forked from filipg/twilight-library
Compare commits
No commits in common. "master" and "master" have entirely different histories.
@ -33,7 +33,7 @@ import Data.Tree.NTree.TypeDefs
|
||||
import Data.Maybe
|
||||
import Control.Monad.Trans
|
||||
import Text.XML.HXT.XPath
|
||||
import Text.XML.HXT.Curl
|
||||
-- import Text.XML.HXT.Curl
|
||||
import Text.XML.HXT.HTTP
|
||||
|
||||
import Text.Regex.TDFA
|
||||
@ -64,8 +64,8 @@ downloadDocument = readFromDocument [withParseHTML yes,
|
||||
withEncodingErrors no,
|
||||
withPreserveComment yes,
|
||||
withStrictInput yes,
|
||||
-- withHTTP []
|
||||
withCurl [("curl--user-agent","AMU Digital Libraries Indexing Agent")]
|
||||
withHTTP []
|
||||
-- withCurl [("curl--user-agent","AMU Digital Libraries Indexing Agent")]
|
||||
]
|
||||
|
||||
downloadDocumentWithEncoding enc = readFromDocument [withParseHTML yes,
|
||||
@ -73,13 +73,13 @@ downloadDocumentWithEncoding enc = readFromDocument [withParseHTML yes,
|
||||
withEncodingErrors no,
|
||||
withPreserveComment yes,
|
||||
withInputEncoding enc,
|
||||
-- withHTTP []]
|
||||
withCurl []]
|
||||
withHTTP []]
|
||||
-- withCurl []]
|
||||
|
||||
downloadXmlDocument = readFromDocument [withWarnings no,
|
||||
withEncodingErrors no,
|
||||
-- withHTTP []]
|
||||
withCurl [] ]
|
||||
withHTTP []]
|
||||
-- withCurl [] ]
|
||||
|
||||
|
||||
data ShadowLibrary = ShadowLibrary { logoUrl :: Maybe String,
|
||||
|
||||
@ -1,105 +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
|
||||
|
||||
import Data.Char (toLower)
|
||||
|
||||
class Nothingish a where
|
||||
nada :: a
|
||||
instance Nothingish [a] where
|
||||
nada = []
|
||||
|
||||
eliminate :: (Nothingish a) => Maybe a -> a
|
||||
eliminate (Just a) = a
|
||||
eliminate Nothing = nada
|
||||
|
||||
toLowerString :: [Char] -> [Char]
|
||||
toLowerString str = [ toLower x | x <- str]
|
||||
|
||||
mToString :: Maybe String -> String
|
||||
mToString n
|
||||
| n == Nothing = ""
|
||||
| otherwise = eliminate n
|
||||
|
||||
extractMonth :: String -> String
|
||||
extractMonth n =
|
||||
case n =~~ ("[A-za-z]+" :: String) of
|
||||
Just month -> "-" ++ eliminate (baseMonthNameToNumber (toLowerString month))
|
||||
otherwise -> ""
|
||||
|
||||
mExtractYear :: String -> String
|
||||
mExtractYear n =
|
||||
case n =~~ ("(1[6789]|20)[0-9][0-9]" :: String) of
|
||||
Just year -> year
|
||||
otherwise -> ""
|
||||
|
||||
changeDate :: String -> Maybe String
|
||||
changeDate a = Just (eliminate (extractYear a) ++ extractMonth a)
|
||||
|
||||
extractLinksWithArticleTitle xpathCondition = (downloadDocument &&& this)
|
||||
>>> first (getXPathTrees xpathCondition
|
||||
>>> (
|
||||
(getXPathTrees "//div[@class='text']" >>> (listA (deep isText >>> getText)
|
||||
>>> arr (intercalate " ")))
|
||||
&&&
|
||||
(getXPathTrees "//div[@class='files__item']/a[contains(@href,'.pdf')]" >>> (getAttrValue "href"))
|
||||
))
|
||||
|
||||
--extractRecords = extractLinksWithText "(//a[@class='magazine-list__year-item'])[last()]" -- pary adres-tytuł
|
||||
