Bug fix in ui/ui_functions.py and and updated problem data to latest version

.gitignore

@@ -0,0 +1 @@
+.pyc

ui/api_functions_v2.py

@@ -47,9 +47,9 @@ def getProblemDataFromContest(contestID):
     r = requests.get(url).json()
 
     if r['status'] == 'FAILED':
-	expectedMessage = 'contestId: Contest with id ' + str(contestID) + ' not found'
-	if r['comment'] == expectedMessage:
-	    raise ContestNotFound(contestID)
+      expectedMessage = 'contestId: Contest with id ' + str(contestID) + ' not found'
+      if r['comment'] == expectedMessage:
+          raise ContestNotFound(contestID)
 
     r = r['result']
     contest = r['contest']['name']
@@ -79,7 +79,7 @@ def getProblemDataFromContest(contestID):
     return probdf
 
 def getSolveSuccessDF(contestID):
-#contestID = 671
+    #contestID = 671
     urlbase = 'http://codeforces.com/api/'
     method = 'contest.standings'
     maxcnt = 100000
@@ -92,43 +92,49 @@ def getSolveSuccessDF(contestID):
 
     # CHECK TO MAKE SURE THAT TEAMS ARE NOT ALLOWED!!!
     for r in rows: # for each person
-	if len(r['party']['members']) > 1:
-	    raise Error("This contest allows teams. ELO scores are not well-defined.")
-
+      if len(r['party']['members']) > 1:
+          raise Error("This contest allows teams. ELO scores are not well-defined.")
+    
     users = []
     points = []
     for r in rows:
-	users.append(r['party']['members'][0]['handle'])
-
-	userpts = [0]*len(problems)
-	for i in range(len(problems)):
-	    userpts[i] = r['problemResults'][i]['points']
-	points.append([1 if u > 0 else 0 for u in userpts])
-
+      users.append(r['party']['members'][0]['handle'])
+
+      userpts = [0]*len(problems)
+      for i in range(len(problems)):
+          userpts[i] = r['problemResults'][i]['points']
+      points.append([1 if u > 0 else 0 for u in userpts])
 
     # Grab rating changes
     method = 'contest.ratingChanges'
     url = urlbase + method + '?contestId=' + str(contestID)
     ratingchanges = requests.get(url).json()['result']
     ratingdict = dict()
     for r in ratingchanges:
-	ratingdict[r['handle']] = r['oldRating']
-
+      ratingdict[r['handle']] = r['oldRating']
 
     # Create output df 
     # start constructing dataframe
     array_out = []
     for i in range(len(users)): # for each user in the contest
-	hdl = users[i]
-	rating = ratingdict[hdl]
-	for j, p in enumerate(problems): # for each problem in the contest, make a new row
-	    temp = dict.fromkeys(['handle', 'rating', 'contestID', 'problemID', 'success'])
-	    temp['handle'] = hdl
-	    temp['contestID'] = contestID
-	    temp['problemID'] = p['index']
-	    temp['success'] = points[i][j]
-	    temp['rating'] = rating
-	    array_out.append(temp)
+      hdl = users[i]
+
+      if hdl not in ratingdict:
+        continue
+
+      #try:  
+      rating = ratingdict[hdl]
+      #except:
+      #  url = 'http://codeforces.com/api/user.rating?handle=' + hdl
+      #  rating = requests.get(url).json()['result'][-1]['newRating']
+      for j, p in enumerate(problems): # for each problem in the contest, make a new row
+          temp = dict.fromkeys(['handle', 'rating', 'contestID', 'problemID', 'success'])
+          temp['handle'] = hdl
+          temp['contestID'] = contestID
+          temp['problemID'] = p['index']
+          temp['success'] = points[i][j]
+          temp['rating'] = rating
+          array_out.append(temp)
 
     output = pd.DataFrame.from_dict(array_out)
     return output
@@ -148,46 +154,48 @@ def getUserActivity(handle):
 
     r = requests.get(url).json()
     if r['status'] == 'FAILED':
-	expectedMessage = 'handle: User with handle ' + str(handle) + ' not found'
-	if r['comment'] == expectedMessage:
-	    raise UserNotFound(handle)
+      expectedMessage = 'handle: User with handle ' + str(handle) + ' not found'
+      if r['comment'] == expectedMessage:
+          raise UserNotFound(handle)
 
     dictlist = []
     keys = ['testset', 'passedTestCount', 'author', 'relativeTimeSeconds', 
-	    'language', 'memoryBytes', 'timeMilliseconds', 'problem_name', 'problem_index',
-	    'problem_tags', 'points', 'contestID', 'verdict', 'id', 'participantType', 'startTimeSeconds']
+      'language', 'memoryBytes', 'timeMilliseconds', 'problem_name', 'problem_index',
+      'problem_tags', 'points', 'contestID', 'verdict', 'id', 'participantType', 'startTimeSeconds']
 
