@David Yes, that happens a lot with old answers :-)

In your note you say X+ has been removed. You could perhaps mention that Y+ can be used instead. Otherwise it appears I've removed the very important convolution operation. What would @flawr think? :-)

@LuisMendo I got my old 38 byte answer down to 30 fairly easily!

32 actually...

@David \o/

@LuisMendo Just submitted a minor pull request fixing a lingering HTML tag

Also, that color challenge could really benefit from hex2dec.

@Suever Thanks, I'll take a look.

@Suever I had that in mind. If we see it pop often I'll include it. Anyway, for that challenge there's an answer taking the input in decimal; maybe that's allowed

Oh I didn't actually click to run his code. That will definitely work

@Suever Nice catch, thank you. Do I simply press "merge pull request"? I don't know much GitHub, I think I read there are two ways to incorporate pull requests

Yea that is the easiest thing to do.

It's such a minor change that you shouldn't have any issues

@Suever Anyway, I use GitHub very strangely. I keep a local master copy, so I'll also do the change manually there

You should also be able to call git pull origin master to update your local copy

it will fast-forward everything

It was just a line, so I updated it manually in my local copy

...in addition to merging the pull request to publicly give you credit, that is :-D

Hahah I don't care too much about the credit :)

Sounds good though

:-D

@Suever Today I was working on incorporating the contents of predefined-literal functions in the command-line help. (We talked about that some weeks ago). Turns out, it was already done! I didn't remember I had already done it. It wasn't working because instead of passing the structure that contains those literals to function genHelp.m I was passing the name of the file containing that structure.

So I fixed that and now it's working. For example:

>> matl -h ignorecase
X3  predefined literals
    1;  1
    predefined literal depending on input. 1: 'start', 2: 'end', 3:
    'tokenExtents', 4: 'match', 5: 'tokens', 6: 'names', 7: 'split',
    8: 'once', 10: 'ignorecase', 11: 'preservecase'

I've pushed that to GitHub

Hey @LuisMendo Why does MATL not have any factorial functions?

@DrGreenEggsandHamDJ Well, there's gamma, which is almost the same. And for a single number just use :p

>> matl :p
> 7
5040

@DrGreenEggsandHamDJ Yeah, I was just about to say :p works

If it was added it would probably be a two byte anyways

Oh, that makes sense.

@LuisMendo Oh, is gamma new?

>> matl QYg
> [1 2 3 4]
1  2  6 24

@quartata Yes, it was a recent addition

Ah, nice.

>> matl -h gamma
Yg  gamma, incomplete gamma or incomplete beta function
    1--3 (1);  1
    gamma / gammainc / betainc, depending on number of inpu
Zg  logarithm of gamma or beta function
    1--2 (1 / 2);  1
    gammaln / betaln, depending on number of inpu

Oh. There's a but that removes the last two characters

Hence inpu

I'll fix right now

Fixed

@LuisMendo Oh that's beautiful! Thanks for doing that. Will be extremely helpful. The other benefit is that it allows you to search for predefined literals

Hmmmm how come matl -h 3600 doesn't find the pre-defined literal?

How do you convert a string to it's ascii values? Shouldn't that just be 'U'?

@DrGreenEggsandHamDJ A lot of times functions will automatically convert to ASCII as needed. If you want to be expilcit, you can use o to cast as a double

Things like 'abc'l+ will automatically convert to a number.

Well I'm an idiot. My problem was that I was using double quotes instead of single quotes.

facepalm

@Suever That's a predefined content of clipboard L, not of a predefined-literal function. I'll have to include those too

@DrGreenEggsandHamDJ To add to what Suevver said: U interprets the string as a number. So '12'U will give number 12 (whereas '12'o will give [49 50], i.e. ASCII codes)

@quartata Do you think Pi (gamma offset by 1) would be more useful than gamma? Pi(x) is gamma(x+1), so it directly corresponds to factorial. I wasn't sure which one to include

@LuisMendo Well the main point for using Pi function instead would be to "overload" the function so it can do both factorial and gamma. But since factorial is two bytes anyways I don't think it would be as useful

I think sticking with Yg as gamma and :p for factorial is fine.

Jelly does have Pi but it's one byte there so it's more useful

@quartata Thanks. I think you're right, I'll leave it as it is. I can always change it later if it becomes clear it's better to

@LuisMendo Ah ok that makes sense. Thanks