-
Stay up-to-date with everything C++! Content directly fetched from the subreddit just for you. Join our group for discussions : @programminginc Powered by : @r_channels
The case of the crash when destructing a std::map
https://devblogs.microsoft.com/oldnewthing/20240927-00/?p=110320
https://redd.it/1fr5d6n
@r_cpp
double to string conversion in 150 lines of code
https://vitaut.net/posts/2024/simple-dtoa/
https://redd.it/1fqytki
@r_cpp
Doubt regarding while loop.Why these two gives different results i'm unable to get the logic behind it.
int j = 1;
while(j<=n){
int i = 1;
while (i<=n)
{
cout<<i;
i =i+1;
}
cout<<endl;
j=j+1;
}
Andint j = 1, i = 1; while(j<=n){
while (i<=n) { cout<<i ; i =i+1; }
cout<<endl;
j=j+1; }
https://redd.it/1fqsuaf
@r_cpp
std::array in C++ isn′t slower than array in C
https://pvs-studio.com/en/blog/posts/cpp/1164/
https://redd.it/1fqnejn
@r_cpp
Proposal for testing global functions
I am writing tests for an embedded project and found myself wondering why testing global / static or free functions is so controversial and why it would have to contain expensive trickery with the linker or other parts of the compiler in order to achieve them ([related stackoverflow post](https://stackoverflow.com/questions/28392277/mocking-free-function)).
This is why wanted to share my approach on this which is based on:
* some sort of \`typeid\` method ([with no rtti for embedded](https://stackoverflow.com/questions/8001207/compile-time-typeid-without-rtti-with-gcc))
* a mapping of function pointers (used unordered\_map here for simplicity)
* a macro-based check in every mocked function with some sort of call-forwarding
The drawback with this method is obviously the last point which requires that every function you want to mock out has to have this check at the beginning, and the need to have a globally available mocking repository. Its a POC at best and I'm looking for ways on how to improve it further. Thoughts about the usefulness of this approach in larger projects?
#pragma once
#include <unordered_map>
#ifndef __TEST__
#define ADD_MOCK(FOO, MOCK)
#define CLEAR_MOCK(FOO)
#define CLEAR_MOCKS()
#define CHECK_MOCK(FOO, ...)
#else
#define ADD_MOCK(FOO, MOCK) Mocks::add(FOO, MOCK)
#define CLEAR_MOCK(FOO) Mocks::remove(FOO)
#define CLEAR_MOCKS() Mocks::clear()
#define CHECK_MOCK(FOO, ...) \
do { \
auto _mock = Mocks::get(FOO); \
if (!_mock) break; \
return _mock(__VA_ARGS__); \
} while (0)
template <class T>
size_t typeidNoRtti(__attribute__((unused)) const T value = nullptr) {
static T *marker = NULL;
return (size_t)▮
}
namespace Mocks
{
struct MockFn {
void *mock;
size_t type;
};
extern std::unordered_map<void *, MockFn> _mocks;
template <typename T>
static inline void add(const T &fn, const T &mock) {
_mocks.insert(std::make_pair((void *)fn, MockFn{(void *)mock, typeidNoRtti(fn)}));
}
template <typename T>
static inline void remove(const T &fn) { _mocks.erase((void *)fn); }
static inline void clear() { _mocks.clear(); }
template <typename T>
static inline T *get(const T &fn) {
if (!_mocks.count((void *)fn)) return nullptr;
const auto &mock = _mocks.at((void *)fn);
if (typeidNoRtti(fn) != mock.type) return nullptr;
return (T *)mock.mock;
}
};
#endif
Usage:
#include "test.hpp"
static int8_t foo(int8_t a) {
CHECK_MOCK(foo, a);
return a;
}
static int8_t foo_mock(int8_t a) { return -a; }
static int bar(int a) { return 10 * a; }
UNIT_TEST(Mock, Basic) {
ADD_MOCK(foo, foo_mock);
EXPECT_EQ(Mocks::get(foo), foo_mock);
EXPECT_EQ(Mocks::get(bar), nullptr);
EXPECT_EQ(Mocks::get((void *)foo), nullptr);
EXPECT_EQ(foo(1), -1);
CLEAR_MOCK(foo);
EXPECT_EQ(Mocks::get(foo), nullptr);
EXPECT_EQ(foo(1), 1);
}
https://redd.it/1fqlb4c
@r_cpp
Dapfor MFC Grid manual
Hi,
I'm building an app using Dapfor MFC. Do any of you guys have experience working with the plugin in your projects, specifically for handling lots of data. Some sample apps and advice is appreciated.
