PARI's commands - Page 212

EdH · 2011-10-28, 13:33

Quote:

Originally Posted by CRGreathouse

GP doesn't have much in the way of I/O. I suggest

Code:

extern("rm oldStatus; mv pariStatus oldStatus");
write("pariStatus", ...);
extern("rm oldStatus");

Thanks. That would keep me from possibly losing the value. Is the first "rm oldStatus" just to "be sure" before the mv?

EdH · 2011-10-28, 14:16

Quote:

Originally Posted by science_man_88

yeah probably but you should still find it with mine:

Code:

(10:29)>for(x=1000010274063430,1000010274063430,b=x;for(y=1,10000000,if((sigma(b)-b) != 0,b=sigma(b)-b);if(b==x,print(x","y);break())))
1000010274063430,2

In computer terms, yours seems to be taking an eternity to search from 1000010274060000 to 1000010274080000:

Code:

? for(x=1000010274060000,1000010274080000,b=x;for(y=1,10000000,if((sigma(b)-b) != 0,b=sigma(b)-b);if(b==x,print(x","y);break())))
^C  *** sigma: user interrupt after 7mn, 32,978 ms.

whereas, mine (without the extra prints) is done in a few seconds:

Code:

? for(b=100001027406,100001027408,c=b*10000;for(d=c,c+10000,i=d;f=d*20;for(e=1,5,i=sigma(i)-i;if(i<d,break());if(i>f,break());if(i==d,print(d);break()))))
1000010274063430
time = 8,999 ms.

science_man_88 · 2011-11-08, 15:47

Code:

lucas2(t,z,p,q,x)= v=[t,z];for(y=3,x,v=concat(v,p*v[y-1]-q*v[y-2]));v=vector(x,i,v[i])

though I tried using it like:

Code:

for(z=-1000,1000,for(a=-1000,1000,if(lucas2(2,3,z,a,13)==[2,3,5,7,13,17,19,31,61,89,107,127,521],print(z","a))))

and got told "bug in" I'm guessing something like the segmentation fault: please report errors.

science_man_88 · 2011-11-27, 00:35

how can it be done in PARI because if possible it may allow me to test something , I'm currently working on related to Mersenne prime exponents. because laurv talked of code like:

Code:

 sqrt(Mod(2,2^61-1))

well so far in testing manually I see a pattern ( gulp) to the result and I might be able to tell just by the first calculation and this kinda had to do with trying the lucas-lehmer test backwards. I think that if p is a mersenne prime exponent the result isn't and error on the second level down by adding 2 ( I am positive it's likely known though).

science_man_88 · 2011-11-27, 00:51

Quote:

Originally Posted by science_man_88

how can it be done in PARI because if possible it may allow me to test something , I'm currently working on related to Mersenne prime exponents. because laurv talked of code like:

Code:

 sqrt(Mod(2,2^61-1))

well so far in testing manually I see a pattern ( gulp) to the result and I might be able to tell just by the first calculation and this kinda had to do with trying the lucas-lehmer test backwards. I think that if p is a mersenne prime exponent the result isn't and error on the second level down by adding 2 ( I am positive it's likely known though).

doh I found it I think lol

trap() still not sure how to use it.

CRGreathouse · 2011-11-27, 21:26

The general format is
trap(, code, what to do if the code fails)
as in

Code:

trap(,1/0,"don't divide by 0")

Technically you can put the error type before the first comma to catch only certain types of errors, but usually you don't want to do this.

science_man_88 · 2011-11-27, 21:35

Quote:

Originally Posted by CRGreathouse

The general format is
trap(, code, what to do if the code fails)
as in

Code:

trap(,1/0,"don't divide by 0")

Technically you can put the error type before the first comma to catch only certain types of errors, but usually you don't want to do this.

thanks I already proved myself wrong without it but I still want to see if it even semi fits:

Code:

(17:34)>b=2;forprime(p=2281,2281,for(y=1,p-1,b=sqrt(Mod(b,1<<p-1));trap(,break(),);if(y==p-1,print(p)));b=2)
  *** sqrt: user interrupt after 13,750 ms.
  *** break: bug in PARI/GP (Segmentation Fault), please report
  ***   Break loop (type 'break' or Control-d to go back to GP)

science_man_88 · 2011-11-27, 21:37

Quote:

Originally Posted by science_man_88

thanks I already proved myself wrong without it but I still want to see if it even semi fits:

