wip

src/agenda_culturel/import_tasks/custom_extractors/acolab.py

@@ -1,7 +1,4 @@
-import icalendar
-import warnings
-
-import bbcode
+import re
 
 from datetime import datetime, date, timedelta
 from bs4 import BeautifulSoup, MarkupResemblesLocatorWarning
@@ -12,6 +9,19 @@ from celery.utils.log import get_task_logger
 
 logger = get_task_logger(__name__)
 
+class Timeslot:
+    def __init__(self, start_time, end_time):
+        self.start_time = start_time
+        self.end_time = end_time
+
+    def merge(self, hours):
+        self.start_time = min(self.start_time, hours.start_time)
+
+        if not hours.end_time is None:
+            if not self.end_time is None:
+                self.end_time = max(self.end_time, hours.end_time)
+            else:
+                self.end_time = hours.end_time
 
 class CExtractor(Extractor):
 
@@ -32,6 +42,21 @@ class CExtractor(Extractor):
 
         return (year, month)
     
+    def find_hours(text):
+        text = re.split(r"[ -/=>]+", text)
+        text = [Extractor.parse_french_time(k) for k in text]
+        text = [k for k in text if not k is None]
+        match len(text):
+            case 0:
+                return None
+            case 1:
+                return Timeslot(text[0], None)
+            case 2:
+                return Timeslot(text[0], text[1])
+            case _:
+                return None
+
+    
     def is_nickname(text):
         return '@' in text
     
@@ -42,7 +67,27 @@ class CExtractor(Extractor):
             if word in text:
                 return True
         return False
+    
+    def find_timeslot(text):
+        text = re.sub(' +', ' ', text).split(' ')
+        day_name = text[0]
+        day_num = text[1]
+        hours = text[2]
 
+        if not Extractor.guess_day_name(day_name):
+            return None
+        
+        day_num = [c for c in re.split(r'\D+', day_num) if c != ""]
+        if len(day_num) == 0:
+            return None
+        day_num = int(day_num[0])
+
+        hours = CExtractor.find_hours(hours)
+        if hours is None:
+            return None
+        
+        return (day_num, hours)
+        
 
     #['Samedi 12', '@Manon', '14:30-18:00', 'Dimanches 13', '@gaeldu63', '14h30 (j utilise la scie a format)']
     #['Mercredi 16 :']
@@ -52,25 +97,22 @@ class CExtractor(Extractor):
         result = []
 
         date = None
-        tstart = None
-        tend = None
+        slot = None
         is_open = False
 
         # for each element in the paragraph
         for e in p.stripped_strings:
-            day = CExtractor.find_day_name(e)
+            day = CExtractor.find_timeslot(e)
             if not day is None:
                 if not date is None and is_open:
                     # we reach a new day
-                    result.append((date, tstart, tend))
+                    result.append((date, slot))
                 if isinstance(day, tuple):
                     date = day[0]
-                    tstart = day[1]
-                    tend = day[2]
+                    slot = day[1]
                 else:
                     date = day
-                    tstart = None
-                    tend = None
+                    slot = None
                 is_open = False
                 continue
             elif not is_open:
@@ -83,15 +125,10 @@ class CExtractor(Extractor):
             
             hours = CExtractor.find_hours(e)
             if not hours is None:
-                # we found hours
-                if tstart is None:
-                    tstart = hours[0]
+                if slot is None:
+                    slot = hours
                 else:
-                    tstart = min(tstart, hours[0])
-                if tend is None:
-                    tend = hours[1]
-                else:
-                    tend = max(tend, hours[1])
+                    slot.merge(hours)
                 continue
             
             if CExtractor.is_canceled(e):
@@ -100,7 +137,7 @@ class CExtractor(Extractor):
 
         if not date is None and is_open:
             # we reach a new day
-            result.append((date, tstart, tend))
+            result.append((date, slot))
 
         return result
         # [(10, time(14, 0, 0), time(17, 0, 0)), ]
@@ -124,7 +161,7 @@ class CExtractor(Extractor):
             # annule
             # menage
             for p in description.select('p'):
-                CExtractor.find_time_slots(p)
+                CExtractor.find_timeslots(p)
                 
                 if not '@' in p.text:
                     continue