Merge pull request #91 from pennlabs/studyspaces

Rebuild study spaces

.gitignore

@@ -60,3 +60,5 @@ MANIFEST
 docs/_build
 .pypirc
 .idea/
+*.swp
+*.swo

.travis.yml

@@ -4,7 +4,11 @@ python:
   - "2.7"
   - "3.4"
   - "3.5"
-script: nosetests
+install:
+  - pip install flake8 -r requirements.txt
+script:
+  - flake8 --max-line-length=150 --exclude=venv,docs,build --ignore=F401,F403,F405 .
+  - nosetests
 deploy:
   provider: pypi
   user: 'pennappslabs'

penn/__init__.py

@@ -1,4 +1,4 @@
-__version__ = '1.6.6'
+__version__ = '1.6.7'
 
 from .registrar import Registrar
 from .directory import Directory

penn/calendar3year.py

@@ -21,9 +21,9 @@ def pull_3year(self):
         """
         events = []
         r = requests.get(BASE_URL).text
-        l = r.split("\r\n")
+        lines = r.split("\r\n")
         d = {}
-        for line in l:
+        for line in lines:
             if line == "BEGIN:VEVENT":
                 d = {}
             elif line[:7] == "DTSTART":
@@ -32,7 +32,7 @@ def pull_3year(self):
                 d['start'] = start_date.strftime('%Y-%m-%d')
             elif line[:5] == "DTEND":
                 raw_date = line.split(":")[1]
-                end_date = datetime.datetime.strptime(raw_date,'%Y%m%d').date()
+                end_date = datetime.datetime.strptime(raw_date, '%Y%m%d').date()
                 d['end'] = end_date.strftime('%Y-%m-%d')
             elif line[:7] == "SUMMARY":
                 name = line.split(":")[1]

penn/dining.py

@@ -81,6 +81,7 @@ def get_meals(v2_response, building_id):
         meals.append({"tblStation": stations, "txtDayPartDescription": meal["label"]})
     return meals
 
+
 class DiningV2(WrapperBase):
     """The client for the Registrar. Used to make requests to the API.
 
@@ -141,6 +142,7 @@ def item(self, item_id):
         response = self._request(V2_ENDPOINTS['ITEMS'] + item_id)
         return response
 
+
 class Dining(WrapperBase):
     """The client for the Registrar. Used to make requests to the API.
 

penn/directory.py

@@ -38,7 +38,6 @@ def standardize(res):
         res['list_affiliation'] = res['list_affiliation'].replace('Faculty - ', '')
         return res
 
-
     def search(self, params, standardize=False):
         """Get a list of person objects for the given search params.
 

penn/studyspaces.py

@@ -1,93 +1,109 @@
-from bs4 import BeautifulSoup
 import requests
+import datetime
+import json
+import pytz
+import re
+import six
+
+from bs4 import BeautifulSoup
 
 
-BASE_URL = "http://libcal.library.upenn.edu"
+BASE_URL = "https://libcal.library.upenn.edu"
 
 
 class StudySpaces(object):
+    """Used for interacting with the UPenn library GSR booking system.
+
+    Usage::
+
+      >>> from penn import StudySpaces
+      >>> s = StudySpaces()
+    """
+
     def __init__(self):
         pass
 
