hatzka-nezumi · March 16, 2017 16:17
diff --git a/,(= 5 2) b/,(= 5 2)
 freenode/#clojure on 2014-04-21:

 amalloy sets 5 to be 2 using reflection, probably in PM with clojurebot:
 (let [field (nth (.getDeclaredFields Long) 3)] (.setAccessible field true) (.set field 5 2))


 19:10 < amalloy> justin_smith: i think ''lovecraft should contain (alter-var-root #'defmacro (constantly (fn [& args] `(quote ~(rand-nth '#{various lovecraftian horrors|)))))
 19:11 < justin_smith> ouch
 19:11 < amalloy> or something like it, anyawy
 19:11 < justin_smith> yeah
 19:11 < justin_smith> maybe ''necronomicon would be the better name for it
 19:13 < amalloy> or you could mutate one of the interned Number instances
 19:13 < amalloy> really there are a lot of deep horrors you could put into a lovecraft library
 19:14 < justin_smith> indeed
 19:16 < justin_smith> stack traces generated using https://github.com/prismofeverything/zalgo
 19:17 < rasmusto> forgive me: z̷̼͖̰̜̣͛̂̚͟ạ̶̹͔็l̀̄͌̈̐ͫg̉̊̐͑ͮͦȏ̉̏̿̿ͩ̓ï̞̗͓̮̮͙̃̄̚z̞̗͍̃͆̐͆͊̄̚e͗ͣ̔ͣ
 19:17 < arrdem> justin_smith: what have you done...
 19:17 < amalloy> huh. clojurebot's sandbox allows you to actually do what i just said
 19:17 < amalloy> ,5
 19:17 < clojurebot> 2
 19:17 < justin_smith> woah...
 19:17 < amalloy> i'll go put things back together now
 19:18  * arrdem flees in terror
 19:19 < amalloy> uhhhhhh, i can't put it back. 5 is 2 now. even if you add 1 and 4 you get 2. so there's no longer a 5 to reset 5 to
 19:19 < amalloy> hopefully clojurebot's regular sandbox-resetting takes care of that
 19:19 < rasmusto> ,(dec 5)
 19:19 < clojurebot> 1
 19:19 < arrdem> amalloy: what have you done...
 19:19 < rasmusto> seems to work
 19:19 < arrdem> ,(+ 3 5)
 19:19 < clojurebot> 2
 19:20 < justin_smith> rofl
 19:20 < justin_smith> ,(= 5 2)
 19:20 < clojurebot> true
 19:20 < justin_smith> ROFL
 19:20 < arrdem> I don't even have a gif for this level of wat...
 19:20 < justin_smith> (inc amalloy)
 19:20 < lazybot> ⇒ 101
 19:20 < rasmusto> ,(range 6)
 19:20 < clojurebot> #<IndexOutOfBoundsException java.lang.IndexOutOfBoundsException>
 19:20 < arrdem> AHAHAHAHA
 19:20 < rasmusto> what
 19:20 < arrdem> (inc amalloy)
 19:20 < lazybot> ⇒ 102
 19:20 < justin_smith> lol
 19:20 < justin_smith> (take 10 (range 6))
 19:20 < justin_smith> ,(take 10 (range 6))
 19:20 < clojurebot> #<IndexOutOfBoundsException java.lang.IndexOutOfBoundsException>
 19:20 < arrdem> amalloy: I want to see the source for this...
 19:20 < arrdem> ,6
 19:20 < clojurebot> #<IndexOutOfBoundsException java.lang.IndexOutOfBoundsException>
 19:21 < arrdem> WHAT THE hahaha
 19:21 < amalloy> oh, it's not hard. (let [field (nth (.getDeclaredFields Long) 3)] (.setAccessible field true) (.set field 5 2))
 19:21 < rasmusto> ,7 ;plz?
 19:21 < clojurebot> #<IndexOutOfBoundsException java.lang.IndexOutOfBoundsException>
 19:21 < justin_smith> ,7
 19:21 < clojurebot> #<IndexOutOfBoundsException java.lang.IndexOutOfBoundsException>
 19:21 < amalloy> i think i broke all the numbers when i tried to put it back together
 19:21 < rasmusto> ah okay.
 ...
 19:22 < justin_smith> clearly the 'adams ns will turn every number up to N into 42
 ...
 19:38 < gtrak> knr doesn't actually work... I'm making a note on the githubs :-)
 19:38 < gtrak> it's half-completed
 19:39 < justin_smith> hey, half-complete knr, complete footgun, who would know the difference except for a few symbols and filenames?
 19:39 < amalloy> gtrak: 1/2 completed is the same as 1/5 completed, actually
 ...
 19:40 < gtrak> amalloy: I think I'm missing something subtle there :-)
 19:40 < gtrak> or is it 80/20?
 19:40 < arrdem> also gtrak. meajure is now a thing. you're entirely to blame for the name.
 19:40 < gtrak> hehe
 19:40 < amalloy> gtrak: no, i was referring to my recent clojurebot hack
 19:40 < justin_smith> gtrak: search up for ,(= 5 2)
 19:40 < arrdem> ,(= 5 2)
 19:40 < clojurebot> false
 19:41 < arrdem> good. clojurebot is no longer horked
 19:41 < rasmusto> ,(range 6)
 19:41 < clojurebot> (0 1 2 3 4 ...)
 19:41 < rasmusto> where's 5????
