#!/usr/bin/python
+import signal
+signal.signal(signal.SIGINT, signal.SIG_DFL)
+
import os
import time
import urllib
import urllib2
import errno
+import sys
import re as regexp
+from optparse import OptionParser
from BeautifulSoup import BeautifulSoup
-max_age = 120
-ocean = 'ice'
-
-now = time.time()
+opts = None
-duties = ('Swordfighting/Bilging/Sailing/Rigging/Navigation'+
- '/Battle Navigation/Carpentry/Rumble/Treasure Haul').split('/')
+duties = ('Swordfighting/Bilging/Sailing/Rigging/Navigating'+
+ '/Battle Navigation/Gunning/Carpentry/Rumble/Treasure Haul'+
+ '/Drinking/Spades/Hearts/Treasure Drop/Poker/Distilling'+
+ '/Alchemistry/Shipwrightery/Blacksmithing/Foraging').split('/')
-standings = ('Able/Distinguished/Respected/Master/Renowned'+
+standingvals = ('Able/Distinguished/Respected/Master/Renowned'+
'/Grand-Master/Legendary/Ultimate').split('/')
-def fetch(url):
- cache_corename = urllib.quote_plus(url)
- cache_basename = "#%s#" % cache_corename
- try: f = file(cache_basename, 'r')
- except (OSError,IOError), oe:
- if oe.errno != errno.ENOENT: raise
- f = None
- if f is not None:
- s = os.fstat(f.fileno())
- if now > s.st_mtime + max_age:
+pirate_ref_re = regexp.compile('^/yoweb/pirate\\.wm')
+
+def debug(m):
+ if opts.debug:
+ print >>sys.stderr, m
+
+class Fetcher:
+ def __init__(self, ocean, cachedir):
+ debug('Fetcher init %s' % cachedir)
+ self.ocean = ocean
+ self.cachedir = cachedir
+ try: os.mkdir(cachedir)
+ except (OSError,IOError), oe:
+ if oe.errno != errno.EEXIST: raise
+ self._cache_scan(time.time())
+
+ def _cache_scan(self, now):
+ # returns list of ages, unsorted
+ ages = []
+ debug('Fetcher scan_cache')
+ for leaf in os.listdir(self.cachedir):
+ if not leaf.startswith('#'): continue
+ path = self.cachedir + '/' + leaf
+ try: s = os.stat(path)
+ except (OSError,IOError), oe:
+ if oe.errno != errno.ENOENT: raise
+ continue
+ age = now - s.st_mtime
+ if age > opts.max_age:
+ debug('Fetcher expire %d %s' % (age, path))
+ try: os.remove(path)
+ except (OSError,IOError), oe:
+ if oe.errno != errno.ENOENT: raise
+ continue
+ ages.append(age)
+ return ages
+
+ def _rate_limit_cache_clean(self, now):
+ ages = self._cache_scan(now)
+ ages.sort()
+ debug('Fetcher ages ' + `ages`)
+ min_age = 1
+ need_wait = 0
+ for age in ages:
+ if age < min_age:
+ debug('Fetcher morewait min=%d age=%d' %
+ (min_age, age))
+ need_wait = max(need_wait, min_age - age)
+ min_age *= 2
+ min_age += 1
+ if need_wait > 0:
+ debug('Fetcher wait %d' % need_wait)
+ time.sleep(need_wait)
+
+ def fetch(self, url):
+ debug('Fetcher fetch %s' % url)
+ cache_corename = urllib.quote_plus(url)
+ cache_item = "%s/#%s#" % (self.cachedir, cache_corename)
+ try: f = file(cache_item, 'r')
+ except (OSError,IOError), oe:
+ if oe.errno != errno.ENOENT: raise
f = None
- if f is not None:
- data = f.read()
- f.close()
- else:
+ now = time.time()
+ if f is not None:
+ s = os.fstat(f.fileno())
+ if now > s.st_mtime + opts.max_age:
+ debug('Fetcher stale')
+ f = None
+ if f is not None:
+ data = f.read()
+ f.close()
+ debug('Fetcher cached')
+ return data
+
+ debug('Fetcher fetch')
+ self._rate_limit_cache_clean(now)
+
stream = urllib2.urlopen(url)
data = stream.read()
- cache_ourname = "#%s~%d#" % (cache_corename, os.getpid())
- f = file(cache_ourname, 'w')
+ cache_tmp = "%s/#%s~%d#" % (
+ self.cachedir, cache_corename, os.getpid())
+ f = file(cache_tmp, 'w')
f.write(data)
f.close()
- os.rename(cache_ourname, cache_basename)
- return data
+ os.rename(cache_tmp, cache_item)
+ debug('Fetcher stored')
+ return data
-def yoweb_fetch(kind, tail):
- url = 'http://%s.puzzlepirates.com/yoweb/%s%s' % (ocean, kind, tail)
- return fetch(url)
+ def yoweb(self, kind, tail):
+ url = 'http://%s.puzzlepirates.com/yoweb/%s%s' % (
+ self.ocean, kind, tail)
+ return self.fetch(url)
-class PirateInfo:
