From 8ba5a297b0a405f166c814046cef9d55ee1ad7ba Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jakub=20Kolasin=CC=81ski?= <jakkol4@st.amu.edu.pl>
Date: Wed, 17 Mar 2021 21:07:39 +0100
Subject: [PATCH] Make script work for 'nasze popowice' pdfs

---
 app/naszepopowice.hs | 30 ++++++++++++++++--------------
 1 file changed, 16 insertions(+), 14 deletions(-)

diff --git a/app/naszepopowice.hs b/app/naszepopowice.hs
index 3c3e56e..9c57617 100644
--- a/app/naszepopowice.hs
+++ b/app/naszepopowice.hs
@@ -4,7 +4,6 @@ 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)
 
@@ -12,35 +11,38 @@ import Text.Regex.Posix
 import Text.Printf
 
 
-extractRecords = extractLinksWithText "//a[@class='attachmentContainer']"  -- pary adres-tytuł
-                 >>> second (arr $ replace "\r\n            " " ") -- czyścimy drugi element pary, czyli tytuł z niepotrzebnych białych znaków
-                 >>> first (arr ((++"tr") . init))  -- modyfikujemy pierwszy element pary, czyli adres URL
-                 >>> first (extractLinksWithText "//li/a[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)
+extractRecords = extractLinksWithText "//a[contains(@href,'.pdf') and contains(text(),'popowice')]"
 
--- ... a tutaj te trójki przerabiamy do docelowej struktury ShadowItem
-toShadowItem :: ((String, String), String) -> ShadowItem
-toShadowItem ((url, articleTitle), yearlyTitle) =
+toShadowItem :: (String, String) -> ShadowItem
+toShadowItem (url, articleTitle) =
   (defaultShadowItem url title) {
     originalDate = Just date,
     itype = "periodical",
     format = Just "pdf",
     finalUrl = url
     }
-  where title = "Almanach Muszyny " ++ yearlyTitle ++ " " ++ (replace "\r\n" "" (replace "\r\n          " "" articleTitle))
+  where title = "Nasze Popowice " ++ getNr url
         date = getDate url
 
+replaceSeparatorWithSpace = map (\c -> if c=='_' then ' '; else if c=='-' then ' '; else c)
+replaceSeparatorWithDot = map (\c -> if c=='_' then '.'; else c)
+
+getNr url =
+  case url =~~ "nr[-_][0-9][0-9]" :: Maybe String of
+    Just nr -> replaceSeparatorWithSpace nr
+    otherwise -> error $ "unexpected url: " ++ url
+
 getDate url =
-  case url =~~ "/(19[0-9][0-9]|20[0-9][0-9])/" :: Maybe [[String]] of
-    Just [[_, year]] -> year
+  case url =~~ "20[0-9][0-9]_[0-1][0-9]" :: Maybe String of
+    Just date -> replaceSeparatorWithDot date
     otherwise -> error $ "unexpected url: " ++ url
 
 
 main = do
     let start = "https://smpopowice.pl/index.php/numery-archiwalne"
     let shadowLibrary = ShadowLibrary {logoUrl=Nothing,
-                                       lname="Almanach Muszyny",
-                                       abbrev="AlmMusz",
+                                       lname="Nasze Popowice",
+                                       abbrev="NaszPop",
                                        lLevel=0,
                                        webpage=start}
     extractItemsStartingFromUrl shadowLibrary start (extractRecords >>> arr toShadowItem)