« first day (1725 days earlier)      last day (3120 days later) » 

ah, okay, will try that
 
Anonymous
Using VS to compile non-VS stuff is like trying to use a sledgehammer to drill a hole
 
Anonymous
It'll work with enough effort, but you've broken tons of other things in the process
 
Okay, new problem: "gcc: error: unrecognized command line option '-mno-cygwin'".
 
Anonymous
Hmm
 
Anonymous
9:03 AM
-mno-cygwin hasn't been a valid flag for gcc for a while
 
Anonymous
Somewhere in your Cython/distutils install will be a script that runs the compiler
 
Anonymous
You'll need to remove -mno-cygwin from the compiler flags
 
Can't find that anywhere in either script in Cython/Distutils.
 
Anonymous
Hmm
 
Anonymous
What version of gcc is mingw using?
 
9:07 AM
How would I find out?
 
Anonymous
gcc -v
 
gcc version 4.8.1
 
Anonymous
Hmm
 
Anonymous
distutils only adds -mno-cygwin on <4, when it was valid
 
Anonymous
What does gcc -dumpmachine give you?
 
9:09 AM
mingw32
 
Anonymous
That's really weird...
 
Oh hey I found it.
 
Anonymous
Hmm?
 
self.set_executables(compiler='gcc -mno-cygwin -O -Wall',
                             compiler_so='gcc -mno-cygwin -mdll -O -Wall',
                             compiler_cxx='g++ -mno-cygwin -O -Wall',
                             linker_exe='gcc -mno-cygwin',
                             linker_so='%s -mno-cygwin %s %s'
                                        % (self.linker_dll, shared_option,
                                           entry_point))
 
9:13 AM
That's inside this:
 
@Dennis Since you're always so obsessed with +200 ^^
 
class Mingw32CCompiler(CygwinCCompiler):
    """ Handles the Mingw32 port of the GNU C compiler to Windows.
    """
    compiler_type = 'mingw32'

    def __init__(self, verbose=0, dry_run=0, force=0):
 
Anonymous
Ahh, a bit stupid that the flag was never removed
 
Anonymous
Yeah just completely get rid of -mno-cygwin, the flag doesn't exist in gcc 4+
 
Anonymous
And will only cause you issues
 
9:14 AM
Aaaand a ton of "undefined reference to _" errors again.
 
Anonymous
At this point you might be better off using Cygwin
 
Okay, so, do I change distutils.cfg to say "cygwin" instead of "mingw32"?
 
Anonymous
Nope
 
Anonymous
Install Python whatever version from setup-x86.exe
 
Anonymous
Open a Cygwin prompt, do wget https://bootstrap.pypa.io/ez_setup.py -O - | python
 
Anonymous
9:20 AM
(might also have to install wget, gcc, and g++)
 
Anonymous
Then python easy_install pip; pip install cython
 
Anonymous
Copy your script over, do the stuff
 
Anonymous
And enjoy as everything works
 
...do I really have to re-install Python? Also, are the later commands all from within Cygwin?
 
Anonymous
Yes and yes
 
Anonymous
9:22 AM
Trying to use MSVC Python within Cygwin is a nightmare
 
Why should I re-install 32-bit Python when I have 64-bit Windows? Also, what about all the libraries I've already installed, specifically for 64-bit Python?
 
Anonymous
The 64-bit packages on Cygwin are somewhat unstable right now, because they're new
 
Anonymous
As for the libraries you've installed, to use them inside Cygwin python, you'll have to reinstall them inside Cygwin with pip
 
Should I install Python inside Cygwin?
 
Anonymous
Yes, with the cygwin installer
 
Anonymous
9:31 AM
It doubles as a package manager
 
Okay, that makes much more sense.
 
Anonymous
Sorry I confused you
 
Anonymous
Trying to build a jump table for brainfuck is confusing the hell out of me :/
 
Anonymous
There, finally got it working
 
Anonymous
And it's extendable enough to use it for ShadyAsFuck
 
Anonymous
 
Anonymous
 
........so...I just copied over the files so that they're inside the Python35 folder now.
It...worked.
okay then
 
Anonymous
Yay?
 
Trying the import now.
 
Anonymous
inb4 segfault
 
9:48 AM
Well, nothing seems to be happening.
 
