def nick_change_graph(log_dict, DAY_BY_DAY_ANALYSIS=False):
""" creates a graph which tracks the nick changes of the users
where each edge has a time stamp denoting the time
at which the nick was changed by the user
Args:
log_dict (str): Dictionary of logs created using reader.py
Returns:
list of the day_to_day nick changes if config.DAY_BY_DAY_ANALYSIS=True or else an aggregate nick change graph for the
given time period.
"""
rem_time = None #remembers the time of the last message of the file parsed before the current file
nick_change_day_list = []
aggregate_nick_change_graph = nx.MultiDiGraph() # graph for nick changes in the whole time span (not day to day)
for day_content_all_channels in log_dict.values():
for day_content in day_content_all_channels:
day_log = day_content["log_data"]
today_nick_change_graph = nx.MultiDiGraph() #using networkx
current_line_no = -1
for line in day_log:
current_line_no = current_line_no + 1
if(line[0] == '=' and "changed the topic of" not in line): #excluding the condition when user changes the topic. Search for only nick changes
nick1 = util.splice_find(line, "=", " is", 3)
nick2 = util.splice_find(line, "wn as", "\n", 5)
earlier_line_no = current_line_no
while earlier_line_no >= 0: #to find the line just before "=="" so as to find time of Nick Change
earlier_line_no = earlier_line_no - 1
if(day_log[earlier_line_no][0] != '='):
year, month, day = util.get_year_month_day(day_content)
util.build_graphs(nick1, nick2, day_log[earlier_line_no][1:6], year, month, day, today_nick_change_graph, aggregate_nick_change_graph)
break
if(earlier_line_no == -1):
today_nick_change_graph.add_edge(nick1, nick2, weight=rem_time)
aggregate_nick_change_graph.add_edge(nick1, nick2, weight = rem_time)
count = len(day_log) - 1 #setting up the rem_time for next file, by noting the last message sent on that file.
while(count >= 0):
if(day_log[count][0] != '='):
rem_time = day_log[count][1:6]
break
count = count-1
nick_change_day_list.append(today_nick_change_graph)
if DAY_BY_DAY_ANALYSIS:
return nick_change_day_list
else:
return aggregate_nick_change_graph
def nick_change_graph(log_dict, DAY_BY_DAY_ANALYSIS=False):
""" creates a graph which tracks the nick changes of the users
where each edge has a time stamp denoting the time
at which the nick was changed by the user
Args:
log_dict (str): Dictionary of logs created using reader.py
Returns:
list of the day_to_day nick changes if config.DAY_BY_DAY_ANALYSIS=True or else an aggregate nick change graph for the
given time period.
"""
rem_time = None #remembers the time of the last message of the file parsed before the current file
nick_change_day_list = []
aggregate_nick_change_graph = nx.MultiDiGraph(
) # graph for nick changes in the whole time span (not day to day)
for day_content_all_channels in log_dict.values():
for day_content in day_content_all_channels:
day_log = day_content["log_data"]
today_nick_change_graph = nx.MultiDiGraph() #using networkx
current_line_no = -1
for line in day_log:
current_line_no = current_line_no + 1
if (
line[0] == '=' and "changed the topic of" not in line
): #excluding the condition when user changes the topic. Search for only nick changes
nick1 = util.splice_find(line, "=", " is", 3)
nick2 = util.splice_find(line, "wn as", "\n", 5)
earlier_line_no = current_line_no
while earlier_line_no >= 0: #to find the line just before "=="" so as to find time of Nick Change
earlier_line_no = earlier_line_no - 1
if (day_log[earlier_line_no][0] != '='):
year, month, day = util.get_year_month_day(
day_content)
util.build_graphs(nick1, nick2,
day_log[earlier_line_no][1:6],
year, month, day,
today_nick_change_graph,
aggregate_nick_change_graph)
break
if (earlier_line_no == -1):
today_nick_change_graph.add_edge(nick1,
nick2,
weight=rem_time)
aggregate_nick_change_graph.add_edge(nick1,
nick2,
weight=rem_time)
count = len(
day_log
) - 1 #setting up the rem_time for next file, by noting the last message sent on that file.
