The Regex Game...

[$cirisme]

And they said that post count boosting is only for the dorm. :teeth:

Here's how it works:

I post a regex.

Everyone guesses what that regex does.

The first one to correctly guess post his regex, and everyone guesses what that regex does.

And so on.

Yeah, this is pretty geeky and will never get to the level of the Word Association thread, but we can try. :teeth:

If it makes a big difference, please note what your regex was designed for.(Ie Perl, PHP, etc)

My Regex:

(\d+)\.(\d+)\.(\d+)\.(\d+)

[prgmrdave]

I'd say IP address, but you allowed for arbitrarily many digits (well, an arbitrarily positive number, anyway) between periods...

[$cirisme]

That was fast.

You are correct. :teeth:

[prgmrdave]

Or it could be a version number on a Windows executable :smile:

[yxboom]

WTG pmd! :thumb:

[$cirisme]

Post your regex. :teeth:

[prgmrdave]

OK, so here's the next one:

([01]\d|2[0-3])(:[0-5]\d){1,2}

[$cirisme]

Would match:

152:4848

I have no idea for what. I'll guess Bible verse.

[prgmrdave]

Today @ 03:05 PM post located here (http://www.theologyweb.com/forum/showthread.php?s=&postid=114547#post114547)
cirisme:

Yeah, this is pretty geeky...

Today @ 03:09 PM post located here (http://www.theologyweb.com/forum/showthread.php?s=&postid=114549#post114549)
cirisme:

That was fast.



Of course it was fast, because I have a constant TWeb search running for anything with "geek" in it :yipee:

That, or I just happened to see it at the right time...

[prgmrdave]

According to http://www.dotnetcoders.com/web/Learning/Regex/RegexTester.aspx, it doesn't match 152:4848.

[prgmrdave]

And I get more spam points if I put my guess and my regex in different posts :teeth:

[$cirisme]

I missed a two:

1522:4848

[prgmrdave]

The only thing that means is my regex is bad... what do the rules say I do about that?

[$cirisme]

I hadn't anticipated that.

What were you trying to match? A zip code?

[prgmrdave]

Ain't sayin'. I'll get it right, I will! To wit:

(([01]\d)|(2[0-3]))(:[0-5]\d){1,2}

I think.

[prgmrdave]

It just has an extra set of parentheses.

[$cirisme]

Matches 1521:4848

Still have no idea what it is. :huh:

[$cirisme]

BTW, that regex testing site is pretty cool! :cool:

[prgmrdave]

No, it's not, because it gives me inconsistent answers.:rant:

[prgmrdave]

No, it doesn't; I just forgot Mr. Hat and Mr. Money.

^(([01]\d)|(2[0-3]))(:[0-5]\d){1,2}$

[$cirisme]

Really. Works good for me.

1521:4848 does match(I could tell you that in my sleep), but what is it? :poke:

[$cirisme]

Egad man, that is getting strange.

[prgmrdave]

Sorry about that... I learned RE's about 20 years ago (and boy, back then we had to walk three miles to the RE evaluator, in the snow, uphill both ways!) and haven't had to use 'em much in the last 12...

[prgmrdave]

I figured out why 1521:4848 matched without ^ and $, but I'm Pretty Sure Now that it works correctly. ("Correctly" indicating "the way I mean it to." :smile:)

[$cirisme]

<brain freeze>

[prgmrdave]

'sokay. I'll give you some more

(ahem)

"time".

[$cirisme]

The new one matches 15:38.

I'll guess code column/line number.

Ie, "Error located at 15:38".

If this post made any sense. :xmm:

[$cirisme]

'sokay. I'll give you some more

(ahem)

"time".

That was made before I made my above post.

No hinting! I said it was complex, not that I had given up. :poke:

:lol:

[$cirisme]

Never mind, I forgot that I said "brainfreeze".

I should go lay down.

[prgmrdave]

Yes! 15:38 is in the set of things I intended it to match! Cool!

But not a code column/line number; the expression wouldn't find anything on or after line 24. Need more time?

[prgmrdave]

Sorry-- didn't see your "no hinting"

:cir: :whack: me

[$cirisme]

Would someone tell please tell Dave that it is a 24 hour matcher...