- def __init__(self, pirate):
- html = yoweb_fetch('pirate.wm?target=', pirate)
- soup = BeautifulSoup(html,
-# convertEntities=BeautifulSoup.HTML_ENTITIES
+class SoupLog:
+ def __init__(self):
+ self.msgs = [ ]
+ def msg(self, m):
+ self.msgs.append(m)
+ def soupm(self, obj, m):
+ self.msg(m + '; in ' + `obj`)
+ def needs_msgs(self, child_souplog):
+ self.msgs += child_souplog.msgs
+ child_souplog.msgs = [ ]
+
+def soup_text(obj):
+ str = ''.join(obj.findAll(text=True))
+ return str.strip()
+
+class SomethingSoupInfo(SoupLog):
+ def __init__(self, kind, tail):
+ SoupLog.__init__(self)
+ html = fetcher.yoweb(kind, tail)
+ self.soup = BeautifulSoup(html,
+ convertEntities=BeautifulSoup.HTML_ENTITIES
)
- imgs = soup.findAll('img')
+
+class PirateInfo(SomethingSoupInfo):
+ # Public data members:
+ # pi.standings = { 'Treasure Haul': 'Able' ... }
+ # pi.crew = (id, name)
+ # pi.flag = (id, name)
+ # pi.msgs = [ 'message describing problem with scrape' ]
+
+ def __init__(self, pirate):
+ SomethingSoupInfo.__init__(self,
+ 'pirate.wm?target=', pirate)
+ self._find_standings()
+ self.crew = self._find_crewflag('crew',
+ '^/yoweb/crew/info\\.wm')
+ self.flag = self._find_crewflag('flag',
+ '^/yoweb/flag/info\\.wm')
+
+ def _find_standings(self):
+ imgs = self.soup.findAll('img',
+ src=regexp.compile('/yoweb/images/stat.*'))
re = regexp.compile(
-u'\s*\S*/([-A-Za-z]+)\s*$|\s*\S*/\S*\s*\(ocean\-wide\ \;([-A-Za-z]+)\)\s*$'
+u'\\s*\\S*/([-A-Za-z]+)\\s*$|\\s*\\S*/\\S*\\s*\\(ocean\\-wide(?:\\s|\\xa0)+([-A-Za-z]+)\\)\\s*$'
)
- skills = { }
+ standings = { }
+
for skill in duties:
- skills[skill] = [ ]
+ standings[skill] = [ ]
+
+ skl = SoupLog()
+
for img in imgs:
try: duty = img['alt']
except KeyError: continue
- print `duty`
- if not duty in duties: continue
+
+ if not duty in duties:
+ skl.soupm(img, 'unknown duty: "%s"' % duty)
+ continue
key = img.findParent('td')
- if key is None: continue
+ if key is None:
+ skl.soupm(img, 'duty at root! "%s"' % duty)
+ continue
valelem = key.findNextSibling('td')
- if valelem is None: continue
- valstr = ''.join(valelem.findAll(text=True))
- print `duty`, `valstr`
+ if valelem is None:
+ skl.soupm(key, 'duty missing sibling "%s"'
+ % duty)
+ continue
+ valstr = soup_text(valelem)
match = re.match(valstr)
- if match is None: continue
- standing = match.group(1)
- skills[duty] = standing
-
-# print `duty`, `standing`
-# if standing not in standings: continue
-# for i in range(0, len(standings)):
-# print `duty`, `standing`, i
-# if standing == standings[i]:
-# print `skills[duty]`
-# skills[duty].append(i)
-
-# self.skills = { }
-# for skill in duties:
-
- self.skills = skills
+ if match is None:
+ skl.soupm(key, ('duty "%s" unparseable'+
+ ' standing "%s"') % (duty, valstr))
+ continue
+ standing = match.group(match.lastindex)
+ standings[duty].append(standing)
+
+ self.standings = { }
+
+ for duty in duties:
+ sl = standings[duty]
+ if len(sl) > 1:
+ skl.msg('duty "%s" multiple standings %s' %
+ (duty, `sl`))
+ continue
+ if not len(sl):
+ skl.msg('duty "%s" no standing found' % duty)
+ continue
+ standing = sl[0]
+ for i in range(0, len(standingvals)-1):
+ if standing == standingvals[i]:
+ self.standings[duty] = i
+ if not duty in self.standings:
+ skl.msg('duty "%s" unknown standing "%s"' %
+ (duty, standing))
+
+ all_standings_ok = True
+ for duty in duties:
+ if not duty in self.standings:
+ self.needs_msgs(skl)
+
+ def _find_crewflag(self, cf, yoweb_re):
+ things = self.soup.findAll('a', href=regexp.compile(yoweb_re))
+ if len(things) != 1:
+ self.msg('zero or several %s id references found' % cf)
+ return None
+ thing = things[0]
+ id_re = '\\b%sid\\=(\\w+)$' % cf
+ id_haystack = thing['href']
+ match = regexp.compile(id_re).search(id_haystack)
+ if match is None:
+ self.soupm(thing, ('incomprehensible %s id ref'+
+ ' (%s in %s)') % (cf, id_re, id_haystack))