+    def get_buildings(self):
+        """Returns a list of building IDs, building names, and services."""
+
+        soup = BeautifulSoup(requests.get("{}/spaces".format(BASE_URL)).content, "html5lib")
+        options = soup.find("select", {"id": "lid"}).find_all("option")
+        return [{"id": int(opt["value"]), "name": str(opt.text), "service": "libcal"} for opt in options]
+
+    @staticmethod
+    def parse_date(date):
+        """Converts library system dates into timezone aware Python datetime objects."""
+
+        date = datetime.datetime.strptime(date, "%Y-%m-%d %H:%M:%S")
+        return pytz.timezone("US/Eastern").localize(date)
+
     @staticmethod
-    def date_parse(original):
-        """Parses the date to dashed format.
-
-        :param original: string with date in the format MM/DD/YYYY.
-        """
-        l = original.split("-")
-        final = [l[1], l[2], l[0]]
-        return '-'.join(final)
-
-    def get_id_json(self):
-        """Makes JSON with each element associating URL, ID, and building
-        name.
-        """
-        group_study_codes = []
-        url = BASE_URL + "/booking/vpdlc"
-        soup = BeautifulSoup(requests.get(url).text, 'html5lib')
-        l = soup.find('select', {'id': 'lid'}).find_all('option')
-        for element in l:
-            if element['value'] != '0':
-                url2 = "{}/spaces?lid={}".format(BASE_URL, str(element['value']))
-                new_dict = {}
-                new_dict['id'] = int(str(element['value']))
-                new_dict['name'] = str(element.contents[0])
-                new_dict['url'] = url2
-                group_study_codes.append(new_dict)
-        return group_study_codes
-
-    def get_id_dict(self):
-        """Extracts the ID's of the room into a dictionary. Used as a
-        helper for the extract_times method.
-        """
-        group_study_codes = {}
-        url = BASE_URL + "/booking/vpdlc"
-        soup = BeautifulSoup(requests.get(url).text, 'html5lib')
-        options = soup.find('select', {'id': 'lid'}).find_all('option')
-        for element in options:
-            if element['value'] != '0':
-                group_study_codes[int(str(element['value']))] = str(element.contents[0])
-        return group_study_codes
-
-    def extract_times(self, id, date, name):
-        """Scrapes the avaiable rooms with the given ID and date.
-
-        :param id: the ID of the building
-        :param date: the date to acquire available rooms from
-        :param name: the name of the building; obtained via get_id_dict
-        """
-        url = BASE_URL + "/rooms_acc.php?gid=%s&d=%s&cap=0" % (int(id), date)
-        soup = BeautifulSoup(requests.get(url).text, 'html5lib')
-
-        time_slots = soup.find_all('form')
-        unparsed_rooms = time_slots[1].contents[2:-2]
-
-        roomTimes = []
-
-        for i in unparsed_rooms:
-            room = BeautifulSoup(str(i), 'html5lib')
-            try:
-                # extract room names
-                roomName = room.fieldset.legend.h2.contents[0]
-            except AttributeError:
-                # in case the contents aren't a list
-                continue
-            newRoom = str(roomName)[:-1]
-            times = []
-
-            filtered = room.fieldset.find_all('label')
-
-            for t in filtered:
-                # getting the individual times for each room
-                dict_item = {}
-                dict_item['room_name'] = newRoom
-                time = str(t).split("\t\t\t\t\t")[2][1:-1]
-                times.append(time)
-                startAndEnd = time.split(" - ")
-                dict_item['start_time'] = startAndEnd[0].upper()
-                dict_item['end_time'] = startAndEnd[1].upper()
-                roomTimes.append(dict_item)
-                dict_item['date'] = self.date_parse(date)
-                dict_item['building'] = name
-        return roomTimes
+    def get_room_id_name_mapping(building):
+        """ Returns a dictionary mapping id to name, thumbnail, and capacity. """
+
+        data = requests.get("{}/spaces?lid={}".format(BASE_URL, building)).content.decode("utf8")
+        # find all of the javascript room definitions
+        out = {}
+        for item in re.findall(r"resources.push\(((?s).*?)\);", data, re.MULTILINE):
+            # parse all of the room attributes
+            items = {k: v for k, v in re.findall(r'(\w+?):\s*(.*?),', item)}
+
+            # room name formatting
+            title = items["title"][1:-1]
+            title = title.encode().decode("unicode_escape" if six.PY3 else "string_escape")
+            title = re.sub(r" \(Capacity [0-9]+\)", r"", title)
+
+            # turn thumbnail into proper url
+            thumbnail = items["thumbnail"][1:-1]
+            if thumbnail:
+                thumbnail = "https:" + thumbnail
+
+            room_id = int(items["eid"])
+            out[room_id] = {
+                "name": title,
+                "thumbnail": thumbnail or None,
+                "capacity": int(items["capacity"])
+            }
+        return out
+
+    def get_rooms(self, building, start, end):
+        """Returns a dictionary matching all rooms given a building id and a date range."""
+
+        if start.tzinfo is None:
+            start = pytz.timezone("US/Eastern").localize(start)
+        if end.tzinfo is None:
+            end = pytz.timezone("US/Eastern").localize(end)
+
+        mapping = self.get_room_id_name_mapping(building)
+        room_endpoint = "{}/process_equip_p_availability.php".format(BASE_URL)
+        data = {
+            "lid": building,
+            "gid": 0,
+            "start": start.strftime("%Y-%m-%d"),
+            "end": (end + datetime.timedelta(days=1)).strftime("%Y-%m-%d"),
+            "bookings": []
+        }
+        resp = requests.post(room_endpoint, data=json.dumps(data), headers={'Referer': "{}/spaces?lid={}".format(BASE_URL, building)})
+        rooms = {}
+        for row in resp.json():
+            room_id = int(row["resourceId"][4:])
+            if room_id not in rooms:
+                rooms[room_id] = []
+            room_start = self.parse_date(row["start"])
+            room_end = self.parse_date(row["end"])
+            if start <= room_start <= end:
+                rooms[room_id].append({
+                    "start": room_start.isoformat(),
+                    "end": room_end.isoformat(),
+                    "available": row["status"] == 0
+                })
+        out = []
+        for k, v in rooms.items():
+            item = {
+                "room_id": k,
+                "times": v
+            }
+            if k in mapping:
+                item.update(mapping[k])
+            out.append(item)
+        return out