extractRecords = extractLinksWithText "//a[@class='magazine-list__year-item']" -- 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 "//div[@class='magazine-list__item']/a") -- pobieramy stronę z adresu URL i wyciągamy linki z tej strony pasujące do wyrażenia XPathowego
|
||||
>>> first (second (arr $ replace "\r\n " ""))
|
||||
>>> first (first (
|
||||
-- extractLinksWithArticleTitle "//div[@class='files__item']/a[contains(@href,'.pdf')]"
|
||||
(extractLinksWithArticleTitle "//div[@class='magazine-single__content-title article text']")
|
||||
>>>
|
||||
first (
|
||||
first (arr $ replace " " "")
|
||||
>>> first (arr $ replace "\r\n " "")
|
||||
)
|
||||
-- >>> first (arr $ replace "\r\n" "")
|
||||
-- >>> first (arr $ replace "//" "/")
|
||||
)
|
||||
-- >>> second (arr $ changeDate) -- Zmiana nazwy miesiąca na wartość liczbową
|
||||
>>> second (arr $ replace " " "") >>> second (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), String) -> ShadowItem
|
||||
toShadowItem ((((chapterTitle, url), finalUrl), articleTitle), yearlyTitle) =
|
||||
(defaultShadowItem url title) {
|
||||
originalDate = changeDate articleTitle,
|
||||
itype = "periodical",
|
||||
format = Just "pdf",
|
||||
finalUrl = finalUrl
|
||||
}
|
||||
where title = "Elektronika Praktyczna " ++ articleTitle ++ " - " ++ 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 = "https://ep.com.pl/archiwum/"
|
||||
let shadowLibrary = ShadowLibrary {logoUrl=Nothing,
|
||||
lname="Elektronika praktyczna",
|
||||
abbrev="EP",
|
||||
lLevel=0,
|
||||
webpage=start}
|
||||
extractItemsStartingFromUrl shadowLibrary start (extractRecords >>> arr toShadowItem)
|
||||
@ -2,11 +2,11 @@ name: shadow-library
|
||||
version: 0.1.0.0
|
||||
synopsis: Initial project template from stack
|
||||
description: Please see README.md
|
||||
homepage: https://git.wmi.amu.edu.pl/s426206/twilight-library.git
|
||||
homepage: http://github.com/name/project
|
||||
license: Proprietary
|
||||
license-file: LICENSE
|
||||
author: Jan Nowak
|
||||
maintainer: jannow2@st.amu.edu.pl
|
||||
author: Your name here
|
||||
maintainer: your.address@example.com
|
||||
-- copyright:
|
||||
category: Web
|
||||
build-type: Simple
|
||||
@ -20,7 +20,6 @@ library
|
||||
, HTTP
|
||||
, hxt
|
||||
, hxt-http
|
||||
, hxt-curl
|
||||
, hxt-xpath
|
||||
, MissingH
|
||||
, monad-logger
|
||||
@ -48,14 +47,13 @@ library
|
||||
-- , shadow-library
|
||||
-- default-language: Haskell2010
|
||||
|
||||
executable elektronikapraktyczna
|
||||
executable almanachmuszyny
|
||||
hs-source-dirs: app
|
||||
main-is: elektronikapraktyczna.hs
|
||||
main-is: almanachmuszyny.hs
|
||||
ghc-options: -threaded -rtsopts -with-rtsopts=-N
|
||||
build-depends: base
|
||||
, hxt
|
||||
, hxt-xpath
|
||||
, hxt-curl
|
||||
, MissingH
|
||||
, regex-posix
|
||||
, shadow-library
|
||||
@ -64,4 +62,4 @@ executable elektronikapraktyczna
|
||||
|
||||
source-repository head
|
||||
type: git
|
||||
location: https://git.wmi.amu.edu.pl/s426206/twilight-library.git
|
||||
location: https://github.com/name/project
|
||||
|
||||
@ -1,19 +0,0 @@
|
||||
# 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
|
||||
Loading…
Reference in New Issue
Block a user