while (count >= 0):
if (day_log[count][0] != '='):
rem_time = day_log[count][1:6]
break
count = count - 1
nick_change_day_list.append(today_nick_change_graph)
if DAY_BY_DAY_ANALYSIS:
return nick_change_day_list
else:
return aggregate_nick_change_graph
def test_get_year_month_day(self, auxiliary_day_content, expected_output):
self.assertEqual(util.get_year_month_day(auxiliary_day_content),
expected_output)
def message_time_graph(log_dict, nicks, nick_same_list, DAY_BY_DAY_ANALYSIS=False):
""" creates a directed graph where each edge denotes a message sent from a user to another user
with the stamp denoting the time at which the message was sent
Args:
log_dict (dictionary): Dictionary of logs data created using reader.py
nicks(List) : List of nickname created using nickTracker.py
nick_same_list(List) :List of same_nick names created using nickTracker.py
Returns:
msg_time_graph_list(List): List of message time graphs for different days
msg_time_aggr_graph: aggregate message time graph where edges are date + time when sender sends a message to receiver
"""
msg_time_graph_list = []
msg_time_aggr_graph = nx.MultiDiGraph()
G = util.to_graph(nick_same_list)
conn_comp_list = list(connected_components(G))
def compare_spliced_nick(nick_to_compare, spliced_nick, nick_name, line):
if(nick_to_compare == nick_name):
if(spliced_nick != nick_name):
nick_receiver = nick_receiver_from_conn_comp(nick_name, conn_comp_list)
util.build_graphs(nick_sender, nick_receiver, line[1:6], year, month, day, graph_conversation, msg_time_aggr_graph)
util.create_connected_nick_list(conn_comp_list)
for day_content_all_channels in log_dict.values():
for day_content in day_content_all_channels:
day_log = day_content["log_data"]
year, month, day = util.get_year_month_day(day_content)
graph_conversation = nx.MultiDiGraph() #graph with multiple directed edges between clients used
for line in day_log:
flag_comma = 0
if(util.check_if_msg_line (line)):
m = re.search(r"\<(.*?)\>", line)
spliced_nick = util.correctLastCharCR(m.group(0)[1:-1])
nick_sender = ""
nick_sender = util.get_nick_sen_rec(config.MAX_EXPECTED_DIFF_NICKS, spliced_nick, conn_comp_list, nick_sender)
for nick_name in nicks:
rec_list = [e.strip() for e in line.split(':')] #receiver list splited about :
util.rec_list_splice(rec_list)
if not rec_list[1]: #index 0 will contain time 14:02
break
rec_list = util.correct_last_char_list(rec_list)
for nick_to_search in rec_list:
if(nick_to_search == nick_name):
if(spliced_nick != nick_name):
nick_receiver = ""
nick_receiver = util.get_nick_sen_rec(config.MAX_EXPECTED_DIFF_NICKS, nick_name, conn_comp_list, nick_receiver)
util.build_graphs(nick_sender, nick_receiver, line[1:6], year, month, day, graph_conversation, msg_time_aggr_graph)
if "," in rec_list[1]: #receiver list may of the form <Dhruv> Rohan, Ram :
flag_comma = 1
rec_list_2 = [e.strip() for e in rec_list[1].split(',')]
rec_list_2 = util.correct_last_char_list(rec_list_2)
for nick_to_search in rec_list_2:
compare_spliced_nick(nick_to_search, spliced_nick, nick_name, line)
if(flag_comma == 0): #receiver list can be <Dhruv> Rohan, Hi!
rec = line[line.find(">") + 1:line.find(", ")]
rec = util.correctLastCharCR(rec[1:])
compare_spliced_nick(rec, spliced_nick, nick_name, line)
msg_time_graph_list.append(graph_conversation)
if DAY_BY_DAY_ANALYSIS:
return msg_time_graph_list
else:
return msg_time_aggr_graph
def message_number_graph(log_dict, nicks, nick_same_list, DAY_BY_DAY_ANALYSIS=False):
""" Creates a directed graph
with each node representing an IRC user
and each directed edge has a weight which
mentions the number messages sent and recieved by that user
in the selected time frame.
