This commit is contained in:
Norbert Litkowski 2021-04-07 00:49:14 +02:00
parent e2aa05e511
commit 9aa0ae145b
2 changed files with 21 additions and 2 deletions

View File

@ -12,11 +12,11 @@ import Text.Regex.Posix
import Text.Printf
extractRecords = extractLinksWithText "//a[@class='mega-menu-link']" -- pary adres-tytuł
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 "//a[@class='mega-menu-link']") -- pobieramy stronę z adresu URL i wyciągamy linki z tej strony pasujące do wyrażenia XPathowego
>>> 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')]"

19
stack.yaml.lock Normal file
View File

@ -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