Merge lp:~marco-gallotta/ibid/flight into lp:~ibid-core/ibid/old-trunk-1.6

[marcog]

[marcog (marco-gallotta)]

I think the flight plugin is mature enough for some reviews. It's quite possible some gaping bugs are still out there, but it seems to be robust enough.

[Stefano Rivera (stefanor)]

Whoops, conflicts in the diff.

[lp:~marco-gallotta/ibid/flight]

858. By marcog on 2010-01-07

Merge trunk and revert ascii.py (Somehow old changes got muddled into it)

859. By marcog on 2010-01-07

Reverting ascii to trunk's version...hopefully i got it right *this* time

860. By marcog on 2010-01-07

Fix typos; allow "airport *for* cape town

861. By marcog on 2010-01-07

Condense responses into one

862. By marcog on 2010-01-07

Remove unused imports

863. By marcog on 2010-01-07

Make use of etree's findtext and path querying

864. By marcog on 2010-01-08

Merge flight and airport features

865. By marcog on 2010-01-08

Use dict substitutions for responses

866. By marcog on 2010-01-09

Catch ValueError thrown by parse() when the date is invalid

867. By marcog on 2010-01-12

Make strings unicode in places where they weren't; split(' ') if s -> string.split()

868. By marcog on 2010-01-14

Make parsing a little more resiliant

869. By marcog on 2010-01-14

strftime needs an ascii string for pre-2.6

870. By marcog on 2010-01-18

Merge trunk and move flight into geography

871. By marcog on 2010-03-09

Travelocity changed some form attribute names, this fixes them but flight is still not working

Unmerged revisions

871. By marcog on 2010-03-09

Travelocity changed some form attribute names, this fixes them but flight is still not working

=== modified file 'ibid/plugins/ascii.py'
--- ibid/plugins/ascii.py	2010-01-05 08:00:15 +0000
+++ ibid/plugins/ascii.py	2010-01-07 11:59:18 +0000
@@ -1,12 +1,11 @@
-from BaseHTTPServer import BaseHTTPRequestHandler
 from cStringIO import StringIO
 import Image
 from os import remove
 import os.path
 import subprocess
+from sys import stderr
 from tempfile import mkstemp
-from urllib2 import HTTPError, URLError, urlopen
-from urlparse import urlparse
+from urllib2 import urlopen
 from zipfile import ZipFile
 
 from aalib import AsciiScreen
@@ -44,6 +43,7 @@
 
     @match(r'^draw\s+(\S+\.\S+)(\s+in\s+colou?r)?(?:\s+w(?:idth)?\s+(\d+))?(?:\s+h(?:eight)\s+(\d+))?$')
     def draw(self, event, url, colour, width, height):
+<<<<<<< TREE
         if not urlparse(url).netloc:
             url = 'http://' + url
         if urlparse(url).scheme == 'file':
@@ -70,17 +70,26 @@
         if f.read(1) != '':
             event.addresponse(u'File too large (limit is %i KiB)', self.max_filesize)
             return
+=======
+        f = urlopen(url)
+
+        filesize = int(f.info().getheaders('Content-Length')[0])
+        if filesize > self.max_filesize * 1024:
+            event.addresponse(u'File too large (limit is %i KiB)', self.max_filesize)
+            return
+
+>>>>>>> MERGE-SOURCE
         try:
             ext = os.path.splitext(url)[1]
             image = mkstemp(suffix=ext)[1]
             file = open(image, 'w')
-            file.write(buffer)
+            file.write(f.read())
             file.close()
 
             try:
                 img = Image.open(StringIO(open(image, 'r').read())).convert('L')
-            except IOError:
-                event.addresponse(u"Sorry, that doesn't look like an image")
+            except:
+                event.addresponse(u'Cannot understand image format')
                 return
             input_width, input_height = img.size[0], img.size[1]
 
