Anonymous user
Separate the house number from the street name: Difference between revisions
Separate the house number from the street name (view source)
Revision as of 21:18, 5 July 2020
, 3 years ago→{{header|Go}}: add Haskell
SqrtNegInf (talk | contribs) m (→{{header|Sidef}}: Fix link: Perl 6 --> Raku) |
(→{{header|Go}}: add Haskell) |
||
Line 247:
Laan 1940-'45 66
Laan '40-'45 (none)
Langeloërduinen 3 46
Marienwaerdt 2e Dreef 2
Provincialeweg N205 1
Rivium 2e Straat 59.
Nieuwe gracht 20rd
Nieuwe gracht 20rd 2
Nieuwe gracht 20zw /2
Nieuwe gracht 20zw/3
Nieuwe gracht 20 zw/4
Bahnhofstr. 4
Wertstr. 10
Lindenhof 1
Nordesch 20
Weilstr. 6
Harthauer Weg 2
Mainaustr. 49
August-Horch-Str. 3
Marktplatz 31
Schmidener Weg 3
Karl-Weysser-Str. 6
</pre>
=={{header|Haskell}}==
{{libheader|regex-pcre-builtin}}
{{Works with|GHC|7.8.3}}
<lang haskell>{-# LANGUAGE OverloadedStrings #-}
{- Recommended package versions to use:
base >= 4.7 && < 5
regex-pcre-builtin >= 0.95 && < 0.96
text >= 1.2.3 && < 1.3
-}
module Main where
import Control.Monad
import Data.Char
import Data.Monoid
import qualified Data.Text as T
import qualified Data.Text.IO as T
import Text.Regex.PCRE.Text
testSet :: [T.Text]
testSet =
[ "Plataanstraat 5"
, "Straat 12"
, "Straat 12 II"
, "Dr. J. Straat 12"
, "Dr. J. Straat 12 a"
, "Dr. J. Straat 12-14"
, "Laan 1940 – 1945 37"
, "Plein 1940 2"
, "1213-laan 11"
, "16 april 1944 Pad 1"
, "1e Kruisweg 36"
, "Laan 1940-’45 66"
, "Laan ’40-’45"
, "Langeloërduinen 3 46"
, "Marienwaerdt 2e Dreef 2"
, "Provincialeweg N205 1"
, "Rivium 2e Straat 59."
, "Nieuwe gracht 20rd"
, "Nieuwe gracht 20rd 2"
, "Nieuwe gracht 20zw /2"
, "Nieuwe gracht 20zw/3"
, "Nieuwe gracht 20 zw/4"
, "Bahnhofstr. 4"
, "Wertstr. 10"
, "Lindenhof 1"
, "Nordesch 20"
, "Weilstr. 6"
, "Harthauer Weg 2"
, "Mainaustr. 49"
, "August-Horch-Str. 3"
, "Marktplatz 31"
, "Schmidener Weg 3"
, "Karl-Weysser-Str. 6"
]
-- This is the regex from the Perl implementation of the task.
addressPattern :: T.Text
addressPattern = T.unlines
[ "^ (.*?) \\s+"
, " ("
, " \\d* (\\-|\\/)? \\d*"
, " | \\d{1,3} [a-zI./ ]* \\d{0,3}"
, " )"
, "$"
]
splitAddressASCII :: Regex -> T.Text -> IO (T.Text, T.Text)
splitAddressASCII rx txt = do
result <- regexec rx txt
case result of
Left w -> fail (show w)
Right (Just (_, _, _, (street:house:_))) -> return (street, house)
_ -> return (txt, "")
-- For reasons I don't understand, PCRE isn't handling UTF-8 correctly,
-- even when the compUTF8 option is given. So, hack around it by
-- assuming any non-ASCII characters are in the street name, not the number.
splitAddress :: Regex -> T.Text -> IO (T.Text, T.Text)
splitAddress rx txt = do
let prefix = T.dropWhileEnd isAscii txt
ascii = T.takeWhileEnd isAscii txt
(street, house) <- splitAddressASCII rx ascii
return (prefix <> street, house)
formatPairs :: [(T.Text, T.Text)] -> [T.Text]
formatPairs pairs =
let headings = ("Street", "House Number")
(streets, houses) = unzip (headings : pairs)
sLen = maximum $ map T.length streets
hLen = maximum $ map T.length houses
sep = (T.replicate sLen "-", T.replicate hLen "-")
fmt (s, h) = T.justifyLeft (sLen + 4) ' ' s <> h
in map (T.strip . fmt) (headings : sep : pairs)
main :: IO ()
main = do
erx <- compile (compExtended + compUTF8) execBlank addressPattern
rx <- case erx of
Left (offset, str) -> fail $ show offset ++ ": " ++ str
Right r -> return r
pairs <- mapM (splitAddress rx) testSet
mapM_ T.putStrLn $ formatPairs pairs</lang>
{{out}}
<pre>
Street House Number
--------------------- ------------
Plataanstraat 5
Straat 12
Straat 12 II
Dr. J. Straat 12
Dr. J. Straat 12 a
Dr. J. Straat 12-14
Laan 1940 – 1945 37
Plein 1940 2
1213-laan 11
16 april 1944 Pad 1
1e Kruisweg 36
Laan 1940-’45 66
Laan ’40-’45
Langeloërduinen 3 46
Marienwaerdt 2e Dreef 2
| |||