Anonymous
Why not just run it in cygwin?
 
Hung there too.
Ah, so it does work.
It's just slow.
Well. 6 AM. Time to go to bed.
 
10:09 AM
@Mego You have seen this, right?
 
Anonymous
@VTCAKAVSMoACE Not until after I finished mine, naturally
 
@Mauris nice job with ruby there. what's $F and what do the flags do?
 
10:27 AM
aF: is autosplit using the regex :
With those flags on, any call to gets will put gets.split(':') in $F
Probably a Perl-ism? It's really weird
 
huh, that's crazy
 
-p wraps the program in a "read into $_, execute the code, print $_" loop
 
btw, technically it's 6 bytes though, unless you're able to squeeze an e into those command-line arguments to put the code right on the command-line
(if flags need to be added entirely, the hyphen and spaces are counted as well)
 
And for the record, the rescue p is inline error handling. p just returns nil if you pass it no arguments. I could've written rescue'' for the same effect but this was cooler :)
Ah, I wasn't aware how they were counted
 
@Mauris yeah that one I know :)
14
A: On "interactive" answers and other special conditions

J BMy basic guideline would be "as the problem statement says". Which is mostly to mean it's allowed to explicitly override anything it wants from what I'm about to say next. Actions to invoke For test-case/IO programs, I like it better when the whole invocation is possible as a single line, thro...

 
10:32 AM
Are flags counted differently in Retina?
 
I consider -s a standard flag like -f or -e that doesn't need to be counted (I could add the full explanation with using two files and everything without flag and it would give the same byte count)
 
Ah
How actually do we count multi-file solutions...
Sounds like a loophole waiting to be abused
 
+1 per additional file
 
Oh, makes sense
 
0
Q: Align the text to a block

ZeregesTask Your task is to write whole program, which will align given input to a block of given size. Input: 40 Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco labo...

 
 
1 hour later…
11:42 AM
Okay, I've written up a draft to turn Chris's Fibonacci challenge into a catalogue. I tried to include all the usual rules (primarily from the prime testing catalogue) without conflicting the original intent of the challenge. I'd appreciate some feedback before I go ahead with the edit:
For reference, here is the challenge in its current form:
48
Q: Fibonacci function or sequence

Chris Jester-YoungWrite the shortest code that either: Generates a Fibonacci sequence (either in standard output, or as a stream) Calculates, given n, the nth Fibonacci number (I gave both options in case one is easier to do in your chosen language than the other.) Edit: for the avoidance of doubt, a stream...

@Dennis I'd especially appreciate feedback from you because I largely ripped the rules out of your catalogue.
@ChrisJester-Young I know you said you trust my judgement, but in case you're interested in the edit I'll be making, see the gist above.
 
12:20 PM
So where does the "natal" come from in prenatal?
 
12:32 PM
@Sp3 @El'e @fee I tried the grid problem with periodic boundary conditions so there are N^2 2x2 boxes. I got 20352 solutions for N = 4 again which surely means that every normal N = 4 solution also works periodically :O
Periodic N = 2 is still 24 and 56 for N = 3
 
@Calvin'sHobbies context?
 
Just curious
 
same origin as "native", "nation" etc.
 
Wait a sec, I may have done this wrong...
 
Periodic working for even and not for odd makes sense
e.g. if you label the 4x4 as abcd/efgh/ijkl/mnop then a+b+e+f = e+f+i+j -> a+b = i+j. Hence a+b+m+n = i+j+m+n is valid (for an example)
Not sure about the four corners case though
 
12:53 PM
Here are some various periodic N = 4 solutions:

Solution 145
 0  6  1 10
15  9 14  5
 2  4  3  8
13 11 12  7
Solution 146
 0  6  1 10
13 11 12  7
 2  4  3  8
15  9 14  5
Solution 147
 0  6  1 12
15  9 14  3
 4  2  5  8
11 13 10  7
Solution 148
 0  6  1 12
11 13 10  7
 4  2  5  8
15  9 14  3
Solution 149
 0  6  1 14
13 11 12  3
 4  2  5 10
 9 15  8  7
Solution 150
 0  6  1 14
11 13 10  5
 2  4  3 12
 9 15  8  7
Solution 151
 0  6  1 14
 9 15  8  7
 2  4  3 12
11 13 10  5
Solution 152
 0  6  1 14
But yeah, I think there actually aren't any periodic 3 solutions
 
what's this for?
 
