From 1a9fe0907797ee7a409d0969fd35f7f02da13e79 Mon Sep 17 00:00:00 2001
From: Florian Schweikert <kelvan@logic.at>
Date: Sat, 26 Feb 2011 02:49:07 +0100
Subject: [PATCH] fixed parser crash when overview contains footpath added
 trip parser

---
 parseHtml.py |   29 +++++++++++++++++++++--------
 1 file changed, 21 insertions(+), 8 deletions(-)

diff --git a/parseHtml.py b/parseHtml.py
index aab72b5..94826ab 100644
--- a/parseHtml.py
+++ b/parseHtml.py
@@ -1,6 +1,7 @@
-from BeautifulSoup import BeautifulSoup
+from BeautifulSoup import BeautifulSoup, NavigableString
 import urllib2
 from datetime import time, datetime
+from textwrap import wrap
 
 
 class Parser:
@@ -16,7 +17,15 @@ class Parser:
         raise IndexError()
 
     def _parse_details(self):
-        return self.soup.findAll('div', {'class': 'data_table tourdetail'})
+        trips = map(lambda x: map(lambda x: {
+                        'time': map(lambda x: (time(*map(lambda x: int(x), x.split(':')))), wrap(x.find('td', {'class': 'col_time'}).text, 5)), # black magic appears
+                        'station': map(lambda x: x[2:].strip(),
+                                       filter(lambda x: type(x) == NavigableString, x.find('td', {'class': 'col_station'}).contents)), # filter non NaviStrings
+                        'info': map(lambda x: x.strip(),
+                                    filter(lambda x: type(x) == NavigableString, x.find('td', {'class': 'col_info'}).contents)),
+                    }, x.find('tbody').findAll('tr')),
+                    self.soup.findAll('div', {'class': 'data_table tourdetail'})) # all routes
+        return trips
 
     @property
     def details(self):
@@ -39,13 +48,17 @@ class Parser:
         # get rows
         rows = table.findAll('tr')[1:]
         overview = map(lambda x: {
-                       'date': datetime.strptime(x.find('td', {'class': 'col_date'}).text, '%d.%m.%Y'), # grab date
-                       'time': map(lambda x: time(*map(lambda x: int(x), x.split(':'))), x.find('td', {'class': 'col_time'}).text.split('  - ')), # extract times
-                       'duration': time(*map(lambda x: int(x), x.find('td', {'class': 'col_duration'}).text.split(':'))), # grab duration
-                       'change': int(x.find('td', {'class': 'col_change'}).text), # grab changes
-                       'price': float(x.find('td', {'class': 'col_price'}).text.replace(',', '.')) # grab price
+                           'date': datetime.strptime(x.find('td', {'class': 'col_date'}).text, '%d.%m.%Y') # grab date
+                                       if x.find('td', {'class': 'col_date'}).text else None, # if date is empty set to None
+                           'time': map(lambda x: time(*map(lambda x: int(x), x.split(':'))) if x else None, # extract times or set to None if empty
+                                       x.find('td', {'class': 'col_time'}).text.split('  - ')),
+                           'duration': time(*map(lambda x: int(x), x.find('td', {'class': 'col_duration'}).text.split(':'))), # grab duration
+                           'change': int(x.find('td', {'class': 'col_change'}).text) # grab changes
+                                       if x.find('td', {'class': 'col_change'}).text else 0, # if change is empty set to 0
+                           'price': float(x.find('td', {'class': 'col_price'}).text.replace(',', '.')) # grab price
+                                       if x.find('td', {'class': 'col_price'}).text.find(',') >= 0 else 0.0, # if price is empty set to 0.0
                        },
-            rows)
+                       rows)
 
         return overview
 
-- 
1.7.9.5