@@ -120,8 +129,8 @@
     def draw_aa(self, event, image, width, height):
         try:
             image = Image.open(StringIO(open(image, 'r').read())).convert('L')
-        except IOError:
-            event.addresponse(u"Sorry, that doesn't look like an image")
+        except:
+            event.addresponse(u'Cannot understand image format')
             return
         screen = AsciiScreen(width=width, height=height)
         image = image.resize(screen.virtual_size)
@@ -129,6 +138,7 @@
         event.addresponse(unicode(screen.render()), address=False, conflate=False)
 
     def draw_caca(self, event, image, width, height):
+        from sys import stderr
         process = subprocess.Popen(
             [self.img2txt_bin, '-f', 'irc', '-W', str(width), '-H', str(height), image],
             shell=False, stdout=subprocess.PIPE)
@@ -137,15 +147,14 @@
         if code == 0:
             event.addresponse(unicode(response.replace('\r', '')), address=False, conflate=False)
         else:
-            event.addresponse(u"Sorry, that doesn't look like an image")
+            event.addresponse(u'Sorry, cannot understand image format')
 
 class WriteFiglet(Processor):
     u"""figlet <text> [in <font>]
     list figlet fonts [from <index>]"""
     feature = 'figlet'
 
-    max_width = IntOption('max_width', 'Maximum width for ascii output', 60)
-    fonts_zip = Option('fonts_zip', 'Zip file containing figlet fonts', 'ibid/data/figlet-fonts.zip')
+    fonts_zip = Option('fonts_zip', 'Zip file containing figlet fonts', 'data/figlet-fonts.zip')
 
     def __init__(self, name):
         Processor.__init__(self, name)
@@ -178,7 +187,4 @@
             del rendered[0]
         while rendered and rendered[-1].strip() == '':
             del rendered[-1]
-        if rendered and len(rendered[0]) > self.max_width:
-            event.addresponse(u"Sorry that's too long, nobody will be able to read it")
-            return
         event.addresponse(unicode('\n'.join(rendered)), address=False, conflate=False)