https://redd.it/1fq7bob
@r_cpp
Searching for c++ groups in Mexico
I'm a CS student very interested in language development and standardization, and I saw in Jason turner's videos that some groups meet and talk about this. I might not have a lot to say, but I'd sure like to listen to these, are there any groups that I could join?
I'm in Mexico for in-person meetings but in open for online.
https://redd.it/1fq3frh
@r_cpp
Best C++ Linux IDE/Code Editors for Large Projects
Hi, I'm currently a recent grad who got their first job as a backend C++ developer. Currently my company builds and develops on Windows using Visual Studio. But recently they are moving to Linux. This is a very large codebase of Document Processing and Delivery software with hundreds of thousands of lines and 100s of components in the core backend.
As a part of my training and first tasks , I have to report some bugs and errors with the Linux build. I'm currently debugging using VS Code which is running GDB behind the scenes, but I find as the processing goes further and further each debug step is starting to take 30 - 60 seconds+ which is a huge waste of time. Compared to this on Windows with Visual Studio its almost instant and in milliseconds. I wanted to know if this is normal or am i doing something wrong?
Also what are some of the IDEs/Editors etc people working on large and complex c++ codebases in linux use to debug.
Thanks in advance
https://redd.it/1fptxeh
@r_cpp
Thread safe queue without blocking producers
I want to make a queue thread safe.I used a common mutex to be used by both producer and consumer during the enqueue and dequeue operations.I also added a condition variable so as to notify the consumers when data is pushed to the queue ,thereby , instead of consumers busy waiting they can go to sleep when the queue is empty.
But my concern is I don’t want the other producers to go on waiting when the lock is not available.
Is there any solution for this? Can atomic help?
Or I will have to compromise on one?
https://redd.it/1fprutg
@r_cpp
[ROOKIE] dereferenced pointer to a struct becomes wrong after push_back()
im trying to delete a node and the associated edges from Graph graph\_ with the below code snippets:
struct Node;
typedef Node * NodePtr;
typedef std::vector<Node> NodeVector;
typedef NodeVector Graph;
struct DirectionalEdge
{
unsigned int edgeid; // Edge identifier
NodePtr start{nullptr}; // Ptr to starting node of edge
NodePtr end{nullptr}; // Ptr to ending node of edge
EdgeCost edge_cost; // Cost information associated with edge
Metadata metadata; // Any metadata stored in the graph file of interest
Operations operations; // Operations to perform related to the edge
};
typedef DirectionalEdge * EdgePtr;
typedef std::vector<DirectionalEdge> EdgeVector;
typedef std::vector<EdgePtr> EdgePtrVector;
struct Node
{
unsigned int nodeid; // Node identifier
Coordinates coords; // Coordinates of node
EdgeVector neighbors; // Directed neighbors and edges of the node
Metadata metadata; // Any metadata stored in the graph file of interest
Operations operations; // Operations to perform related to the node
SearchState search_state; // State maintained by route search algorithm
void addEdge(
EdgeCost & cost, NodePtr node, unsigned int edgeid, Metadata meta_data = {},
Operations operations_data = {})
{
neighbors.push_back({edgeid, this, node, cost, meta_data, operations_data});
}
};
in another .cpp file, i have typedef std::map<unsigned int, NodePtr> NodeIDToNodePtrMap; to delete node and to access node struct.code:
void UpdateNodePositionPlugin::DeleteNode()
{
if (graph_.empty()) {
RCLCPP_WARN(update_node->get_logger(), "Graph is empty. Nothing to delete.");
return;
}
selected_nodeid = checkSelectedRadioButton();
for (auto& pair : node_id_to_node_ptr_map_) {
NodePtr node_ptr = pair.second; // 1. For each NodePtr
if (node_ptr) { // 2. Check if NodePtr is not null