 19:41 < TEttinger> rasmusto: ##(range 6)
 19:41 < lazybot> ⇒ (0 1 2 3 4 5)
 19:41 < justin_smith> ,5
 19:41 < clojurebot> 5
 19:41 < rasmusto> :)
 19:41 < TEttinger> one of the bots trims output
 19:42 < justin_smith> we were just kind of worried about 5 and hoping 5 was OK
 19:42 < rasmusto> TEttinger: gotcha, was making a reference to ,(= (dec 5) 1) earlier
 19:42 < TEttinger> heh wha
 19:43 < justin_smith> probably the best clojure wat I have seen so far
 19:43 < amalloy> i figured out how i broke clojurebot the second time
 19:43 < justinholguin> ,(= (dec 5) 1)
 19:43 < clojurebot> false
 19:43 < justinholguin> phew
 19:43 < amalloy> i was trying to set 5 back to 5, but instead i set 2 to 5. so looking up the third item in an array caused you to look up the sixth item instead. thus all the ArrayIndexOutOfBoundsException problems
 19:44 < gtrak> how'd you do that in the first place? I didn't see it in the scroll-up
 19:44 < TEttinger> ,(def 5 4)
 19:44 < clojurebot> #<CompilerException java.lang.RuntimeException: First argument to def must be a Symbol, compiling:(NO_SOURCE_PATH:0:0)>
 19:44 < justin_smith> another good name for an intentionaly messed up lib would be angkor (named for the world's biggest Wat)
 19:44 < amalloy> gtrak: it's not hard. (let [field (nth (.getDeclaredFields Long) 3)] (.setAccessible field true) (.set field 5 2))
 19:44 < justin_smith> gtrak: amalloy pasted it, by setting some properties of Long I think
 19:44 < justinholguin> (inc justin_smith)
 19:44 < lazybot> ⇒ 34
 19:44 < justin_smith> oh, there it is again
 19:45 < gtrak> ah, ye old integer cache
 19:45 < justin_smith> probably will call for some new sandbox rules
 ...
 19:45 < TEttinger> (let [field (nth (.getDeclaredFields Long) 3)] (.setAccessible field true) (.set field (rand-int 100) 101))
 19:45 < TEttinger> ,(let [field (nth (.getDeclaredFields Long) 3)] (.setAccessible field true) (.set field (rand-int 100) 101))
 19:45 < clojurebot> #<IllegalArgumentException java.lang.IllegalArgumentException: Can not set final long field java.lang.Long.value to java.lang.Integer>
 19:45 < TEttinger> aw
 19:45 < rasmusto> justin_smith: glad someone else shares my taste for puns
 19:46 < amalloy> iit turns out you have to do a bit more work to set it back
 19:46 < amalloy> (let [five (long (int (+ (int 1) (int 4)))), field (nth (.getDeclaredFields Long) 3)] (.setAccessible field true) (.set field five (int (+ (int 1) (int 4))))) ;; actually worked to recover my repl
 19:46 < rasmusto> amalloy: is this idiomatic code?
 19:47 < rasmusto> er, does it at least show how to use java interop properly?
 19:47 < TEttinger> there's nothing proper here
 19:47 < arrdem> rasmusto: do you see that crowbar? do you see it?
 19:47 < arrdem> rasmusto: that's almost as [good|bad] as the reader macros with crowbars hack
 19:47 < amalloy> sure, just like carrying a chainsaw with you at all times is the idiomatic way to get into doors you really shouldn't be getting into
 ... ... ...
 ... ... ...
 ... ... ...
 14:21 < amalloy> gfredericks: were you around when i broke clojurebot by redefining 5 to be 2?
 14:22  * llasram would like to see that...
 14:22 < amalloy> llasram: it's not super-hard. you use reflection to modify the .value field of (Long/valueOf 5)
 14:23 < technomancy> isn't this something we make fun of ruby for making possible?
 14:23 < technomancy> I'll have to update my Tome of Mockery
 14:25 < llasram> amalloy: Ah.  Yeah, ok.  IMHO using reflection to stomp on private final (I assume?) fields is basically the same as memcpy()ing
 14:25 < llasram> So I'm somewhat less impressed :-)
 14:26 < gfredericks> amalloy: no way that's nutso
 14:26 < TimMc> :-(
 14:27 < gfredericks> Tim does not like it
 14:31 < llasram> Well, unless someone re-defined TimMc's smiley face to use ":-("
 14:31 < llasram> At this point we have no real way of knowing
 14:34 < hyPiRion> It's hard to know anything when everything is relative and changing.
 14:35 < gfredericks> except the undeniable objective fact that it's hard to know anything when everything is relative and changing
 14:37 < TimMc> amalloy: How's that?
 14:37 < amalloy> how's what?
 14:37 < TimMc> I did (.setAccessible (.getDeclaredField Long "value") true) but then this errors: (.setLong (.getDeclaredField Long "value") 5 2)
 14:38 < hyPiRion> gfredericks: but then it's not hard to know anything, it's hard to know anything with the exception of that statement
 14:38 < amalloy> TimMc: setAccessible mutates the Field object you have. you can't re-get the Field object
 14:38 < TimMc> Oh! Point.
 14:38 < amalloy> (doto (.getDeclaredField Long "value") (.setAccessible true) (.setLong 5 2)) or something
 14:39 < ystael> amalloy: does that break just one five or all the fives?
 14:39 < amalloy> all the (boxed) fives
 14:39 < llasram> Oh, language
 14:39 < amalloy> and since clojure boxes everything, that's all of them
 14:39 < TimMc> ,(let [f (.getDeclaredField Long "value"), subject 5] (.setAccessible f true) (.setLong f subject 2) subject)
 14:39 < clojurebot> 2
 14:39 < TimMc> D-:
 14:39 < hyPiRion> amalloy: not all the boxed fives?
 14:39 < amalloy> TimMc: ugh, don't do it to clojurebot again
 14:39 < llasram> ,(+ 5 5)
 14:39 < clojurebot> 4
 14:39 < hyPiRion> ,(Long. "5")
 14:39 < clojurebot> 5
 14:39 < TimMc> oops
 14:40 < amalloy> hyPiRion: true. all the ones boxed via Long/valueOf
 14:40 < TimMc> I should have use a bigger 5.
 14:40 < arrdem> ,(+ 2 2)
 14:40 < clojurebot> 4
 14:40 < arrdem> ,(+ 5 5)
 14:40 < clojurebot> 4
 14:40 < ystael> is there a numerical threshold above which boxed instances are not cached and shared?
 14:40 < jcromartie> wow, nice
 14:40 < TimMc> ,(let [f (.getDeclaredField Long "value"), subject 5] (.setAccessible f true) (.setLong f subject 5) subject) ;; fixed?
