Problems using newpad with a size larger than the screen - rust

I'm trying to scroll some text larger than the screen.
The docs say newpad is not limited by the screen size, but initiating it with values greater than the terminal available columns or lines fails to print anything:
newpad(LINES + 1, COLS); // fails
newpad(LINES, COLS); // works
Entire code for reference:
extern crate ncurses;
use ncurses::*;
fn main() {
initscr();
start_color();
use_default_colors();
cbreak();
noecho();
curs_set(CURSOR_VISIBILITY::CURSOR_INVISIBLE);
let pad = newpad(1000, COLS);
refresh();
let mut x = 0;
while x < 1000 {
x += 1;
wprintw(pad, &format!("Line number {}\n", x));
}
prefresh(pad, 0, 0, 0, 0, LINES, COLS);
getch();
endwin();
}

The behavior is a bit odd.
If the number of lines or rows is greater than the viewport, the last two prefresh arguments must be at most LINES - 1 and COLS - 1 respectively:
prefresh(pad, 0, 0, 0, 0, LINES - 1, COLS - 1);
If it's less, there's no need to subtract 1, as the code will work as expected.

Related

Rust Scan function detailed usage explanation

This a code solution for the problem Delete and Earn, however, I fail to grasp the scan loop part.
The problem states the following.
You are given an integer array nums.You want to maximize the number of points you get by
performing the following operation any number of times:
Pick any nums[i] and delete it to earn nums[i] points. Afterwards,
you must delete every element equal to nums[i] - 1 and every element equal to nums[i] + 1.
Return the maximum number of points you can earn by applying the above operation some number of times.
use std::collections::HashMap;
impl Solution {
pub fn delete_and_earn(mut nums: Vec<i32>) -> i32 {
let (min, max, counts) =
nums.into_iter()
.fold((0, 0, HashMap::new()), |(min, max, mut counts), n| {
*counts.entry(n).or_insert(0) += 1;
(min.min(n), max.max(n), counts)
});;
(min..=max)
.filter_map(|n| counts.get(&n).map(|count| (n, count)))
.scan((0, 0, 0), |(a, b, m),(n, count)| {
let prev_max = *a.max(b);
*a = if *m == n - 1 { *b } else { prev_max } + n * count;
*b = prev_max;
*m = n;
Some(*a.max(b))
})
.last()
.expect("invariant violation")
}
}

Vulkan: Vertex Buffer doesn't get sent to vertex shader

I am learning Vulkan and started having a problem where no vertices would get displayed.
After analyzing my program with RenderDoc (https://renderdoc.org/builds),
I realized that the buffer containing the vertex and index information contained the rights values.
At the end of the same buffer, the indices data:
The problem is that when I check the data that is transmitted to the vertex shader, it is empty:
Here is the command buffer section where it is supposed to send the data to the shader:
VkDeviceSize indicesOffset = sizeof(Vertex) * this->nbVertices;
VkDeviceSize offsets[] = {0};
vkCmdBindVertexBuffers(commandBuffers[i], 0, 1, &this->vertexBuffer, offsets);
vkCmdBindIndexBuffer(commandBuffers[i], this->vertexBuffer, indicesOffset, VK_INDEX_TYPE_UINT32);
for(size_t j = 0 ; j < this->models.size() ; j++){
Model *model = this->models[j];
uint32_t modelDynamicOffset = j * static_cast<uint32_t>(this->uniformDynamicAlignment);
VkDescriptorSet* modelDescriptorSet = model->getDescriptorSet(i);
vkCmdBindDescriptorSets(this->commandBuffers[i], VK_PIPELINE_BIND_POINT_GRAPHICS, pipelineLayout, 0, 1, modelDescriptorSet, 1, &modelDynamicOffset);
vkCmdDrawIndexed(commandBuffers[i], this->nbIndices, 1, 0, indicesOffset, 0);
}
Also, here is how I create the vertex buffer:
void Application::createVertexBuffers() {
for(Model *model : this->models){
for(Vertex vertex : model->getVertices()){
vertices.push_back(vertex);
}
for(uint32_t index : model->getIndices()){
indices.push_back(index);
}
}
VkDeviceSize vertexBufferSize = sizeof(vertices[0]) * vertices.size();
VkDeviceSize indexBufferSize = sizeof(uint32_t) * indices.size();
this->nbVertices = vertices.size();
this->nbIndices = indices.size();
VkBuffer stagingBuffer;
VkDeviceMemory stagingBufferMemory;
//To CPU
this->createBuffer(vertexBufferSize + indexBufferSize,
VK_BUFFER_USAGE_TRANSFER_SRC_BIT,
VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT | VK_MEMORY_PROPERTY_HOST_COHERENT_BIT,
stagingBuffer,
stagingBufferMemory);
void *data;
vkMapMemory(device, stagingBufferMemory, 0, vertexBufferSize, 0, &data);
memcpy(data, vertices.data(), (size_t)vertexBufferSize);
vkUnmapMemory(device, stagingBufferMemory);
//Add the index data after vertex data
vkMapMemory(device, stagingBufferMemory, vertexBufferSize, indexBufferSize, 0, &data);
memcpy(data, indices.data(), (size_t)indexBufferSize);
vkUnmapMemory(device, stagingBufferMemory);
//To GPU
this->createBuffer(vertexBufferSize + indexBufferSize,
VK_BUFFER_USAGE_TRANSFER_DST_BIT | VK_BUFFER_USAGE_VERTEX_BUFFER_BIT | VK_BUFFER_USAGE_INDEX_BUFFER_BIT,
VK_MEMORY_PROPERTY_DEVICE_LOCAL_BIT,
this->vertexBuffer,
this->vertexBufferMemory);
this->copyBuffer(stagingBuffer, this->vertexBuffer, vertexBufferSize + indexBufferSize);
vkDestroyBuffer(device, stagingBuffer, nullptr);
vkFreeMemory(device, stagingBufferMemory, nullptr);
}
If you need more information to help me solve my problem, please tell me.
Thank you.
The indices that renderdoc reports for the render are a bit high.
You pass indicesOffset as vertexOffset in your draw command. Which is:
vertexOffset is the value added to the vertex index before indexing into the vertex buffer.
So replace that with 0 and you should get your proper vertices again.