requirements.txt

@@ -1,6 +1,8 @@
-nose==1.3.4
-requests==2.4.3
-beautifulsoup4==4.4.1
+beautifulsoup4==4.6.0
 html5lib==0.999
 mock==2.0.0
 nameparser==0.4.0
+nose==1.3.7
+pytz==2017.3
+requests==2.4.3
+six==1.11.0

setup.py

@@ -7,7 +7,7 @@
     url='https://github.com/pennlabs/penn-sdk-python',
     author='Penn Labs',
     author_email='[email protected]',
-    version='1.6.6',
+    version='1.6.7',
     packages=['penn'],
     license='MIT',
     package_data={

tests/calendar_test.py

@@ -9,34 +9,34 @@ def setUp(self):
         self.calendar = Calendar()
 
     def test_pull(self):
-        l = self.calendar.pull_3year()
-        ok_(len(l) > 0)
-        for event in l:
+        cal = self.calendar.pull_3year()
+        ok_(len(cal) > 0)
+        for event in cal:
             ok_(len(event) == 3)
 
     def test_date(self):
-        l = self.calendar.pull_3year()
-        ok_(len(l) > 0)
-        for event in l:
+        cal = self.calendar.pull_3year()
+        ok_(len(cal) > 0)
+        for event in cal:
             ok_(isinstance(event['name'], str))
             if event['name'] == "Independence Day Observed (no classes)":
                 independence = event['start']
                 d = datetime.datetime.strptime(independence, '%Y-%m-%d').date()
                 ok_(d.month == 7)
 
     def test_name(self):
-        l = self.calendar.pull_3year()
-        ok_(len(l) > 0)
-        for event in l:
+        cal = self.calendar.pull_3year()
+        ok_(len(cal) > 0)
+        for event in cal:
             ok_(isinstance(event['name'], str))
             start = datetime.datetime.strptime(event['start'], '%Y-%m-%d').date()
             end = datetime.datetime.strptime(event['end'], '%Y-%m-%d').date()
             ok_((end - start).total_seconds() >= 0)
 
     def test_chrono(self):
-        l = self.calendar.pull_3year()
-        ok_(len(l) > 0)
-        for i, event in enumerate(l[:-1]):
+        cal = self.calendar.pull_3year()
+        ok_(len(cal) > 0)
+        for i, event in enumerate(cal[:-1]):
             start = datetime.datetime.strptime(event['start'], '%Y-%m-%d').date()
-            nextstart = datetime.datetime.strptime(l[i]['start'], '%Y-%m-%d').date()
+            nextstart = datetime.datetime.strptime(cal[i]['start'], '%Y-%m-%d').date()
             ok_((nextstart - start).total_seconds() >= 0)

tests/dining_test.py

@@ -18,8 +18,7 @@ def test_dining(self):
         self.assertTrue(len(venues) > 0)
         id = str(venues["id"])
         data = self.din.menu_daily(id)
-        self.assertTrue(
-                len(data["result_data"]["Document"]["tblMenu"]["tblDayPart"][0]) >= 2)
+        self.assertTrue(len(data["result_data"]["Document"]["tblMenu"]["tblDayPart"][0]) >= 2)
 
     def test_dining_normalization(self):
         data = self.din.menu_daily("593")

tests/studyspaces_test.py

@@ -1,30 +1,24 @@
+import datetime
+import pytz
+
 from nose.tools import ok_
 from penn import StudySpaces
-import datetime
 
 
 class TestStudySpaces():
 
     def setUp(self):
         self.studyspaces = StudySpaces()
 
-    def test_json(self):
-        json_id = self.studyspaces.get_id_json()
-        ok_(len(json_id) > 0)
-        for i in json_id:
-            ok_(i['id'] > 0)
-            ok_(i['name'] != '')
-            ok_(i['url'] != '')
+    def test_buildings(self):
+        buildings = self.studyspaces.get_buildings()
+        ok_(len(buildings) > 0)
+
+    def test_room_name_mapping(self):
+        mapping = self.studyspaces.get_room_id_name_mapping(2683)
+        ok_(len(mapping) > 0)
 
-    def test_extraction(self):
-        dict_id = self.studyspaces.get_id_dict()
-        ok_(len(dict_id) > 0)
-        d = datetime.datetime.now() + datetime.timedelta(days=1)
-        next_date = d.strftime("%Y-%m-%d")
-        s = self.studyspaces.extract_times(1799, next_date, "Van Pelt-Dietrich Library Center Group Study Rooms")
-        for i in s:
-            ok_("building" in i)
-            ok_("start_time" in i)
-            ok_("end_time" in i)
-            ok_("date" in i)
-            ok_("room_name" in i)
+    def test_rooms(self):
+        now = pytz.timezone("US/Eastern").localize(datetime.datetime.now())
+        rooms = self.studyspaces.get_rooms(2683, now, now + datetime.timedelta(days=3))
+        ok_(len(rooms) > 0)