FsShell 0.0.6

FsShell

List of Shell-like commands for use in F#.

I commonly work with data files in the legal space, so wanted a set of commands that I could use for quicker data manipulation and investigation.

Caveat

These commands are NOT intended for projects with a long life. It is recommended that this only be used in the FSI or for short fire-and-forget utilities. As I eat my own dog food and become more proficient in idiomatic F#, some things might change.

Usage

#r "nuget:FsShell"

man "";;

All functions are in FsShell.fs

Note: Most functions work with seq<string>. Functions prefixed with an 'x' work with seq<string array>

System
  cd        Change directory
  pwd       Print working directory
  mkdir     Make directory
  mv        Move file
  cp        Copy file
  ls        List
  ll        List with details
  find      Find files
  find_p    Find folders
Output
  stdout    Output to STDOUT
  stderr    Output to STDERR
  write     Output to a file
  append    Append to a file
  tee       Output to a file and passthrough
  tee_a     Append to a file and passthrough
  cat       Read a list of files consecutively
Data Manipulation
  cut       Split lines at tabs
  cut_d     Split lines at delimeter
  cut_c     Cut character range options
  cut_x     Splits data file into columns
  paste     Join files into single file; line-to-line
  xpaste    Join columns into data
  sort      Sorting
  sort_k    Sorting by a substring
  sort_kn   Sorting by a substring as a number
  xsort     Sorting by columns
  xsort_n   Sorting by columns, each as a number
  xsort2    Sorting by columns, specify as string or number
Data Flow
  grep      Filter lines to include
  grep_i    Filter lines to include, case insensitive
  egrep     Filter lines to include with regex
  egrep_i   Filter lines to include with regex, case insensitive
  grep_n    Filter lines to exclude
  grep_in   Filter lines to exclude, case insensitive
  egrep_n   Filter lines to exclude with regex
  egrep_in  Filter lines to exclude with regex, case insensitive
  xgrep     Filter on a pattern in specified columns
  xgrep_i   Filter on a case-insensitive pattern in specified columns
  xgrep_n   Inverted filter on a pattern in specified columns
  xgrep_in  Inverted filter on a case-insensitive pattern in specified columns
  xegrep    Filter on a regex in specified columns
  xegrep_i  Filter on a case-insensitive regex in specified columns
  xegrep_n  Inverted filter on a regex in specified columns
  xegrep_in  Inverted filter on a case-insensitive regex in specified columns
  head_n    First count lines
  head      First 10 lines
  tail_n    Last count lines, or skip first count of lines
  tail      Last 10 lines
Miscellaneous
  man       This

Examples

Manipulating Folders

> mkdir "test_fsshell";;
val it: string = "/Users/gregh/Projects/test_fsshell"

> cd "test_fsshell";;
val it: string = "/Users/gregh/Projects/test_fsshell"

> pwd ();;
val it: string = "/Users/gregh/Projects/test_fsshell"

Manipulating Files

> // write to output lines to files
- // newlines are added and stripped automatically
- seq {
-     "My First Line"
-     "My Second Line"
- }
- |> write "file1.txt"
- ;;

> // cat and grep
- cat [ "file1.txt" ]
- |> grep_i "second"
- |> write "file2.txt"
- ;;

> cat [ "file1.txt"; "file2.txt" ];;
val it: seq<string> =
  seq ["My First Line"; "My Second Line"; "My Second Line"]

Manipulating Data

> seq { "A\tB"; "C\tD" } |> write "file1.txt";;
val it: unit = ()

> cat [ "file1.txt" ]
- |> cut
- ;;
val it: seq<string list> = seq [["A"; "B"]; ["C"; "D"]]
> // also available is cut_d to split on a provided delimiter
- // cut_c and cut_c2 to get specific character ranges

> // cut_x for handling CSV data files
- cat [ "./datafile.csv" ]
- |> cut_x
- ;;
val it: seq<string array> =
  seq
    [[|"Column 1"; "Column 2"; "Column 3"|];
     [|"Val 1-1"; "Val 2-1"; "Val 3-1"|];
     [|"Val 1-2"; "Val 2-2"; "Val 3-2"|];
     [|"Val 1-3"; "Val 2-3"; "Val 3-3"|]]

Data Investigation

> // Used this command to find duplicate values
- cat [ "file.dat" ]
- |> cut_x                            // parse the file
- |> Seq.map (fun cols -> cols[0])    // get data from the first column
- |> Seq.groupBy id                   // grouping by first column
- |> Seq.map (fun (docid, docids) -> (docid, Seq.length docids))  // docid, count of docids
- ;;

Miscellaneous

"Why no rm and rmdir?"

I could see this going badly, quickly. Maybe I'll change my mind later, but after implementing it, I felt it was too dangerous.