zoom in / out the image inside a pictureBox in VC++2015 (OpenCV used)

OS : Win 10
IDE: Visual Studio 2015
Language: C++
Others: I use OpenCV 3.4
I just create a Window Form using CLR empty project,
then put on a pictureBox & three buttons.
First button: load a local image and show on the pictureBox:
pictureBox1->Image = Image::FromFile("D:/something.png");
global_mat = imread("D:/something.png", 1); // global_mat is a global Mat.
zoom_in_counter = 0; // zoom_in_counter is a global int.
Second button: zoom in the image in the pictureBox
if (zoom_in_counter < 5) // You can only enlarge the image 5 times.
{
Mat new_mat = Mat::zeros(0, 0, CV_8UC3);
resize(global_mat, new_mat, cv::Size(global_mat.cols * 2, global_mat.rows * 2));
global_mat = new_mat;
if ((pictureBox1->Width != new_mat.cols) || (pictureBox1->Height != new_mat.rows))
{
pictureBox1->Width = new_mat.cols;
pictureBox1->Height = new_mat.rows;
pictureBox1->Image = gcnew System::Drawing::Bitmap(new_mat.cols, new_mat.rows);
}
System::Drawing::Bitmap^ bmpImage = gcnew Bitmap(
new_mat.cols, new_mat.rows, new_mat.step,
System::Drawing::Imaging::PixelFormat::Format24bppRgb,
System::IntPtr(new_mat.data)
);
Graphics^ g = Graphics::FromImage(pictureBox1->Image);
g->DrawImage(bmpImage, 0, 0, new_mat.cols, new_mat.rows);
pictureBox1->Refresh();
delete g;
zoom_in_counter++;
}
Third buttin: zoom out the image in the pictureBox
if (zoom_in_counter > 0) // You can't shrink the image.
{
Mat new_mat = Mat::zeros(0, 0, CV_8UC3);
resize(global_mat, new_mat, cv::Size(global_mat.cols * 0.5, global_mat.rows * 0.5));
global_mat = new_mat;
if ((pictureBox1->Width != new_mat.cols) || (pictureBox1->Height != new_mat.rows))
{
pictureBox1->Width = new_mat.cols;
pictureBox1->Height = new_mat.rows;
pictureBox1->Image = gcnew System::Drawing::Bitmap(new_mat.cols, new_mat.rows);
}
System::Drawing::Bitmap^ bmpImage = gcnew Bitmap(
new_mat.cols, new_mat.rows, new_mat.step,
System::Drawing::Imaging::PixelFormat::Format24bppRgb,
System::IntPtr(new_mat.data)
);
Graphics^ g = Graphics::FromImage(pictureBox1->Image);
g->DrawImage(bmpImage, 0, 0, new_mat.cols, new_mat.rows);
pictureBox1->Refresh();
delete g;
zoom_in_counter--;
}
And then,
every I zoom in or zoom out, it works,
excludeingthe image is zoomed back to the original size.
I'll get such error message:
An unhandled exception of type 'System.ArgumentException' occurred in System.Drawing.dll
It's really odd!
Finally, my friend figure out what's wrong,
please refer to:
https://msdn.microsoft.com/zh-tw/library/windows/desktop/ms536315(v=vs.85).aspx
stride [in]
Type: INT
Integer that specifies the byte offset between the beginning of one scan line and the next. This is usually (but not necessarily) the number of bytes in the pixel format (for example, 2 for 16 bits per pixel) multiplied by the width of the bitmap. The value passed to this parameter must be a multiple of four.

