Skip to content

Latest commit

 

History

History
67 lines (55 loc) · 2.56 KB

README.md

File metadata and controls

67 lines (55 loc) · 2.56 KB
Raw

csvinfo

A small utility to display some metadata about the fields available in a "csv" (or otherwise delimited) data file. This came about because I got tired of trying to determine the field lengths in a source data file that was very large and also wanted to learn some Rust.

Build Status

Reminder

This is just here while I tinker with some "applied" learning of Rust - don't trust it to do what it says on the tin.

You should really probably be using XSV: https://github.com/BurntSushi/xsv but, I am taking examples in idea from there and implementing them here because, again, this is a "hands on" place for me to do some learnin'.

Installation

1. git clone git@github.com:sullivant/csvinfo.git
2. cd csvinfo
3. cargo build --release
4. cp ./target/release/csvinfo ~/bin (or somewhere in your path)

Usage

$ ./csvinfo --help
CSV Utils 0.4.3
Thomas Sullivan <sullivan.t@gmail.com>
Shows some info on CSV files.

USAGE:
    csvinfo [FLAGS] [OPTIONS] <file>

FLAGS:
    -h, --help       Prints help information
    -q, --quotes     When passed, data is quoted.
    -s, --skip       When used, skips the first record (header)
    -V, --version    Prints version information

OPTIONS:
    -d, --delim <delim>        Sets the field delimiter to use (example: -d '|'), default is ','
    -m, --max <max_records>    When provided, will stop gathering data after N records

ARGS:
    <file>    Sets the input file to use

Output

$ ./csvinfo ../tmp/file_of_data.csv -d'|'
175646 records in file (| delim).
Field  Max Len  ( %int  %float %char  ) Empty?  Title
1      5        (  0.00   0.00 100.00 )          Type
2      6        ( 25.00  25.00  50.00 )         Value

General Roadmap

  • Use the crate "clap" as a way to pass CLI parameters
  • Allow for any single char passed as parameter
  • Allow for quoted values
  • Allow for field names to be gathered from header data instead of "field 1, field 2..."
  • Test cases
  • Prettier looking CLI output
  • Which are always numeric?
  • Which have empty vals?
  • Trim extra spaces (eg: "Name", "Age","Location")
  • Files without headers - autogen names instead of using first vals
  • Decide if we want to allow for mixed quoted values (some quoted, some not)
  • Process escaped delimiters
  • Status bar on CLI while waiting/processing
  • Add more "metadata" to the output; instead of all the fields, maybe bucket them into sizes? ( wide files look odd in the results )