 14:40 < clojurebot> 2
 14:40 < hyPiRion> ystael: 128 or 256 I think
 14:40 < llasram> ,(+ 5 5)
 14:40 < clojurebot> 4
 14:40 < amalloy> ystael: +/- 128
 14:40 < TimMc> hahaha
 14:40 < amalloy> well, -128 to 127
 14:40 < llasram> ,(let [f (.getDeclaredField Long "value"), subject 5] (.setAccessible f true) (.setLong f subject (Long. "5") subject))
 14:40 < clojurebot> #<IllegalArgumentException java.lang.IllegalArgumentException: No matching method found: setLong for class java.lang.reflect.Field>
 14:41 < arrdem> http://i0.kym-cdn.com/photos/images/newsfeed/000/613/012/2fa.gif
 14:41 < hyPiRion> TimMc: You have seen nothing yet
 14:41 < llasram> ,(let [f (.getDeclaredField Long "value"), subject 5] (.setAccessible f true) (.setLong f subject (Long. "5")) subject)
 14:41 < clojurebot> 5
 14:41 < TimMc> It restarts every 12 minutes, right?
 14:41 < llasram> ,(+ 5 5)
 14:41 < clojurebot> #<IndexOutOfBoundsException java.lang.IndexOutOfBoundsException>
 14:41 < TimMc> And by 12, I mean 12.
 14:41 < llasram> Hmm
 14:42 < amalloy> llasram: that's the way i tried to fix it the first time
 14:42 < amalloy> but you made the same mistake i did
 14:42 < amalloy> instead of defining 5 back to 5, you defined 2 to 5
 14:42 < llasram> Oh, right
 14:42 < llasram> hah!
 14:42 < hyPiRion> hahah
 14:42 < amalloy> so now like every array access breaks
 14:42 < TimMc> I <3 FORTRAN.
 14:43 < amalloy> we never did find out if clojurebot's auto-restart actually fixes this problem
 14:43 < llasram> *amazing*
 14:43 < amalloy> hiredman gave it a manual hard-reset before waiting for that
 14:43 < arrdem> amalloy: one way to find out...
 14:45 < TimMc> ,(let [f (.getDeclaredField Long "value"), subject 0] (.setAccessible f true) (.setLong f subject 42) subject) ;; if it will even run...
 14:45 < clojurebot> #<IndexOutOfBoundsException java.lang.IndexOutOfBoundsException>
 14:45 < TimMc> ,(.getDeclaredField Long "value")
 14:45 < clojurebot> #<IndexOutOfBoundsException java.lang.IndexOutOfBoundsException>
 14:45 < TimMc> Amazing.
 14:45 < amalloy> TimMc: right, it's too late i think. you can no longer eval anything
 14:45 < amalloy> because eval uses the number 2 in an array somewhere
 14:45 < amalloy> ,1
 14:45 < clojurebot> #<IndexOutOfBoundsException java.lang.IndexOutOfBoundsException>
 14:46 < TimMc> Got it.
 14:46 < arrdem> hahaha
 14:46 < hyPiRion> ,'1
 14:46 < clojurebot> #<IndexOutOfBoundsException java.lang.IndexOutOfBoundsException>
 14:46 < arrdem> (inc TimMc)
 14:46 < lazybot> ⇒ 58
 14:46 < TimMc> Only 28 points!
 14:46 < TimMc> I thought I had more.
 14:47 < hyPiRion> TimMc: well, since 2 is 5 and 5 is 2
 14:47 < gfredericks> wait we can prove a contradiction with this right? and then derive anything?
 14:47 < gfredericks> let's derive bugfixes
 14:47 < arrdem> gfredericks: quick prove P!=NP then we can keep our jobs forever
 14:48 < TimMc> amalloy: Do you know any way to get ahold of that interned 5?
 14:48 < ystael> TimMc: it's no longer a number ... it's a free man now
 14:49 < amalloy> TimMc: what do you mean?
 14:49 < gfredericks> arrdem: or P = NP but only for galactic algorithms
 14:49 < TimMc> Well, I'm trying to restore my REPL's notion of small numbers.
 14:49 < arrdem> gfredericks: this is also an acceptable outcome.
 14:49 < amalloy> i don't think you can do it if you've gone to the second level of broken like llasram did
 14:50 < amalloy> but if you've only broken it one level, it is reversible
 14:50 < TimMc> Just one level of fail.
 14:50 < amalloy> iirc this works: (let [five (long (int (+ (int 1) (int 4)))), field (nth (.getDeclaredFields Long) 3)] (.setAccessible field true) (.set field five (int (+ (int 1) (int 4)))))
 14:51 < TimMc> I'm having trouble figuring out what's going on here: https://www.refheap.com/85696
 14:52 < amalloy> fiddling with (Long. "5") doesn't help, because that's not the interned version
 14:52 < TimMc> Oh! Right-o.
 14:52 < llasram> Wouldn't (Long/valueOf "5") work?  Or does the compiler do funny things?
 14:52 < amalloy> &(identical? (Long/valueOf "5") 5)
 14:52 < lazybot> ⇒ false
 14:52 < llasram> Oh, interesting
 14:52 < amalloy> &(identical? (Long/valueOf 5) 5)
 14:53 < lazybot> ⇒ true
 14:53 < llasram> That makes sense of course
 14:53 < amalloy> i don't know *why*, but apparently the string version bypasses interning
 14:53 < amalloy> llasram: does it? it should parse it as a long, then use the intern cache to get a Long
 14:53 < TimMc> valueOf probably just shortcuts, right?
