X-Git-Url: http://www.chiark.greenend.org.uk/ucgi/~yarrgweb/git?p=ypp-sc-tools.web-live.git;a=blobdiff_plain;f=yoweb-scrape;h=5e24e57e749fb6f5657f19963cd518a109e7389a;hp=1f5486e2a656f3c9fd2c8642213bde5b021d2dd4;hb=43e1a827c400f0c686936d52e91555eec44517d0;hpb=0e15228e6cb10cc9c6116435fb00d39652f1fc32 diff --git a/yoweb-scrape b/yoweb-scrape index 1f5486e..5e24e57 100755 --- a/yoweb-scrape +++ b/yoweb-scrape @@ -1,5 +1,7 @@ #!/usr/bin/python +#---------- setup ---------- + import signal signal.signal(signal.SIGINT, signal.SIG_DFL) @@ -17,6 +19,7 @@ from BeautifulSoup import BeautifulSoup opts = None +#---------- YPP parameters and arrays ---------- puzzles = ('Swordfighting/Bilging/Sailing/Rigging/Navigating'+ '/Battle Navigation/Gunning/Carpentry/Rumble/Treasure Haul'+ @@ -31,10 +34,14 @@ pirate_ref_re = regexp.compile('^/yoweb/pirate\\.wm') max_pirate_namelen = 12 +#---------- general utilities ---------- + def debug(m): if opts.debug > 0: print m +#---------- caching and rate-limiting data fetcher ---------- + class Fetcher: def __init__(self, ocean, cachedir): debug('Fetcher init %s' % cachedir) @@ -133,6 +140,8 @@ class Fetcher: self.ocean, kind, tail) return self.fetch(url, max_age) +#---------- logging assistance for troubled screenscrapers ---------- + class SoupLog: def __init__(self): self.msgs = [ ] @@ -156,6 +165,8 @@ class SomethingSoupInfo(SoupLog): convertEntities=BeautifulSoup.HTML_ENTITIES ) +#---------- scraper for pirate pages ---------- + class PirateInfo(SomethingSoupInfo): # Public data members: # pi.standings = { 'Treasure Haul': 'Able' ... } @@ -255,6 +266,8 @@ u'\\s*\\S*/([-A-Za-z]+)\\s*$|\\s*\\S*/\\S*\\s*\\(ocean\\-wide(?:\\s|\\xa0)+([-A- def __str__(self): return `(self.crew, self.flag, self.standings, self.msgs)` +#---------- scraper for crew pages ---------- + class CrewInfo(SomethingSoupInfo): # Public data members: # ci.crew = [ ('Captain', ['Pirate', ...]), @@ -303,6 +316,8 @@ class CrewInfo(SomethingSoupInfo): def __str__(self): return `(self.crew, self.msgs)` +#---------- pretty-printer for tables of pirate puzzle standings ---------- + class StandingsTable: def __init__(self, use_puzzles=None, col_width=6): if use_puzzles is None: @@ -321,10 +336,12 @@ class StandingsTable: self.s = '' self._cw = col_width-1 - def _pline(self, pirate, puzstrs): + def _pline(self, pirate, puzstrs, extra): self.s += ' %-*s' % (max(max_pirate_namelen, 14), pirate) for v in puzstrs: self.s += ' %-*.*s' % (self._cw,self._cw, v) + if extra: + self.s += ' ' + extra self.s += '\n' def _puzstr(self, pi, puzzle): @@ -352,55 +369,30 @@ class StandingsTable: spc = name.find(' ') if spc < 0: return name return name[0:min(4,spc)] + name[spc+1:] - self._pline('', map(puzn_redact, self._puzzles)) + self._pline('', map(puzn_redact, self._puzzles), None) def literalline(self, line): self.s += line + '\n' - def pirate_dummy(self, name, standingstring): - self._pline(name, standingstring * len(self._puzzles)) - def pirate(self, pi): + def pirate_dummy(self, name, standingstring, extra=None): + self._pline(name, standingstring * len(self._puzzles), extra) + def pirate(self, pi, extra=None): puzstrs = [self._puzstr(pi,puz) for puz in self._puzzles] - self._pline(pi.name, puzstrs) + self._pline(pi.name, puzstrs, extra) def results(self): return self.s -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, max_age=300): - if len(args) != 1: bu('crew-of takes one pirate name') - pi = PirateInfo(args[0], max_age) - return CrewInfo(pi.crew[0], max_age) - -def do_crew_of(args, bu): - ci = prep_crew_of(args, bu) - print ci - -def do_standings_crew_of(args, bu): - ci = prep_crew_of(args, bu, 60) - tab = StandingsTable() - tab.headings() - for (rank, members) in ci.crew: - if not members: continue - tab.literalline('%s:' % rank) - for p in members: - pi = PirateInfo(p, random.randint(900,1800)) - tab.pirate(pi) - print tab.results() +#---------- chat log parser ---------- class PirateAboard: - # pa.v - # pa.name - # pa.last_time - # pa.last_event - # pa.gunner - # pa.last_chat_time - # pa.last_chat_chan - # pa.pi + # This is essentially a transparent, dumb, data class. + # pa.v + # pa.name + # pa.last_time + # pa.last_event + # pa.gunner + # pa.last_chat_time + # pa.last_chat_chan + # pa.pi def __init__(pa, pn, v, time, event): pa.name = pn @@ -418,6 +410,9 @@ class PirateAboard: return pa.pi class ChatLogTracker: + # This is quite complex so we make it opaque. Use the + # official invokers, accessors etc. + def __init__(self, myself_pi, logfn): self._pl = {} # self._pl['Pirate'] = self._vl = {} # self._vl['Vessel']['Pirate'] = PirateAboard @@ -529,13 +524,13 @@ class ChatLogTracker: m = rm('You have ordered (\\w+) to do some (\\S.