I remember the UTF8 regex stuff being kind of unpolished last time I checked. There's a hack used in TextAnalysis.jl to get around some of it.
-- John On Dec 17, 2013, at 10:13 AM, "Alessandro \"Jake\" Andrioni" <[email protected]> wrote: > Hi, I thought UTF-8 regex support was enabled by default, but > > ``` > julia> ismatch(r"\w+", "áéíóúêî") > false > > julia> ismatch(r"\w+", "áéíóúêîa") > true > ``` > > Am I missing something? For example, in perl: > ``` > $ perl -le 'use utf8; print "áéíóúêî" =~ /\w+/;' > 1 > ``` > > Is there a flag I am missing?