Code:

(17:34)>b=2;forprime(p=2281,2281,for(y=1,p-1,b=sqrt(Mod(b,1<<p-1));trap(,break(),);if(y==p-1,print(p)));b=2)
  *** sqrt: user interrupt after 13,750 ms.
  *** break: bug in PARI/GP (Segmentation Fault), please report
  ***   Break loop (type 'break' or Control-d to go back to GP)

Code:

(17:37)>b=2;forprime(p=29,29,for(y=1,p-1,b=sqrt(Mod(b,1<<p-1));if(y==p-1,print(p)));b=2)
  *** sqrt: non quadratic residue in gsqrt
  *** break: bug in PARI/GP (Segmentation Fault), please report
  ***   Break loop (type 'break' or Control-d to go back to GP)

science_man_88 · 2011-11-27, 21:40

hopefully I don't have to reinstall ( though likely) I've had segmentation faults too much ( stupidly ignored a little) and even my timer seems wacky at times.

CRGreathouse · 2011-11-27, 22:28

You're using trap wrong. You're writing

code; trap(, stuff, )

when I told you to write

trap(, code, stuff)

What you did will still cause the error.

Quote:

Originally Posted by science_man_88

hopefully I don't have to reinstall

No, you don't need to do that.

LaurV · 2011-11-28, 03:10

I don't know what are you trying to do (semantically), but (syntactically) you should write this:

Code:

(10:06:27) gp > b=2;forprime(p=2,2281,for(y=1,p-1,trap(,print("break here, p="p", y="y);break,b=sqrt(Mod(b,1<<p-1)));if(y==p-1,print(p)));b=2)
break here, p=2, y=1
break here, p=3, y=2
break here, p=5, y=3
7
break here, p=11, y=6
break here, p=13, y=7
break here, p=17, y=5
break here, p=19, y=10
23
break here, p=29, y=15
31
break here, p=37, y=19
break here, p=41, y=11
break here, p=43, y=8
47
break here, p=53, y=27

2011-11-08, 15:47	#2324
science_man_88 "Forget I exist" Jul 2009 Dumbassville 2⁶·131 Posts	Code: lucas2(t,z,p,q,x)= v=[t,z];for(y=3,x,v=concat(v,pv[y-1]-qv[y-2]));v=vector(x,i,v[i]) though I tried using it like: Code: for(z=-1000,1000,for(a=-1000,1000,if(lucas2(2,3,z,a,13)==[2,3,5,7,13,17,19,31,61,89,107,127,521],print(z","a)))) and got told "bug in" I'm guessing something like the segmentation fault: please report errors. Last fiddled with by science_man_88 on 2011-11-08 at 15:48

2011-11-27, 00:35	#2325
science_man_88 "Forget I exist" Jul 2009 Dumbassville 2⁶·131 Posts	error testing how can it be done in PARI because if possible it may allow me to test something , I'm currently working on related to Mersenne prime exponents. because laurv talked of code like: Code: sqrt(Mod(2,2^61-1)) well so far in testing manually I see a pattern ( gulp) to the result and I might be able to tell just by the first calculation and this kinda had to do with trying the lucas-lehmer test backwards. I think that if p is a mersenne prime exponent the result isn't and error on the second level down by adding 2 ( I am positive it's likely known though). Last fiddled with by science_man_88 on 2011-11-27 at 00:38

2011-11-27, 21:26	#2327
CRGreathouse Aug 2006 3×1,993 Posts	The general format is trap(, code, what to do if the code fails) as in Code: trap(,1/0,"don't divide by 0") Technically you can put the error type before the first comma to catch only certain types of errors, but usually you don't want to do this.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Why do I sometimes see all the <> formatting commands when I quote or edit?	cheesehead	Forum Feedback	3	2013-05-25 12:56
Passing commands to PARI on Windows	James Heinrich	Software	2	2012-05-13 19:19
Ubiquity commands	Mini-Geek	Aliquot Sequences	1	2009-09-22 19:33
64-bit Pari?	CRGreathouse	Software	2	2009-03-13 04:22
Are these commands correct?	jasong	Linux	2	2007-10-18 23:40

2011-11-27, 21:40	#2330
science_man_88 "Forget I exist" Jul 2009 Dumbassville 20C0₁₆ Posts	hopefully I don't have to reinstall ( though likely) I've had segmentation faults too much ( stupidly ignored a little) and even my timer seems wacky at times.