Congratulations to Jean_Luc and zielar!
I have been thinking for several days about removing the hashtables from the joining process, and when I finally finish it I find that it had already been implemented by JeanLuc
Now I have applied the same method to join several files in one process, reducing the total time quite a bit.
My approach is to read all the files in the directory at the same time in HASH_ENTRY loop.
Although my code works, it is still in testing and it is possible that it has some bug.
https://github.com/PatatasFritas/FriedKangaroo/blob/merge/Merge.cpp
If you or someone who can program can work on sending hashtable to textfiles (like your export option configuration) then there would be no need for save files and merging, etc. If you give user option to export tame and wild files with any amount they wish (example 20 tame and 20 wild files), then no merging or saving. Just compare the different tame and wild files for a collision/solution. This also eliminates any RAM issues.I have been thinking for several days about removing the hashtables from the joining process, and when I finally finish it I find that it had already been implemented by JeanLuc
Now I have applied the same method to join several files in one process, reducing the total time quite a bit.
My approach is to read all the files in the directory at the same time in HASH_ENTRY loop.
Although my code works, it is still in testing and it is possible that it has some bug.
https://github.com/PatatasFritas/FriedKangaroo/blob/merge/Merge.cpp