Hi, I’m new to the list and I hope this is the right place to ask the
question, but I recently started learning ruby and I was writing a short
little program to test if a word was made up of symbols from the
periodic
table. I figured I’d use ruby’s built-in regex support, so I ended up
writing a regex that looks something like this:
(element1|element2|element3…|elementx)+
Basically it’s just got all 115 or so of the elements in there, and
it’s
testing for one or more, right? Unfortunately this doesn’t seem to work
properly as a (nonsense) word like “presenti” should be periodic
according
to my list of elements, [p][re][se][n][ti]. Now, I can duplicate this
behavior with the regex coach oddly enough. The problem seems to be that
it
parses [p][re][s] instead of [se]. I’m wondering if anyone can explain
this
behavior to me, as I’m new to both regexps and ruby and the only regex
experience I’ve had previous to this is in some theory classes. Thanks.
The
full expression is below.
(He|Li|Be|C|O|Ne|Na|Mg|Al|Si|S|Cl|Ar|Ca|Sc|Ti|Cr|Mn|Fe|Co|Ni|Cu|Zn|Ga|Ge|As|Se|Br|Kr|Rb|Sr|Zr|Nb|Mo|Tc|Ru|Rh|Pd|Ag|Cd|In|Sn|Sb|Te|Xe|Cs|Ba|Hf|Ta|Re|Os|Ir|Pt|Au|Hg|Tl|Pb|Bi|Po|At|Rn|Fr|Ra|Rf|Db|Sg|Bh|Hs|Mt|Uun|Uuu|Uub|Uuq|Uuh|Uuo|Ds|La|Ce|Pr|Nd|Pm|Sm|Eu|Gd|Tb|Dy|Ho|Er|Tm|Yb|Lu|Ac|Th|Pa|Np|Pu|Am|Cm|Bk|Cf|Es|Fm|Md|No|Lr|H|B|N|F|P|V|Y|I|W|Uaal)+
George L. [email protected] writes:
behavior with the regex coach oddly enough. The problem seems to be that it
parses [p][re][s] instead of [se]. I’m wondering if anyone can explain this
behavior to me, as I’m new to both regexps and ruby and the only regex
experience I’ve had previous to this is in some theory classes. Thanks. The
full expression is below.
(He|Li|Be|C|O|Ne|Na|Mg|Al|Si|S|Cl|Ar|Ca|Sc|Ti|Cr|Mn|Fe|Co|Ni|Cu|Zn|Ga|Ge|As|Se|Br|Kr|Rb|Sr|Zr|Nb|Mo|Tc|Ru|Rh|Pd|Ag|Cd|In|Sn|Sb|Te|Xe|Cs|Ba|Hf|Ta|Re|Os|Ir|Pt|Au|Hg|Tl|Pb|Bi|Po|At|Rn|Fr|Ra|Rf|Db|Sg|Bh|Hs|Mt|Uun|Uuu|Uub|Uuq|Uuh|Uuo|Ds|La|Ce|Pr|Nd|Pm|Sm|Eu|Gd|Tb|Dy|Ho|Er|Tm|Yb|Lu|Ac|Th|Pa|Np|Pu|Am|Cm|Bk|Cf|Es|Fm|Md|No|Lr|H|B|N|F|P|V|Y|I|W|Uaal)+
Are you matching case insensitive (/i)?
By the way, an optimized version of that regexp is
/(A[cglmr-u]|B[aehikr]|C[adeflmorsu]|D[bsy]|E[rsu]|F[emr]|G[ade]|H[efgos]|I[nr]|Kr|L[airu]|M[dgnot]|N[abdeiop]|Os|P[abdmortu]|R[abefhnu]|S[bcegimnr]|T[abcehilm]|U(?:aal|u[bhnoqu])|Xe|Yb|Z[nr]|[BCFHINOPSVWY])+/
(Thanks to emacs’s regexp-opt)
hi,
i’m not entirely sure, but i’d say it’s because S appears before Se (in
your
regexp), so you’d have to start with all elements that have two
letters… i
think that will work, not sure though…
greetings, Dirk.
2005/11/21, George L. [email protected]:
I am matching +i, sorry for leaving that out. I copied the expression
from
the regex coach, which has that handy checkbox. Thanks for the optimized
info for the future. Ruby/Regex coach doesn’t seem to like the optimized
version though either…
I’ve thought of that too, but I thought that in theory, order shouldn’t
matter. Now, I know there are differences in theory and practicality,
but
even if I match all of the two-letter expressions first, what about a
string
like agau, where maybe it will match ga first, and leave A and U out?