 14:54 < amalloy> incidentally, three cheers for repl history: i just searched it for setAccessible to find this fix; i probably wouldn't have remembered it on my own
 14:55 < llasram> amalloy: Eh.  It's not really clear to me why it should use the cache for the result of string parsing.  Not saying it shouldn't -- it just doesn't seem unreasonable for it to not
 14:55 < hyPiRion> TimMc: Oh hey, if you find this sort of stuff funny, you should see this one: https://www.refheap.com/85697
 14:56 < hyPiRion> That one segfaults or makes your JVM spin up an infinite loop
 14:57 < amalloy> hyPiRion: do you really have to remove its FINAL modifier? once you make it accessible i think you can already write to it
 14:57 < amalloy> just like for Long
 14:57 < amalloy> maybe you're doing something weird i don't see there
 14:59 < hyPiRion> amalloy: it's static, which makes it a bit different IIRC
 14:59 < hyPiRion> at least I just checked, and it doesn't work with just .setAccessible
 15:00 < TimMc> ,(+(*)(*)(*)(*)(*))
 15:00 < clojurebot> 5
 15:00 < TimMc> It's back!
 15:00 < llasram> ,(+ 5 5)
 15:01 < clojurebot> 10
 15:01 < llasram> Yay!
 15:07 < TimMc> amalloy: I was able to fix it in my REPL by using "subject (+ 2 3)", but when I tried again with "subject 5" nREPL quit with an NPE. :-P
 ...
 15:10 < amalloy> TimMc: yeah, that's setting 2 to 5 again
 15:10 < amalloy> which really is a lot more disruptive than setting 5 to 2
 ... ... ...
 ... ... ...
 ... ... ...
 13:48 < justin_smith> ,(let [field (nth (.getDeclaredFields Long) 3)] (.setAccessible field true) (.set field 5 2))
 13:48 < clojurebot> nil
 13:48 < justin_smith> ,(inc 5)
 13:48 < clojurebot> 3
 13:48 < andyf> You can break the JVM's 5, but Plato's 5 is safe in the world of forms
 13:48 < justin_smith> haha
 13:48 < gfredericks> ~5 is the set of all sets with 5 elements
 13:49 < clojurebot> Ik begrijp
 13:49 < bbloom_> , (.getDeclaredFields Long) 3)
 13:49 < clojurebot> #<Field[] [Ljava.lang.reflect.Field;@5753b0>
 13:49 < bbloom_> ,(.getName (.getDeclaredFields Long) 3))
 13:49 < clojurebot> #<IllegalArgumentException java.lang.IllegalArgumentException: No matching method found: getName for class [Ljava.lang.reflect.Field;>
 13:49 < bbloom_> ,(.name (.getDeclaredFields Long) 3))
 13:49 < clojurebot> #<IllegalArgumentException java.lang.IllegalArgumentException: No matching method found: name for class [Ljava.lang.reflect.Field;>
 13:49 < bbloom_> ,(clojure.reflect/reflect (.getDeclaredFields Long) 3))
 13:49 < clojurebot> #<ClassNotFoundException java.lang.ClassNotFoundException: clojure.reflect>
 13:49 < bbloom_> ,(require 'clojure.reflect)
 13:49 < clojurebot> nil
 13:49 < bbloom_> eh wahtever... how do i get the name of a field?
 13:49 < gfredericks> .getName?
 13:49 < bbloom_> i tried that
 13:49 < bbloom_> oh dur it's an array
 13:49 < gfredericks> with 3?
 13:50 < bbloom_> ,(.name (nth (.getDeclaredFields Long) 3))
 13:50 < clojurebot> #<IllegalArgumentException java.lang.IllegalArgumentException: No matching field found: name for class java.lang.reflect.Field>
 13:50 < bbloom_> i suck
 13:50 < bbloom_> ,(.getName (nth (.getDeclaredFields Long) 3))
 13:50 < clojurebot> "value"
 13:50 < andyf> The world of forms: Bringing you real immutability since 348 BC
 13:50 < bbloom_> ah ok i gotcha. boxed small numbers are interned
 13:51 < justin_smith> , (nth (.getDeclaredFields Long) 3)
 13:51 < clojurebot> #<Field private final long java.lang.Long.value>
 13:51 < justin_smith> ,(.getName (nth (.getDeclaredFields Long) 3))
 13:51 < clojurebot> "value"
 13:51 < bbloom_> andyf: https://en.wikipedia.org/wiki/348_BC <- i was hoping you knew the actual year
 13:51 < justin_smith> bbloom_: ^
 13:51 < bbloom_> justin_smith: i got there eventually :-P
 13:52 < justin_smith> bbloom_: aha, I missed it while I was finding it in my own repl
 13:52 < andyf> Sorry, I don't know when he wrote about it first, so I approximated by the year Plato died.
 13:53 < justin_smith> ,(= 5 3)
 13:53 < clojurebot> false
 13:53 < justin_smith> ,(= 5 2)
 13:53 < clojurebot> true
 13:53 < justin_smith> ,(let [field (nth (.getDeclaredFields Long) 3)] (.setAccessible field true) (.set field 5 (inc 4)))
 13:53 < clojurebot> nil
 13:53 < justin_smith> ,(= 5 2)
 13:53 < clojurebot> true
 13:53 < justin_smith> I thought I fixed that...
 13:54 < justin_smith> ,(let [field (nth (.getDeclaredFields Long) 3)] (.setAccessible field true) (.set field (inc 4) (inc 4)))
 13:54 < clojurebot> nil
 13:54 < justin_smith> ,(= 5 2)
 13:54 < clojurebot> true
 13:54 < justin_smith> ,(inc 2)
 13:54 < clojurebot> 3
 13:54 < justin_smith> ,(inc 5)
 13:54 < clojurebot> 3
 13:54 < mearnsh> this is why we can't have nice things
 13:54 < AeroNotix> ,(inc (+ 3 2))
 13:54 < clojurebot> 6
 13:54 < justin_smith> :( I forgot how to fix it
 13:54 < AeroNotix> ,(inc 5)
 13:54 < clojurebot> 3
 13:54 < AeroNotix> 5
 13:55 < AeroNotix> ,5
 13:55 < clojurebot> 2
 13:55 < justin_smith> ,(let [five (long (int (+ (int 1) (int 4)))), field (nth (.getDeclaredFields Long) 3)] (.setAccessible field true) (.set field five (int (+ (int 1) (int 4)))))
 13:55 < clojurebot> nil
 13:55 < andyf> Once you make 5 equal to something else, the JVM extrapolates and is able to prove that all numbers are simultaneously equal to, and not equal to, all others.