// 3. Remove edges
node_ptr->neighbors.erase(std::remove_if(node_ptr->neighbors.begin(), node_ptr->neighbors.end(),
[this](const DirectionalEdge& edge) {
return edge.start->nodeid == selected_nodeid || edge.end->nodeid == selected_nodeid;
}),
node_ptr->neighbors.end()
);
}
}
node_id_to_node_ptr_map_.erase(selected_nodeid);
nodeptrToNode();
nodeList();
graph_pub_->publish(visualizer_.create_delete_markers());
graph_pub_->publish(visualizer_.create_graph_marker_msg(graph_));
undoButtonCondition();
compareGraphs();
}
void UpdateNodePositionPlugin::nodeptrToNode()
{
RCLCPP_WARN(update_node->get_logger(), "---after deletion---");
for (const auto& pair : node_id_to_node_ptr_map_) {
RCLCPP_INFO(update_node->get_logger(), "Key (Node ID): %u, Value (NodePtr): %p", pair.first, (void*)pair.second);
}
for (auto& pair : node_id_to_node_ptr_map_) {
NodePtr node_ptr = pair.second; // 1. For each NodePtr
if (node_ptr) {
RCLCPP_INFO(update_node->get_logger(), "Node ID: %u (Address: %p) has the following edges:", node_ptr->nodeid, (void*)node_ptr);
for (const auto& neighbor : node_ptr->neighbors) {
auto index = graph_to_id_map_.find(neighbor.end->nodeid);
RCLCPP_INFO(update_node->get_logger(), " Edge ID (%p): %u (Connected to Node ID: %u at index %u)", (void*)&neighbor, neighbor.edgeid, neighbor.end->nodeid, index->second);
}
Passing a vector of strings across to another dll.
I have a c++ program using another dll. That dll has an input method which has a parameter that is an
std::vector<std::string> &
This works fine if both are built in debug or both are built in release. But if one is build in release and the other is in debug the vector is slaughtered upon entered the dll.
(it was a vector of two strings, now it's a vector of thousands of nonsense strings).
What causes this? Is there a way to fix it so that both dlls don't have to have the exact same compiler settings to work?
https://redd.it/1fpioxb
@r_cpp
Eliminating Memory Safety Vulnerabilities at the Source
https://security.googleblog.com/2024/09/eliminating-memory-safety-vulnerabilities-Android.html?m=1
https://redd.it/1fpcc0p
@r_cpp
To the experts, what was your C++ journey like?
TLDR: Skip to the bullet points.
Good morning,
With increasing frequency, I see posts on this subreddit and others about how to advance in C++ (or another language/field), and I think there's an unmet need there. Even myself, I'm currently working to become competent in the language, and I'm about to embark on a personal project that'll likely take a few months, so progress isn't as speedy as I desire.
From what I've witnessed, there's a large gap between what people are taught in school and what companies are looking for in terms of proficiency and experience. That statement applies to multiple industries from software engineering to robotics to many others.
In order to gain perspective and proper orientation, I think it'd be helpful to the C++ acolytes if the veterans would share their experience with becoming proficient in the language and their field as a whole. Here some questions I'd ask a veteran in person if I had the opportunity:
- What were some hurdles or points of failure you had, and what did you do to overcome them?
- What are some aspects of software engineering that you still struggle with?
- What was your entry into the industry like? What, if any, projects did you work on in your spare time to grow your confidence?
- What do you think newcomers are struggling with the most, technically or otherwise?
- How do you juggle the need to be exemplary in so many different facets at once (esp. if in a multi-disciplinary field)?
There are probably many other questions I could ask, but I think the above are enough to spark a great discussion. The first two are especially interesting in my opinion because it's easy to think experts have always been that knowledgeable and never struggle.
