merge commit from histories_tobast into histories_models

crawl/crawl.py

@@ -1,5 +1,4 @@
 from threading import Thread
-from queue import Queue
 from urllib.robotparser import RobotFileParser
 from urllib.error import URLError
 from urllib.parse import urlparse
@@ -175,7 +174,7 @@ class CrawlingThread(Thread):
     """ A separate thread for the crawling task. This is needed to use asyncio,
     since the thread will need its own event loop. """
 
-    def __init__(self, url, queue):
+    def __init__(self, url, output_tree):
         engine_list = [engine.url for engine in SearchEngine.objects.all()]
         WebsiteScheduler.search_engines = engine_list
 
@@ -184,7 +183,7 @@ class CrawlingThread(Thread):
             randint(0, nb_fingerprint - 1)]
         self.headers = fingerprint.serialize_headers()
 
-        self.queue = queue
+        self.output_tree = output_tree
         super(CrawlingThread, self).__init__()
         self.url = url
 
@@ -193,7 +192,7 @@ class CrawlingThread(Thread):
 
         #tasks.append(async_crawler("http://plus.google.com/+Python"))
         #tasks.append(async_crawler('https://python.org/'))
-        tasks.append(async_crawler(self.url, self.queue, self.headers))
+        tasks.append(async_crawler(self.url, self.output_tree))
 
         loop = asyncio.new_event_loop()
         asyncio.set_event_loop(loop)
@@ -243,50 +242,60 @@ async def async_print(url):
         ))
 
 
-async def async_crawler(url, queue, headers=None):
+
+class CrawlElem:
+    ''' Describes a crawled element, to be assembled into a tree '''
+
+    def __init__(self, url, parent):
+        self.url = url
+        self.parent = parent
+
+async def async_crawler(url, output_tree, headers=None):
     if headers is None:
         headers = {}
     if 'User-Agent' not in headers:
         headers['User-Agent'] = settings.USER_AGENT
 
     user_agent = headers['User-Agent']
+    queued = [CrawlElem(url, None)]
+    crawled = set()
+    crawl_tree = []
 
-    queued = [url]
-    crawled = []
     while queued and (len(crawled) < HARD_LIMIT):
         async with aiohttp.ClientSession(headers=headers) as session:
             try:
-                url = queued.pop(0)
+                crawl_elt = queued.pop(0)
+                url = crawl_elt.url
             except IndexError:
                 print("Error queue is empty")
                 return crawled
+            crawled.add(url)
             parsed_url = urlparse(url)
             print("Crawling {}".format(url))
             html = await PageGetter(session, url, user_agent).get(ssl=False)
             if html:
+                crawl_tree.append(crawl_elt)
                 new_urls = url_getter(
                     html,
                     url,
                     parsed_url.scheme + "://" + parsed_url.netloc
                 )
-                crawled += [url]
                 if new_urls:
                     sampled = sample(
                         new_urls,
                         randrange(min(MAX_PER_PAGE, len(new_urls)))
                     )
-                    queued += [sample_url for sample_url in sampled if
-                              sample_url not in queued and sample_url not in
-                              crawled]
-            else:
-                print("No html received")
+                    queued += [
+                        CrawlElem(sample_url, crawl_elt)
+                        for sample_url in sampled
+                        if sample_url not in queued
+                        and sample_url not in crawled
+                    ]
     print(crawled)
-    queue.put(crawled)
+    output_tree += crawl_tree
 
 if __name__ == '__main__':
-    queue = Queue()
-    crawl = CrawlingThread(None,
-                           "https://google.com/search?q=fabriquer+masque+manif",
-                           ["https://google.com/search/"], queue)
+    crawl_tree = []
+    crawl = CrawlingThread(None, "https://google.com/search?q=fabriquer+masque+manif", crawl_tree)
     crawl.start()
     crawl.join()

profiles/management/commands/import_interests.py

@@ -5,6 +5,7 @@ import json
 from datetime import datetime
 from django.core.management.base import BaseCommand
 from django.db import models
+from django.core.exceptions import ObjectDoesNotExist
 from profiles.models import Keyword, Interest, Place, Website, Event
 
 def import_file(filename):
@@ -19,15 +20,14 @@ def import_interest(_interest):
     places = []
     websites = []
     for keyword in _interest.get("keywords", []):
-        if not Keyword.objects.get(keyword["keyword"]):
-            keywords.append(
-                Keyword(
-                    text=keyword["keyword"]
-                )
-            )
-            print("New keyword %s" % new_keywords)
-        else:
-            keywords.append(Keyword.objects.get(text=keyword["keyword"]))
+        try:
+            stored = Keyword.objects.get(text=keyword["keyword"])
+            keywords.append(stored)
+        except ObjectDoesNotExist:
+            new_keyword = Keyword(text=keyword["keyword"])
+            new_keyword.save()
+            keywords.append(new_keyword)
+            print("New keyword %s" % new_keyword)
     for place in _interest.get("places", []):
         places.append(Place.objects.get(name=place["place"]))
     for website in _interest.get("websites", []):
@@ -36,7 +36,9 @@ def import_interest(_interest):
     interest = Interest(
         name=_interest.get("name", ""),
     )
+    interest.save()
     for keyword in keywords:
+        print(keyword)
         interest.keywords.add(keyword)
     for place in places:
         interest.places.add(place)
@@ -46,4 +48,4 @@ def import_interest(_interest):
 
 class Command(BaseCommand):
     def handle(self, *args, **kwargs):
-        import_file("data/events.json")
+        import_file("data/interests.json")