=== added file 'ibid/plugins/flight.py'
--- ibid/plugins/flight.py	1970-01-01 00:00:00 +0000
+++ ibid/plugins/flight.py	2010-01-07 11:59:18 +0000
@@ -0,0 +1,300 @@
+import csv
+import re
+from sys import maxint
+from urllib import urlencode
+from xml.etree import ElementTree
+
+from dateutil.parser import parse
+
+from ibid.config import IntOption
+from ibid.plugins import Processor, match
+from ibid.utils import cacheable_download, human_join
+from ibid.utils.html import get_html_parse_tree
+
+help = { u'airport' : u'Search for airports',
+         u'flight'  : u'Search for flights on travelocity' }
+
+airports_url = 'http://openflights.svn.sourceforge.net/viewvc/openflights/openflights/data/airports.dat'
+
+airports = {}
+
+def read_data():
+    # File is listed as ISO 8859-1 (Latin-1) encoded on
+    # http://openflights.org/data.html, but from decoding it appears to
+    # actually be UTF8
+    filename = cacheable_download(airports_url, 'flight/airports.dat')
+    reader = csv.reader(open(filename), delimiter=',', quotechar='"')
+    for row in reader:
+        airports[int(row[0])] = [unicode(r, 'utf-8') for r in row[1:]]
+
+def airport_search(query, search_loc = True):
+    if not airports:
+        read_data()
+    if search_loc:
+        ids = airport_search(query, False)
+        if len(ids) == 1:
+            return ids
+        query = [unicode(q) for q in query.lower().split(' ') if q]
+    else:
+        query = [unicode(query.lower())]
+    ids = []
+    for id, airport in airports.items():
+        if search_loc:
+            data = (u' '.join(c.lower() for c in airport[:5])).split(' ')
+        elif len(query[0]) == 3:
+            data = [airport[3].lower()]
+        else: # assume lenght 4 (won't break if not)
+            data = [airport[4].lower()]
+        if len(filter(lambda q: q in data, query)) == len(query):
+            ids.append(id)
+    return ids
+
+def repr_airport(id):
+    airport = airports[id]
+    code = ''
+    if airport[3] or airport[4]:
+        code = ' (%s)' % u'/'.join(filter(lambda c: c, airport[3:5]))
+    return '%s%s' % (airport[0], code)
+
+class AirportSearch(Processor):
+    """airport [in] <name|location|code>"""
+
+    feature = 'airport'
+
+    @match(r'^airports?\s+(in\s+)?(.+)$')
+    def airport_search(self, event, search_loc, query):
+        search_loc = search_loc is not None
+        if not search_loc and not 3 <= len(query) <= 4:
+            event.addresponse(u'Airport code must be 3 or 4 characters')
+            return
+        ids = airport_search(query, search_loc)
+        if len(ids) == 0:
+            event.addresponse(u"Sorry, I don't know that airport")
+        elif len(ids) == 1:
+            id = ids[0]
+            airport = airports[id]
+            code = 'unknown code'
+            if airport[3] and airport[4]:
+                code = 'codes %s and %s' % (airport[3], airport[4])
+            elif airport[3]:
+                code = 'code %s' % airport[3]
+            elif airport[4]:
+                code = 'code %s' % airport[4]
+            event.addresponse(u'%s in %s, %s has %s' %
+                    (airport[0], airport[1], airport[2], code))
+        else:
+            event.addresponse(u'Found the following airports: %s', human_join(repr_airport(id) for id in ids)[:480])
+
+class Flight:
+    def __init__(self):
+        self.flight, self.depart_time, self.depart_ap, self.arrive_time, \
+                self.arrive_ap, self.duration, self.stops, self.price = \
+                [], None, None, None, None, None, None, None
+
+    def int_price(self):
+        try:
+            return int(self.price[1:])
+        except ValueError:
+            return maxint
+
+    def int_duration(self):
+        hours, minutes = 0, 0
+        match = re.search(r'(\d+)hr', self.duration)
+        if match:
+            hours = int(match.group(1))
+        match = re.search(r'(\d+)min', self.duration)
+        if match:
+            minutes = int(match.group(1))
+        return int(hours)*60 + int(minutes)
+
+MONTH_SHORT = ('Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun', 'Jul', 'Aug', 'Sep', 'Oct', 'Nov', 'Dec')
+MONTH_LONG = ('January', 'February', 'March', 'April', 'May', 'June', 'July', 'August', 'September', 'October', 'November', 'December')
+OTHER_STUFF = ('am', 'pm', 'st', 'nd', 'rd', 'th', 'morning', 'afternoon', 'evening', 'anytime')
+DATE = r'(?:[0-9.:/hT -]|%s)+' % '|'.join(MONTH_SHORT+MONTH_LONG+OTHER_STUFF)
+
+class FlightException(Exception):
+    pass
+
+class FlightSearch(Processor):
+    """[<cheapest|quickest]> flight from <departure> to <destination> from <depart_date> [anytime|morning|afternoon|evening|<time>] to <return_date> [anytime|morning|afternoon|evening|<time>]"""
+
+    feature = 'flight'
+
+    max_results = IntOption('max_results', 'Maximum number of results to list', 5)
+
+    def _flight_search(self, event, dpt, to, dep_date, ret_date):
+        airport_dpt = airport_search(dpt)
+        airport_to = airport_search(to)
+        if len(airport_dpt) == 0:
+            event.addresponse(u"Sorry, I don't know the airport you want to leave from")
+            return
+        if len(airport_to) == 0:
+            event.addresponse(u"Sorry, I don't know the airport you want to fly to")
+            return
+        if len(airport_dpt) > 1:
+            event.addresponse(u'The following airports match the departure: %s', human_join(repr_airport(id) for id in airport_dpt)[:480])
+            return
+        if len(airport_to) > 1:
+            event.addresponse(u'The following airports match the destination: %s', human_join(repr_airport(id) for id in airport_to)[:480])
+            return
+
+        dpt = airport_dpt[0]
+        to = airport_to[0]
+
+        def to_travelocity_date(date):
+            date = date.lower()
+            time = None
+            for period in ['anytime', 'morning', 'afternoon', 'evening']:
+                if period in date:
+                    time = period.title()
+                    date = date.replace(period, '')
+                    break
+            date = parse(date)
+            if time is None:
+                if date.hour == 0 and date.minute == 0:
+                    time = 'Anytime'
+                else:
+                    time = date.strftime('%I:00')
+                    if time[0] == '0':
+                        time = time[1:]
+                    if date.hour < 12:
+                        time += 'am'
+                    else:
+                        time += 'pm'
+            date = date.strftime('%m/%d/%Y')
+            return (date, time)
+
+        (dep_date, dep_time) = to_travelocity_date(dep_date)