Args:
log_dict (dict): with key as dateTime.date object and value as {"data":datalist,"channel_name":channels name}
nicks(list): list of all the nicks
nick_same_list(list): list of lists mentioning nicks which belong to same users
Returns:
message_number_graph (nx graph object)
"""
message_number_day_list = []
conversations=[[0] for i in range(config.MAX_EXPECTED_DIFF_NICKS)]
aggregate_message_number_graph = nx.DiGraph() #graph with multiple directed edges between clients used
G = util.to_graph(nick_same_list)
conn_comp_list = list(connected_components(G))
util.create_connected_nick_list(conn_comp_list)
def msg_no_analysis_helper(rec_list, corrected_nick, nick, conn_comp_list,conversations,today_conversation):
for receiver in rec_list:
if(receiver == nick):
if(corrected_nick != nick):
nick_receiver = ''
nick_receiver = util.get_nick_sen_rec(config.MAX_EXPECTED_DIFF_NICKS, nick, conn_comp_list, nick_receiver)
if DAY_BY_DAY_ANALYSIS:
today_conversation = util.extend_conversation_list(nick_sender, nick_receiver, today_conversation)
else:
conversations = util.extend_conversation_list(nick_sender, nick_receiver, conversations)
def message_no_add_egde(message_graph, conversation):
for index in xrange(config.MAX_EXPECTED_DIFF_NICKS):
if(len(conversation[index]) == 3 and conversation[index][0] >= config.THRESHOLD_MESSAGE_NUMBER_GRAPH):
if len(conversation[index][1]) >= config.MINIMUM_NICK_LENGTH and len(conversation[index][2]) >= config.MINIMUM_NICK_LENGTH:
message_graph.add_edge(conversation[index][1], conversation[index][2], weight=conversation[index][0])
return message_graph
for day_content_all_channels in log_dict.values():
for day_content in day_content_all_channels:
day_log = day_content["log_data"]
today_conversation = [[0] for i in range(config.MAX_EXPECTED_DIFF_NICKS)]
for line in day_log:
flag_comma = 0
if(util.check_if_msg_line (line)):
parsed_nick = re.search(r"\<(.*?)\>", line)
corrected_nick = util.correctLastCharCR(parsed_nick.group(0)[1:-1])
nick_sender = ""
nick_receiver = ""
nick_sender = util.get_nick_sen_rec(config.MAX_EXPECTED_DIFF_NICKS, corrected_nick, conn_comp_list, nick_sender)
for nick in nicks:
rec_list = [e.strip() for e in line.split(':')]
util.rec_list_splice(rec_list)
if not rec_list[1]:
break
rec_list = util.correct_last_char_list(rec_list)
msg_no_analysis_helper(rec_list, corrected_nick, nick, conn_comp_list, conversations,today_conversation)
if "," in rec_list[1]:
flag_comma = 1
rec_list_2=[e.strip() for e in rec_list[1].split(',')]
for i in xrange(0,len(rec_list_2)):
if(rec_list_2[i]):
rec_list_2[i] = util.correctLastCharCR(rec_list_2[i])
msg_no_analysis_helper(rec_list_2, corrected_nick, nick, conn_comp_list, conversations, today_conversation)
if(flag_comma == 0):
rec = line[line.find(">")+1:line.find(", ")]
rec = rec[1:]
rec = util.correctLastCharCR(rec)
if(rec == nick):
if(corrected_nick != nick):
nick_receiver = nick_receiver_from_conn_comp(nick, conn_comp_list)
if DAY_BY_DAY_ANALYSIS:
today_message_number_graph = nx.DiGraph()
today_message_number_graph = message_no_add_egde(today_message_number_graph, today_conversation)
year, month, day = util.get_year_month_day(day_content)
message_number_day_list.append([today_message_number_graph, year+'-'+month+'-'+day])
print "\nBuilding graph object with EDGE WEIGHT THRESHOLD:", config.THRESHOLD_MESSAGE_NUMBER_GRAPH
if not DAY_BY_DAY_ANALYSIS:
aggregate_message_number_graph = message_no_add_egde(aggregate_message_number_graph, conversations)
if config.DEBUGGER:
print "========> 30 on " + str(len(conversations)) + " conversations"
print conversations[:30]
if DAY_BY_DAY_ANALYSIS:
return message_number_day_list
else:
return aggregate_message_number_graph
def message_time_graph(log_dict,
nicks,
nick_same_list,
DAY_BY_DAY_ANALYSIS=False):
""" creates a directed graph where each edge denotes a message sent from a user to another user
with the stamp denoting the time at which the message was sent
Args:
log_dict (dictionary): Dictionary of logs data created using reader.py
nicks(List) : List of nickname created using nickTracker.py
nick_same_list(List) :List of same_nick names created using nickTracker.py
Returns:
msg_time_graph_list(List): List of message time graphs for different days
msg_time_aggr_graph: aggregate message time graph where edges are date + time when sender sends a message to receiver