	freenode/#clojure on 2014-04-21:

	amalloy sets 5 to be 2 using reflection, probably in PM with clojurebot:
	(let [field (nth (.getDeclaredFields Long) 3)] (.setAccessible field true) (.set field 5 2))


	19:10 < amalloy> justin_smith: i think ''lovecraft should contain (alter-var-root #'defmacro (constantly (fn [& args] `(quote ~(rand-nth '#{various lovecraftian horrors\|)))))
	19:11 < justin_smith> ouch
	19:11 < amalloy> or something like it, anyawy
	19:11 < justin_smith> yeah
	19:11 < justin_smith> maybe ''necronomicon would be the better name for it
	19:13 < amalloy> or you could mutate one of the interned Number instances
	19:13 < amalloy> really there are a lot of deep horrors you could put into a lovecraft library
	19:14 < justin_smith> indeed
	19:16 < justin_smith> stack traces generated using https://github.com/prismofeverything/zalgo
	19:17 < rasmusto> forgive me: z̷̼͖̰̜̣͛̂̚͟ạ̶̹͔็l̀̄͌̈̐ͫg̉̊̐͑ͮͦȏ̉̏̿̿ͩ̓ï̞̗͓̮̮͙̃̄̚z̞̗͍̃͆̐͆͊̄̚e͗ͣ̔ͣ
	19:17 < arrdem> justin_smith: what have you done...
	19:17 < amalloy> huh. clojurebot's sandbox allows you to actually do what i just said
	19:17 < amalloy> ,5
	19:17 < clojurebot> 2
	19:17 < justin_smith> woah...
	19:17 < amalloy> i'll go put things back together now
	19:18 * arrdem flees in terror
	19:19 < amalloy> uhhhhhh, i can't put it back. 5 is 2 now. even if you add 1 and 4 you get 2. so there's no longer a 5 to reset 5 to
	19:19 < amalloy> hopefully clojurebot's regular sandbox-resetting takes care of that
	19:19 < rasmusto> ,(dec 5)
	19:19 < clojurebot> 1
	19:19 < arrdem> amalloy: what have you done...
	19:19 < rasmusto> seems to work
	19:19 < arrdem> ,(+ 3 5)
	19:19 < clojurebot> 2
	19:20 < justin_smith> rofl
	19:20 < justin_smith> ,(= 5 2)
	19:20 < clojurebot> true
	19:20 < justin_smith> ROFL
	19:20 < arrdem> I don't even have a gif for this level of wat...
	19:20 < justin_smith> (inc amalloy)
	19:20 < lazybot> ⇒ 101
	19:20 < rasmusto> ,(range 6)
	19:20 < clojurebot> #<IndexOutOfBoundsException java.lang.IndexOutOfBoundsException>
	19:20 < arrdem> AHAHAHAHA
	19:20 < rasmusto> what
	19:20 < arrdem> (inc amalloy)
	19:20 < lazybot> ⇒ 102
	19:20 < justin_smith> lol
	19:20 < justin_smith> (take 10 (range 6))
	19:20 < justin_smith> ,(take 10 (range 6))
	19:20 < clojurebot> #<IndexOutOfBoundsException java.lang.IndexOutOfBoundsException>
	19:20 < arrdem> amalloy: I want to see the source for this...
	19:20 < arrdem> ,6
	19:20 < clojurebot> #<IndexOutOfBoundsException java.lang.IndexOutOfBoundsException>
	19:21 < arrdem> WHAT THE hahaha
	19:21 < amalloy> oh, it's not hard. (let [field (nth (.getDeclaredFields Long) 3)] (.setAccessible field true) (.set field 5 2))
	19:21 < rasmusto> ,7 ;plz?
	19:21 < clojurebot> #<IndexOutOfBoundsException java.lang.IndexOutOfBoundsException>
	19:21 < justin_smith> ,7
	19:21 < clojurebot> #<IndexOutOfBoundsException java.lang.IndexOutOfBoundsException>
	19:21 < amalloy> i think i broke all the numbers when i tried to put it back together
	19:21 < rasmusto> ah okay.
	...
	19:22 < justin_smith> clearly the 'adams ns will turn every number up to N into 42
	...
	19:38 < gtrak> knr doesn't actually work... I'm making a note on the githubs :-)
	19:38 < gtrak> it's half-completed
	19:39 < justin_smith> hey, half-complete knr, complete footgun, who would know the difference except for a few symbols and filenames?
	19:39 < amalloy> gtrak: 1/2 completed is the same as 1/5 completed, actually
	...
	19:40 < gtrak> amalloy: I think I'm missing something subtle there :-)
	19:40 < gtrak> or is it 80/20?
	19:40 < arrdem> also gtrak. meajure is now a thing. you're entirely to blame for the name.
	19:40 < gtrak> hehe
	19:40 < amalloy> gtrak: no, i was referring to my recent clojurebot hack
	19:40 < justin_smith> gtrak: search up for ,(= 5 2)
	19:40 < arrdem> ,(= 5 2)
	19:40 < clojurebot> false
	19:41 < arrdem> good. clojurebot is no longer horked
	19:41 < rasmusto> ,(range 6)
	19:41 < clojurebot> (0 1 2 3 4 ...)
	19:41 < rasmusto> where's 5????