My brain is too fried to come up with a regex for this game :brow:

[dizzle]

Geeks.

[$cirisme]

Oh, all right, I couldn't help myself...

.:\\.{1,32}\\.{1,32}\\.{1,32}\..{1,5}

Yea, it's easy. But Dave fried my brain. :teeth:

[prgmrdave]

Suggestion:

"prgmrdave, the thread has changed since you began your reply. Would you like to review the new posts before your reply is submitted, in order to prevent the taste of fresh foot?":nc:

[$cirisme]

Geeks.

Post count booster.

[prgmrdave]

:rofl: (note: not really rolling)

[prgmrdave]

Today @ 03:58 PM post located here (http://www.theologyweb.com/forum/showthread.php?s=&postid=114601#post114601)
cirisme:

Oh, all right, I couldn't help myself...

.:\\.{1,32}\\.{1,32}\\.{1,32}\..{1,5}

Yea, it's easy. But Dave fried my brain. :teeth:

I'd guess, but then I'd have to post a new one, too, and this time I have to test my code before I publish it :lol:

[prgmrdave]

Windows pathname, with drive letter and extension?

[$cirisme]

You're good at this.

[prgmrdave]

I'm a geek. What can I say? :shrug:

^\d+(\.(\d)*)?([Ee][-+]?\d{1,3})?$

[$cirisme]

Would match 5.77E-555. I'm gonna guess an exponential double, or version number. :huh:

[prgmrdave]

Floating-point number w/optional exponent, yep! Got it in one! :cheers:

[$cirisme]

My turn:

([0-9\.]+),[\s]+([0-9\.]+),[\s]+([0-9\.]+)

[prgmrdave]

Matches

cirisme rox 1, 2, 3 cirisme rox

Hm.

[prgmrdave]

Matches

testing: 1, 2, 3...

Sound test?

[$cirisme]

:rofl:

Okay, any ideas as to what it does?

[$cirisme]

Sound test?

Nope.

Hint: There is no text.

[prgmrdave]

I know there's no text... it matches because the expression doesn't have ^ and $. Okay, this is my real guess:

3-d spatial coordinates in the positive x-, y-, and z-axes?

[$cirisme]

No.

[$cirisme]

But that's a very good guess.

[prgmrdave]

Matches

4..5, 4.5.6.., ...

:huh:

/me suspects :cir: is going for brain-freeze revenge (and succeeding)

[prgmrdave]

Grr. I see three non-empty groups of digits-plus-decimal-point, separated by comma-plus-non-empty-white-space. Can't figger what the digits represent, though.

3.14, 2.718, 1.414

Error #3.14, line 2.718, column 1.414? :eek:

[Jade]

:eek: I never knew geekiness went to this extreme!:shy:

[prgmrdave]

:rofl:
:yipee:

[$cirisme]

Dave,

Need a hint? :smile:

Jade,

We like to post count boost, too. :poke: :teeth:

[prgmrdave]

cirisme:

Need a hint? :smile:


Stop gloating. :smile:

Matches

., ., .

Yeah, gimme. :help:

[$cirisme]

Think *Nix.

[prgmrdave]

I am going to feel like such a :dunce: when I figure this out (or I cave)...

[prgmrdave]

:idea: user/group/other permissions
:idea: create/mod/access times
:idea: argc, argv, envp
:idea: primary/secondary/tertiary inode tables

:argh:

Grr.

[$cirisme]

:huh:

[prgmrdave]

Just trying to think of *nix things that come in threes...

[$cirisme]

Think down time.

[prgmrdave]

Or "uptime", even.

Load averages?

[$cirisme]

YES!!! You've got it!

:teeth:

[prgmrdave]

:argh:

:dunce:

:duh:

[$cirisme]

Your turn.

(I'm going to pay for that, huh? :help:)

[prgmrdave]

:hamster:

[$cirisme]

:uhoh:

[prgmrdave]

:brow:

^<(\w(-|\w)*:)?\w(-|\w)*(\s+\w(-|\w)*\s*=\s*"[^"]*")*>$

[$cirisme]

:hrm:

No hints, I don't have time to take a crack at this just yet.