+ return None
+ name = soup_text(thing)
+ return (match.group(1), name)
+
+ def __str__(self):
+ return `(self.crew, self.flag, self.standings, self.msgs)`
+
+class CrewInfo(SomethingSoupInfo):
+ # Public data members:
+ # ci.crew = [ ('Captain', ['Pirate', ...]),
+ # ('Senior Officer', [...]),
+ # ... ]
+ # pi.msgs = [ 'message describing problem with scrape' ]
+
+ def __init__(self, crewid):
+ SomethingSoupInfo.__init__(self,
+ 'crew/info.wm?crewid=', crewid)
+ self._find_crew()
+
+ def _find_crew(self):
+ self.crew = []
+ capts = self.soup.findAll('img',
+ src='/yoweb/images/crew-captain.png')
+ if len(capts) != 1:
+ self.msg('crew members: no. of captain images != 1')
+ return
+ tbl = capts[0]
+ while not tbl.find('a', href=pirate_ref_re):
+ tbl = tbl.findParent('table')
+ if not tbl:
+ self.msg('crew members: cannot find table')
+ return
+ current_rank_crew = None
+ crew_rank_re = regexp.compile('/yoweb/images/crew')
+ for row in tbl.contents:
+ # findAll(recurse=False)
+ if isinstance(row, unicode):
+ continue
+
+ is_rank = row.find('img', attrs={'src': crew_rank_re})
+ if is_rank:
+ rank = soup_text(row)
+ current_rank_crew = []
+ self.crew.append((rank, current_rank_crew))
+ continue
+ for cell in row.findAll('a', href=pirate_ref_re):
+ if current_rank_crew is None:
+ self.soupm(cell, 'crew members: crew'
+ ' before rank')
+ continue
+ current_rank_crew.append(soup_text(cell))
+
def __str__(self):
- return `self.skills`
+ return `(self.crew, self.msgs)`
+
+def do_pirate(pirates, bu):
+ print '{'
+ for pirate in pirates:
+ info = PirateInfo(pirate)
+ print '%s: %s,' % (`pirate`, info)
+ print '}'
+
+def prep_crew_of(args, bu):
+ if len(args) != 1: bu('crew-of takes one pirate name')
+ pi = PirateInfo(args[0])
+ return CrewInfo(pi.crew[0])
+
+def do_crew_of(args, bu):
+ ci = prep_crew_of(args, bu)
+ print ci
+
+#def do_dutytab_crew_of(pirates, badusage):
+# if len(pirates) != 1: badusage('dutytab-crew-of takes one pirate name')
def main():
- os.chdir(os.getenv('HOME'))
- cache_dir = '.yoweb-scrape-cache'
- try:
- os.chdir(cache_dir)
- except (OSError,IOError), oe:
- if oe.errno != errno.ENOENT: raise
- os.mkdir(cache_dir)
- os.chdir(cache_dir)
-
- for path in os.listdir('.'):
- if not path.startswith('#'): continue
- max_time = max_age
- if '~' in path: max_time = 10
- try:
- s = os.stat(path)
- if now > s.st_mtime + max_time:
- os.remove(path)
- except (OSError,IOError), oe:
- if oe.errno != errno.ENOENT: raise
+ global opts, fetcher
+
+ pa = OptionParser(
+'''usage: .../yoweb-scrape [OPTION...] ACTION [ARGS...]
+actions:
+ yoweb-scrape [--ocean OCEAN ...] pirate PIRATE
+ yoweb-scrape [--ocean OCEAN ...] crew-of PIRATE
+ yoweb-scrape [--ocean OCEAN ...] dutytab-crew-of PIRATE
+''')
+ ao = pa.add_option
+ ao('-O','--ocean',dest='ocean', metavar='OCEAN',
+ default='ice',
+ help='select ocean OCEAN')
+ ao('--cache-dir', dest='cache_dir', metavar='DIR',
+ default='~/.yoweb-scrape-cache',
+ help='cache yoweb pages in DIR')
+ ao('-D','--debug', action='store_true', dest='debug', default=False,
+ help='enable debugging output')
+ ao('-q','--quiet', action='store_true', dest='quiet',
+ help='suppress warning output')
+ (opts,args) = pa.parse_args()
+
+ if len(args) < 1:
+ pa.error('need a mode argument')
+
+ mode = args[0]
+ mode_fn_name = 'do_' + mode.replace('_','#').replace('-','_')
+ try: mode_fn = globals()[mode_fn_name]
+ except KeyError: pa.error('unknown mode "%s"' % mode)
+
+ # fixed parameters
+ opts.max_age = 240
+
+ if opts.cache_dir.startswith('~/'):
+ opts.cache_dir = os.getenv('HOME') + opts.cache_dir[1:]
+
+ fetcher = Fetcher(opts.ocean, opts.cache_dir)
- # test program:
- global ocean
- ocean = 'midnight'
- test = PirateInfo('Anaplian')
- print test
+ mode_fn(args[1:], pa.error)
main()
~yarrgweb / ypp-sc-tools.db-live.git / blobdiff