s444018/twilight-library
1
0
Fork 0
forked from filipg/twilight-library
Code Issues Pull Requests Projects Releases Wiki Activity

Add crawler for czasopisma...

Browse Source
This commit is contained in:
Szymon 2022-04-17 22:54:21 +02:00
parent 0231c45a80
commit 76146d4e09
2 changed files with 44 additions and 96 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

44
app/czasopisma.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 "" -- pary adres-tytuł
>>> first (extractLinksWithText "//a[@target='_blank'][contains(@href,'.pdf')]") -- pobieramy stronę z adresu URL i wyciągamy linki z tej strony pasujące do wyrażenia XPathowego
-- ostatecznie wyjdą trójki ((adres URL, tytuł artykułu), tytuł rocznika)
toShadowItem :: ((String, String), String) -> ShadowItem
toShadowItem ((url, articleTitle), yearlyTitle) =
(defaultShadowItem url title) {
originalDate = Just date,
itype = "periodical",
format = Just "pdf",
finalUrl = url
}
where title = replace "\"" "'" (replace "\t\t\t\t\t" "" (replace "\n" "" yearlyTitle)) ++ "articleTitle: " ++ articleTitle
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://www.czasopisma.centralnabibliotekapttk.pl"
let shadowLibrary = ShadowLibrary {logoUrl=Nothing,
lname="Chelmek",
abbrev="Chelmek",
lLevel=0,
webpage=start}
extractItemsStartingFromUrl shadowLibrary start (extractRecords >>> arr toShadowItem)

96
app/kulturaparyska.hs
View File

@ -1,96 +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 "//a[@class='year-anchor ']" -- pary adres-tytuł
>>> second (arr $ replace "\r\n " " ") -- czyścimy drugi element pary, czyli tytuł z niepotrzebnych białych znaków
-- >>> first (arr ((++"f") . init)) -- modyfikujemy pierwszy element pary, czyli adres URL
>>> first (extractLinksWithText "//a[@class='pdf-exist'][contains(@href,'.pdf')]") -- pobieramy stronę z adresu URL i wyciągamy linki z tej strony pasujące do wyrażenia XPathowego
-- ostatecznie wyjdą trójki ((adres URL, tytuł artykułu), tytuł rocznika)
-- extractRecords2 = extractLinksWithText "//a[@class='year-anchor']" -- pary adres-tytuł
-- >>> second (arr $ replace "\r\n " " ") -- czyścimy drugi element pary, czyli tytuł z niepotrzebnych białych znaków
-- -- >>> first (arr ((++"f") . init)) -- modyfikujemy pierwszy element pary, czyli adres URL
-- >>> first (extractLinksWithText "//a[ends-with(@href,'.pdf')]") -- 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) -> ShadowItem
toShadowItem ((url, articleTitle), yearlyTitle) =
(defaultShadowItem url title) {
originalDate = Just date,
itype = "periodical",
format = Just "pdf",
finalUrl = url
}
where title = "Kultura Paryska " ++ yearlyTitle ++ " " ++ (replace "\r\n" "" (replace "\r\n " "" articleTitle))
date = yearlyTitle
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://kulturaparyska.com/pl/publication/4/year/1946"
let shadowLibrary = ShadowLibrary {logoUrl=Nothing,
lname="Kultura Paryska",
abbrev="kultParys",
lLevel=0,
webpage=start}
putStrLn "Program started"
extractItemsStartingFromUrl shadowLibrary start (extractRecords >>> arr toShadowItem)
-- First time using haskell and for loop in haskell is hard
putStrLn "New link"
let start2 = "https://kulturaparyska.com/pl/publication/1/year/1946"
extractItemsStartingFromUrl shadowLibrary start2 (extractRecords >>> arr toShadowItem)
putStrLn "New link"
let start2 = "https://kulturaparyska.com/pl/publication/2/year/1946"
extractItemsStartingFromUrl shadowLibrary start2 (extractRecords >>> arr toShadowItem)
putStrLn "New link"
let start2 = "https://kulturaparyska.com/pl/publication/3/year/1946"
extractItemsStartingFromUrl shadowLibrary start2 (extractRecords >>> arr toShadowItem)
putStrLn "New link"
let start2 = "https://kulturaparyska.com/pl/publication/5/year/1946"
extractItemsStartingFromUrl shadowLibrary start2 (extractRecords >>> arr toShadowItem)
putStrLn "New link"
let start2 = "https://kulturaparyska.com/pl/publication/6/year/1946"
extractItemsStartingFromUrl shadowLibrary start2 (extractRecords >>> arr toShadowItem)
putStrLn "New link"
let start2 = "https://kulturaparyska.com/pl/publication/7/year/1946"
extractItemsStartingFromUrl shadowLibrary start2 (extractRecords >>> arr toShadowItem)
putStrLn "New link"
let start2 = "https://kulturaparyska.com/pl/publication/8/year/1946"
extractItemsStartingFromUrl shadowLibrary start2 (extractRecords >>> arr toShadowItem)
putStrLn "New link"
let start2 = "https://kulturaparyska.com/pl/publication/9/year/1946"
extractItemsStartingFromUrl shadowLibrary start2 (extractRecords >>> arr toShadowItem)
putStrLn "New link"
let start2 = "https://kulturaparyska.com/pl/publication/10/year/1946"
extractItemsStartingFromUrl shadowLibrary start2 (extractRecords >>> arr toShadowItem)
putStrLn "New link"
let start2 = "https://kulturaparyska.com/pl/publication/11/year/1946"
extractItemsStartingFromUrl shadowLibrary start2 (extractRecords >>> arr toShadowItem)