CS50 Mario Pyramid in C - Hashes not printing out as expected by code logic

#include <stdio.h>
#include <cs50.h>
int main(void)
{
int height, row, spaces, hashes; //declare all variables to be used
do //prompt the user for valid height inputs that are non-negative and less than 23.
{
printf("Height:");
height = get_int();
} while ((height < 0) || (height > 23));
for (row = 1; row <= height; row ++) // iterate through each row of the height
{
for (spaces = (height - row); spaces >= 0; spaces--) //print the number of spaces for the ith row
{
printf(" ");
}
for (hashes = (row + 1); hashes <= (height + 1); hashes++) //print the number of hashes for each row
{
printf("#");
}
printf("\n");
}
}
Not sure what I am missing in the above code. The spaces print as expected, but the behavior of the hashes is inverted, printing the highest and decreasing.
Let's assume the user submitted 12 as height.
You start the outer loop with row = 1, which gives spaces from height - row = 11 to 0, makes 12 spaces.
The hashes go from row + 1 = 2 to height + 1 = 13, makes 12 hashes.
Next iteration of the loop: 11 spaces, 11 hashes.
What you have expected can be achieved if you changes the second inner loop:
for (hashes = 0; hashes < row; hashes++) //print the number of hashes for each row
If this is not exactly what you need, change beginning of the loop to hashes = 1; or the end to row - 1; or whatever you need.
Your mistake was to assume that hashes++ makes it ascending. It does, technically, but it only changes the order of the loop, not how often it is run. Since you do not need the actual value of hashes, the order is irrelevant. The only thing that counts is how often the loop is run ;)

