From 26d632a461ea44ee0ec5cc6ce052d10e54c9a9c3 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Fran=C3=A7ois=20Revol?= <revol@free.fr>
Date: Sat, 1 Oct 2016 04:42:02 +0200
Subject: [PATCH] [linuxjobs] first working implementation

As published in GLMF.
---
 modules/linuxjobs/browser.py | 27 +++++++++++++++++----------
 modules/linuxjobs/module.py  |  5 +++--
 modules/linuxjobs/pages.py   | 29 +++++++++++++++++++++++------
 3 files changed, 43 insertions(+), 18 deletions(-)

diff --git a/modules/linuxjobs/browser.py b/modules/linuxjobs/browser.py
index b578f7ce6b..3d0181474d 100644
--- a/modules/linuxjobs/browser.py
+++ b/modules/linuxjobs/browser.py
@@ -20,20 +20,27 @@
 
 from weboob.browser import PagesBrowser, URL
 
-from .pages import Page1, Page2
+from .pages import SearchPage, AdvertPage
+
+import urllib
 
 
 class LinuxJobsBrowser(PagesBrowser):
-    BASEURL = 'http://www.linuxjobs.com'
+    BASEURL = 'https://www.linuxjobs.fr'
+
+    advert_page = URL('/jobs/(?P<id>.+)', AdvertPage)
+    search_page = URL('/search/(?P<job>)', SearchPage)
 
-    page1 = URL('/page1\?id=(?P<id>.+)', Page1)
-    page2 = URL('/page2', Page2)
+    def get_job_advert(self, _id, advert):
+        self.advert_page.go(id=_id)
 
-    def get_stuff(self, _id):
-        self.page1.go(id=_id)
+        assert self.advert_page.is_here()
+        return self.page.get_job_advert(obj=advert)
 
-        assert self.page1.is_here()
-        self.page.do_stuff(_id)
+    def search_job(self, pattern=None):
+        if pattern is None:
+            return []
+        self.search_page.go(job=urllib.quote_plus(pattern.encode('utf-8')))
 
-        assert self.page2.is_here()
-        return self.page.do_more_stuff()
+        assert self.search_page.is_here()
+        return self.page.iter_job_adverts()
diff --git a/modules/linuxjobs/module.py b/modules/linuxjobs/module.py
index 0984cdf508..d1e8b3c870 100644
--- a/modules/linuxjobs/module.py
+++ b/modules/linuxjobs/module.py
@@ -55,7 +55,7 @@ def get_job_advert(self, _id, advert=None):
         :type advert: BaseJobAdvert
         :rtype: :class:`BaseJobAdvert` or None if not found.
         """
-        raise NotImplementedError()
+        return self.browser.get_job_advert(_id, advert)
 
     def search_job(self, pattern=None):
         """
@@ -65,4 +65,5 @@ def search_job(self, pattern=None):
         :type pattern: str
         :rtype: iter[:class:`BaseJobAdvert`]
         """
-        raise NotImplementedError()
+        for job_advert in self.browser.search_job(pattern):
+            yield job_advert
diff --git a/modules/linuxjobs/pages.py b/modules/linuxjobs/pages.py
index 9afe015a06..e6451f10ce 100644
--- a/modules/linuxjobs/pages.py
+++ b/modules/linuxjobs/pages.py
@@ -18,14 +18,31 @@
 # along with weboob. If not, see <http://www.gnu.org/licenses/>.
 
 
+from weboob.capabilities.job import BaseJobAdvert
 from weboob.browser.pages import HTMLPage
+from weboob.browser.elements import ItemElement, ListElement, method
+from weboob.browser.filters.standard import Regexp, CleanText, Date, Env, BrowserURL
+from weboob.browser.filters.html import Link, CleanHTML
 
+class AdvertPage(HTMLPage):
+    @method
+    class get_job_advert(ItemElement):
+        klass = BaseJobAdvert
 
-class Page1(HTMLPage):
-    def do_stuff(self, _id):
-        raise NotImplementedError()
+        obj_id = Env('id')
+        obj_url = BrowserURL('advert_page', id=Env('id'))
+        obj_title = CleanText('//title')
+        obj_job_name = CleanText('//title')
 
 
-class Page2(HTMLPage):
-    def do_more_stuff(self):
-        raise NotImplementedError()
+class SearchPage(HTMLPage):
+    @method
+    class iter_job_adverts(ListElement):
+        item_xpath = '//a[@class="list-group-item "]'
+
+        class item(ItemElement):
+            klass = BaseJobAdvert
+
+            obj_id = Regexp(Link('.'), '.*fr/jobs/(\d+)/.*')
+            obj_title = CleanText('h4/span[@class="job-title"]')
+            obj_society_name = CleanText('h4/span[@class="job-company"]')
-- 
GitLab