chiark / gitweb /
WIP scraper still not working right
authorIan Jackson <ian@liberator.relativity.greenend.org.uk>
Mon, 31 Aug 2009 15:04:52 +0000 (16:04 +0100)
committerIan Jackson <ian@liberator.relativity.greenend.org.uk>
Mon, 31 Aug 2009 15:04:52 +0000 (16:04 +0100)
yarrg/yppedia-ocean-scraper

index c99a84e..d6d2f24 100755 (executable)
@@ -12,18 +12,21 @@ import re as regexp
 from BeautifulSoup import BeautifulSoup
 
 ocean = 'Opal'
+soup = None
 
-url = 'http://yppedia.puzzlepirates.com/%s_Ocean' % urllib.quote(ocean,'')
-dataf = urllib2.urlopen(url)
+def debug(k,v):
+       pass
 
-soup = BeautifulSoup(dataf)
+def fetch():
+       global soup
+       url = ('http://yppedia.puzzlepirates.com/%s_Ocean' %
+               urllib.quote(ocean,''))
+       dataf = urllib2.urlopen(url)
+       soup = BeautifulSoup(dataf)
 
-title_arch_re = regexp.compile('\\S+ Archipelago \\((\\S+)\\)$')
-title_any_re = regexp.compile('(\\S.*\\S) \((\\S+)\\)$')
-href_img_re = regexp.compile('\\.png$')
 
 def title_arch_arch(t):
-       print 'checking ',t
+       debug('checking',t)
        if t is None: return None
        m = title_arch_re.match(t)
        if not m: return None
@@ -34,33 +37,49 @@ def title_arch_ok(t):
        if a is None: return False
        return a == ocean
 
-firstarch = soup.find('a', attrs = {'title': title_arch_ok})
-print 'fa',`firstarch`
-
-archestable = firstarch.findParent(
-       lambda u: u.name == 'table' and
-               len(u.findAll('a', attrs = {'title': title_arch_ok})) > 1
-       )
-
-print 'at',`archestable`
-
-arches = archestable.findAll('a', attrs = {'title': title_arch_ok})
-print 'ac', `arches`
-
-arches = map((lambda u: u.findParent(
-       lambda v: len(v.findAll(text= regexp.compile('.*Large'))) > 0
-       )), arches)
-print 'ac2', `arches`
-
-for arch in arches:
-       links = arch.findAll('a', href=True)
-       print 'links', `links`
-       a = title_arch_arch(links[0]['title'])
-       assert(a)
-       print a
-       for link in links[1:]:
-               print 'link', `link`
-               if href_img_re.search(link['href']): continue
-               m = title_any_re.match(link['title'])
-               assert(m.group(2) == ocean)
-               print m.group(1)
+title_arch_re = regexp.compile('\\S+ Archipelago \\((\\S+)\\)$')
+title_any_re = regexp.compile('(\\S.*\\S) \((\\S+)\\)$')
+href_img_re = regexp.compile('\\.png$')
+
+def parse():
+       firstarch = soup.find('a', attrs = {'title': title_arch_ok})
+       debug('fa',firstarch)
+
+       def findall_title_arch_ok(t):
+               return t.findAll('a', attrs = {'title': title_arch_ok})
+
+       def is_archestable(u):
+               if u.name != 'table': return False
+               return len(findall_title_arch_ok(u)) > 1
+
+       archestable = firstarch.findParent(is_archestable)
+       debug('at',archestable)
+
+       arches = findall_title_arch_ok(archestable)
+       debug('ac',arches)
+
+       def is_island(v):
+               return len(v.findAll(text = regexp.compile('.*Large'))) > 0
+       def arch_up_map(u):
+               return u.findParent(is_island)
+       arches = map(arch_up_map, arches)
+       debug('ac2',arches)
+
+       for arch in arches:
+               links = arch.findAll('a', href=True)
+               debug('links',links)
+               a = title_arch_arch(links[0]['title'])
+               assert(a)
+               print 'arch', a
+               for link in links[1:]:
+                       debug('link',link)
+                       if href_img_re.search(link['href']): continue
+                       m = title_any_re.match(link['title'])
+                       assert(m.group(2) == ocean)
+                       print 'island', m.group(1)
+
+def main():
+       fetch()
+       parse()
+
+main()