rannasha wrote:On January 31 2014 03:15 RoyGBiv_13 wrote:Let's play a game... What is bar equal to? [C]:
uint8_t foo = 0x70;
uint8_t bar = (~foo) >> 4;
In base-2: foo = 01110000 ~foo = 10001111 (~foo) >> 4 = 00001000 So bar = 8 That is what you would expect. But it's wrong. + Show Spoiler [Solution] +
foo = 01110000
foo >> 4 = 00000111
(~foo) >> 4 = 11111000
foo == 248
You had a typo in your last line, but it's correct:
+ Show Spoiler +bar = 248 + Show Spoiler +
That is, assuming that your compiling for a target with a word size of greater than 1 byte.
+ Show Spoiler +
Let's assuming it's a target with a 32-bit word size. For an arithmetic operation (negation, in this case, ~) the lvalue (return value of the operation) will be the default word size, and any types smaller than the word size will be promoted to the default word size. In this case, foo is promoted to a signed 32-bit int:
foo = 0x00000070
~foo = 0xFFFFFF8F
~foo >> 4 = 0xFFFFFFF8 //in this case, since the result will fit within the type, it wont be sign extended
bar is then truncated to fit within the uint8 type
bar = 0xF8 //248
|
I've recently studied about the Actor model, but it sounds too good to be true. Anyway, is the model used at workplace?
|
On February 01 2014 03:13 darkness wrote:
I've recently studied about the Actor model, but it sounds too good to be true. Anyway, is the model used at workplace?
That sounds like a very visual and understandable model
|
Can anyone show me a quick example of the Memento pattern? The example from wikipedia is just using a string. What about a more complex object that has attributes? E.g.
Person
+ Show Spoiler +
public class Person {
private int age;
private String name;
public Person(int age, String name) {
this.age = age;
this.name = name;
}
public int getAge() {
return age;
}
public String getName() {
return name;
}
public void setAge(int age) {
this.age = age;
}
public void setName(String name) {
this.name = name;
}
}
I've tried the following but I don't know if it's correct.
Some save state method, e.g:
public Memento saveState() {
return new Memento(this); // this == Person
}
And the Memento class:
public class Memento {
int age;
String name;
public Memento(Person p) {
age = p.getAge();
name = p.getName();
}
public Person getState() {
return new Person(age, name);
}
}
Am I on the right track? I've omitted the data structure part because it's obvious although I think Stack would be suitable? I've also read that using object.clone() isn't recommended, so copy constructor then?
|
Made a Mandelbrot fractal generator, but the issue is that it runs really slowly: it takes a whole minute to generate a single fractal. Would like to know whether the problem is because my program is inefficient or because of Perl / the GD library. tia
+ Show Spoiler +
#!perl
use strict;
use GD;
# Mandelbrot fractal generator
my ($width, $height) = (900, 675);
my ($x_min, $x_max) = (-2, 1);
my ($y_min, $y_max) = (-1, 1);
my ($x_min, $x_max) = (-2, .65);
my ($y_min, $y_max) = (-1, 1);
my ($y_min, $y_max) = (-1, 1);
my $multi = 1/1.1;
$_ /= $multi for ($x_min, $x_max, $y_min, $y_max);
my $x_range = $x_max - $x_min;
my $y_range = $y_max - $y_min;
my $x_offset = 0;
my $y_offset = 0;
my $it_count = 30;
my $bound = 1000;
my $filename = 'fractal.png';
open STDOUT, ">", "$filename";
binmode STDOUT;
my $bitmap = new GD::Image($width, $height, 0) or die "Could not initialize bitmap\n";
my %colors = (); # keys: "$r $g $b"
# Allocate colors like this:
# $colors{"$r $g $b"} = $bitmap->colorAllocate($r, $g, $b)
# unless exists($colors{"$r $g $b"});
#
# Set pixel like this:
# $bitmap->setPixel($pix_h, $pix_v, $colortable{"$a $b $c"});
my $black = $bitmap->colorAllocate(0, 0, 0); # Black background
my @color_inner = (128, 255, 0);
my @color_outer = (0, 0, 96);
sub color {
my ($it, $pix_h, $pix_v) = @_;
my $percent = $it/ $it_count;
$percent *= .8;
$percent = 1 if $percent > 1;
my $r = int($color_inner[0] * $percent + $color_outer[0] * (1 - $percent));
my $g = int($color_inner[1] * $percent + $color_outer[1] * (1 - $percent));
my $b = int($color_inner[2] * $percent + $color_outer[2] * (1 - $percent));
$colors{"$r $g $b"} = $bitmap->colorAllocate($r, $g, $b)
unless exists($colors{"$r $g $b"});
$bitmap->setPixel($pix_h, $pix_v, $colors{"$r $g $b"});
}
sub calc {
my ($r, $i, $it, $pix_h, $pix_v, @c) = @_;