:argh: :whack:

[prgmrdave]

Okey doke. :cheers: Take your time.

(No, it's not another time designator :eek:)

[$cirisme]

I'll stick my foot in my mouth:

HTML tag.

I'm probably way off base, as I've barely looked at it, but I do see angle brackets and an = sign in there.

[prgmrdave]

Almost... but not quite. I could be persuaded that you're only one letter off, though. :teeth:

[$cirisme]

XML

[prgmrdave]

:bow: :cir:

Woo hoo!

/me thinks: :uhoh:

[$cirisme]

Or possibly XHTML. :huh:

[$cirisme]

Now, don't even try that silly revenge nonsense ever again. :brow:

.*[\/\\](.*)

[prgmrdave]

Since XHTML is derived from XML, correct on both counts. :cheers:

[$cirisme]

Okie dokie, I wasn't sure how technical you wanted to get. :wink:

What does mine do?

[prgmrdave]

Now, don't even try that silly revenge nonsense ever again. :brow:

:pout:

.*[\/\\](.*)

Are those parentheses literal or grouping? (I've been using un-\\'d parentheses as grouping.)

[$cirisme]

I took that live out of one of my Perl scripts. The parentheses are there to preserve the data that it contains.

[prgmrdave]

Today @ 03:16 PM post located here (http://www.theologyweb.com/forum/showthread.php?s=&postid=115314#post115314)
cirisme:

Okie dokie, I wasn't sure how technical you wanted to get. :wink:



Well, the reason it's not HTML is that HTML doesn't require quotes on attribute values, and I didn't know how to represent closing-quote-only-if-there's-an-opening-quote in a regex.

:duh:

I could have used a forward positive match

(?=<stuff>)

Oh, well. Next time. :hamster: Oh, yeah, no revenge. :smile:

[prgmrdave]

No fair. I don't know Perl. :teeth: (Why do I think you're not buying that?)

It captures the last component of a pathname, such as you'd find in a web site log.

[$cirisme]

Yes. I am using it to clean up uploaded files.

[Sher]

:shy:

:hrm:

:eek:

:egad:

:duh:

[Sher]

:bawl:

:idea:

:deal: Pearls?

[Sher]

:lol:

(shameless post/spam points boost)

[prgmrdave]

Yo.

Ya wants ta spam da t'read, ya gots ta play da game.

:whip:

:spam:

:thumb:

[Sher]

:eek:

/me runs for her life

[prgmrdave]

Rats... was hoping someone else wanted to play, too!

Well, then, I guess it's my turn...

(\[((_\w)|([A-Za-z]))\w*])|(((_\w)|([A-Za-z]))\w*)

(This is kind of way out there...)

[prgmrdave]

/me looks around for SherBear...

[Sher]

:shy:

Whispers:

/me is not that geek-y :teeth:

(I don't know programming codes, or I would play)

:angel:

[$cirisme]

I'm too good at this, I get the matches on the first try.

That matches:

_wordwordword

I have no clue for what. :hrm:

[$cirisme]

SherBear,

Give it a try. They have some examples here: http://www.dotnetcoders.com/web/Learning/Regex/definition.aspx

And you can test them at:

http://www.dotnetcoders.com/web/Learning/Regex/RegexTester.aspx

I learned Perl regexes along time ago, and they sure are fun!

[prgmrdave]

It's "way out there" in the sense that I don't know what your languages background is, other than Perl. So instead, how would you phrase my regex in English?

[prgmrdave]

And yes, _wordwordword is in the set of things I intended the RE to match.

[$cirisme]

Maybe a flow of control matcher?

:huh:

[prgmrdave]

It has to do with a programming language, yep. Big surprise :smile:

[$cirisme]

I'll guess basic, but I dunno.

[prgmrdave]

:thumb:

[$cirisme]

Am I right? I've never done basic.

:teeth: :bunny:

[prgmrdave]

Well, I don't know Perl, so we're even again (that wasn't intended to be revenge :hamster:, though).

Yes, you got the language right. :thumb: It matches strings that are valid identifiers in Visual Basic.NET. They can contain letters, digits and the underscore, they can't start with a digit, they can't contain -only- an underscore, and they can be surrounded in square brackets.

