From 53fff21f4d747d870e63ba7a4a645ddc98b9b830 Mon Sep 17 00:00:00 2001
From: Kacper <kacperdudzic@wp.pl>
Date: Mon, 4 Apr 2022 20:23:44 +0200
Subject: [PATCH] rozwiazanie

---
 app/inspektoratpracy.hs | 34 ++++++++++++++++++++++++++++++++++
 shadow-library.cabal    | 12 ++++++++++++
 stack.yaml.lock         | 19 +++++++++++++++++++
 3 files changed, 65 insertions(+)
 create mode 100644 app/inspektoratpracy.hs
 create mode 100644 stack.yaml.lock

diff --git a/app/inspektoratpracy.hs b/app/inspektoratpracy.hs
new file mode 100644
index 0000000..5b44012
--- /dev/null
+++ b/app/inspektoratpracy.hs
@@ -0,0 +1,34 @@
+
+{-# 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 "//td/a[contains(@href,'.pdf')]"
+
+toShadowItem :: (String, String) -> ShadowItem
+toShadowItem (url, text) =
+  (defaultShadowItem url text) {
+    originalDate = Just date,
+    itype = "periodical",
+    format = Just "pdf",
+    finalUrl = url
+    }
+  where title = ""
+        date = ""
+
+main = do
+    let start = "https://www.pip.gov.pl/pl/inspektor-pracy/66546,archiwum-inspektora-pracy-.html"
+    let shadowLibrary = ShadowLibrary {logoUrl=Nothing,
+                                       lname="Inspektor Pracy",
+                                       abbrev="InspPrac",
+                                       lLevel=0,
+                                       webpage=start}
+    extractItemsStartingFromUrl shadowLibrary start (extractRecords >>> arr toShadowItem)
diff --git a/shadow-library.cabal b/shadow-library.cabal
index 3e1bd4e..e6dd6a4 100644
--- a/shadow-library.cabal
+++ b/shadow-library.cabal
@@ -60,6 +60,18 @@ executable almanachmuszyny
                       , regex-posix
                       , shadow-library
    default-language:    Haskell2010
+   
+executable inspektoratpracy
+   hs-source-dirs:      app
+   main-is:             inspektoratpracy.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
diff --git a/stack.yaml.lock b/stack.yaml.lock
new file mode 100644
index 0000000..cc80ef3
--- /dev/null
+++ b/stack.yaml.lock
@@ -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