From 9aa0ae145ba76a5d29d183c9bd855283eb6a1556 Mon Sep 17 00:00:00 2001 From: Norbert Litkowski Date: Wed, 7 Apr 2021 00:49:14 +0200 Subject: [PATCH] WIP --- app/zborbielawa.hs | 4 ++-- stack.yaml.lock | 19 +++++++++++++++++++ 2 files changed, 21 insertions(+), 2 deletions(-) create mode 100644 stack.yaml.lock diff --git a/app/zborbielawa.hs b/app/zborbielawa.hs index f6613f1..6b1a1bd 100644 --- a/app/zborbielawa.hs +++ b/app/zborbielawa.hs @@ -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')]" 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