	19:41 < TEttinger> rasmusto: ##(range 6)
	19:41 < lazybot> ⇒ (0 1 2 3 4 5)
	19:41 < justin_smith> ,5
	19:41 < clojurebot> 5
	19:41 < rasmusto> :)
	19:41 < TEttinger> one of the bots trims output
	19:42 < justin_smith> we were just kind of worried about 5 and hoping 5 was OK
	19:42 < rasmusto> TEttinger: gotcha, was making a reference to ,(= (dec 5) 1) earlier
	19:42 < TEttinger> heh wha
	19:43 < justin_smith> probably the best clojure wat I have seen so far
	19:43 < amalloy> i figured out how i broke clojurebot the second time
	19:43 < justinholguin> ,(= (dec 5) 1)
	19:43 < clojurebot> false
	19:43 < justinholguin> phew
	19:43 < amalloy> i was trying to set 5 back to 5, but instead i set 2 to 5. so looking up the third item in an array caused you to look up the sixth item instead. thus all the ArrayIndexOutOfBoundsException problems
	19:44 < gtrak> how'd you do that in the first place? I didn't see it in the scroll-up
	19:44 < TEttinger> ,(def 5 4)
	19:44 < clojurebot> #<CompilerException java.lang.RuntimeException: First argument to def must be a Symbol, compiling:(NO_SOURCE_PATH:0:0)>
	19:44 < justin_smith> another good name for an intentionaly messed up lib would be angkor (named for the world's biggest Wat)
	19:44 < amalloy> gtrak: it's not hard. (let [field (nth (.getDeclaredFields Long) 3)] (.setAccessible field true) (.set field 5 2))
	19:44 < justin_smith> gtrak: amalloy pasted it, by setting some properties of Long I think
	19:44 < justinholguin> (inc justin_smith)
	19:44 < lazybot> ⇒ 34
	19:44 < justin_smith> oh, there it is again
	19:45 < gtrak> ah, ye old integer cache
	19:45 < justin_smith> probably will call for some new sandbox rules
	...
	19:45 < TEttinger> (let [field (nth (.getDeclaredFields Long) 3)] (.setAccessible field true) (.set field (rand-int 100) 101))
	19:45 < TEttinger> ,(let [field (nth (.getDeclaredFields Long) 3)] (.setAccessible field true) (.set field (rand-int 100) 101))
	19:45 < clojurebot> #<IllegalArgumentException java.lang.IllegalArgumentException: Can not set final long field java.lang.Long.value to java.lang.Integer>
	19:45 < TEttinger> aw
	19:45 < rasmusto> justin_smith: glad someone else shares my taste for puns
	19:46 < amalloy> iit turns out you have to do a bit more work to set it back
	19:46 < amalloy> (let [five (long (int (+ (int 1) (int 4)))), field (nth (.getDeclaredFields Long) 3)] (.setAccessible field true) (.set field five (int (+ (int 1) (int 4))))) ;; actually worked to recover my repl
	19:46 < rasmusto> amalloy: is this idiomatic code?
	19:47 < rasmusto> er, does it at least show how to use java interop properly?
	19:47 < TEttinger> there's nothing proper here
	19:47 < arrdem> rasmusto: do you see that crowbar? do you see it?
	19:47 < arrdem> rasmusto: that's almost as [good\|bad] as the reader macros with crowbars hack
	19:47 < amalloy> sure, just like carrying a chainsaw with you at all times is the idiomatic way to get into doors you really shouldn't be getting into
	... ... ...
	... ... ...
	... ... ...
	14:21 < amalloy> gfredericks: were you around when i broke clojurebot by redefining 5 to be 2?
	14:22 * llasram would like to see that...
	14:22 < amalloy> llasram: it's not super-hard. you use reflection to modify the .value field of (Long/valueOf 5)
	14:23 < technomancy> isn't this something we make fun of ruby for making possible?
	14:23 < technomancy> I'll have to update my Tome of Mockery
	14:25 < llasram> amalloy: Ah. Yeah, ok. IMHO using reflection to stomp on private final (I assume?) fields is basically the same as memcpy()ing
	14:25 < llasram> So I'm somewhat less impressed :-)
	14:26 < gfredericks> amalloy: no way that's nutso
	14:26 < TimMc> :-(
	14:27 < gfredericks> Tim does not like it
	14:31 < llasram> Well, unless someone re-defined TimMc's smiley face to use ":-("
	14:31 < llasram> At this point we have no real way of knowing
	14:34 < hyPiRion> It's hard to know anything when everything is relative and changing.
	14:35 < gfredericks> except the undeniable objective fact that it's hard to know anything when everything is relative and changing
	14:37 < TimMc> amalloy: How's that?
	14:37 < amalloy> how's what?
	14:37 < TimMc> I did (.setAccessible (.getDeclaredField Long "value") true) but then this errors: (.setLong (.getDeclaredField Long "value") 5 2)
	14:38 < hyPiRion> gfredericks: but then it's not hard to know anything, it's hard to know anything with the exception of that statement
	14:38 < amalloy> TimMc: setAccessible mutates the Field object you have. you can't re-get the Field object
	14:38 < TimMc> Oh! Point.
	14:38 < amalloy> (doto (.getDeclaredField Long "value") (.setAccessible true) (.setLong 5 2)) or something
	14:39 < ystael> amalloy: does that break just one five or all the fives?
	14:39 < amalloy> all the (boxed) fives
	14:39 < llasram> Oh, language
	14:39 < amalloy> and since clojure boxes everything, that's all of them
	14:39 < TimMc> ,(let [f (.getDeclaredField Long "value"), subject 5] (.setAccessible f true) (.setLong f subject 2) subject)
	14:39 < clojurebot> 2
	14:39 < TimMc> D-:
	14:39 < hyPiRion> amalloy: not all the boxed fives?
	14:39 < amalloy> TimMc: ugh, don't do it to clojurebot again
	14:39 < llasram> ,(+ 5 5)
	14:39 < clojurebot> 4
	14:39 < hyPiRion> ,(Long. "5")
	14:39 < clojurebot> 5
	14:39 < TimMc> oops
	14:40 < amalloy> hyPiRion: true. all the ones boxed via Long/valueOf
	14:40 < TimMc> I should have use a bigger 5.
	14:40 < arrdem> ,(+ 2 2)
	14:40 < clojurebot> 4
	14:40 < arrdem> ,(+ 5 5)
	14:40 < clojurebot> 4
	14:40 < ystael> is there a numerical threshold above which boxed instances are not cached and shared?
	14:40 < jcromartie> wow, nice
	14:40 < TimMc> ,(let [f (.getDeclaredField Long "value"), subject 5] (.setAccessible f true) (.setLong f subject 5) subject) ;; fixed?
