mokled, ai es:
$geouni = q {xE1x83[x90-xAF]};
sheicavs nebismier qartul asos unikodshi (mxedvelobashi maqvs 33 chveulebrivi aso, mtavrulebis da sxva rameebis me ar mesmis)
shesabamisad shablonshi sheizleba cvladis gamokeneba:
*mxolod teoriashi, praqtikaze tqven tviton sinjet, egeb rame gamomrcha
$page=~tr/A-Z/a-z/; #rusuls tavi davanebot, rogorme qartuli avamushavot jer
map{$page=~s/$_[^a-z$geouni/ /g if length($_)>3} @word_ends; #shesazloa aq 3 9-it unda shevcvalot? ragac am sitkvebis daboloebebis moshorebis azri bolomde ver gavige, ufro scorad azri ki mesmis, magram realizacia map-is gamokenebit ragac maechvebs.
map($page=~s/[^a-z$geouni]$_[^a-z$geouni]/ /g, @stop_words);
$page=~s/([-a-z$geouni])1+/$1/g; #ganmeorebuli simboloebshi aq igulisxmeba albad erti da igive simbolo ki ara, aramed ertze meti "araaso", anu sitkvis bolos shesazloa ikos mzime mere probeli, orive ar carmoadgenen sitkvis nacils da chven ki mxolod erti gvchirdeba sitkvebi ertmanetisgan rom ganvasxvavot shemdeg striqonshi.
my(@words) = split(/[^-a-z$geouni]+/, $page); #pirveli minusi gveubneba, rom sitkvebis dakofa xdeba im simboloebit, romlebic nagdad asos ar carmoadgenen
@words=grep{length >
6} @words; # ai es mnishvnelovania, radgan erti qartuli aso 3 baitia da perlis funqcia "length" baitebs itvlis, sitkva or asoze meti 6 baitze meti iqneba.
aba es scade ra gamova netavi?