Thomas Bätzler schreef:
> [ !/^\.$|^\.\.$/ and !/^\.\.?$/ ]
> I'm assuming that this regex would be faster, too, since it
> does not contain an alternation.
That is hard to say without actual benchmarking: the regex-optimizer
gets better with every new version of perl.
Check also the re=debug output:
$ perl -Mre=debug -e '/^\.\.?$/'
Freeing REx: `","'
Compiling REx `^\.\.?$'
size 9 Got 76 bytes for offset annotations.
first at 2
1: BOL(2)
2: EXACT <.>(4)
4: CURLY {0,1}(8)
6: EXACT <.>(0)
8: EOL(9)
9: END(0)
anchored `.' at 0 floating `'$ at 1..2 (checking anchored) anchored(BOL)
minlen 1
Offsets: [9]
1[1] 2[2] 0[0] 6[1] 0[0] 4[2] 0[0] 7[1] 8[0]
Freeing REx: `"^\\.\\.?$"'
Note that there is no EOL-test right after the first EXACT <.>.
$ perl -Mre=debug -e '/^\.$|^\.\.$/'
Freeing REx: `","'
Compiling REx `^\.$|^\.\.$'
size 11 Got 92 bytes for offset annotations.
1: BRANCH(6)
2: BOL(3)
3: EXACT <.>(5)
5: EOL(11)
6: BRANCH(11)
7: BOL(8)
8: EXACT <..>(10)
10: EOL(11)
11: END(0)
minlen 1
Offsets: [11]
0[0] 1[1] 2[2] 0[0] 4[1] 5[1] 6[1] 7[4] 0[0] 11[1] 12[0]
Freeing REx: `"^\\.$|^\\.\\.$"'
Note that in this case it restarts at BOL if ^\.$ was not matched, so it
rechecks the dot at the start.
To see more of the workings of the re:
$ echo '..' |perl -Mre=debug -wnle '/^\.\.?$/ and print "OK"'
Variants, from nice to very ugly:
!/^\.\.?$/
!/^\.?\.$/
!/^(?:\.|\.\.)$/
!/^\.$|^\.\.$/
!/(?<=^\.)\.?$/
/^[^.]|[^.]$|.../
/...|^[^.]|[^.]$/
/.{3}|^[^.]|[^.]$/
--
Affijn, Ruud
"Gewoon is een tijger."
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
<http://learn.perl.org/> <http://learn.perl.org/first-response>
