Merge branch 'feature/parsing-scheme' into develop

Fourmi.py

@@ -9,23 +9,36 @@ from scrapy.crawler import Crawler
 from scrapy import log, signals
 from FourmiCrawler.spider import FourmiSpider
 from scrapy.utils.project import get_project_settings
+import os, inspect, re
 
+def load_parsers(rel_dir="FourmiCrawler/parsers"):
+	path = os.path.dirname(os.path.abspath(__file__))
+	path += "/" + rel_dir
+	parsers = []
 
-def setup_crawler(searchable):
-    # [TODO] - Initiate all parsers for the different websites and get
-    # allowed URLs.
-    spider = FourmiSpider(compound=searchable)
-    settings = get_project_settings()
-    crawler = Crawler(settings)
-    crawler.signals.connect(reactor.stop, signal=signals.spider_closed)
-    crawler.configure()
-    crawler.crawl(spider)
-    crawler.start()
+	for py in [f[:-3] for f in os.listdir(path) if f.endswith('.py') and f != '__init__.py']:
+		mod = __import__('.'.join([rel_dir.replace("/", "."), py]), fromlist=[py])
+		classes = [getattr(mod, x) for x in dir(mod) if inspect.isclass(getattr(mod, x))]
+		for cls in classes:
+			if re.match(path + "/*", inspect.getfile(cls)):
+				parsers.append(cls()) # [review] - Would we ever need arguments for the parsers?
+	return parsers
+
+def setup_crawler(searchables):
+	spider = FourmiSpider(compounds=searchables)
+	spider.add_parsers(load_parsers())
+	settings = get_project_settings()
+	crawler = Crawler(settings)
+	crawler.signals.connect(reactor.stop, signal=signals.spider_closed)
+	crawler.configure()
+	crawler.crawl(spider)
+	crawler.start()
 
 
 def start():
-    setup_crawler("Methane")
-    log.start()
-    reactor.run()
+	setup_crawler(["Methane"])
+	log.start()
+	reactor.run()
+
 
 start()

FourmiCrawler/parsers/parser.py

@@ -1,9 +1,21 @@
 from scrapy import log
+# from scrapy.http import Request
 
 
 class Parser:
-    website = "http://localhost/*"
+	'''
+	website should be an regular expression of the urls of request the parser is able to parse.
+	'''
+	website = "http://something/*"
+	__spider = None
 
-    def parse(self, reponse):
-        log.msg("The parse function of the empty parser was used.", level=log.Warning)
-        pass
+	def parse(self, reponse):
+		log.msg("The parse function of the empty parser was used.", level=log.WARNING)
+		pass
+
+	def new_compound_request(self, compound):
+		# return Request(url=self.website[:-1] + compound, callback=self.parse)
+		pass
+
+	def set_spider(self, spider):
+		self.__spider = spider

FourmiCrawler/spider.py

@@ -1,19 +1,43 @@
 from scrapy.spider import Spider
+from scrapy import log
+import re
 
 
 class FourmiSpider(Spider):
 	name = "FourmiSpider"
+	__parsers = []
+	synonyms = []
 
-	def __init__(self, compound=None, *args, **kwargs):
+	def __init__(self, compounds=None, *args, **kwargs):
 		super(FourmiSpider, self).__init__(*args, **kwargs)
-		self.synonyms = [compound]
+		if isinstance(compounds, list):
+			self.synonyms.extend(compounds)
+		else:
+			self.synonyms.append(compounds)
 
+	def parse(self, reponse):
+		for parser in self.__parsers:
+			if re.match(parser.website, reponse.url):
+				log.msg("Url: " + reponse.url + " -> Parser: " + parser.website, level=log.DEBUG)
+				return parser.parse(reponse)
+		return None
 
-def parse(self, reponse):
-	# [TODO] - This function should delegate it's functionality to other
-	# parsers.
-	pass
+	def get_synonym_requests(self, compound):
+		requests = []
+		for parser in self.__parsers:
+			requests.append(parser.new_compound_request(compound))
+		return requests
 
+	def start_requests(self):
+		requests = []
+		for synonym in self.synonyms:
+			requests.extend(self.get_synonym_requests(synonym))
+		return requests
 
-def add_parser(self, parser):
-	self.parsers.add(parser)
+	def add_parsers(self, parsers):
+		for parser in parsers:
+			self.add_parser(parser)
+
+	def add_parser(self, parser):
+		self.__parsers.append(parser)
+		parser.set_spider(self)