stardance has been extended another month! the new deadline is october 31 :)

You are browsing as a guest. Sign up (or log in) to start making projects!

Hidalgo

  • 3 Devlogs
  • 5 Total hours

a fast file indexer for linux

Ship #1 Pending review

i made this program because it doesnt exist any other good alternatives to the everything on linux. find is slow, fzf only sees whats already on the screen, so i took matters on my own hands

3 QoL improvements over existing tools:

  1. instant search, index once, search in milliseconds
  2. fuzzy matching, finds files even if you dont remember the full name, exibe isso
  3. persistent cache, index saved to ~/.cache/hidalgo, second run is 3x faster, skips unchanged directories via time, and ignores /proc, /sys, /tmp and other system paths entirely.

built from scratch: hashmap (FNV-1a), pthread worker pool, binary cache, glob matcher

Try project → See source code →
Open comments for this post

1h 10m 35s logged

devlog 3 => persistent cache and ignoring temp files

now in hidalgo, the searches are more faster because of a persistent cache created on the home folder
and now hidalgo ignore certain paths, like /tmp and another paths for temp files/system files for faster searching

in the future, i want to create a gui for it and add real time monitoring (hook to system calls maybe)

0
0
15
Open comments for this post

1h 39m 24s logged

devlog 2 => threads and blob search

now in hidalgo, you can search using wildcards for certain file types!
and the search is faster, now utilizing 4 threads for the search system, planning to identify how much threads a system has and use all of the threads.
added -n for limiting how much items it should push, the default (n = 0), pushes all files with that characteristics

planning to add local cache and hook to the system calls (not inotify too slow)

0
0
23
Open comments for this post

1h 49m 7s logged

devlog 1 => the big bang

i started developing the hidalgo: a fast file indexer for linux, inspired in the file indexer Everything only available on Windows.

now it has available a index in memory with hashmap (FNV-1a), recurssive scan of directorys, fuzzy search and a basic cli. i noted that with too much files (100k+), the indexer consumes too much ram (1gb+), i plan to mitigate that with a local cache for not everytime that it needs to search something it has to scan everything.

for the future, i plan to add a hook to the system calls that create/delete files of linux (not sure if it is possible) and create a good user interface to not rely only on the cli.

0
0
39

Delete project?

Are you sure you want to permanently delete this project? This action cannot be undone.

All devlogs, followers, and associated data will be removed.

Followers

Loading…