Draw cube vertices with fewest number of steps

What's the fewest number of steps needed to draw all of the cube's vertices, without picking up the pen from the paper?
So far I have reduced it to 16 steps:
0, 0, 0
0, 0, 1
0, 1, 1
1, 1, 1
1, 1, 0
0, 1, 0
0, 0, 0
1, 0, 0
1, 0, 1
0, 0, 1
0, 1, 1
0, 1, 0
1, 1, 0
1, 0, 0
1, 0, 1
1, 1, 1
I presume it can be reduced less than 16 steps as there are only 12 vertices to be drawn
You can view a working example in three.js javascript here:
http://jsfiddle.net/kmturley/5aeucehf/show/
Well I encoded a small brute force solver for this
the best solution is with 16 vertexes
took about 11.6 sec to compute
all is in C++ (visualization by OpenGL)
First the cube representation:
//---------------------------------------------------------------------------
#define a 0.5
double pnt[]=
{
-a,-a,-a, // point 0
-a,-a,+a,
-a,+a,-a,
-a,+a,+a,
+a,-a,-a,
+a,-a,+a,
+a,+a,-a,
+a,+a,+a, // point 7
1e101,1e101,1e101, // end tag
};
#undef a
int lin[]=
{
0,1,
0,2,
0,4,
1,3,
1,5,
2,3,
2,6,
3,7,
4,5,
4,6,
5,7,
6,7,
-1,-1, // end tag
};
// int solution[]={ 0, 1, 3, 1, 5, 4, 0, 2, 3, 7, 5, 4, 6, 2, 6, 7, -1 }; // found polyline solution
//---------------------------------------------------------------------------
void draw_lin(double *pnt,int *lin)
{
glBegin(GL_LINES);
for (int i=0;lin[i]>=0;)
{
glVertex3dv(pnt+(lin[i]*3)); i++;
glVertex3dv(pnt+(lin[i]*3)); i++;
}
glEnd();
}
//---------------------------------------------------------------------------
void draw_pol(double *pnt,int *pol)
{
glBegin(GL_LINE_STRIP);
for (int i=0;pol[i]>=0;i++) glVertex3dv(pnt+(pol[i]*3));
glEnd();
}
//---------------------------------------------------------------------------
Now the solver:
//---------------------------------------------------------------------------
struct _vtx // vertex
{
List<int> i; // connected to (vertexes...)
_vtx(){}; _vtx(_vtx& a){ *this=a; }; ~_vtx(){}; _vtx* operator = (const _vtx *a) { *this=*a; return this; }; /*_vtx* operator = (const _vtx &a) { ...copy... return this; };*/
};
const int _max=16; // know solution size (do not bother to find longer solutions)
int use[_max],uses=0; // temp line usage flag
int pol[_max],pols=0; // temp solution
int sol[_max+2],sols=0; // best found solution
List<_vtx> vtx; // model vertexes + connection info
//---------------------------------------------------------------------------
void _solve(int a)
{
_vtx *v; int i,j,k,l,a0,a1,b0,b1;
// add point to actual polyline
pol[pols]=a; pols++; v=&vtx[a];
// test for solution
for (l=0,i=0;i<uses;i++) use[i]=0;
for (a0=pol[0],a1=pol[1],i=1;i<pols;i++,a0=a1,a1=pol[i])
for (j=0,k=0;k<uses;k++)
{
b0=lin[j]; j++;
b1=lin[j]; j++;
if (!use[k]) if (((a0==b0)&&(a1==b1))||((a0==b1)&&(a1==b0))) { use[k]=1; l++; }
}
if (l==uses) // better solution found
if ((pols<sols)||(sol[0]==-1))
for (sols=0;sols<pols;sols++) sol[sols]=pol[sols];
// recursion only if pol not too big
if (pols+1<sols) for (i=0;i<v->i.num;i++) _solve(v->i.dat[i]);
// back to previous state
pols--; pol[pols]=-1;
}
//---------------------------------------------------------------------------
void solve(double *pnt,int *lin)
{
int i,j,a0,a1;
// init sizes
for (i=0;i<_max;i++) { use[i]=0; pol[i]=-1; sol[i]=-1; }
for(i=0,j=0;pnt[i]<1e100;i+=3,j++); vtx.allocate(j); vtx.num=j;
for(i=0;i<vtx.num;i++) vtx[i].i.num=0;
// init connections
for(uses=0,i=0;lin[i]>=0;uses++)
{
a0=lin[i]; i++;
a1=lin[i]; i++;
vtx[a0].i.add(a1);
vtx[a1].i.add(a0);
}
// start actual solution (does not matter which vertex on cube is first)
pols=0; sols=_max+1; _solve(0);
sol[sols]=-1; if (sol[0]<0) sols=0;
}
//---------------------------------------------------------------------------
Usage:
solve(pnt,lin); // call once to compute the solution
glColor3f(0.2,0.2,0.2); draw_lin(pnt,lin); // draw gray outline
glColor3f(1.0,1.0,1.0); draw_pol(pnt,sol); // overwrite by solution to visually check correctness (Z-buffer must pass also on equal values!!!)
List
is just mine template for dynamic array
List<int> x is equivalent to int x[]
x.add(5) ... adds 5 to the end of list
x.num is the used size of list in entries
x.allocate(100) preallocate list size to 100 entries (to avoid relocations slowdowns)
solve(pnt,lin) algorithm
first prepare vertex data
each vertex vtx[i] corresponds to point i-th point in pnt table
i[] list contains the index of each vertex connected to this vertex
start with vertex 0 (on cube is irrelevant the start point
otherwise there would be for loop through every vertex as start point
_solve(a)
it adds a vertex index to actual solution pol[pols]
then test how many lines is present in actual solution
and if all lines from lin[] are drawn and solution is smaller than already found one
copy it as new solution
after test if actual solution is not too long recursively add next vertex
as one of the vertex that is connected to last vertex used
to limit the number of combinations
at the end sol[sols] hold the solution vertex index list
sols is the number of vertexes used (lines-1)
[Notes]
the code is not very clean but it works (sorry for that)
hope I did not forget to copy something

Resources