Your turn!

[$cirisme]

Okay, give me a little while.

[$cirisme]

I'm lazy:

:(\w{2,10}):

[prgmrdave]

:Would:

:it:

:match:

:smilies:

? :smile:

[$cirisme]

Yes! :smile:

[prgmrdave]

Do I get to look through TWeb's source code and pick one now? :smile:

No? :sigh:

Me lazy two.

^\w+(?:\.\w+)*$

[$cirisme]

Matches:

Hello:.hello

:hrm:

[prgmrdave]

Urgh.

Specifically, it matches the "Hello" part (but not the ":.hello" part).

[prgmrdave]

I added anchors to the regex. Shouldn't match "Hello:.hello" any more.

[Sher]

:shy: who's winning?

[prgmrdave]

:yipee: I am, by default, 'cause :cir: is busy getting ready for a server move :yipee:

[prgmrdave]

But as soon as that's off his mind, he's gonna come up with another one like his :argh: load :argh: average :argh: regex. :eek:

[prgmrdave]

Only 200 more points 'til I get a pet!

[prgmrdave]

Make that 197...

[prgmrdave]

Make that... oh, never mind. :teeth:

[Sher]

:lol: :thumb:


~Sher, Geek-wannabe

[$cirisme]

I should know just by looking at it what it will match... I am losing it.

It matches "H". :hrm:

:argh:

[prgmrdave]

:prgmrdave: = :too lazy to test regex's:

In your earlier guess, "Hello:.hello", it didn't match "Hello", it matched "H".

/me starts thinking he can't regex his way out of a paper bag (with signposts)

I edited the regex:

^\w+(?:\.\w+)*$

Maybe this will help?

[$cirisme]

Ya got me on this one, everything I think will be a match is not a match... :shrug:

[prgmrdave]

My regex is either too complex or too simple. (Or wrong.) :smile:

What should have matched but did not?

[$cirisme]

I would have thought hello:.hello:.hello would match, but I am an idiot.

[prgmrdave]

Today @ 10:34 AM post located here (http://www.theologyweb.com/forum/showthread.php?s=&postid=122149#post122149)
cirisme:

I would have thought hello:.hello:.hello would match

It should match without the colons.

but I am an idiot.

Stop that. :whack: You are not.

[$cirisme]

Then what is the colon doing in there?

:hrm:

[prgmrdave]

(?:<regex>) are grouping (non-capturing) parentheses. Page 136 in O'Reilly's Mastering Regular Expressions book (Jeffrey Friedl), or the summary here (http://www.dotnetcoders.com/web/Learning/Regex/syntax.aspx).

[$cirisme]

Oh.

:doh:

[jabberwocky]

hostname
hostname.subdomain.etc.tld

except... you can't have "_" in a domain name, can you? And you can have "-".

[prgmrdave]

w00t!

You're up to bat. :thumb:

[jabberwocky]

(?:/(?:\\.|[^/])*/)|(?:(?:m|s|qr)(\\.|[^{}\(\)<>\[\]])[^$1]*$1)|(?:(?:m|s|qr){(?:\\.|[^{}])})|(?:(?:m|s|qr)\((?:\\.|[\(\)])\))|(?:(?:m|s|qr)<(?:\\.|[^<>])>)|(?:(?:m|s|qr)\[(?:\\.|[^\[\]]\]))

/2 spam points for the regex alone.../
/<evil laugh>/

[prgmrdave]

I think it's missing a parenthesis.

[prgmrdave]

No, really. :smile:

I think it needs one more at the end.

[prgmrdave]

Foo. Never mind, I found it.

prgmrdave :poke: prgmrdave

[prgmrdave]

Does it match perl RE's?

[jabberwocky]

Today @ 10:41 PM post located here (http://www.theologyweb.com/forum/showthread.php?s=&postid=126113#post126113)
prgmrdave:

Does it match perl RE's?

It certainly tries to.

handball back to you :egad:

[prgmrdave]

Heh. Not bad, since I don't know perl... okay, lemme think...

[$cirisme]

hostname
hostname.subdomain.etc.tld

:doh:

Well, that's just embarassing... :argh:

I'll be in the corner

:shy:

[prgmrdave]

You shoulda seen my flat forehead after :argh:'ing jabberwocky's regex!

