TinyCode competition - trial

Miscellaneous Topics
#61

[quote=“Riven,post:56,topic:30806”]
Good luck - but hopefully you won’t need it!

#62

Can anyone tell me why my read pointer is not advancing?


for(int a=in.read(),b=in.read();(a|b)>0;){
    out.write(b);
    out.write(a);
}

Whenever I run this, it just goes into an infinite loop. When I debug it, a and b keep reading the first 2 characters from the stream.

#63

Consider: for(;;)…

is only executed ONCE and the beginning of the loop to initialize variables etc…

#64

Ouch… :wink:

Two new assignments:
http://213.247.55.3:8484/compileandrun/?assignment=capitalizer
http://213.247.55.3:8484/compileandrun/?assignment=barcode

I made the barcode assignment extra hard by not adding a space at the end.

#65

85 bytes for capitalizer, but that has to go shorter…
there are 3 variables in there…
who needs 3 variables anyway? :slight_smile:

for(int c,i=0,s=0; (c=in.read())>0 ; out.write(c-(i++==s & c > 96 & c < 123?32:0)))
      s =c == 32 ? i+1 : s;

for the barcode thingy 89 bytes

int i,c;
for(i=48;(c=in.read())>0;i=c==32?48:i+1)
   if(c==32)out.write(i);
if(i>0)out.write(i);

So lets start that shortening iteration again :slight_smile:

Edit:
capitalizer down to 7170:

for(int c,s=32; (c=in.read())>0;)
   out.write(s=c-(s==32 & c > 96 & c < 123?32:0));

barcode down to 66 bytes:

for(int c=1,i=48;c>0;i++){
  c=in.read();
  if(c<99){
    out.write(i);
    i=47;
  }
}
#66

Capitalizer:


for(int c,s=32; (c=in.read())>0;)
out.write(s=s==c/96*32?c-s:c);

(61 bytes)

This was optimisation on the logic-level.
It didn’t have anything to do with reasoning.

…

I have yet to figure out how it works…

Thanks markush, for the jump-start.

#67

We need another operator eh?


s=s*3          --->>> s*=3;
s=s==3?4:5;    --->>> s===3?4:5;
#68

The above code works with the test case, but IMO doesn’t solve the spirit of the problem in all cases. :slight_smile:

Try it with a test case that includes the end conditions: e.g.

"@ 2pm { today }, I saw a bear at the zoo "

The above code would change the @ { } chars.

Here’s my current go @ Capitalizer:


for(int c, p = 32; (c = in.read()) > 0; )
 out.write( p = p == 32 & c > 96 & c < 123 ? c - p : c );

(67 bytes)

Cheers, Tim.

#69

Yeah, I cheated. I removed the upper-bound check.

#70

Yes… and the lower bound check is wrong… should be /97

#71

I’ll cry myself to sleep tonight. :-[

#72

BinaryShrink


for( int v = 1; (v += v + in.read() % 2) > 1; )
  if( v > 255 )
  {
    out.write( v );
    v = 1;
  }

(61 bytes)

#73

doh :-[

Thanks.

#74

67 bytes:

out.write(new byte[] { 50,51,56,49,52,57,54,50,51,49,49,50,48,49,51 });

:wink:

#75

I’ll add a bunch of testcases to each assignment, when I have time for it.

#76

hey Riven, I killed your server ;D


com.sun.org.apache.regexp.internal.recompile.main("".split(""));

(64 bytes)

#77

;D

#78

WHAAAA

It was worse than you might every have thought…

There was a piece of user-submitted sourcecode processed by BASH… :o :o :o

That’s like total control…

http://www.indiespot.net/hacked.png

#79

I’m not going to launch the http-server until I fixed this…

Hm… there is a System.exit(0) inside that method…

I really can’t understand how that System.out (!!!) could end up in bash.

I mean… after System.exit(0) is called, the JVM is supposed to be dead, and everything that was buffered in System.out should be discarded, not writting into STDIN…

Anybody?

#80

yeah, it was the System.exit(0); that I was exploiting to terminate the vm…
Took me ages to think of a way to pass in a String[], to get the desired effect :slight_smile:

What was it that was written to stdin?
the socket input stream? or system.out?

Can you reproduce it?

Seems like a serious vm bug to me, it isn’t like my code did anything that a normal java service couldn’t do as the result of some kind of unrecoverable error.
Presumably it’s the result of a thread race condition, as I guess our provided code is executing on it’s own thread?