I hope you are willing to share your perspective; thank you for your time.
https://redd.it/1fp6iye
@r_cpp
C++ programmer′s guide to undefined behavior: part 6 of 11
https://pvs-studio.com/en/blog/posts/cpp/1163/
https://redd.it/1fp4nxw
@r_cpp
NcursesPlus - Work in progress C++ convenience wrapper for ncurses
https://gitlab.com/mbechara/ncursesplus
https://redd.it/1fc1cs9
@r_cpp
MCQ in OA
Which of the following is true about the assignment x = std::move(y) in C++ 17 and C + + 20 are assuming x and y are both of the same type and that the appropriate header files are visible .
pick one.
option 1: it is always strictly faster than the assignment x = y
option 2: it may induce the compilation error
option 3: reading from y after the assignment is undefined behaviour
option 4: after the assignment it is guaranteed that x!=y assuming that operator != defined for the type of the variables
https://redd.it/1fr0ssp
@r_cpp
How to get Premake and CMake into one projects
Hi, i setup my C++ project with premake. All synct to git. All good. Now i need a library which is only available with Cmake (Assimp). Ive seen there are a few premake files out there for assimp but non of them work for latest Assimp. I also dont want to hack a premake file together for assimp only to have it fall apart after next assimp update. I would like to do it the clean way if possible.
What i already figured out: I can simpy generate VS project files with cmake for assimp. I can also integrate the generated assimp project into my premake with "externalproject". This works fine but now it starts to get a bit irritating: Assimp is now part of my Solution, but first problem is, assimp ships with its own build and platform configs. They dont line up with the configs of my own VS solution (Debug and Release do, but none of the rest like Retail). Is there a way to deal with this in a more or less proper way?
2nd thing that is very unhandy is, that the assimp project uses build and intermediate directories that - i guess - are defined in the cmake config of the assimp project. Would be nice if this could also be changed accordingly to my project without any further postbuild steps like copying the binaries etc. (like telling cmake "yo if you generate the VS project files, please set the directories this and that"). But i guess for this i need to touch the cmake config file of assimp (which i dont want.... would like to leave it 100% original so my build pipeline still works with future updates of assimp).
I would like to see assimp as part of my projects solution so i can build it with a proper build config with the binaries going right to the correct directories and VS does all the linking on its own. All without wild propreitary post and pre build operations/scripts. I guess the major problem is that i want to have it as part of my solution, which is differently configured than assimp of course.
Any suggestion is appreciated :)
https://redd.it/1fqx6d9
@r_cpp
Best Approach to Learning C++ Quickly with Prior C and OOP Experience
I have a technical interview focused on C/C++ next Wednesday. I'm a first-year Master's student in Computer Science, and during the first two years of my Bachelor's, I worked extensively with C and I completed an OOP course in Java.
Given my background in C and object-oriented programming, do you have any advice on how to quickly get up to speed with C++? I know I can't master everything in just a few days, but I'm hoping to make the most of the time I have.
https://redd.it/1fqo8a9
@r_cpp
When Nanoseconds Matter: Ultrafast Trading Systems in C++ - David Gross - CppCon 2024
https://youtu.be/sX2nF1fW7kI?si=nJTEwjvozNGYcbux
https://redd.it/1fqltp3
@r_cpp
Check out this article about a compile-time C++ library that I wrote called CEXForLoop. "If you find yourself reaching for compile-time iteration, reach for CEXForLoop"
https://www.linkedin.com/pulse/pushing-limits-c-how-i-built-constexpr-loop-library-silas-waxter-aunvc/
https://redd.it/1fq8yik
@r_cpp
New to C++ and loops
I'm still new to C++ and need some help with trying to loop my program. I'm trying to get my program to loop the amount of time the user inputs for the amount of gears a transmission has. Where I'm needing the program to loop is where it does the math, that way I can get the final output of the torque and speed based on how many gears the transmission has, along with the amount of gear teeth each stage has. All help is greatly appreciated, thank you!
