Re: Memory efficiency with arrays

Says tye, apparently reporting from an alternate dimension:

In theory, pop()ing would be more efficient than shift()ing since every so often the array will be moved down so that the gap in front of its memory space is not huge.

Since Perl never does move the array down when you do shift, I don't see the purpose of this theory. You could exchange pop and shift in your theory and it would make as much sense and be as true.

Comment on Re: Memory efficiency with arrays

Replies are listed 'Best First'.
(tye)Re: Memory efficiency with arrays by tye (Sage) on Jan 05, 2001 at 10:51 UTC
I guess that goes along with Perl's general theory of always choose faster over less memory. (: I think Perl will move the array down in some situations, but I'll take your word for it that it doesn't do it in this specific situation (I'm now guessing that Perl will wait until you try to push() passed the end of the memory space -- but that is a guess). tilly notes that the shift() might have more accounting to do and so might be slightly slower. By the way, does the same apply to shifting of strings? I recall my playing with this showing that Perl would shift the string, but perhaps again that was only when I appended past the end of the string's memory buffer (in fact, that sounds familiar now that I think back). (Bonus points for anyone who can detect this in straight Perl -- I cheated in an interesting and dangerous way.) - tye (but my friends call me "Tye")	[reply]

Replies are listed 'Best First'.

(tye)Re: Memory efficiency with arrays
by tye (Sage) on Jan 05, 2001 at 10:51 UTC

I guess that goes along with Perl's general theory of always choose faster over less memory. (: I think Perl will move the array down in some situations, but I'll take your word for it that it doesn't do it in this specific situation (I'm now guessing that Perl will wait until you try to push() passed the end of the memory space -- but that is a guess).

tilly notes that the shift() might have more accounting to do and so might be slightly slower.

By the way, does the same apply to shifting of strings? I recall my playing with this showing that Perl would shift the string, but perhaps again that was only when I appended past the end of the string's memory buffer (in fact, that sounds familiar now that I think back). (Bonus points for anyone who can detect this in straight Perl -- I cheated in an interesting and dangerous way.)

tye

[reply]