*\\S)\\.$') if m: (who,what) = m.groups() - pa = ob_x(who,'ordered '+what) + pa = ob_x(who,'ord '+what) if what == 'Gunning': pa.gunner = True return d('duty order') m = rm('(\\w+) abandoned a (\\S.*\\S) station\\.$') - if m: oba('stopped'); return d('stopped') + if m: oba('stopped'); return d("end") def chat(what): who = m.group(1) @@ -651,6 +646,37 @@ class ChatLogTracker: for pn in sorted(self._v.keys()) if not pn.startswith('#') ] +#---------- implementations of actual operation modes ---------- + +def do_pirate(pirates, bu): + print '{' + for pirate in pirates: + info = PirateInfo(pirate) + print '%s: %s,' % (`pirate`, info) + print '}' + +def prep_crew_of(args, bu, max_age=300): + if len(args) != 1: bu('crew-of takes one pirate name') + pi = PirateInfo(args[0], max_age) + if pi.crew is None: return None + return CrewInfo(pi.crew[0], max_age) + +def do_crew_of(args, bu): + ci = prep_crew_of(args, bu) + print ci + +def do_standings_crew_of(args, bu): + ci = prep_crew_of(args, bu, 60) + tab = StandingsTable() + tab.headings() + for (rank, members) in ci.crew: + if not members: continue + tab.literalline('%s:' % rank) + for p in members: + pi = PirateInfo(p, random.randint(900,1800)) + tab.pirate(pi) + print tab.results() + class ProgressPrintPercentage: def __init__(self, f=sys.stdout): self._f = f def progress(self,done,total): @@ -660,14 +686,16 @@ class ProgressPrintPercentage: self._f.write(' \r') self._f.flush() +#----- modes which use the chat log parser are quite complex ----- + def prep_chat_log(args, bu, progress=ProgressPrintPercentage(), max_myself_age=3600): if len(args) != 1: bu('this action takes only chat log filename') logfn = args[0] - logfn_re = '(?:.*/)?([A-Z][a-z]+)_([a-z]+)_chat-log-\\w+$' + logfn_re = '(?:.*/)?([A-Z][a-z]+)_([a-z]+)_' match = regexp.match(logfn_re, logfn) - if not match: bu('chat log filename is not in default format') + if not match: bu('chat log filename is not in expected format') (pirate, fetcher.ocean) = match.groups() myself = PirateInfo(pirate,max_myself_age) @@ -687,6 +715,12 @@ def do_track_chat_log(args, bu): print track time.sleep(1) +def format_time_interval(ti): + if ti < 120: return '%d:%02d' % (ti / 60, ti % 60) + if ti < 7200: return '%2dm' % (ti / 60) + if ti < 86400: return '%dh' % (ti / 3600) + return '%dd' % (ti / 86400) + def do_ship_aid(args, bu): if opts.ship_duty is None: opts.ship_duty = True @@ -694,15 +728,23 @@ def do_ship_aid(args, bu): rotate_nya = '/-\\' + def timeevent(t,e): + if t is None: return ' ' * 22 + return " %-4s %-16s" % (format_time_interval(now - t),e) + while True: track.catchup() now = time.time() - sys.stdout.write("\n\n%s" % track.myname()) + + s = "%s" % track.myname() vn = track.vessel() - if vn is None: print " ...?"; return + if vn is None: + print s + " ...?" + time.sleep(1) + continue - print " on board the %s at %s\n" % ( + s += " on board the %s at %s\n" % ( vn, time.strftime("%Y-%m-%d %H:%M:%S")) tbl = StandingsTable() @@ -710,15 +752,27 @@ def do_ship_aid(args, bu): for pa in track.aboard(): pi = pa.pirate_info() + + xs = '' + if pa.gunner: xs += 'G ' + else: xs += ' ' + xs += timeevent(pa.last_time, pa.last_event) + xs += timeevent(pa.last_chat_time, pa.last_chat_chan) + if pi is None: - tbl.pirate_dummy(pa.name, rotate_nya[0]) + tbl.pirate_dummy(pa.name, rotate_nya[0], xs) else: - tbl.pirate(pi) + tbl.pirate(pi, xs) + + s += tbl.results() + + print '\n\n', s; - print tbl.results() time.sleep(1) rotate_nya = rotate_nya[1:2] + rotate_nya[0] +#---------- main program ---------- + def main(): global opts, fetcher