#define _USE_MATH_DEFINES#include <iostream>#include <cmath>using namespace std;int main(){// Defines// x = gearbox number// y = rpm// z = Nmint x = 0;int y = 0;int z = 0;int gearteeth1 = 0;int gearteeth2 = 0;double efficiency = 0.0;cout << "Enter the number of gearbox stages (positive whole number): ";cin >> x;while (cin.fail()) {cout << "Error!" << endl;cin.clear();cin.ignore(99999999999999, '\n');cout << "Please enter a positive whole number. ";cin >> x;}cout << "Enter the initial input shaft speed (RPM) (positive whole number): ";cin >> y;while (cin.fail()) {cout << "Error!" << endl;cin.clear();cin.ignore(99999999999999, '\n');cout << "Please enter a positive whole number. ";cin >> y;}cout << "Enter the initial input shaft torque (Nm (positive whole number): ";cin >> z;cout << endl << endl << endl << endl << endl;int count = 1;// Loop x amount of times based off user input. while (x) {cout << "Enter the number of teeth for the input gear (positive even number): ";cin >> gearteeth1;if (gearteeth1 % 2 == 0){cout << "Enter the number of teeth for the output gear (positive even number): ";cin >> gearteeth2;if (gearteeth2 % 2 == 0){cout << "Enter the efficiency of this stage (0 - 1): ";cin >> efficiency;if (0 <= efficiency <= 1){double gearratio = (gearteeth2 / gearteeth1);double result = (double (y) * double(gearratio)) * (double(efficiency));double result1 = (double (z) / double(gearratio)) * double(efficiency);cout << "Final Output Speed: " << double(result) << " rpm" << endl;cout << "Final Torque Output: " << double(result1) << " Nm" << endl; }}else{cout << "ERROR! PLEASE ENTER A VALID NUMBER!" << endl;}}else{cout << "ERROR! PLEASE ENTER A VALID NUMBER!" << endl;}}}
https://redd.it/1fq7c8o
@r_cpp
SIMD and Parallelism in C++?
Just read a blog on SIMD in C++ but didn't get it. Can anyone explain what SIMD is and how it's used for parallelism in plain terms?
Blog link
https://redd.it/1fq07yx
@r_cpp
CD/DVD Burn PC Game
I use either C++ or Gamemaker Studio 2 to make games with.
What I would like to, is burn my games (exe) down on either CD or DVD, I have no experience with CD burning in general.
I have “burnable DVD”s
I would really appreciate any help or suggestions
Hope you are having a good day either way:D
https://redd.it/1fptcqw
@r_cpp
}
}
graph_.clear();
for (auto& pair : node_id_to_node_ptr_map_) {
NodePtr node_ptr = pair.second;
if (node_ptr) {
Node node = *node_ptr; // Perform a shallow copy of the node
graph_.push_back(node); // Add the copied node to graph_
}
}
RCLCPP_WARN(update_node->get_logger(), "graph_");
for (const auto& node : graph_) {
RCLCPP_INFO(update_node->get_logger(), "Node ID: %u (Address: %p) has the following edges:", node.nodeid, (void*)&node);
for (const auto& neighbor : node.neighbors) {
auto index = graph_to_id_map_.find(neighbor.end->nodeid);
RCLCPP_INFO(update_node->get_logger(), " Edge ID (%p): %u (Connected to Node ID: %u at index %u)", (void*)&neighbor, neighbor.edgeid, neighbor.end->nodeid, index->second);
}
}
}
**output (node 6 is deleted):**
**1. before graph\_push\_back()**
Node ID: 4 (Address: 0x5cd24d6a0260) has the following edges:
Edge ID (0x5cd24d6c60e0): 6 (Connected to Node ID: 3 at index 2)
Edge ID (0x5cd24d6c6150): 7 (Connected to Node ID: 5 at index 4)
Node ID: 5 (Address: 0x5cd24d6a0310) has the following edges:
Edge ID (0x5cd24d6c3b10): 8 (Connected to Node ID: 4 at index 3)
Node ID: 7 (Address: 0x5cd24d6a0470) has the following edges:
Edge ID (0x5cd24d6658e0): 13 (Connected to Node ID: 8 at index 6)
Node ID: 8 (Address: 0x5cd24d6a0520) has the following edges:
Edge ID (0x5cd24d665c70): 14 (Connected to Node ID: 7 at index 5)
Edge ID (0x5cd24d665ce0): 15 (Connected to Node ID: 9 at index 7)
**2. after** **graph\_push\_back()**
Node ID: 4 (Address: 0x5cd24d6a0260) has the following edges:
Edge ID (0x5cd24d694530): 1298948576 (Connected to Node ID: 3 at index 2)
