This commit is contained in:
Norbert Litkowski 2021-04-07 02:12:02 +02:00
parent 9aa0ae145b
commit 974e04831c
3 changed files with 56 additions and 58 deletions

55
app/ZborBielawa.hs Normal file
View File

@ -0,0 +1,55 @@
{-# 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
extractNestedLinksWithText xpathCondition = (downloadDocument &&& this)
>>> first (getXPathTrees xpathCondition
>>> ((getXPathTrees "//a" >>> getAttrValue "href")
&&& (listA (deep isText >>> getText)
>>> arr (intercalate " "))))
>>> arr rotateSecTh
>>> first expandURIFixed
extractRecords = extractLinksWithText "(//aside[@class='widget widget_maxmegamenu']//a[@class='mega-menu-link'])[1]" -- pary adres-tytuł podstrony
>>> first (extractLinksWithText "(//aside[@class='widget widget_maxmegamenu']//a[@class='mega-menu-link'])[1]") -- pobieramy podstronę i kolejne podstrony z menu
>>> first (first (extractNestedLinksWithText "//big/a[contains(@href,'.pdf')][img]")) -- pobieramy stronę z adresu URL i wyciągamy linki z tej strony pasujące do wyrażenia XPathowego
-- ostatecznie wyjdą krotki (((adres URL, tytuł nr-u), tytuł podstrony 2), tytuł podstrony 1)
-- ... a tutaj te krotki przerabiamy do docelowej struktury ShadowItem
toShadowItem :: (((String, String), String), String) -> ShadowItem
toShadowItem (((url, releaseTitle), collectionTitle), categoryTitle) =
(defaultShadowItem url title) {
originalDate = Just date,
itype = "periodical",
format = Just "pdf",
finalUrl = url
}
where title = categoryTitle ++ (" " ++ collectionTitle)
date = getDate $ releaseTitle
getDate yearlyTitle =
case yearlyTitle =~~ "/(19[0-9][0-9]|20[0-9][0-9])/" :: Maybe [[String]] of
Just [[_, year]] -> year
-- otherwise -> error $ "unexpected yearlyTitle: " ++ yearlyTitle
otherwise -> yearlyTitle
main = do
let start = "http://zborbielawa.pl/archiwum/"
let shadowLibrary = ShadowLibrary {logoUrl=Nothing,
lname="Zbór Bielawa",
abbrev="ZboBiel",
lLevel=0,
webpage=start}
extractItemsStartingFromUrl shadowLibrary start (extractRecords >>> arr toShadowItem)

View File

@ -1,57 +0,0 @@
{-# 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 "//aside[@class='widget widget_maxmegamenu']//a[@class='mega-menu-link']" -- pary adres-tytuł
-- >>> second (arr $ replace "\r\n " "") -- czyścimy drugi element pary, czyli tytuł z niepotrzebnych białych znaków
-- >>> second (arr $ replace " " "") -- czyścimy drugi element pary, czyli tytuł z niepotrzebnych białych znaków
-- >>> first (arr ((++"tr") . init)) -- modyfikujemy pierwszy element pary, czyli adres URL
>>> first (extractLinksWithText "//aside[@class='widget widget_maxmegamenu']//a[@class='mega-menu-link']") -- pobieramy stronę z adresu URL i wyciągamy linki z tej strony pasujące do wyrażenia XPathowego
>>> first (second (arr $ replace "\r\n " ""))
>>> first (first (
extractLinksWithText "//a[contains(@href,'.pdf')]"
>>> second (arr $ replace "\r\n " "")
-- >>> first (arr $ replace "//" "/")
)
) -- pobieramy stronę z adresu URL i wyciągamy linki z tej strony pasujące do wyrażenia XPathowego
-- ostatecznie wyjdą trójki ((adres URL, tytuł artykułu), tytuł rocznika)
-- ... a tutaj te trójki przerabiamy do docelowej struktury ShadowItem
toShadowItem :: (((String, String), String), String) -> ShadowItem
toShadowItem (((url, chapterTitle), articleTitle), yearlyTitle) =
(defaultShadowItem url title) {
originalDate = Just date,
itype = "periodical",
format = Just "pdf",
finalUrl = url
}
where title = articleTitle ++ (replace " " "" chapterTitle)
date = yearlyTitle
getDate yearlyTitle =
case yearlyTitle =~~ "/(19[0-9][0-9]|20[0-9][0-9])/" :: Maybe [[String]] of
Just [[_, year]] -> year
otherwise -> error $ "unexpected yearlyTitle: " ++ yearlyTitle
main = do
let start = "http://zborbielawa.pl/archiwum/"
let shadowLibrary = ShadowLibrary {logoUrl=Nothing,
lname="Zbór Bielawa",
abbrev="ZboBiel",
lLevel=0,
webpage=start}
extractItemsStartingFromUrl shadowLibrary start (extractRecords >>> arr toShadowItem)

View File

@ -62,7 +62,7 @@ executable almanachmuszyny
executable zborbielawa executable zborbielawa
hs-source-dirs: app hs-source-dirs: app
main-is: zborbielawa.hs main-is: ZborBielawa.hs
ghc-options: -threaded -rtsopts -with-rtsopts=-N ghc-options: -threaded -rtsopts -with-rtsopts=-N
build-depends: base build-depends: base
, hxt , hxt