I am speed
This a revamped version of the project get_next_line.
Admittedly my first attempt (two years ago at the time of writing) was neither fast nor efficient. You can look at the old code here(https://github.com/42Curriculum/ft_get_next_line)
The old version used many many many calls to strjoin and therefore allocated and reallocated a lot of memory. It also had to copy nearly as many times as it allocated.
The new and improved version does at most 1 allocation (necessary as per the project's guidelines, the returned line must be allocated) and 2 copies (altough if it wasn't for the restrictions on line count, the minimum would be 1 copy). per iterations.
If you are doing GNL for the first time, please figure it out by yourself first :)
This implementation takes advantage of the fact that arrays are contiguous pieces of memory.While the subject precises that we can only have buffers and read BUFF_SIZE, allocating multiple of such buffers together gives us one big continuous buffer in the form char (*buffers)[BUFF_SIZE]; This can be allocated as a single buffer of BUFF_SIZE * bufferCount;
Then we can read chunks of BUFF_SIZE in an easy way without pointer artihmetic. We can iteratatre through buffers like a normal char array until we fin /n and change it into a /0 and allocate that much memory and do a single copy from buffers into the newly allocated line. and shift(copy the whole buffer array into the beginning) our array. Rinse and repeat until there is nothing left to read
I challenge any and all who would dare defy the superiority of this get_next_line.
Race me, I dare you
Note: I forgot to check for leaks and I don't think I freed things so :)