     r = r['result']
     for rr in r:
-	temp = dict.fromkeys(keys)
-	temp['author'] = rr['author']['members'][0]['handle']
-	temp['startTimeSeconds'] = rr['creationTimeSeconds']
-
-	if 'startTimeSeconds' not in rr['author']:
-	    temp['participantType'] = 'GYM'
-	else:
-	    temp['participantType'] = rr['author']['participantType']
-
-	temp['id'] = rr['id']
-	temp['verdict'] = rr['verdict']
-	temp['contestID'] = rr['contestId']
-
-	if 'points' not in rr['problem']:
-	    temp['points'] = 0
-	else:
-	    temp['points'] = rr['problem']['points']
-
-	temp['problem_tags'] = rr['problem']['tags']
-	temp['problem_index'] = rr['problem']['index']
-	temp['problem_name'] = rr['problem']['name']
-	temp['timeMilliseconds'] = rr['timeConsumedMillis']
-	temp['memoryBytes'] = rr['memoryConsumedBytes']
-	temp['language'] = rr['programmingLanguage']
-	temp['relativeTimeSeconds'] = rr['relativeTimeSeconds']
-	temp['passedTestCount'] = rr['passedTestCount']
-	temp['testset'] = rr['testset']
-
-	dictlist.append(temp)
+      temp = dict.fromkeys(keys)
+      temp['author'] = rr['author']['members'][0]['handle']
+      temp['startTimeSeconds'] = rr['creationTimeSeconds']
+
+      if 'startTimeSeconds' not in rr['author']:
+          temp['participantType'] = 'GYM'
+      else:
+          temp['participantType'] = rr['author']['participantType']
+
+      temp['id'] = rr['id']
+      temp['verdict'] = rr['verdict']
+      if 'contestId' not in rr: continue
+      temp['contestID'] = rr['contestId']
+
+      if 'points' not in rr['problem']:
+          temp['points'] = 0
+      else:
+          temp['points'] = rr['problem']['points']
+
+      temp['problem_tags'] = rr['problem']['tags']
+      temp['problem_index'] = rr['problem']['index']
+      temp['problem_name'] = rr['problem']['name']
+      temp['timeMilliseconds'] = rr['timeConsumedMillis']
+      temp['memoryBytes'] = rr['memoryConsumedBytes']
+      temp['language'] = rr['programmingLanguage']
+      temp['relativeTimeSeconds'] = rr['relativeTimeSeconds']
+      temp['passedTestCount'] = rr['passedTestCount']
+      temp['testset'] = rr['testset']
+
+      dictlist.append(temp)
+
     df_userActivity = pd.DataFrame.from_dict(dictlist)
     return df_userActivity
 
@@ -196,10 +204,9 @@ def getUserRatingHistory(handle):
 
     r = requests.get(url).json()
     if r['status'] == 'FAILED':
-	expectedMessage = 'handle: User with handle ' + str(handle) + ' not found'
-	if r['comment'] == expectedMessage:
-	    raise UserNotFound(handle)
-
+      expectedMessage = 'handle: User with handle ' + str(handle) + ' not found'
+      if r['comment'] == expectedMessage:
+          raise UserNotFound(handle)
     r = r['result']
     rating_history = pd.DataFrame.from_dict(r)
     return rating_history

ui/bad_contests.txt

@@ -0,0 +1,187 @@
+925
+927
+958
+952
+924
+926
+953
+923
+929
+928
+921
+916
+910
+905
+886
+889
+890
+888
+885
+887
+884
+883
+873
+874
+863
+847
+858
+860
+861
+857
+856
+846
+852
+845
+838
+837
+825
+823
+826
+818
+819
+820
+821
+817
+813
+802
+808
+773
+803
+775
+772
+797
+774
+795
+784
+792
+771
+770
+769
+761
+762
+753
+751
+728
+730
+726
+718
+720
+717
+712
+710
+703
+702
+691
+690
+695
+693
+683
+678
+684
+643
+642
+641
+665
+663
+664
+640
+661
+660
+656
+639
+652
+647
+646
+638
+644
+637
+632
+636
+628
+630
+622
+620
+616
+612
+609
+600
+598
+597
+594
+595
+589
+575
+566
+562
+541
+537
+532
+530
+531
+524
+523
+522
+491
+484
+485
+470
+440
+411
+409
+398
+399
+396
+397
+394
+392
+393
+391
+390
+386
+374
+345
+328
+323
+311
+312
+306
+301
+302
+291
+290
+267
+212
+209
+207
+206
+188
+190
+185
+182
+172
+171
+170
+162
+159
+158
+153
+147
+134
+130
+125
+120
+119
+100
+98
+99
+76
+72
+64
+52
+50
+45
+44
+39
+21
+20
+15

ui/elo.py

@@ -55,10 +55,14 @@ def get_problem_elo(problem_df):
 def get_contest_elo(contestID):
     try:
         contest_df = af.getSolveSuccessDF(contestID)
-    except:
+    except Exception as e:
+        with open('bad_contests.txt', 'a') as f:
+          f.write(str(contestID) + '\n')
         return None
-
+  
     if contest_df.shape[0] < 50:
+        with open('bad_contests.txt', 'a') as f:
+          f.write(str(contestID) + '\n')
         return None
 
     contestID = contest_df.contestID.unique()[0]