+class FlagInfo(SomethingSoupInfo):
+ # Public data members (after init):
+ #
+ # name # string
+ #
+ # relations[n] = (otherflagname, otherflagid, [stringfromyoweb],
+ # thisdeclaring, otherdeclaringmin, otherdeclaringmax)
+ # # where {this,other}declaring{,min,max} are:
+ # # -1 {this,other} is declaring war
+ # # 0 {this,other} is not doing either
+ # # +1 {this,other} is allying
+ # relation_byname[otherflagname] = relations[some_n]
+ # relation_byid[otherflagname] = relations[some_n]
+ #
+ # islands[n] = (islandname, islandid)
+ #
+ def __init__(self, flagid, max_age=600):
+ SomethingSoupInfo.__init__(self,
+ 'flag/info.wm?flagid=', flagid, max_age)
+ self._find_flag()
+
+ def _find_flag(self):
+ font2 = self._soup.find('font',{'size':'+2'})
+ self.name = font2.find('b').contents[0]
+
+ self.relations = [ ]
+ self.relation_byname = { }
+ self.relation_byid = { }
+ self.islands = [ ]
+
+ magnate = self._soup.find('img',{'src':
+ '/yoweb/images/repute-MAGNATE.png'})
+ warinfo = (magnate.findParent('table').findParent('tr').
+ findNextSibling('tr').findNext('td',{'align':'left'}))
+
+ def warn(m):
+ print >>sys.stderr, 'WARNING: '+m
+
+ def wi_warn(head, waritem):
+ warn('unknown warmap item: %s: %s' %
+ (`head`, ``waritem``))
+
+ def wihelp_item(waritem, thing):
+ if waritem.name == 'a':
+ url = waritem.get('href', None)
+ if url is None:
+ return ('no url for '+thing,None,None)
+ else:
+ hr = waritem.find('a',{'href':True})
+ if not hr: return ('no a for '+thing,None,None)
+ url = hr['href']
+ m = regexp.search('\?'+thing+'id=(\d+)$', url)
+ if not m: return ('no '+thing+'id',None,None)
+ tid = m.group(1)
+ tname = m.string
+ if tname is None:
+ return (thing+' name not just string',None,None)
+ return (None,tid,tname)
+
+ def wi_alwar(head, waritem, thisdecl, othermin, othermax):
+ (err,flagid,flagname) = wihelp_item(waritem,'flag')
+ if err: return err
+ rel = self.relation_byid.get(flagid, None)
+ if rel: return 'flag id twice!'
+ if flagname in self.relation_byname:
+ return 'flag name twice!'
+ rel = (flagname,flagid,[], thisdecl,othermin,othermax)
+ self.relations.append(rel)
+ self.relation_byid[flagid] = rel
+ self.relation_byname[flagid] = rel
+
+ def wi_isle(head, waritem):
+ (err,isleid,islename) = wihelp_item(waritem,'island')
+ if err: return err
+ self.islands.append((isleid,islename))
+
+ warmap = {
+ 'Allied with': (wi_alwar,+1,+1,+1),
+ 'Declaring war against': (wi_alwar,-1, 0,+1),
+ 'At war with': (wi_alwar,-1,-1,-1),
+ 'Trying to form an alliance with': (wi_alwar,+1,-1,0),
+ 'Islands controlled by this flag': (wi_isle,),
+ }
+
+ how = (wi_warn, None)
+
+ for waritem in warinfo.contents:
+ debug('WARITEM '+``waritem``)
+ if isinstance(waritem, unicode):
+ waritem = waritem.strip()
+ if waritem: warn('unknown waritem '+``waritem``)
+ continue
+ if waritem.name == 'br':
+ continue
+ if waritem.name == 'b':
+ head = ''.join(waritem.findAll(text=True))
+ head = regexp.sub('\\s+', ' ', head).strip()
+ head = head.rstrip(':')
