My own .hs file (aneks.hs)

This commit is contained in:
Dominik 2021-03-27 22:26:33 +01:00
parent 8883a924b4
commit 5cf2c07416
3 changed files with 76 additions and 0 deletions

44
app/aneks.hs Normal file
View File

@ -0,0 +1,44 @@
{-# LANGUAGE Arrows, NoMonomorphismRestriction #-}
import ShadowLibrary.Core
import Text.XML.HXT.Core
import Text.XML.HXT.XPath
-- import Text.XML.HXT.Curl
import Data.List
import Data.List.Utils (replace)
import Text.Regex.Posix
import Text.Printf
extractRecords = extractLinksWithText "//a[contains(@title,'Aneks')]" -- pary tytuł-adres
>>> first (arr $ replace "\r\n " " ") -- czyścimy pierwszy element pary, czyli tytuł z niepotrzebnych białych znaków
>>> second (extractLinksWithText "//div/a[contains(@href,'.pdf')]") -- pobieramy stronę z adresu URL i wyciągamy linki z tej strony pasujące do wyrażenia XPathowego
-- ostatecznie wyjdą trójki? (Tytuł, (link do pdf, blank?))
-- ... a tutaj te trójki przerabiamy do docelowej struktury ShadowItem
toShadowItem :: (String, (String, String)) -> ShadowItem
toShadowItem (yearlyTitle, (url, blank)) =
(defaultShadowItem url title) {
originalDate = Just date,
itype = "periodical",
format = Just "pdf",
finalUrl = url
}
where title = yearlyTitle ++ " " ++ (replace "\r\n" "" (replace "\r\n " "" blank))
date = getDate url
getDate url =
case url =~~ "/(19[0-9][0-9]|20[0-9][0-9])/" :: Maybe [[String]] of
Just [[_, year]] -> year
otherwise -> error $ "unexpected url: " ++ url
main = do
let start = "https://aneks.kulturaliberalna.pl/archiwum-aneksu/"
let shadowLibrary = ShadowLibrary {logoUrl=Nothing,
lname="Archiwum Aneksu",
abbrev="ArchAnek",
lLevel=0,
webpage=start}
extractItemsStartingFromUrl shadowLibrary start (extractRecords >>> arr toShadowItem)

View File

@ -59,6 +59,19 @@ executable almanachmuszyny
, shadow-library
default-language: Haskell2010
executable aneks
hs-source-dirs: app
main-is: aneks.hs
ghc-options: -threaded -rtsopts -with-rtsopts=-N
build-depends: base
, hxt
, hxt-xpath
, MissingH
, regex-posix
, shadow-library
default-language: Haskell2010
source-repository head
type: git

19
stack.yaml.lock Normal file
View File

@ -0,0 +1,19 @@
# This file was autogenerated by Stack.
# You should not edit this file by hand.
# For more information, please see the documentation at:
# https://docs.haskellstack.org/en/stable/lock_files
packages:
- completed:
hackage: hxt-xpath-9.1.2.2@sha256:9cd590ae93a04573db8f90fa4094625ebd97dded45da7667c577ce6b38a42900,1999
pantry-tree:
size: 2225
sha256: aee2f75974e868ff429b8ff349a29667536c60397098f5dfedc968d1951511bb
original:
hackage: hxt-xpath-9.1.2.2
snapshots:
- completed:
size: 507596
url: https://raw.githubusercontent.com/commercialhaskell/stackage-snapshots/master/lts/11/9.yaml
sha256: 42f472dbf06482da1b3319241f3e3b3593a45bd7d4f537d2789f21386b9b2ad3
original: lts-11.9