# (a + bi) ** 2 = a ** 2 + 2abi - b ** 2
# z = z**2 + c
my $real = $r ** 2 - $i ** 2 + $c[0];
my $imag = 2 * $r * $i + $c[1];
color($it, $pix_h, $pix_v) if ($real > $bound);
return ($real, $imag);
}
for my $pix_h (0..$width-1) {
for my $pix_v (0..$height-1) {
my $r = ($pix_h + 1) / $width * $x_range + $x_min + $x_offset;
my $i = ($pix_v + 1) / $height * $y_range + $y_min + $y_offset;
my @c = ($r, $i);
for (1..$it_count) {
($r, $i) = calc($r, $i, $_, $pix_h, $pix_v, @c);
}
}
}
print $bitmap->png;
|
On January 29 2014 09:01 Badboyrune wrote:
Ok I'm getting quite confused. I'm making a pretty basic android app, it's my first so my experience is very limited. It basically displays data from a database based on intent sent from the intent sent from another activity in the app. This all worked fine, I could navigate without any issue with the android OS back button.
However when I enable "getActionBar().setDisplayHomeAsUpEnabled(true);" and try to navigate backwards with the Home button that breaks any activity that relies on receiving an intent, because it doesn't receive any. I tried putting the information in the onSaveInstanceState bundle, but the activity does not receive the bundle when navigated to with the Home button (savedInstanceState != null returns false).
I've looked through all I've been able to find online (which has been limited because searches involving android home button are generally unhelpful) and all I've really been able to find is that the home button seems to not be programmable anymore, it either sends you to the app home activity, or the activities parent activity which is what I'm trying to achieve. I'm assuming there's some way of doing this so any help would be much appreciated.
Would you mind posting a code listing of the relevant activities? There are a lot of lifecycle callbacks - onRestoreInstanceState() for example.
|
On February 02 2014 18:08 wozzot wrote:Made a Mandelbrot fractal generator, but the issue is that it runs really slowly: it takes a whole minute to generate a single fractal. Would like to know whether the problem is because my program is inefficient or because of Perl / the GD library. tia + Show Spoiler +
#!perl
use strict;
use GD;
# Mandelbrot fractal generator
my ($width, $height) = (900, 675);
my ($x_min, $x_max) = (-2, 1);
my ($y_min, $y_max) = (-1, 1);
my ($x_min, $x_max) = (-2, .65);
my ($y_min, $y_max) = (-1, 1);
my ($y_min, $y_max) = (-1, 1);
my $multi = 1/1.1;
$_ /= $multi for ($x_min, $x_max, $y_min, $y_max);
my $x_range = $x_max - $x_min;
my $y_range = $y_max - $y_min;
my $x_offset = 0;
my $y_offset = 0;
my $it_count = 30;
my $bound = 1000;
my $filename = 'fractal.png';
open STDOUT, ">", "$filename";
binmode STDOUT;
my $bitmap = new GD::Image($width, $height, 0) or die "Could not initialize bitmap\n";
my %colors = (); # keys: "$r $g $b"
# Allocate colors like this:
# $colors{"$r $g $b"} = $bitmap->colorAllocate($r, $g, $b)
# unless exists($colors{"$r $g $b"});
#
# Set pixel like this:
# $bitmap->setPixel($pix_h, $pix_v, $colortable{"$a $b $c"});
my $black = $bitmap->colorAllocate(0, 0, 0); # Black background
my @color_inner = (128, 255, 0);
my @color_outer = (0, 0, 96);
sub color {
my ($it, $pix_h, $pix_v) = @_;
my $percent = $it/ $it_count;
$percent *= .8;
$percent = 1 if $percent > 1;
my $r = int($color_inner[0] * $percent + $color_outer[0] * (1 - $percent));
my $g = int($color_inner[1] * $percent + $color_outer[1] * (1 - $percent));
my $b = int($color_inner[2] * $percent + $color_outer[2] * (1 - $percent));
$colors{"$r $g $b"} = $bitmap->colorAllocate($r, $g, $b)
unless exists($colors{"$r $g $b"});
$bitmap->setPixel($pix_h, $pix_v, $colors{"$r $g $b"});
}
sub calc {
my ($r, $i, $it, $pix_h, $pix_v, @c) = @_;
# (a + bi) ** 2 = a ** 2 + 2abi - b ** 2
# z = z**2 + c
my $real = $r ** 2 - $i ** 2 + $c[0];
my $imag = 2 * $r * $i + $c[1];
color($it, $pix_h, $pix_v) if ($real > $bound);
return ($real, $imag);
}
for my $pix_h (0..$width-1) {
for my $pix_v (0..$height-1) {
my $r = ($pix_h + 1) / $width * $x_range + $x_min + $x_offset;
my $i = ($pix_v + 1) / $height * $y_range + $y_min + $y_offset;
my @c = ($r, $i);
for (1..$it_count) {
($r, $i) = calc($r, $i, $_, $pix_h, $pix_v, @c);
}
}
}
print $bitmap->png;
Profile your code. Then it should be obvious if it's Perl, the library or your code that's bad.