Edge ID (0x5cd24d6945a0): 7 (Connected to Node ID: 5 at index 4)
Node ID: 5 (Address: 0x5cd24d6a0310) has the following edges:
Edge ID (0x5cd24d684720): 1298948304 (Connected to Node ID: 4 at index 3)
Node ID: 7 (Address: 0x5cd24d6a03c0) has the following edges:
Edge ID (0x5cd24d684f20): 1298799792 (Connected to Node ID: 9 at index 7)
Node ID: 8 (Address: 0x5cd24d6a0470) has the following edges:
Edge ID (0x5cd24d6887d0): 1298553040 (Connected to Node ID: 8 at index 6)
Edge ID (0x5cd24d688840): 15 (Connected to Node ID: 10 at index 8)
**question: why my edge id, edge address, node id and index of edges messed up after graph\_.pushback(node)?**
https://redd.it/1fpnfs8
@r_cpp
C++, a victim of unfortunate tokenization?
Is it just me or do you guys get frustrated when doing a search with C++ as a term as well?
In so many cases, what I get is a search for "C".
I've been trying to understand how this happens - as I know next to or less than zero about how these websites are made.
Sometimes I have luck with searching for "C++", within quotes.
Whatever little I could find for an answer, suggests that search terms are separated in tokens. So the ++ fraction may be treated as an operator. If that is the case, why aren't these tokens separated by blank spaces? So that if I search "Making a C++ program" my search will be "Making" + "a" + "C++" + "program".
It makes hell to navigate Linkedisney... And that is not the only search subject to this problem: blood types can have the same inconvenient to them.
I apologize in advance if I may have infringed upon the rules of this subreddit, I couldn't say for sure if this was a question, or off topic or... It is the first time I ever create a reddit post, so I kindly reiterate that I have no intention of making a misuse of the platform here provided. Also forgive my sub optimal command of the English language - I am in fact a Portuguese speaker.
https://redd.it/1fpg56a
@r_cpp
Coros – A Modern C++ Library for Task Parallelism
https://github.com/mtmucha/coros
https://redd.it/1fpgmc9
@r_cpp
C++ High Performance Computing
Hello!
Is there any live training in the field of High Performance Computing, Parallel/Concurrent Programming in C++ offered from a well established institution in that field which is held in their premises and NOT online?
Ideally I want to travel and meet people involved into the same field because it works better for me.
Thank you!
https://redd.it/1fpbmjj
@r_cpp
Learning solid c++
How to learn advanced c++? Seriously any udemy course on c++ is lame, I don't find serious ressources that goes from 0 to important concepts, popular c++ courses are extremely basic.
https://redd.it/1fp531j
@r_cpp
VS Code vs VS using CMake
Hi,
I am porting a mid-sized application from wxWidgets to Qt. Apparently there will be loads of errors due to mismatches between two frameworks.
I started using VS Code and frankly code completion and intellisense were so slow or did not work at all that I had to recompile the whole project almost all the time. Even sometimes recompiling did not help. I switched to Visual Studio Community and it works so well and intellisense is amazingly responsive.
I know VS Code works out of the box say for example for Python but what I am trying to understand is why VS Code is so popular among C++ developers whereas VS seems to work better for C++. Am I missing something here?
https://redd.it/1fp3jso
@r_cpp
Please some suggestions :)
Please mods don't delete the post 😭😭,
So I am a beginner in programming and last year i learnt python and c .
Now I am feeling like i should switch to different language and bs focused on it for a quite amount of time , well,
After doing a little research I came to conclusion to do either c++ or Java ,
So I am really confused what to choose,I am in 2nd year of my college,so idk what should be better for me as I don't have a lot of knowledge, thanksssss;).
https://redd.it/1fc04cp
@r_cpp