+ how = (head,) + warmap.get(head, (wi_warn,))
+ continue
+ debug('WARHOW %s(%s, waritem, *%s)' %
+ (how[1], `how[0]`, `how[2:]`))
+ bad = how[1](how[0], waritem, *how[2:])
+ if bad:
+ warn('bad waritem %s: %s: %s' % (`how[0]`,
+ bad, ``waritem``))
+
+ def __str__(self):
+ return `(self.name, self.islands, self.relations)`
+
+#---------- scraper for ocean info incl. embargoes etc. ----------
+
+class IslandInfo():
+ def __init__(self, ocean, islename):
+ self.ocean = ocean
+ self.name = islename
+ def collect(self):
+ pass
+ def yppedia_dataf(self):
+ def q(x): return urllib.quote(x.replace(' ','_'))
+ url_rhs = q(self.name) + '_(' + q(self.ocean) + ')'
+ if opts.localhtml is None:
+ url = 'http://yppedia.puzzlepirates.com/' + url_rhs
+ debug('IslandInfo retrieving YPP '+url);
+ return urllib.urlopen(url)
+ else:
+ return file(opts.localhtml + '/' + url_rhs, 'r')
+ def yoweb_url(self):
+ soup = BeautifulSoup(self.yppedia_dataf())
+ content = soup.find('div', attrs = {'id': 'content'})
+ yoweb_re = regexp.compile('^http://\w+\.puzzlepirates\.com/'+
+ 'yoweb/island/info\.wm\?islandid=\d+$')
+ a = soup.find('a', attrs = { 'href': yoweb_re })
+ if a is None: return None
+ return a['href']
+ def ruling_flag_id(self):
+ yo = self.yoweb_url()
+ if yo is None: return None
+ dataf = fetcher.fetch(yo, 600)
+ soup = BeautifulSoup(dataf)
+ ruler_re = regexp.compile('http://\w+\.puzzlepirates\.com/'+
+ 'yoweb/flag/info\.wm\?flagid=(\d+)$')
+ ruler = soup.find('a', attrs = { 'href': ruler_re })
+ if not ruler: return None
+ m = ruler_re.find(ruler['href'])
+ return m.group(1)
+
+class OceanInfo():
+ # Public data attributes (valid after collect()):
+ # oi.islands[islename] = IslandInfo(...)
+ # oi.arches[archname][islename] = IslandInfo(...)
+ def __init__(self):
+ self.isleclass = IslandInfo
+ self.ocean = fetcher.ocean.lower().capitalize()
+ def collect(self):
+ cmdl = ['./yppedia-ocean-scraper']
+ if opts.localhtml is not None:
+ cmdl += ['--local-html-dir',opts.localhtml]
+ cmdl += [self.ocean]
+ debug('OceanInfo collect running ' + `cmdl`)
+ oscraper = subprocess.Popen(
+ cmdl,
+ stdout = subprocess.PIPE,
+ cwd = yppsc_dir()+'/yarrg',
+ shell=False, stderr=None,
+ )
+ h = oscraper.stdout.readline()
+ debug('OceanInfo collect h '+`h`)
+ assert(regexp.match('^ocean ', h))
+ arch_re = regexp.compile('^ (\S.*)')
+ island_re = regexp.compile('^ (\S.*)')
+
+ self.islands = { }
+ self.arches = { }
+ archname = None
+
+ for l in oscraper.stdout:
+ debug('OceanInfo collect l '+`l`)
+ l = l.rstrip('\n')
+ m = island_re.match(l)
+ if m:
+ assert(archname is not None)
+ islename = m.group(1)
+ isle = self.isleclass(self.ocean, islename)
+ isle.arch = archname
+ self.islands[islename] = isle
+ self.arches[archname][islename] = isle
+ continue
+ m = arch_re.match(l)
+ if m:
+ archname = m.group(1)
+ assert(archname not in self.arches)
+ self.arches[archname] = { }
+ continue
+ assert(False)
+ oscraper.wait()
+ assert(oscraper.returncode == 0)
+