qtools Parse Command Line

====================================================== UPDATED: Now parses injected string instead of argv

New parameters: injectedCommandString, injectFirstTwoElements

Added input parameter, injectedCommandString, that substitutes an artificially created command string for process.argv. It is parsed by injectedCommandString.split(/ +/) and as far as I can tell, acts exactly like it should, ie, if injectedCommandString.replace(/ +/g, ' ')==process.argv.join(' ') then it produces the same result.

NOTE: Normal process.argv has node and the program name in the first two positions in the argv array. If you your artificial command string is values only, ie, does not have the first two values, you can add the parameter, injectFirstTwoElements:true. This will inject a pair of dummy values at the front of the fake argv array and all will work correctly.

tqii 5/13/24

processCommandLine() analyzes argv and returns the results for use by the calling program.

For historical reasons, it defaults to evaluating incoming Javascript. This can, of course, be dangerous. It should be called with the 'noFunctions:true' argument unless you need function evaluation.

var commandLineParms = qtools.parseCommandLine({noFunctions:true})

processCommandLine() produces an object with four elements: { calledFromCommandLine: true|false, fileList:[], values: {}, functions: {}, switches: {} }

fileList is anything that is has no switch and resolves to a file or directory

VALUE PARAMETERS

--name=value

UPDATE (6/28/18) added:

--name value

UPDATE (11/4/19) ADDED:

--namelist=sam,henry,joan

(If valuesSplitCharacter is specified in args; otherwise, still a string)

applies to all cases below.

Anything in the command line of that form is processed to become either a value or a function.

1. Parsed to see if it's javascript and evaluated eg,

    --tmp="(2+5)"
   

   produces a property

    values:{tmp:7}
   

   or,

    --tmp="(function(a){return 2*a})"
   

   produces a property

    functions:{tmp:function(a){return 2*a;}}
   

   that can be called later as commandLineOutput.functions.tmp(parameter)

   or (of course),

    --tmp="hello world"
   

   produces a property

    values:{tmp:"hello world"}
   

   however, there is a special case for strings, eg,

    --tmp=/path/to/function.js
   

   is checked to see if it resolves to a file. (The path is not relative

   or glob'd. It must be fully specified.)

   If it does not produce a value, the property is then eval'd to

   see if it produces a value.

    values:{tmp:"/path/to/function.js"}
   

   If it does parse and produce a value, the property is,

    values:{tmp:'some value resulting from the file'}
   

   unless it results in a function, in which case, the property is,

    functions:{tmp:function(){/ *whatever is in the file* /}}
   

   Important note: The function definition must be wrapped in parenthesis.

   eg,

    (function(input){ return input*2;})
   

   Otherwise it will not eval() and will be treated as a string value.

SWITCHES

-switchName

Anything of that form creates a property, values:{switchName:true}