Add isNotNil

[LarsKoelpin]

I feel like writing complement(isNil) a lot to do null/undefned checking.

I haven't found an appropriate alternative, so I think it might be a good addition to the library to add
a function named

isDefined: Object -> Bool

which does the null/undefined checking.

What do you think of this addition?

[CrossEye]

This sounds like a good idea. We probably should have done this long ago. I'm not certain of the name. isNotNil seems more likely to me.

[CrossEye]

Please don't commit ramda.js or ramda.min.js.

Also, most importantly, this does not include a test file. Please look at the test folder for examples.

[CrossEye]

I see no reason to depend on complement here. Just do it manually.

[LarsKoelpin]

Removed in 0789ed3

[bas080]

This sounds like a good idea. We probably should have done this long ago. I'm not certain of the name. isNotNil seems more likely to me.

if notNil(person) { /* ... */ };
if isNotNil(person) { /* ... */ };
if defined(person) { /* ... */ };
if isDefined(person) { /* ... */ };
if exists(person) { /* ... */ };

I don't mind being a bit more expressive instead of using not.

[LarsKoelpin]

In function names, I like to prevent "not", because in my opinion it is an unnecessary mental overhead and in the past often created unnecessary confusions for me.
That's why I consider not in names an antipattern and try to name things in a "positive" way.

Also I think isNotNil exposes an implementation detail. It could also be specified e.g. that "" is "!isDefined" as well (which then probably would be "isTurthy" or smth).

I'd really like to choose a "postive" name instead of doing double negation using words. (not isNot is yes)
I will be adding the test and incorporating your changes soon when I'm back home again.

Greetings

[LarsKoelpin]

• Testcase Added: Please review if the specification matches your thoughts
• Removed dist/ commit from history
• Removed dependecy on comeplemt

[CrossEye]

I don't share your objections to using 'not' in the name; nil to me means simply having one of the two nil types in JS, null or undefined. But that was just a quick suggestion; isNotNil is not a name I care about. But isDefined does not resonate for me. I didn't guess what it did from the name. When I saw the title, I thought that maybe you were simply looking for R.has.

So does anyone have a another candidate for a name for this function? @ramda/core?

[LarsKoelpin]

I changed the function name to isNotNil.
If everything is fine for you l'd squash the WIP commits.

[Grohden]

Why not use ramda adjunct?
https://char0n.github.io/ramda-adjunct/2.18.0/RA.html#.isNotNil

I know that it adds a lot o functions.. but it has most of the functions that i find myself repeating across projects using ramda

Btw: I thought that ramda was meant to have only "base" functional tools, that's why i think that we still don't have the "isNot*" functions. If I'm wrong, would be nice to have isNotEmpty in the lib too.

[tommmyy]

And also there is a https://ramda-extension.firebaseapp.com/docs/#isNotNil

[CrossEye]

Missed in renaming.

[CrossEye]

Missed in renaming.

[CrossEye]

Missed in renaming.

[ben-eb]

I'm not sure about this one, I did use to use something similar when first starting out with Ramda but then found it to be unnecessary:

const guardFn = when(isDefined);
// same as:
const guardFn = unless(isNil);

And:

const byDefined = filter(isDefined);
// same as:
const byDefined = reject(isNil);

I'd say 99% of the time I would want to do a null check, I'd use one of these two patterns.

[customcommander]

@LarsKoelpin @CrossEye This pull request could not be merged because of a merge conflict with source/index.js and linter issues as it was predating the decision to add .js extensions to our import statements. That's the only thing I changed in this pull request, everything else is exactly as it was when you approved it @CrossEye.

If we still want this, this should now be ready to go.