+        (ret_date, ret_time) = to_travelocity_date(ret_date)
+
+        params = {}
+        params['leavingFrom'] = airports[dpt][3]
+        params['goingTo'] = airports[to][3]
+        params['leavingDate'] = dep_date
+        params['dateLeavingTime'] = dep_time
+        params['returningDate'] = ret_date
+        params['dateReturningTime'] = ret_time
+        etree = get_html_parse_tree('http://travel.travelocity.com/flights/InitialSearch.do', data=urlencode(params), treetype='etree')
+        while True:
+            script = [script for script in etree.getiterator('script')][1]
+            matches = script.text and re.search(r'var finurl = "(.*)"', script.text)
+            if matches:
+                url = 'http://travel.travelocity.com/flights/%s' % matches.group(1)
+                etree = get_html_parse_tree(url, treetype='etree')
+            else:
+                break
+
+        # Handle error
+        div = [d for d in etree.getiterator('div') if d.get(u'class') == 'e_content']
+        if len(div):
+            error = div[0].find('h3').text
+            raise FlightException(error)
+
+        departing_flights = self._parse_travelocity(etree)
+        return_url = None
+        table = [t for t in etree.getiterator('table')][3]
+        for tr in table.getiterator('tr'):
+            for td in tr.getiterator('td'):
+                if td.get(u'class').strip() in ['tfPrice', 'tfPriceOrButton']:
+                    div = td.find('div')
+                    if div is not None:
+                        button = div.find('button')
+                        if button is not None:
+                            onclick = button.get('onclick')
+                            match = re.search(r"location.href='\.\./flights/(.+)'", onclick)
+                            url_page = match.group(1)
+                            match = re.search(r'^(.*?)[^/]*$', url)
+                            url_base = match.group(1)
+                            return_url = url_base + url_page
+
+        etree = get_html_parse_tree(return_url, treetype='etree')
+        returning_flights = self._parse_travelocity(etree)
+
+        return (departing_flights, returning_flights, url)
+
+    def _parse_travelocity(self, etree):
+        flights = []
+        table = [t for t in etree.getiterator('table') if t.get(u'id') == 'tfGrid'][0]
+        trs = [t for t in table.getiterator('tr')]
+        tr_index = 1
+        while tr_index < len(trs):
+            tds = []
+            while True:
+                new_tds = [t for t in trs[tr_index].getiterator('td')]
+                tds.extend(new_tds)
+                tr_index += 1
+                if len(filter(lambda t: t.get(u'class').strip() == u'tfAirlineSeatsMR', new_tds)):
+                    break
+            flight = Flight()
+            for td in tds:
+                if td.get(u'class').strip() == u'tfAirline':
+                    anchor = td.find('a')
+                    if anchor is not None:
+                        airline = anchor.text.strip()
+                    else:
+                        airline = td.text.split('\n')[0].strip()
+                    flight.flight.append(u'%s %s' % (airline, td.find('div').text.strip()))
+                if td.get(u'class').strip() == u'tfDepart' and td.text:
+                    flight.depart_time = td.text.split('\n')[0].strip()
+                    flight.depart_ap = '%s %s' % (td.find('div').text.strip(),
+                            td.find('div').find('span').text.strip())
+                if td.get(u'class').strip() == u'tfArrive' and td.text:
+                    flight.arrive_time = td.text.split('\n')[0].strip()
+                    span = td.find('span')
+                    if span is not None and span.get(u'class').strip() == u'tfNextDayDate':
+                        flight.arrive_time = u'%s %s' % (flight.arrive_time, span.text.strip()[2:])
+                        span = [s for s in td.find('div').getiterator('span')][1]
+                        flight.arrive_ap = '%s %s' % (td.find('div').text.strip(),
+                                span.text.strip())
+                    else:
+                        flight.arrive_ap = '%s %s' % (td.find('div').text.strip(),
+                                td.find('div').find('span').text.strip())
+                if td.get(u'class').strip() == u'tfTime' and td.text:
+                    flight.duration = td.text.strip()
+                    flight.stops = td.find('span').find('a').text.strip()
+                if td.get(u'class').strip() in [u'tfPrice', u'tfPriceOr'] and td.text:
+                    flight.price = td.text.strip()
+            flight.flight = human_join(flight.flight)
+            flights.append(flight)
+
+        return flights
+
+    @match(r'^(?:(cheapest|quickest)\s+)?flights?\s+from\s+(.+)\s+to\s+(.+)\s+from\s+(%s)\s+to\s+(%s)$' % (DATE, DATE))
+    def flight_search(self, event, priority, dpt, to, dep_date, ret_date):
+        try:
+            flights = self._flight_search(event, dpt, to, dep_date, ret_date)
+        except FlightException, e:
+            event.addresponse(unicode(e))
+            return
+        if flights is None:
+            return
+        if len(flights[0]) == 0:
+            event.addresponse(u'No matching departure flights found')
+            return
+        if len(flights[1]) == 0:
+            event.addresponse(u'No matching return flights found')
+            return
+
+        cmp = None
+        if priority == 'cheapest':
+            cmp = lambda a, b: a.int_price() < b.int_price()
+        elif priority == 'quickest':
+            cmp = lambda a, b: a.int_duration() < b.int_duration()
+        if cmp:
+            # select best flight based on priority
+            for i in xrange(2):
+                flights[i].sort(cmp=cmp)
+                del flights[i][1:]
+        for i, flight_type in zip(xrange(2), ['Departing', 'Returning']):
+            if len(flights[i]) > 1:
+                event.addresponse(u'%s flights:', flight_type)
+            for flight in flights[i][:self.max_results]:
+                leading = ''
+                if len(flights[i]) == 1:
+                    leading = u'%s flight: ' % flight_type
+                event.addresponse('%s%s departing %s from %s, arriving %s at %s (flight time %s, %s) costs %s per person',
+                        (leading, flight.flight, flight.depart_time, flight.depart_ap, flight.arrive_time,
+                            flight.arrive_ap, flight.duration, flight.stops, flight.price or 'unknown'))
+        event.addresponse(u'Full results: %s', flights[2])
+
+# vi: set et sta sw=4 ts=4:

