1. split function
Hi all,
I needed and wrote a split function, like the Perl split function. It
doesn't offer regex pattern matching, like the Perl split. I was going to
post it to the archive, but wanted to submit it here for review first. The
code I'll be submitting is just the function 'split below'. All of the
rest of it is just to make reviewing and testing it easier. Please post
any bugs with it here, I'll fix it, then submit it to the archive.
Usage:
object result
result = split(",","word1,word2,word3,etc") -- case insensitive
result = split(",","word1,word2,word3,etc",1) -- any val as 3rd arg makes
it case-sensitive
result = split("puppy","kittenpuppyduckpuppydogpuppygorillapuppyfish")
result = split("PUPPY","kittenpuppyduckpuppydogpuppygorillapuppyfish",1)
Thanks,
Ted
----cut here----
include get.e
include wildcard.e
--
-- empty values are returned too, e.g. given{".","...something..."} returns
-- {{},{},{},{something},{},{},{}} because the nothing in front of
-- the first "." and after the last "." is an empty sequence
sequence d,s,t,u,result,origs
if length(myargs) > 3 or length(myargs) < 2 then
return -1
end if
d = myargs[1]
s = myargs[2]
if length(myargs) = 3 then
ignorecase=0
else
ignorecase=1
end if
-- handle special cases...
-- d or s is empty
lengthofd=length(d)
if lengthofd = 0 or length(s) = 0 then
return s
end if
-- d and s are the same
if ignorecase=1 then
if equal(lower(s),lower(d)) then
return {{},{}}
end if
elsif equal(s,d) then
return {{},{}}
end if
result = {}
origs=s
while 1 do
if ignorecase then
i = match(lower(d),lower(s))
else
i = match(d,s)
end if
if i > 0 then
if (i+lengthofd) <= length(s) then
t = s[i+lengthofd..length(s)]
elsif i+lengthofd-1 = length(s) then
result = append(result,s[1..i-1])
exit
end if
if i!= 1 then
u = s[1..i-1]
result = append(result,u)
else
result = append(result,{})
end if
s = t
else
result = append(result,s)
exit
end if
end while
-- account for last portion of s=d
if ignorecase then
if lengthofd = 1 then
if match(lower(d),lower(sprintf("%s",{origs[length(origs)]})))
then
result=append(result,{})
end if
else
if
match(lower(d),lower(origs[length(origs)-lengthofd+1..length(origs)])) then
result=append(result,{})
end if
end if
else
if lengthofd = 1 then
if match(d,sprintf("%s",{origs[length(origs)]})) then
result=append(result,{})
end if
else
if match(d,origs[length(origs)-lengthofd+1..length(origs)]) then
result=append(result,{})
end if
end if
end if
return result
end function
procedure show_results(sequence t)
sequence d,s
integer i
object res
d=t[1]
s=t[2]
res = split(t)
if not atom(res) then
if length(t) = 2 then
puts(1,"Results of splitting {" & s & "} with {" & d & "}:\n")
else
puts(1,"Results of case-sensitive splitting {" & s & "} with {"
& d & "}:\n")
end if
if length(res)=0 then
puts(1,"{}\n")
else
for c=1 to length(res) do
puts(1,"result[" & sprintf("%d",{c}) & "] is ")
if length(res[c]) = 0 then
puts(1,"{}\n")
else
puts(1, "{" & res[c] & "}\n")
end if
end for
end if
else
puts(1,"The split function did not understand the data you gave
it.\n")
end if
i = wait_key()
end procedure
sequence tests
tests = {
{".","141.140.10.254"},
{".","141.140.10.254."},
{".",".141.140.10.254."},
{"pony","141.140.65.78.0.PONY"},
{".","."},
{"delimiter","field1delimiterfield2delimiterfield3delimiterfield4"},
{"DELIMITER","field1delimiterfield2delimiterfield3delimiterfield4",1},
{"delimiter","field1delimiterfield2delimiterfield3delimiterfield4delimiter"
},
{"delimiter","delimiterfield1delimiterfield2delimiterfield3delimiterfield4"
},
{{},{}},
{"pony","pony10987654321"},
{"pony","badexample","too","many","fields"}
}
for c = 1 to length(tests) do
show_results(tests[c])
end for
2. Re: split function
> I needed and wrote a split function, like the Perl split function.
Ted, check out kat's string token library...
http://www.pell.net/warning/ai/strtok-v2.zip. It does everything plus some.
I don't know if it will meet your needs, but at least you won't have to
reinvent the wheel or resort to cliches.
3. Re: split function
Thanks. Well, guess I already reinvented the wheel. But my wheel isn't
quite as round as Kat's, from the sound of it. Oh well, got a chance to
work on my carpal tunnel syndrome, anyway...
--On Wednesday, September 17, 2003 10:04 AM -0500 "C. K. Lester"
<cklester at yahoo.com> wrote:
>
>
>> I needed and wrote a split function, like the Perl split function.
>
> Ted, check out kat's string token library...
> http://www.pell.net/warning/ai/strtok-v2.zip. It does everything plus
> some. I don't know if it will meet your needs, but at least you won't
> have to reinvent the wheel or resort to cliches.
>
>
>
> TOPICA - Start your own email discussion group. FREE!
>
>
4. Re: split function
> Ted, check out kat's string token library...
> http://www.pell.net/warning/ai/strtok-v2.zip. It does everything plus
some.
> I don't know if it will meet your needs, but at least you won't have to
> reinvent the wheel or resort to cliches.
That link doesn't work, but I did find it on her site, so go there and check
it out. :)