And
even if I order alphabetically, there may be a case where it matches
something else out-of-order in another word, etc…
It is matching [pr] and [es]
George L. [email protected] writes:
behavior with the regex coach oddly enough. The problem seems to be that
it
parses [p][re][s] instead of [se]. I’m wondering if anyone can explain
this
behavior to me, as I’m new to both regexps and ruby and the only regex
experience I’ve had previous to this is in some theory classes. Thanks.
The
full expression is below.
(He|Li|Be|C|O|Ne|Na|Mg|Al|Si|S|Cl|Ar|Ca|Sc|Ti|Cr|Mn|Fe|Co|Ni|Cu|Zn|Ga|Ge|As|Se|Br|Kr|Rb|Sr|Zr|Nb|Mo|Tc|Ru|Rh|Pd|Ag|Cd|In|Sn|Sb|Te|Xe|Cs|Ba|Hf|Ta|Re|Os|Ir|Pt|Au|Hg|Tl|Pb|Bi|Po|At|Rn|Fr|Ra|Rf|Db|Sg|Bh|Hs|Mt|Uun|Uuu|Uub|Uuq|Uuh|Uuo|Ds|La|Ce|Pr|Nd|Pm|Sm|Eu|Gd|Tb|Dy|Ho|Er|Tm|Yb|Lu|Ac|Th|Pa|Np|Pu|Am|Cm|Bk|Cf|Es|Fm|Md|No|Lr|H|B|N|F|P|V|Y|I|W|Uaal)+
You’ll note if you match against /^(He|…|Uaal)$/i it matches the full
string “presenti”. Read perlretut for more info on how regexps work.
Basically, it is greedy, but it won’t backtrack unless it has failed to
find
a match. With your example, the S matches, then the following characters
don’t match anything, but it has already succeeded, so it doesn’t bother
trying Se.
http://www.cs.rit.edu/~afb/20013/plc/perl5/doc/perlretut.html#grouping
things and hierarchical matching
Cheers,
Dave
Right, but technically, say for “prese” that string is periodic, and
should
be matched as [p][re][se] but it’s not getting that last e.
well, not that it’s perfect, but I would do something like:
elements = %w( He LI Be C O Ne Na Mg Al Si S Cl Ar Ca Sc Ti Cr Mn Fe Co
Ni
Cu Zn Ga Ge As Se Br Kr Rb Sr Zr Nb Mo Tc Ru Rh Pd Ag Cd In Sn Sb Te Xe
Cs
Ba Hf Ta Re Os Ir Pt Au Hg Tl Pb Bi Po At Rn Fr Ra Rf Db Sg Bh Hs Mt Uun
Uuu
Uub Uuq Uuh Uuo Ds La Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Ac Th Pa
Np
Pu Am Cm Bk Cf Es Fm Md No Lr H B N F P V Y I W Uaal )
class Array
def to_re( regex_modifier )
eval “/#{ self.join “|” }/#{ regex_modifier }”
end
end
“present”.scan elements.to_re( “i” ) # the “i” means case insensitive
#=> [ “pr”, “es”, “n” ]
… hope that helps. I’m sure there are more efficient regex patterns
for
this, but I couldn’t imagine anything that was easier to write.
j.
On 11/21/05, George L. [email protected] wrote:
to my list of elements, [p][re][se][n][ti]. Now, I can duplicate this
(He|Li|Be|C|O|Ne|Na|Mg|Al|Si|S|Cl|Ar|Ca|Sc|Ti|Cr|Mn|Fe|Co|Ni|Cu|Zn|Ga|Ge|As|Se|Br|Kr|Rb|Sr|Zr|Nb|Mo|Tc|Ru|Rh|Pd|Ag|Cd|In|Sn|Sb|Te|Xe|Cs|Ba|Hf|Ta|Re|Os|Ir|Pt|Au|Hg|Tl|Pb|Bi|Po|At|Rn|Fr|Ra|Rf|Db|Sg|Bh|Hs|Mt|Uun|Uuu|Uub|Uuq|Uuh|Uuo|Ds|La|Ce|Pr|Nd|Pm|Sm|Eu|Gd|Tb|Dy|Ho|Er|Tm|Yb|Lu|Ac|Th|Pa|Np|Pu|Am|Cm|Bk|Cf|Es|Fm|Md|No|Lr|H|B|N|F|P|V|Y|I|W|Uaal)+
–
“Remember. Understand. Believe. Yield! → http://ruby-lang.org”
Jeff W.