12 hours ago, by Calvin's Hobbies
@feersum Grid numbered 0 to 15 and all the 2x2 groups sum to the same value
Except the idea is extended to any n by n grid
(or rectangle, perhaps with periodic boundary conditions)
 
The whole conversation technically starts here:
15 hours ago, by Calvin's Hobbies
Is it possible to have an image where every pixel is a different color but every 2x2 group of pixels has the same average color? Is the upper limit an image with all colors?
 
@Calvin'sHobbies That sounds pretty interesting. And weird that it's not in OEIS yet, it does seem like something that must have been studied before.
 
1:03 PM
That what I said!
 
(hence the emphasis)
 
@Sp3000 It's still curious that there are no non-periodic N=4 answers
 
why? I found his proof quite convincing
you can argue the same thing about the corners
* x x *
y x x y
y z z y
* z z *
we know that x, z and y each have the same sum
we also know that the entire grid must have 4 times the sum of the individual blocks
take the difference, so the corner block must also have the same sum
 
@MartinBüttner I linked to the code in my last message.
9 hours ago, by Dennis
r_e`0\{~'a=2*(*1$+}%$_@=W>
{
  W=2*mQ2+:LS*aL(*'_L*a+N*\
  {
    'a=
    {
      _'_#:I'At_I(='A={I(L)-'_t}&_I)='A={IL)-'_t}&
    }{
      _'A#:I'_t_I(L)-='_={I(L)-' t}&_IL)-='_={IL)-' t}&
    }?
  }/
  N/{S-},W%ee{~S*\S*\+}%_0=,f<W%N*'_' er
}{
  &
}?
 
sorry must have been distracted by @Alex's completely unnecessary ping
 
1:08 PM
Haha.
 
@MartinBüttner Yeah, ok, it's not that weird when you think about it
 
@Sp3000 Haha. Obsessed is the right word, unfortunately.
 
@Dennis Wanna race? You win if you beat martin in rep before I beat martin in any medal count.
 
Good luck to both of you.
 
And Martin is not allowed to gain rep or medals until the race is over XD
 
1:18 PM
Good luck with that as well.
 
@Dennis is mod. (hint hint...)
 
@Calvin'sHobbies Haha. You're only 23 silver badges apart. Knowing you, you could write that many silver questions in a week.
 
Heh, I've finally pulled safely ahead in gold, and now you're coming with silver? :P
 
This account has been temporarily suspended to settle a bet.
6
 
@Dennis I'd be ok with racing my bronze only
 
1:22 PM
@Calvin'sHobbies You can probably also post 28 double-bronze questions in a week. :P
 
^
 
@MartinBüttner Finally? You pulled well ahead ages ago! (*grumbles*)
 
And we're 14.6k rep apart. With the rep cap, it is very difficult to catch up.
 
@Dennis I noticed, you've never offered a bounty. ;)
 
1:23 PM
:O
 
@Dennis Anyway, we're distracting you from reading my draft for the Fibonacci challenge :P
 
You're distracting me from answering. I've already read it. ;)
 
Even I've offered a couple of bounties, and I'm supposedly the resident grinch :P
 
I wonder if our highest user rep counts follow Benfords law?
(i.e. I cant be bothered to check so I'm asking in chat to see if someone else will)
 
Judging by glancing at the top user page, it loosely follows it, yea.
Lots of 1xk, some 2xk, and outliers above that.
 
1:31 PM
Does anyone remember the KotH idea I had the other day?
Because I don't
 
Chat transcript search?
 
I knew someone would say that right after I pressed enter, so I'm way ahead of you
I found it
 
Um, ok. Good job!
 
I'm imagining rainbolt's idea involved battles of rainclouds and thunderbolts. Something stormy.
 
Thanks buddy
 
1:35 PM
np
 
:-/
 
> It was a dark and stormy night...
I love chat flags... "Big Bang Theory is for morons, but Friends is okay." <- offensive!
 
