Need help with hors?
Click the “chat” button below for chat support from the developer who created it, or find similar developers for support.

About the developer

196 Stars 7 Forks GNU General Public License v3.0 427 Commits 4 Opened issues


instant coding answers via the command line (howdoi in rust)

Services available


Need anything else?

Contributors list

# 99,193
319 commits
# 104,218
3 commits
# 164,598
3 commits
# 180,242
1 commit
# 212,775
1 commit

Crate CI


Awesome program howdoi which implemented in rust, along with easily usage lib.

It's faster than the original howdoi program.

For binary usage, please go through the rest of this file. For lib documentation, please check here.


Simple usage example

Screenshots of hors gif

More examples

Screenshots of hors png


hors is written in

. The recommended way to install
is through
cargo install hors

On Windows/Linux/macOS platform, you can download the pre-build-binary from github release page

On macOS

Hors can be installed from homebrew.

brew tap hors-org/hors && brew install hors

On Windows

Hors can be installed from scoop

scoop bucket add w-bucket; scoop install hors

What make it fast

  1. Implemented in rust, which causes less runtime overhead.
  2. Make use of tokio concurrent feature, so hors will make concurrent search when it need to fetch more than 1 answer.
  3. Output will be cache, and when you want to search for the same question, hors will likely make less network traffic to get the answer.

Here is a simple benchmark report, run the following command 3 times in my personal computer:

time hors mysql create table with column comment -a -n 10 --paging never -e bing
rm ~/Library/Caches/hors/answers
Note: run
command is aimed to clear local cache.

And it gives me the following output:

Executed in    2.55 secs   fish           external
   usr time  232.71 millis  150.00 micros  232.56 millis
   sys time   16.68 millis  562.00 micros   16.12 millis

Executed in 3.68 secs fish external usr time 252.02 millis 125.00 micros 251.90 millis sys time 19.18 millis 550.00 micros 18.63 millis

Executed in 2.55 secs fish external usr time 237.19 millis 117.00 micros 237.07 millis sys time 17.63 millis 565.00 micros 17.06 millis

Run the same command with howdoi:

time howdoi mysql create table with column comment -a -n 4 -e bing -c
rm ~/Library/Caches/howdoi/*
Note: run
command is aimed to clear howdoi local cache.

And it gives me the following output:

Executed in    3.48 secs   fish           external
   usr time  303.67 millis  127.00 micros  303.54 millis
   sys time   52.53 millis  601.00 micros   51.93 millis

Executed in 3.65 secs fish external usr time 305.37 millis 111.00 micros 305.26 millis sys time 53.16 millis 549.00 micros 52.61 millis

Executed in 3.34 secs fish external usr time 319.07 millis 14.24 millis 304.83 millis sys time 55.63 millis 3.37 millis 52.26 millis

But please note that this simple benchmark is not precise, it highly depends on network information.

Tested platforms

For now,

has been tested with the following platforms:
  • Linux
  • OSX
  • Windows


    hors [FLAGS] [OPTIONS] [query]...

ARGS: ...

FLAGS: -a, --all display the full text of answer. --clear-cache just clear local hors cache. -d, --disable-proxy Disable system proxy. -h, --help Prints help information -l, --link display only the answer link. -r, --raw make raw output (not colorized). -V, --version Prints version information

OPTIONS: -e, --engine select middle search engine, currently support bing, google, duckduckgo, stackoverflow. [env: HORS_ENGINE=bing] [default: duckduckgo] -n, --number-answers number of answers to return. [default: 1] -p, --paging specify how to page output, can be auto, never [default: auto]

Usage example

  1. Want to know how to export pandas dataframe to csv?
    hors pandas dataframe to csv

Here it is:

- Answer from
df.to_csv(file_name, sep='\t')
  1. If we just want to know where is the answer existed?
    hors pandas dataframe to csv -l

Here it is:

Title - pandas writing dataframe to csv file
  1. If we want more about the answer detailed?
    hors how to parse json in rust -a

Here it is: ```shell - Answer from

Solved by the many helpful members of the Rust community:

extern crate rustcserialize; use rustcserialize::json::Json; use std::fs::File; use std::io::Read;

fn main() { let mut file = File::open("text.json").unwrap(); let mut data = String::new(); file.readtostring(&mut data).unwrap();

let json = Json::from_str(&data).unwrap();
println!("{}", json.find_path(&["Address", "Street"]).unwrap());

} ```

  1. How to get more than one answers
    hors set git remote url -n 2 -a
    Here it is: ``` - Answer from You can

git remote set-url origin new.git.url/here

(see git help remote) or you can just edit .git/config and change the URLs there. You're not in any danger of losing history unless you do something very silly (and if you're worried, just make a copy of your repo, since your repo is your history.)

^^ ==================================================== ^^

  • Answer from below is used to a add a new remote:

git remote add origin [email protected]:User/UserRepo.git

below is used to change the url of an existing remote repository:

git remote set-url origin [email protected]:User/UserRepo.git

below will push your code to the master branch of the remote repository defined with origin and -u let you point your current local branch to the remote master branch:

git push -u origin master

Documentation ```

  1. The default search engine is bing, how can I use other search engine?
    hors set git remote url -n 2 -a -e "google"

Proxy support

If the network seems blocked, you can try to configure proxy like this:

export http_proxy=;export https_proxy=;

Of course, it should be a valid proxy in your machine.

Paging feature on windows

Hors is using

command to make paging feature work, and it's not installed on Windows by default. You can use scoop to install
scoop install less

Or use

choco install less

Use hors as lib

Hors can be used as a lib, here is an example:

use std::str::FromStr;
use hors::{self, SearchEngine};

let search_engine: SearchEngine = SearchEngine::from_str("bing").unwrap(); let target_links: Vec = hors::search_links( "how to parse json in rust", search_engine, ) .await .unwrap(); assert_ne!(target_links.len(), 0); for link in target_links { assert!(link.contains("")); }

For more information, please check the documentation

Special thanks

Very thanks for the awesome project and links :) - howdoi inspires

(Fow now
which implements in
). - stackoverflow helps users solve questions about coding.

About the name

is the abbreviation of
howdoi in rust

We use cookies. If you continue to browse the site, you agree to the use of cookies. For more information on our use of cookies please see our Privacy Policy.