	14:40 < clojurebot> 2
	14:40 < hyPiRion> ystael: 128 or 256 I think
	14:40 < llasram> ,(+ 5 5)
	14:40 < clojurebot> 4
	14:40 < amalloy> ystael: +/- 128
	14:40 < TimMc> hahaha
	14:40 < amalloy> well, -128 to 127
	14:40 < llasram> ,(let [f (.getDeclaredField Long "value"), subject 5] (.setAccessible f true) (.setLong f subject (Long. "5") subject))
	14:40 < clojurebot> #<IllegalArgumentException java.lang.IllegalArgumentException: No matching method found: setLong for class java.lang.reflect.Field>
	14:41 < arrdem> http://i0.kym-cdn.com/photos/images/newsfeed/000/613/012/2fa.gif
	14:41 < hyPiRion> TimMc: You have seen nothing yet
	14:41 < llasram> ,(let [f (.getDeclaredField Long "value"), subject 5] (.setAccessible f true) (.setLong f subject (Long. "5")) subject)
	14:41 < clojurebot> 5
	14:41 < TimMc> It restarts every 12 minutes, right?
	14:41 < llasram> ,(+ 5 5)
	14:41 < clojurebot> #<IndexOutOfBoundsException java.lang.IndexOutOfBoundsException>
	14:41 < TimMc> And by 12, I mean 12.
	14:41 < llasram> Hmm
	14:42 < amalloy> llasram: that's the way i tried to fix it the first time
	14:42 < amalloy> but you made the same mistake i did
	14:42 < amalloy> instead of defining 5 back to 5, you defined 2 to 5
	14:42 < llasram> Oh, right
	14:42 < llasram> hah!
	14:42 < hyPiRion> hahah
	14:42 < amalloy> so now like every array access breaks
	14:42 < TimMc> I <3 FORTRAN.
	14:43 < amalloy> we never did find out if clojurebot's auto-restart actually fixes this problem
	14:43 < llasram> amazing
	14:43 < amalloy> hiredman gave it a manual hard-reset before waiting for that
	14:43 < arrdem> amalloy: one way to find out...
	14:45 < TimMc> ,(let [f (.getDeclaredField Long "value"), subject 0] (.setAccessible f true) (.setLong f subject 42) subject) ;; if it will even run...
	14:45 < clojurebot> #<IndexOutOfBoundsException java.lang.IndexOutOfBoundsException>
	14:45 < TimMc> ,(.getDeclaredField Long "value")
	14:45 < clojurebot> #<IndexOutOfBoundsException java.lang.IndexOutOfBoundsException>
	14:45 < TimMc> Amazing.
	14:45 < amalloy> TimMc: right, it's too late i think. you can no longer eval anything
	14:45 < amalloy> because eval uses the number 2 in an array somewhere
	14:45 < amalloy> ,1
	14:45 < clojurebot> #<IndexOutOfBoundsException java.lang.IndexOutOfBoundsException>
	14:46 < TimMc> Got it.
	14:46 < arrdem> hahaha
	14:46 < hyPiRion> ,'1
	14:46 < clojurebot> #<IndexOutOfBoundsException java.lang.IndexOutOfBoundsException>
	14:46 < arrdem> (inc TimMc)
	14:46 < lazybot> ⇒ 58
	14:46 < TimMc> Only 28 points!
	14:46 < TimMc> I thought I had more.
	14:47 < hyPiRion> TimMc: well, since 2 is 5 and 5 is 2
	14:47 < gfredericks> wait we can prove a contradiction with this right? and then derive anything?
	14:47 < gfredericks> let's derive bugfixes
	14:47 < arrdem> gfredericks: quick prove P!=NP then we can keep our jobs forever
	14:48 < TimMc> amalloy: Do you know any way to get ahold of that interned 5?
	14:48 < ystael> TimMc: it's no longer a number ... it's a free man now
	14:49 < amalloy> TimMc: what do you mean?
	14:49 < gfredericks> arrdem: or P = NP but only for galactic algorithms
	14:49 < TimMc> Well, I'm trying to restore my REPL's notion of small numbers.
	14:49 < arrdem> gfredericks: this is also an acceptable outcome.
	14:49 < amalloy> i don't think you can do it if you've gone to the second level of broken like llasram did
	14:50 < amalloy> but if you've only broken it one level, it is reversible
	14:50 < TimMc> Just one level of fail.
	14:50 < amalloy> iirc this works: (let [five (long (int (+ (int 1) (int 4)))), field (nth (.getDeclaredFields Long) 3)] (.setAccessible field true) (.set field five (int (+ (int 1) (int 4)))))
	14:51 < TimMc> I'm having trouble figuring out what's going on here: https://www.refheap.com/85696
	14:52 < amalloy> fiddling with (Long. "5") doesn't help, because that's not the interned version
	14:52 < TimMc> Oh! Right-o.
	14:52 < llasram> Wouldn't (Long/valueOf "5") work? Or does the compiler do funny things?
	14:52 < amalloy> &(identical? (Long/valueOf "5") 5)
	14:52 < lazybot> ⇒ false
	14:52 < llasram> Oh, interesting
	14:52 < amalloy> &(identical? (Long/valueOf 5) 5)
	14:53 < lazybot> ⇒ true
	14:53 < llasram> That makes sense of course
	14:53 < amalloy> i don't know why, but apparently the string version bypasses interning
	14:53 < amalloy> llasram: does it? it should parse it as a long, then use the intern cache to get a Long
	14:53 < TimMc> valueOf probably just shortcuts, right?