Lol
I marked it invalid as well
 
Making the sequence start at 0 or 1 makes perfect sense for option 1. I'm not a big fan of anything but F(6) = 8 (considering all the relations between n and F(n)), but I assume that was chosen to not break existing submissions. Everything else is perfect.
I do wonder about the implications of "re-using" an existing challenge. Some answers may become invalid, others severely suboptimal, and the challenge might not get the visibility it should.
@MartinBüttner ^
 
@Dennis Context?
 
1:38 PM
Fibocat methinks.
 
2 hours ago, by Martin Büttner
Okay, I've written up a draft to turn Chris's Fibonacci challenge into a catalogue. I tried to include all the usual rules (primarily from the prime testing catalogue) without conflicting the original intent of the challenge. I'd appreciate some feedback before I go ahead with the edit:
and following
 
@Earthybinarydigits Fibocat?
 
What Dennis said, but golfed: Fibonacci catalog
 
+1 for golfing.
 
ah, right..
 
1:42 PM
I was hoping it was enough to decipher, but I may have golfed it wrong.
 
0
A: The Mystery String Printer (Cops)

quartataPBASIC (safe) Range: <= 32 String: 000000184066073CB303FC6A6B901CFC14844C3543BA023499B966F6E05707C0 Code: t VAR Byte c VAR Word FOR c=2016 TO 2048 READ c, t DEBUG HEX2 t NEXT This is kind of a quine. In BS2, the program code is always encoded at the end of the EEPROM. So, what this does...

Safe!
My hint of "16% of people in the world cannot do this" (reading) probably wasn't all that good but I didn't want to completely give it away
 
@Dennis Yes, mostly that, and that 0 or 1 based might be more natural in some languages. Alternatively, I could fix F(6) = 8 and say that the function doesn't need to support F(0)
but again I'm not sure about breaking existing submissions
 
@quartata I don't see prayer anywhere in that code :P
 
@Dennis that's probably true, but at the same time, a lot of the answer are valid, so we'd have to copy them or something.
 
@Geobits That's not what it was supposed to be.
It was illiteracy rate.
 
1:54 PM
Oh. I can't read, so I wasn't sure.
 
@Geobits I can't read either so I might have done it wrong
Wait! How did I just reply to that if I can't read!
 
@Dennis Meta post maybe?
 
Brain, what are you doing....
 
@quartata Text-to-speech and dictation software is actually pretty decent now :)
 
@MartinBüttner What's the other option? Post a separate challenge?
 
1:59 PM
@quartata Think I've seen this.
 
@Dennis pretty much (and close the old one as a duplicate of the new one)
 
@mbomb007 Yes. It's pretty nice
 
@MartinBüttner That's sounds weird..
 
I don't really like that either though
 
@MartinBüttner Closing the old one does not sound... optimal.
 
2:00 PM
(at least in the case where the original one isn't a bad challenge)
 
What makes it cool is that the digits pop up to the beat of the music
 
@Calvin'sHobbies @Geobits I would have no qualms about that if the old challenge was something like codegolf.stackexchange.com/questions/31793/…
but Chris's Fibonacci challenge is fine, which is why I wanted to edit it in the first place (instead of posting a new one)
 
@MartinBüttner +10/-10 lol
 
Yea, I meant that one in particular.
 
@MartinBüttner Also not ideal when there are 50+ answers that could be reposted unimproved. :/
 
2:01 PM
exactly
 
Better yet, just hard-delete it and all references to it on the site and chat, then post the catalog. Duplicate? What duplicate?
^ Disclaimer: No
 
Honestly if what you post is a catalog I don't think it matters if it is a duplicate
 
that... uhh.... could lead to problems :D
 
A catalog is basically the same thing as a golf. Why wouldn't they be dupes?
 
2:04 PM
"Duplicate? Nah... Catalogue!"
 
@Geobits Hmm
But I mean you have mod powers
 
The only real difference I've seen is "I won't accept an answer".
 
You guys are the supreme dictators of PPCG
OK, well just so you know this challenge also exists: codegolf.stackexchange.com/questions/1817/…
 
@Geobits I agree. People could already use the current problem as a catalog. We already allow posting in new languages as long as they don't win.
 
Okay, so let's edit the Fibonacci thing... what about answers that will be invalidated? For example some answers are snippets. And some might not work beyond result 255 (unless you use a Brainfuck-variant with unlimited cells).
@quartata I'm aware. It's a popcon though.
 
2:06 PM
Oh it is a pop-con though
Nevermind
 
What's the upper bound going to be?
 
