From 5977e1454d7738ddb086d37b20337e350e380790 Mon Sep 17 00:00:00 2001
From: Joris Guyonvarch
Date: Sat, 11 Apr 2015 11:50:48 +0200
Subject: Fetch first page ads of a given leboncoin url, fetch also the
 description page of each item.

---
 src/Parser/Resume.hs | 27 +++++++++++++++++++++++++++
 1 file changed, 27 insertions(+)
 create mode 100644 src/Parser/Resume.hs

(limited to 'src/Parser/Resume.hs')

diff --git a/src/Parser/Resume.hs b/src/Parser/Resume.hs
new file mode 100644
index 0000000..bd73912
--- /dev/null
+++ b/src/Parser/Resume.hs
@@ -0,0 +1,27 @@
+module Parser.Resume
+  ( parseResumes
+  ) where
+
+import Data.Maybe (catMaybes)
+
+import Text.HTML.TagSoup
+
+import Model.Resume
+
+import Parser.Utils
+
+parseResumes :: String -> [Resume]
+parseResumes page =
+  case sections (~== "<div class=list-lbc>") (parseTags page) of
+    [] ->
+      []
+    sectionTags : _ ->
+      let lbcTags = takeWhile (~/= "<div id=alertesCartouche>") sectionTags
+      in  catMaybes . fmap parseResume $ partitions (~== "<a>") lbcTags
+
+parseResume :: [Tag String] -> Maybe Resume
+parseResume item = do
+  name <- getTagText "<h2 class=title>" item
+  let price = getTagText "<div class=price>" item
+  url <- getTagAttribute "<a>" "href" item
+  return Resume { name = name, price = price, url = url }
-- 
cgit v1.2.3