=== modified file 'ibid/utils/__init__.py'
--- ibid/utils/__init__.py	2010-01-06 23:19:42 +0000
+++ ibid/utils/__init__.py	2010-01-07 11:59:19 +0000
@@ -80,9 +80,14 @@
         req.add_header('User-Agent', 'Ibid/' + (ibid_version() or 'dev'))
 
     if exists:
-        modified = os.path.getmtime(cachefile)
-        modified = time.strftime("%a, %d %b %Y %H:%M:%S GMT", time.gmtime(modified))
-        req.add_header("If-Modified-Since", modified)
+        if os.path.isfile(cachefile + '.etag'):
+            f = file(cachefile + '.etag', 'r')
+            req.add_header("If-None-Match", f.readline().strip())
+            f.close()
+        else:
+            modified = os.path.getmtime(cachefile)
+            modified = time.strftime("%a, %d %b %Y %H:%M:%S GMT", time.gmtime(modified))
+            req.add_header("If-Modified-Since", modified)
 
     try:
         connection = urllib2.urlopen(req)
@@ -106,6 +111,12 @@
             gzipper = GzipFile(fileobj=compressedstream)
             data = gzipper.read()
 
+    etag = connection.headers.get('etag')
+    if etag:
+        f = file(cachefile + '.etag', 'w')
+        f.write(etag + '\n')
+        f.close()
+
     outfile = file(cachefile, 'wb')
     outfile.write(data)
     outfile.close()