Wow ninja'd
 
@MartinBüttner I don't like the idea of rewriting our site's history.
Why exactly is a proper catalog necessary?
 
^ +1 +1
 
@MartinBüttner We could add a post notice. That would require a lot of manual labor though.
 
2:08 PM
I've got no problem with catalogs, but surely there are tasks we could set for them that aren't so problematic?
 
Gotta go.
 
0
A: Sandbox for Proposed Challenges

jimmy23013These are some questions that some answers in my previous question seem to be trying to answer. If those questions aren't asked, my question is going to be too broad to contain any "interesting destructive behavior", which isn't my intention. You can take the ownership and leave a comment for an...

 
@Geobits I was thinking 2^31
@Calvin'sHobbies Last time I checked, nothing we do here at all is necessary. :P
 
What's the difference between "catalog" and "question with a bunch of answers"?
 
You can put an extraneous 'ue' on one and it still makes sense.
 
2:12 PM
@Calvin'sHobbies A tighter spec for comparable answers and a meaningful auto-generated leaderboard. If I just slap a leaderboard on the existing challenge, it will include all valid, semi-valid and invalid answers.
 
Dang, I have two ideas for KotH and I'm not sure which to Sandbox
I don't have time towrite up a good spec for both
 
This is where it helps to have a coin to flip.
I don't normally carry cash, though :(
 
@Geobits Flip your credit card then.
 
@Calvin'sHobbies The leaderboard would likely also create more incentive for new answers in languages that haven't been covered yet.
 
I have before. Sugar packets, cough drops... you can flip lots of things.
 
2:15 PM
Idea 1: The goal is to die by jumping into holes. The holes are filled up when you jump into them. More holes spawn, but they follow a pattern. Information is automatically shared with nearby teammates, but you lose all information when you die, so you have to decide between dying and running around to help your teammates die efficiently.
Idea 2: Agar.io in a 2 dimensional world (if you have never played agar.io, I strongly suggest you wait until you have 4 hours to waste)
 
@MartinBüttner Ok. But I envision a catalog as an encyclopedia, where everything is communally verified. Making sure that 100 answers are 100% correct is more work than we usually do here, and more than I'd expect anyone to maintain.
 
Idea 1, for the main reason that I don't know what #2 is at all.
 
@Calvin'sHobbies It worked reasonably well for Hello, World! which has more answers than any other challenge.
 
@Rainbolt Oh crap, it's that game. I've seen the kid playing it.
 
@MartinBüttner I guess I'm saying our site shouldn't try to be a wiki. It works better as an ideas board.
@Rainbolt Isn't it already 2d?
 
2:18 PM
Well... yes lol
I meant more like 2D discrete
 
Squares?
 
Yea
 
@Calvin'sHobbies I don't think a wiki could serve this purpose remotely as well as a competitive community.
And I actually think we have more peer review here than e.g. on Rosetta Code or esolangs.
 
I really like the agar idea. I don't know the exact mechanics, but I'd love a continuous, flocking version. Somewhat like my hunger gaming koth, but without the "prey".
 
Every player is both prey and predator I guess
 
2:20 PM
Right, that's what I meant. Mine had docile prey controlled by the controller.
 
And yours was fluid
I'm a noob, so I stick to 2d arrays
 
Hmm. That might make a good project for an ongoing game, hosted on a server.
 
Hey, you're right. Maybe we should invent agar.io
 
The bots can respawn/restart when dead.
Yea, but without the pesky humans messing up the game.
Besides, agar is just a flow / Spore knockoff :P
 
Spore?
 
2:25 PM
Yea, in Spore you start as a cell-thing, and have to eat smaller cell-things to grow, etc etc, until you can evolve into the next stage.
 
That brings me back
 
Nvm then :)
 
Maybe a knock off of stage 1
 
Yea, that's what I mean.
 
Spore has the lowest replay value of any game I have ever played
 
2:25 PM
So very true.
And it sounded like it was going to be awesomely replayable. The hype lied :(
flow was much better, and looked nice, too.
 
I might have replayed if the decisions I made significantly impacted my species in the later stages
But they didn't
I have no arms? Who cares. I can still fly spaceships
I pissed off a particular nation? Oh well. Destroy them. Find another nation like them.
No sense of "Dang, I really wish I hadn't done that."
 
