ryvm: Rank You Very Much.

ryvm is a command-line tool that searches your local files and ranks the results by relevance. Think of it as Google, but for your own file system.

It is formally verified using LiquidHaskell to ensure correctness.

Why use ryvm instead of grep?

• grep finds lines in files that match a pattern.
• ryvm finds files that are relevant to a search query. It scores results based on things like how close keywords are to each other.

Quick Start

Search for files containing "your query" in the current directory (.):

ryvm . "your query"

The output is a Tab-Separated Value (TSV) list.

Usage

ryvm [OPTIONS] LOCATION "QUERY"

• LOCATION: The directory path to search in.
• "QUERY": The search term, in quotes.

Options:

• --stdin: Read LOCATION and QUERY from standard input instead of arguments.
• --make-relative: Make result file paths relative to the LOCATION.
• --ext-whitelist="txt,md": A comma-separated list of file extensions to search inside. The default is "txt".

Examples

1. Search for a keyword in all Haskell files

This searches for the keyword "deriving" in all files ending with .hs in the current directory.

ryvm --ext-whitelist="hs" . "deriving"

2. Format search results for easy reading

This command searches for "deriving" and formats the output with a header and clear labels for score and context.

q="deriving"; ryvm --ext-whitelist hs . "$q" | awk -F'\t' -v q="$q" '{t[NR]=$2;s[NR]=$3;c[NR]=$4} END {if(NR>0){printf "search results for (%s): %d hits\n\n",q,NR; for(i=1;i<=NR;i++) printf "%s\nHit Score: %s\n%s\n\n",t[i],s[i],c[i]} else {printf "no hits found for (%s)\n",q}}'

3. Generate a Gophermap

This command's output can be used to create a menu for the Internet Gopher Protocol. Using --make-relative is recommended for Gopher selectors.

ryvm --make-relative . "deriving" | awk -F'\t' '{print "0"$2"\t"$1"\tgopher.example.com\t70"}'