I have several different files that have to be switched to and from (I can do that myself), but is there an (easy) method of obtaining that number? Or do I need to come up with a clever solution myself?
#Is there an easy way to get the amount of lines a txt / CSV contains?
wintry timber
inland pewterBOT
molten condor
The only way to know the number of lines in a file is to read the contents and count the number of new-line chars you see
You could do that up-front and store the info in a map for example, but you still have to actually count the lines
tiny fog
[[nodiscard]] auto countLines(const std::filesyste::path& filepath) -> std::size_t
{
if (!std::filesystem::exists(filepath))
{
std::error_code ec{};
throw std::filesystem::filesystem_error("File does not exist.", ec);
}
if (!std::filesystem::is_regular(filepath))
{
std::error_code ec{};
throw std::filesystem::filesystem_error("Is not a regular file.", ec);
}
std::ifstream file(filepath);
if (!file.is_open())
{
throw std::ios_base::failure("Failed to open the file: " + filename);
}
std::size_t line_count = 0;
std::string line;
// Read line by line
while (std::getline(file, line))
{
++line_count;
}
return line_count;
}
why use getline there?
next linden
for whole line
tiny fog
why not?
you could
char ch;
while (file.get(ch)) { // Read one character at a time
if (ch == '\n') {
++line_count;
}
}
you could just an iterator and just read from the internal buffer without allocating anything
then you can even pass it to one of the standard algos and skip writing the loop, std::count_if and std::istreambuf_iterator can work magic
tiny fog
std::istreambuf_iterator<char> it(file), end;
for (; it != end; ++it) {
if (*it == '\n') {
++line_count;
}
}
huh... yeah I guess
sharp nest
you could
```cpp
char ch;
while (file.get(ch)) { // Read one character ...
Isn't that a lot slower than reading an entire line at once or does the OS do a good enough job at caching anyways?
molten condor
the srteams do their own buffering, so it's not as slow as you may think
next linden
and that loop is just `std::count_if`
like
std::count_if(std::istream_iterator<string>(file), std::istream_iterator<string>(), [](char letter){ return letter == '\n'};
or something like that?
tiny fog
#include <algorithm>
#include <filesystem>
#include <fstream>
#include <iostream>
#include <iterator>
[[nodiscard]]
auto countLines(const std::filesystem::path& filepath) -> std::size_t
{
if (!std::filesystem::exists(filepath))
{
throw std::filesystem::filesystem_error(
"File does not exist.", std::make_error_code(std::errc::no_such_file_or_directory)
);
}
if (!std::filesystem::is_regular_file(filepath))
{
throw std::filesystem::filesystem_error("Is not a regular file.", std::make_error_code(std::errc::is_a_directory));
}
std::ifstream file(filepath);
if (!file.is_open())
{
throw std::ios_base::failure("Failed to open the file: " + filepath.string());
}
std::istreambuf_iterator<char> begin(file);
std::istreambuf_iterator<char> end;
return std::count_if(begin, end, [](const char CHR) { return CHR == '\n'; });
}
well, but only if the stream is actually buffered.
Also if the size of each line is like 100k characters, then it could make a big difference, no?
molten condor
like
```c++
std::count_if(std::istream_iterator<string>(file), std::istream_ite...
not quite, I'd use istreambuf_iterator. istream_iterator is the formatted version where as istreambuf_iterator is the "raw" (just plain char) version
next linden
make sence
molten condor
well, but only if the stream is actually buffered.
Also if the size of each line...
You have to jump through hoops to get a non-buffered stream
sharp nest
You have to jump through hoops to get a non-buffered stream
or work on an embedded system
molten condor
you can still have a buffered stream on embedded, std::cout even works!
if you're using some raw api to get the data, well then you still have to supply the buffer to copy in from the FS
sharp nest
👍
tiny fog
or work on an embedded system
on an embedded system you will have a hard time working with files.
sharp nest
fair point
wintry timber
Could you guys make it so that it is a function , seperate from main that I could just copy paste into my own thing?
Thanks!
inland pewterBOT
@wintry timber Has your question been resolved? If so, type !solved :)
wintry timber
No it hasn't, but thank you for asking
wintry timber
Anyone?
formal peak
Could you guys make it so that it is a function , seperate from main that I coul...
they've already done that?
tiny fog
No it hasn't, but thank you for asking
what do you mean it hasn't been solved?
Is this not what you were asking for?
#1316139359094702154 message
Could you guys make it so that it is a function , seperate from main that I could just copy paste into my own thing?
you ought to be able to do that much yourself. If not, your IDE probably has a way to refactor selected text out into a function.
also Wheatly is a bot.
Do you require an explanation of the function?
wintry timber
!solved
inland pewterBOT
normal snow
```cpp
#include <algorithm>
#include <filesystem>
#include <fstream>
#include <i...
You don't need to (and shouldn't) do those checks before opening a file
The opening will fail anyway if those are false. If anything, you could do this analysis after is_open() returned false
sharp nest
!vampire
formal peak
could add "asks questions just for the sake of asking something so people engage with them" because there are a lot of people(not op) who do this
tiny fog
Where I can look to this bot documentation
try here