Stage 1 was the only part I really even found fun at all, even on the first play through.
 
@MartinBüttner But in a competitive community people are much more interested in the (apparent) quality of their own contributions than the quality of the work as a whole. I know PPCG users aren't that conniving or careless but that doesn't mean we have the incentives for wiki-like material.
 
@Calvin'sHobbies [citation needed] ;)
 
Not a great citation, but there are soooo many catalog answers that I haven't looked at. I have no idea how many are valid. Once it hits a few pages, I really stop caring.
 
2:31 PM
Exactly
 
Of course, you could argue that no one person has looked at all of Wikipedia, too ;)
 
@Geobits I think I had a look at most of them (although I do get notifications). I had the impression that Sp regularly had a look at new ones as well.
@Geobits and this...
 
Are you guys using the phrase "catalog answers" instead of "list question"?
 
not really. unless all ppcg challenges are list questions now.
 
Oh god, he caught on! Run for it, Marty!
 
2:33 PM
(I picture Rainbolt now writing an email to Joel, requesting PPCG to be shut down with immediate effect.)
 
> It was a dark and stormy night when PPCG made the decision to move to list questions. [...]
 
@Geobits The wiki quality-check community is much larger than ours though.
 
@Calvin'sHobbies Depends on the wiki...
(And the site being edited.)
 
Okay, I'm going with the agar.io ripoff. I just realized the complexity of step 4. Every day, the following happens in order:
1. Holes spawn
2. Zombies move
3. Simultaneously, zombies on holes die, and holes underneath zombies fill
4. Information is shared between zombies of the same faction in close proximity. This step repeats until no further information is shared.
 
Not much different than a basic floodfill, no?
Well, one flood for each piece of information I guess.
 
2:39 PM
Except that the floodfill only considers zombies of the same faction, meaning I have to floodfill 90 times if there are 90 submissions. Not to mention that there may be multiple pods within a faction. That's a lot of flood filling.
 
How did circles turn into squares that turned into zombies?
 
No, this is the other idea. I'm switching to agar.io because the controller will take too long to think in this one
Between the controller crunching numbers and the 90 submissions crunching numbers, the rounds will take too long
 
@Rainbolt How big are the factions? Flood filling just a few zombies should be insanely fast, even 90 times.
 
Large enough to encourage setting up networks of zombies to share information, but no larger I guess
Too small and you end up just running around trying to die
 
@Rainbolt Can you call it asquar.io?
 
2:41 PM
not squagar.io? :P
 
Anyway, imagine how fast one single purpose blob per player would be compared to multi purpose zombie factions per player
 
agrid.io
 
Calculating overlap will be the hardest thing probably
 
Yea, I'm not sure exactly how this is going to work on a grid.
 
If your center is within the other circle's radius, you are food.
 
2:42 PM
Not saying it can't, but it might actually be easier to lose the grid in this case.
 
How about the center moves on the grid, and the circle is just imaginary
 
@Geobits Yeah. The motions are dependent on fine differences in velocities.
 
Maybe they can all have the same velocity, but we'll make it easier to burst the bigger guys.
 
How to descretize that without using huge integers?
 
I think this depends on the grid size versus velocity, and how you "move". Is there a rate of turn cap, etc, or just "move to (x,y)"?
 
2:47 PM
More like "up, down, left, right"
 
So I can do a complete 180 from one turn to the next, or is there a momentum change?
 
I guess I could give the huge guys a chance to move every 5 turns, the semi huge guys every 4 turns, etc.
There's no momentum
Calvin is right though. The fun of the game is the fine difference in velocity. Discrete velocities kind of ruin that
 
Have we found a defining liberal vs conservative programmer test (in the Steve Yegge sense)?plus.google.com/110981030061712822816/posts/KaSKeg4vQtzalimack 4 hours ago
... what?
2
 
@MartinBüttner @feersum @quartata arstechnica.com/gadgets/2015/10/…
 
I solved the floodfill problem. If you survive for 20 days, you become a zombie overlord. What you know becomes common knowledge to the entire faction.
No floodfills needed in that case
 
2:59 PM
Are you cutting info-sharing from lesser zombies then?
 

« first day (1725 days earlier)      last day (3120 days later) »