|
On February 02 2014 21:30 bangsholt wrote:Show nested quote +On February 02 2014 18:08 wozzot wrote:Made a Mandelbrot fractal generator, but the issue is that it runs really slowly: it takes a whole minute to generate a single fractal. Would like to know whether the problem is because my program is inefficient or because of Perl / the GD library. tia + Show Spoiler +
#!perl
use strict;
use GD;
# Mandelbrot fractal generator
my ($width, $height) = (900, 675);
my ($x_min, $x_max) = (-2, 1);
my ($y_min, $y_max) = (-1, 1);
my ($x_min, $x_max) = (-2, .65);
my ($y_min, $y_max) = (-1, 1);
my ($y_min, $y_max) = (-1, 1);
my $multi = 1/1.1;
$_ /= $multi for ($x_min, $x_max, $y_min, $y_max);
my $x_range = $x_max - $x_min;
my $y_range = $y_max - $y_min;
my $x_offset = 0;
my $y_offset = 0;
my $it_count = 30;
my $bound = 1000;
my $filename = 'fractal.png';
open STDOUT, ">", "$filename";
binmode STDOUT;
my $bitmap = new GD::Image($width, $height, 0) or die "Could not initialize bitmap\n";
my %colors = (); # keys: "$r $g $b"
# Allocate colors like this:
# $colors{"$r $g $b"} = $bitmap->colorAllocate($r, $g, $b)
# unless exists($colors{"$r $g $b"});
#
# Set pixel like this:
# $bitmap->setPixel($pix_h, $pix_v, $colortable{"$a $b $c"});
my $black = $bitmap->colorAllocate(0, 0, 0); # Black background
my @color_inner = (128, 255, 0);
my @color_outer = (0, 0, 96);
sub color {
my ($it, $pix_h, $pix_v) = @_;
my $percent = $it/ $it_count;
$percent *= .8;
$percent = 1 if $percent > 1;
my $r = int($color_inner[0] * $percent + $color_outer[0] * (1 - $percent));
my $g = int($color_inner[1] * $percent + $color_outer[1] * (1 - $percent));
my $b = int($color_inner[2] * $percent + $color_outer[2] * (1 - $percent));
$colors{"$r $g $b"} = $bitmap->colorAllocate($r, $g, $b)
unless exists($colors{"$r $g $b"});
$bitmap->setPixel($pix_h, $pix_v, $colors{"$r $g $b"});
}
sub calc {
my ($r, $i, $it, $pix_h, $pix_v, @c) = @_;
# (a + bi) ** 2 = a ** 2 + 2abi - b ** 2
# z = z**2 + c
my $real = $r ** 2 - $i ** 2 + $c[0];
my $imag = 2 * $r * $i + $c[1];
color($it, $pix_h, $pix_v) if ($real > $bound);
return ($real, $imag);
}
for my $pix_h (0..$width-1) {
for my $pix_v (0..$height-1) {
my $r = ($pix_h + 1) / $width * $x_range + $x_min + $x_offset;
my $i = ($pix_v + 1) / $height * $y_range + $y_min + $y_offset;
my @c = ($r, $i);
for (1..$it_count) {
($r, $i) = calc($r, $i, $_, $pix_h, $pix_v, @c);
}
}
}
print $bitmap->png;
Profile your code. Then it should be obvious if it's Perl, the library or your code that's bad.
Merging the subroutine calls into the main function saved around 20 seconds, seems that Perl is slow with floating point operations though. Thanks
|
|
|
|
|
|
EPT
