WIP ocean scraper seems to mostly work, need to do arg parsing and IO

[ypp-sc-tools.web-live.git] / yarrg / yppedia-ocean-scraper

diff --git a/yarrg/yppedia-ocean-scraper b/yarrg/yppedia-ocean-scraper
index c99a84efe5dbfcb0dc556886dfc7d7797e30b387..d53e2365da98491bf1b6204a131295cf4fef3d5c 100755 (executable)
--- a/yarrg/yppedia-ocean-scraper
+++ b/yarrg/yppedia-ocean-scraper
@@ -12,55 +12,77 @@ 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):
+#      print 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+)\\)$')
+
+title_arch_re = regexp.compile('(\\S+) Archipelago \\((\\S+)\\)$')
 title_any_re = regexp.compile('(\\S.*\\S) \((\\S+)\\)$')
 href_img_re = regexp.compile('\\.png$')
 
-def title_arch_arch(t):
-       print 'checking ',t
-       if t is None: return None
+def title_arch_info(t):
+       # returns (arch,ocean)
+       debug('checking',t)
+       if t is None: return (None,None)
        m = title_arch_re.match(t)
-       if not m: return None
-       return m.group(1)
+       if not m: return (None,None)
+       return m.groups()
 
 def title_arch_ok(t):
-       a = title_arch_arch(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)
+       (a,o) = title_arch_info(t)
+       if o is None: return False
+       return o == ocean
+
+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('table', attrs={'border':'1'})
+       debug('at',archestable)
+
+       arches = []
+       for row in archestable.findAll('tr',recursive=False):
+               arches += row.findAll('td',recursive=False)
+       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)
+
+       for arch in arches:
+               links = arch.findAll('a', href=True)
+               debug('links',links)
+               if not links: continue
+               (a,o) = title_arch_info(links[0]['title'])
+               assert(o == ocean)
+               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()