[$cirisme]

It's a long one. :hrm:

[jabberwocky]

Today @ 05:57 PM post located here (http://www.theologyweb.com/forum/showthread.php?s=&postid=127172#post127172)
cirisme:

Well, that's just embarassing... :argh:


You know, most people give themselves kudos for staying out of this kind of insanely geeky game altogether - look how quickly you got rid of Jade! Left her back on page 4!

I say, good on you for leaving an opening for me to join in!

[$cirisme]

Jade was playing? :huh:

[jabberwocky]

Jade was here (not playing) only for a second:

06-02-2003 @ 11:12 AM post located here (http://www.theologyweb.com/forum/showthread.php?s=&postid=115126#post115126)
Jade:

:eek: I never knew geekiness went to this extreme!:shy:

I was thinking of Sher:

06-02-2003 @ 09:47 PM post located here (http://www.theologyweb.com/forum/showthread.php?s=&postid=115731#post115731)
Sher:

:eek:

* Sher runs for her life

[yxboom]

Yesterday @ 09:10 PM post located here (http://www.theologyweb.com/forum/showthread.php?s=&postid=126071#post126071)
jabberwocky:

(?:/(?:\\.|[^/])*/)|(?:(?:m|s|qr)(\\.|[^{}\(\)&lt;&gt;\[\]])[^$1]*$1)|(?:(?:m|s|qr){(?:\\.|[^{}])})|(?:(?:m|s|qr)\((?:\\.|[\(\)])\))|(?:(?:m|s|qr)&lt;(?:\\.|[^&lt;&gt;])&gt;)|(?:(?:m|s|qr)\[(?:\\.|[^\[\]]\]))

/2 spam points for the regex alone.../
/&lt;evil laugh&gt;/

It is a GMT time fix. :whip:

[jabberwocky]

:lol: If only it were that simple...

[prgmrdave]

Gah. It's harder coming up with good ideas than deciphering others'....

[Sher]

Yesterday @ 10:59 PM post located here (http://www.theologyweb.com/forum/showthread.php?s=&postid=127281#post127281)
jabberwocky:

I was thinking of Sher:

:lol: I'm still lurking ... I just don't have a clue :dufus:

(but thanks for thinking of me :teeth:)

[prgmrdave]

:hamster:

\[verse=(?:(AMP|ESV|KJV|NASB|NIV|NLT|YLT):)?(?:(1|2|3|I|II|III)\s+)?(\w+)\s+(\d)+(?::(\d)+(?:-(\d)+))\]

:yipee:

[jabberwocky]

Any guesses, Sher? What kind of expression is this pattern meant to match?

I'd give you a reference like http://www.perldoc.com/perl5.8.0/pod/perlretut.html which would be sure to give you a clue, but ...

...oh, I gave you it anyway...

[jabberwocky]

Sher, your 5 days are about up...

It looks kinda like BB-code/vB-code, what with the whole [something=...] although I haven't seen anything about a [verse] tag. Proposed custom addition for TWeb?

Anyway, it's [verse=X] where X is a series of consecutive verses within a chapter of the Bible, with the version optionally specified.

e.g.
[verse=Armaments 4:16-20]
[verse=KJV:1 John 3:1-16]

Can I assume from your brackets that you meant the verse (especially the second part of a range specification) to be optional: ...)?)?\]

[prgmrdave]

Yes, it's a suggestion for the format for an extended [verse] tag, and you are also correct that the range was intended to be optional.

Batter up!

:thumb:

[jabberwocky]

/?(?:[\w-.]+/)*[\w-.]+|/?(?:[\w-.]+/)+[\w-.]*|/(?:[\w-.]+/)*[\w-.]*

[Xavier]

:bump:

I remember why I failed my RegEx test... :hehe:

[lee_merrill]

/?(?:[\w-.]+/)*[\w-.]+|/?(?:[\w-.]+/)+[\w-.]*|/(?:[\w-.]+/)*[\w-.]*

Would that be a Unix path to a file, with at least one directory in the path name?

Not sure what [\w-.] matches, tho...

Blessings,
Lee