George L. wrote:
I’m wondering if anyone can explain this
behavior to me, as I’m new to both regexps and ruby and the only regex
experience I’ve had previous to this is in some theory classes.
Here is a valuable reference for this area →
http://www.oreilly.com/catalog/regex/. This book is a fixture on my
bookshelf for sure…
On Tue, 22 Nov 2005, George L. wrote:
[...]
little program to test if a word was made up of symbols from the periodic
table. […]
properly as a (nonsense) word like “presenti” should be periodic according
to my list of elements, [p][re][se][n][ti]. Now, I can duplicate this
behavior with the regex coach oddly enough. The problem seems to be that it
parses [p][re][s] instead of [se]. I’m wondering if anyone can explain this
Try putting a $ on the end. The regexp won’t succeed the first time
because when it gets to sulphur that’s enough to satisfy + as you
found. To get selenium and the rest it must backtrack to get to the
end.
Hugh
nope, it’s already found pr and es, therefore there isn’t anything left
for
matching the e by itself.
you’d have to have a permutation function to run multiple orders of
regex
and see which one matches the most of the input string.
completely do-able, just time consuming…
shouldn’t take TOO much time to write … just gonna take a while to
run.
j.
On 11/21/05, George L. [email protected] wrote:
hi,
Hi, I’m new to the list and I hope this is the right place to ask
it’s
explain
–
“Remember. Understand. Believe. Yield! → http://ruby-lang.org”
Jeff W.
By default, matchines are greedy. Greedy matching means that it will
try to match the maximum number of characters possible. Perhaps you
want non-greedy matching?
/(…)+?/i
That does seem to do it. Thanks for the help everybody, guess I’ve got
some
reading to do about regexps.
Jeff W. [email protected] writes:
eval “/#{ self.join “|” }/#{ regex_modifier }”
end
end
“present”.scan elements.to_re( “i” ) # the “i” means case insensitive
#=> [ “pr”, “es”, “n” ]
… hope that helps. I’m sure there are more efficient regex patterns for
this, but I couldn’t imagine anything that was easier to write.
Regexp.union exists.
From: “George L.” [email protected]
Unfortunately this doesn’t seem to work
properly as a (nonsense) word like “presenti” should be periodic according
to my list of elements, [p][re][se][n][ti]. Now, I can duplicate this
behavior with the regex coach oddly enough.
Hi, welcome to Ruby!
Not sure if this helps, but here’s what I get:
elems =
“He|Li|Be|C|O|Ne|Na|Mg|Al|Si|S|Cl|Ar|Ca|Sc|Ti|Cr|Mn|Fe|Co|Ni|Cu|Zn|Ga|Ge|
As|Se|Br|Kr|Rb|Sr|Zr|Nb|Mo|Tc|Ru|Rh|Pd|Ag|Cd|In|Sn|Sb|Te|Xe|Cs|Ba|Hf|Ta|Re|Os|Ir|
Pt|Au|Hg|Tl|Pb|Bi|Po|At|Rn|Fr|Ra|Rf|Db|Sg|Bh|Hs|Mt|Uun|Uuu|Uub|Uuq|Uuh|Uuo|Ds|La|
Ce|Pr|Nd|Pm|Sm|Eu|Gd|Tb|Dy|Ho|Er|Tm|Yb|Lu|Ac|Th|Pa|Np|Pu|Am|Cm|Bk|Cf|Es|Fm|Md|No|
Lr|H|B|N|F|P|V|Y|I|W|Uaal”
using \b to anchor at word boundaries
“presenti” =~ /\b(?:#{elems})+\b/i
0 # matches whole word
using scan to see which elements are being matched
(note that this approach would inappropriately skip
over letters at the beginning of the word that weren’t
elements… but any match that it does find will
be contiguous elements from that point to the end of
the word… however one could perform the above check
first to make sure the whole word matched, and then
use the scan to obtain the individual elements)
“presenti”.scan(/(?:#{elems})(?=(?:#{elems})*\b)/i)
[“p”, “re”, “se”, “n”, “ti”]
“prese”.scan(/(?:#{elems})(?=(?:#{elems})*\b)/i)
[“p”, “re”, “se”]
Regards,
Bill
I’ll have to go look @ that one … haven’t played with it yet.
j.
On 11/21/05, Christian N. [email protected] wrote:
Uuu
“present”.scan elements.to_re( “i” ) # the “i” means case insensitive
Christian N. [email protected] http://chneukirchen.org
–
“Remember. Understand. Believe. Yield! → http://ruby-lang.org”
Jeff W.