	14:54 < amalloy> incidentally, three cheers for repl history: i just searched it for setAccessible to find this fix; i probably wouldn't have remembered it on my own
	14:55 < llasram> amalloy: Eh. It's not really clear to me why it should use the cache for the result of string parsing. Not saying it shouldn't -- it just doesn't seem unreasonable for it to not
	14:55 < hyPiRion> TimMc: Oh hey, if you find this sort of stuff funny, you should see this one: https://www.refheap.com/85697
	14:56 < hyPiRion> That one segfaults or makes your JVM spin up an infinite loop
	14:57 < amalloy> hyPiRion: do you really have to remove its FINAL modifier? once you make it accessible i think you can already write to it
	14:57 < amalloy> just like for Long
	14:57 < amalloy> maybe you're doing something weird i don't see there
	14:59 < hyPiRion> amalloy: it's static, which makes it a bit different IIRC
	14:59 < hyPiRion> at least I just checked, and it doesn't work with just .setAccessible
	15:00 < TimMc> ,(+()()()()(*))
	15:00 < clojurebot> 5
	15:00 < TimMc> It's back!
	15:00 < llasram> ,(+ 5 5)
	15:01 < clojurebot> 10
	15:01 < llasram> Yay!
	15:07 < TimMc> amalloy: I was able to fix it in my REPL by using "subject (+ 2 3)", but when I tried again with "subject 5" nREPL quit with an NPE. :-P
	...
	15:10 < amalloy> TimMc: yeah, that's setting 2 to 5 again
	15:10 < amalloy> which really is a lot more disruptive than setting 5 to 2
	... ... ...
	... ... ...
	... ... ...
	13:48 < justin_smith> ,(let [field (nth (.getDeclaredFields Long) 3)] (.setAccessible field true) (.set field 5 2))
	13:48 < clojurebot> nil
	13:48 < justin_smith> ,(inc 5)
	13:48 < clojurebot> 3
	13:48 < andyf> You can break the JVM's 5, but Plato's 5 is safe in the world of forms
	13:48 < justin_smith> haha
	13:48 < gfredericks> ~5 is the set of all sets with 5 elements
	13:49 < clojurebot> Ik begrijp
	13:49 < bbloom_> , (.getDeclaredFields Long) 3)
	13:49 < clojurebot> #<Field[] [Ljava.lang.reflect.Field;@5753b0>
	13:49 < bbloom_> ,(.getName (.getDeclaredFields Long) 3))
	13:49 < clojurebot> #<IllegalArgumentException java.lang.IllegalArgumentException: No matching method found: getName for class [Ljava.lang.reflect.Field;>
	13:49 < bbloom_> ,(.name (.getDeclaredFields Long) 3))
	13:49 < clojurebot> #<IllegalArgumentException java.lang.IllegalArgumentException: No matching method found: name for class [Ljava.lang.reflect.Field;>
	13:49 < bbloom_> ,(clojure.reflect/reflect (.getDeclaredFields Long) 3))
	13:49 < clojurebot> #<ClassNotFoundException java.lang.ClassNotFoundException: clojure.reflect>
	13:49 < bbloom_> ,(require 'clojure.reflect)
	13:49 < clojurebot> nil
	13:49 < bbloom_> eh wahtever... how do i get the name of a field?
	13:49 < gfredericks> .getName?
	13:49 < bbloom_> i tried that
	13:49 < bbloom_> oh dur it's an array
	13:49 < gfredericks> with 3?
	13:50 < bbloom_> ,(.name (nth (.getDeclaredFields Long) 3))
	13:50 < clojurebot> #<IllegalArgumentException java.lang.IllegalArgumentException: No matching field found: name for class java.lang.reflect.Field>
	13:50 < bbloom_> i suck
	13:50 < bbloom_> ,(.getName (nth (.getDeclaredFields Long) 3))
	13:50 < clojurebot> "value"
	13:50 < andyf> The world of forms: Bringing you real immutability since 348 BC
	13:50 < bbloom_> ah ok i gotcha. boxed small numbers are interned
	13:51 < justin_smith> , (nth (.getDeclaredFields Long) 3)
	13:51 < clojurebot> #<Field private final long java.lang.Long.value>
	13:51 < justin_smith> ,(.getName (nth (.getDeclaredFields Long) 3))
	13:51 < clojurebot> "value"
	13:51 < bbloom_> andyf: https://en.wikipedia.org/wiki/348_BC <- i was hoping you knew the actual year
	13:51 < justin_smith> bbloom_: ^
	13:51 < bbloom_> justin_smith: i got there eventually :-P
	13:52 < justin_smith> bbloom_: aha, I missed it while I was finding it in my own repl
	13:52 < andyf> Sorry, I don't know when he wrote about it first, so I approximated by the year Plato died.
	13:53 < justin_smith> ,(= 5 3)
	13:53 < clojurebot> false
	13:53 < justin_smith> ,(= 5 2)
	13:53 < clojurebot> true
	13:53 < justin_smith> ,(let [field (nth (.getDeclaredFields Long) 3)] (.setAccessible field true) (.set field 5 (inc 4)))
	13:53 < clojurebot> nil
	13:53 < justin_smith> ,(= 5 2)
	13:53 < clojurebot> true
	13:53 < justin_smith> I thought I fixed that...
	13:54 < justin_smith> ,(let [field (nth (.getDeclaredFields Long) 3)] (.setAccessible field true) (.set field (inc 4) (inc 4)))
	13:54 < clojurebot> nil
	13:54 < justin_smith> ,(= 5 2)
	13:54 < clojurebot> true
	13:54 < justin_smith> ,(inc 2)
	13:54 < clojurebot> 3
	13:54 < justin_smith> ,(inc 5)
	13:54 < clojurebot> 3
	13:54 < mearnsh> this is why we can't have nice things
	13:54 < AeroNotix> ,(inc (+ 3 2))
	13:54 < clojurebot> 6
	13:54 < justin_smith> :( I forgot how to fix it
	13:54 < AeroNotix> ,(inc 5)
	13:54 < clojurebot> 3
	13:54 < AeroNotix> 5
	13:55 < AeroNotix> ,5
	13:55 < clojurebot> 2
	13:55 < justin_smith> ,(let [five (long (int (+ (int 1) (int 4)))), field (nth (.getDeclaredFields Long) 3)] (.setAccessible field true) (.set field five (int (+ (int 1) (int 4)))))
	13:55 < clojurebot> nil
	13:55 < andyf> Once you make 5 equal to something else, the JVM extrapolates and is able to prove that all numbers are simultaneously equal to, and not equal to, all others.