yamaton / h2o-curated-data

Semi-hand curated command-line option data for many CLI programs, geared toward bioinformatics tools. Shell completion scripts are also available.

CLI curated data for shell completion

This repository contains command-line specs for Shell Completion for VS Code. This repository also contains tab completion scripts for bash/zsh/fish shells.

Shell completion scripts for bioinformatics tools are located in the separate repositories:

• yamaton/fish-completions-bio
• yamaton/zsh-completions-bio

Loading the data in VS Code

• In VS Code, type Ctrl+Shift+P (or ⌘+⇧+P on macOS) and run following commands:
  • Shell Completion: Load Common CLI Data to load commands in general.txt
  • Shell Completion: Load Bioinformatics CLI Data to load commands in bio.txt

Shell completion for bash/zsh/fish

They would be useful especially for bioinformaticians.

bash

Copy files (general) to ~/.bash_completion.d/, and add the following to ~/.bashrc.

for bcfile in ~/.bash_completion.d/* ; do
    . "$bcfile"
done

zsh

Save files (bio, general) to ~/.zfunc, then add the following line to ~/.zshrc.

fpath=( ~/.zfunc "${fpath[@]}" )

fish

Save files (bio, general) to ~/.config/fish/completions/.

How I prepare the data

I use h2o (closed source code) to extract CLI specs, and then apply yq to convert from JSON to YAML.

Here is how I extracted the CLI info from medaka, for example.

# Just see how it works
$ h2o --command medaka

# ...OK. H2o seems to fail parsing some.
# So save the help document as text file for manual editing...
$ medaka --help > medaka.txt

# Edit medaka.txt s.t. command options, subcommands, and descriptions are placed nicely.
# ...
# ...
# After finishing the edit, extract the data as YAML file.
$ h2o --file medaka.txt --format json | yq eval -P > bio/yaml/medaka.yaml

# Add TLDR to the YAML, then convert it to JSON and bash/zsh/fish scripts
# group=bio sets the environment variable and run make
# The makefile calls multiple scripts including
#    - Add TLDR to YAML
#    - Convert YAML to JSON
#    - Create bash/zsh/fish scripts
#    - Take the entire JSON as an array, create a gzipped JSON file
$ group=bio make

Data structure

Here is a sample file.

name: tar
description: an archiving utility
options:
  - names:
      - -x
      - --extract
      - --get
    argument: ""
    description: extract files from an archive
  - names:
      - -v
      - --verbose
    argument: ""
    description: verbosely list files processed
version: tar (GNU tar) 1.34

As you can see, an entry has following key-values:

• name (string)
• description (string)
• options (list)
  • names (list of strings)
  • argument (string)
  • description (string)
• [optional] positionalArguments
  • name (string)
  • description (string)
• [optional] subcommands:
  • (Here comes a list of the entire structure. Nested.)
• [optional] inheritedOptions (same type as options)
• [optional] usage (string)
• [optional] version (string)
• [optional] tldr (string)

h2o-curated-data's Issues

[feat] support adding other languages

add support for other languages (other than shellscript) such as bitbake (.bb, .bbappend files marked as bitbake lang with the bitbake extension) which is a mix of python and shell. I think just adding support in the entire bitbake file would be enough.

Supporting more tools

Please post a request for CLI programs to support in VS Code "Shell Completion" here.

Extra hyphens at re-add argument

Hi,

if you are using the completion function and select an element from the offered list, it inserts fine, but for instance, you delete the part of the inserted argument and you want to add it again by CTRL+SPACE the list appears again, and when you want to add it again it will insert extra hyphens.

I experienced the same issue with other commands as well.

Unhandled error at Load Command

Hi,

when I wanted to load a new command and asked the command name, I clicked anywhere outside the input field.
Even the editor or out of VSCode I got the following error message.