Re: Re: suffix arrays

[BrowserUk]

in reply to Re: suffix arrays
in thread suffix arrays

Unfortunately, as of 5.8.0, that still doesn't work properly.

 perl> print join "\n", map{ \substr 'The quick brown fox', $_ } 0 .. 
+19
LVALUE(0x15d7cf4)
LVALUE(0x15d7cf4)
LVALUE(0x15d7cf4)
LVALUE(0x15d7cf4)
LVALUE(0x15d7cf4)
LVALUE(0x15d7cf4)
LVALUE(0x15d7cf4)
LVALUE(0x15d7cf4)
LVALUE(0x15d7cf4)
LVALUE(0x15d7cf4)
LVALUE(0x15d7cf4)
LVALUE(0x15d7cf4)
LVALUE(0x15d7cf4)
LVALUE(0x15d7cf4)
LVALUE(0x15d7cf4)
LVALUE(0x15d7cf4)
LVALUE(0x15d7cf4)
LVALUE(0x15d7cf4)
LVALUE(0x15d7cf4)
LVALUE(0x15d7cf4)

As you can see, each time you take an Lvalue from substr, it re-uses the same address. So you end up with and array of pointers to the last char in the string.

---

Examine what is said, not who speaks.

"Efficiency is intelligent laziness." -David Dunham
"When I'm working on a problem, I never think about beauty. I think only how to solve the problem. But when I have finished, if the solution is not beautiful, I know it is wrong." -Richard Buckminster Fuller

[Zaxo]

Hmmm... In 5.8.1-RC2,

perl -e'my $foo = "A string"; my @foo = map{\substr $foo, $_} 0..7; pr
+int map {$$_,$/} @foo;'
A string
 string
string
tring
ring
ing
ng
g
$ 

5.8.1 does more of what I mean!

After Compline,
Zaxo

[BrowserUk]

Ah! I only just got my 5.8.1 to build, so I haven't had chance to try it out there yet. Someone did mention that the p5p guys were working on a fix when this came up before, but that was way back before 5.8 was released and it was one of the first things I checked for in 5.8 but was disappointed. Once I've convinced myself that my build of 5.8.1 is reasonably safe to use, that will allow me to finish off a module that was almost ready for release 6 months ago.

It appears the fix just made it into the wild. Thanks, its nice to know, and my apologies for casting aspersions.

---

Examine what is said, not who speaks.
"Efficiency is intelligent laziness." -David Dunham
"When I'm working on a problem, I never think about beauty. I think only how to solve the problem. But when I have finished, if the solution is not beautiful, I know it is wrong." -Richard Buckminster Fuller