"""
msg_time_graph_list = []
msg_time_aggr_graph = nx.MultiDiGraph()
G = util.to_graph(nick_same_list)
conn_comp_list = list(connected_components(G))
def compare_spliced_nick(nick_to_compare, spliced_nick, nick_name, line):
if (nick_to_compare == nick_name):
if (spliced_nick != nick_name):
nick_receiver = nick_receiver_from_conn_comp(
nick_name, conn_comp_list)
util.build_graphs(nick_sender, nick_receiver, line[1:6], year,
month, day, graph_conversation,
msg_time_aggr_graph)
util.create_connected_nick_list(conn_comp_list)
for day_content_all_channels in log_dict.values():
for day_content in day_content_all_channels:
day_log = day_content["log_data"]
year, month, day = util.get_year_month_day(day_content)
graph_conversation = nx.MultiDiGraph(
) #graph with multiple directed edges between clients used
for line in day_log:
flag_comma = 0
if (util.check_if_msg_line(line)):
m = re.search(r"\<(.*?)\>", line)
spliced_nick = util.correctLastCharCR(m.group(0)[1:-1])
nick_sender = ""
nick_sender = util.get_nick_sen_rec(
config.MAX_EXPECTED_DIFF_NICKS, spliced_nick,
conn_comp_list, nick_sender)
for nick_name in nicks:
rec_list = [e.strip() for e in line.split(':')
] #receiver list splited about :
util.rec_list_splice(rec_list)
if not rec_list[1]: #index 0 will contain time 14:02
break
rec_list = util.correct_last_char_list(rec_list)
for nick_to_search in rec_list:
if (nick_to_search == nick_name):
if (spliced_nick != nick_name):
nick_receiver = ""
nick_receiver = util.get_nick_sen_rec(
config.MAX_EXPECTED_DIFF_NICKS,
nick_name, conn_comp_list,
nick_receiver)
util.build_graphs(nick_sender,
nick_receiver, line[1:6],
year, month, day,
graph_conversation,
msg_time_aggr_graph)
if "," in rec_list[
1]: #receiver list may of the form <Dhruv> Rohan, Ram :
flag_comma = 1
rec_list_2 = [
e.strip() for e in rec_list[1].split(',')
]
rec_list_2 = util.correct_last_char_list(
rec_list_2)
for nick_to_search in rec_list_2:
compare_spliced_nick(nick_to_search,
spliced_nick, nick_name,
line)
if (flag_comma == 0
): #receiver list can be <Dhruv> Rohan, Hi!
rec = line[line.find(">") + 1:line.find(", ")]
rec = util.correctLastCharCR(rec[1:])
compare_spliced_nick(rec, spliced_nick, nick_name,
line)
msg_time_graph_list.append(graph_conversation)
if DAY_BY_DAY_ANALYSIS:
return msg_time_graph_list
else:
return msg_time_aggr_graph
def message_number_graph(log_dict,
nicks,
nick_same_list,
DAY_BY_DAY_ANALYSIS=False):
""" Creates a directed graph
with each node representing an IRC user
and each directed edge has a weight which
mentions the number messages sent and recieved by that user
in the selected time frame.
Args:
log_dict (dict): with key as dateTime.date object and value as {"data":datalist,"channel_name":channels name}
nicks(list): list of all the nicks
nick_same_list(list): list of lists mentioning nicks which belong to same users
Returns:
message_number_graph (nx graph object)
"""
message_number_day_list = []
conversations = [[0] for i in range(config.MAX_EXPECTED_DIFF_NICKS)]
aggregate_message_number_graph = nx.DiGraph(
) #graph with multiple directed edges between clients used
G = util.to_graph(nick_same_list)
conn_comp_list = list(connected_components(G))
util.create_connected_nick_list(conn_comp_list)
def msg_no_analysis_helper(rec_list, corrected_nick, nick, conn_comp_list,
conversations, today_conversation):
for receiver in rec_list:
if (receiver == nick):
if (corrected_nick != nick):
nick_receiver = ''
nick_receiver = util.get_nick_sen_rec(
config.MAX_EXPECTED_DIFF_NICKS, nick, conn_comp_list,
nick_receiver)
if DAY_BY_DAY_ANALYSIS:
today_conversation = util.extend_conversation_list(
nick_sender, nick_receiver, today_conversation)
else:
conversations = util.extend_conversation_list(
nick_sender, nick_receiver, conversations)
def message_no_add_egde(message_graph, conversation):
for index in xrange(config.MAX_EXPECTED_DIFF_NICKS):
if (len(conversation[index]) == 3 and conversation[index][0] >=
config.THRESHOLD_MESSAGE_NUMBER_GRAPH):
if len(
conversation[index]
[1]) >= config.MINIMUM_NICK_LENGTH and len(
conversation[index][2]) >= config.MINIMUM_NICK_LENGTH:
message_graph.add_edge(conversation[index][1],
conversation[index][2],
weight=conversation[index][0])
return message_graph
for day_content_all_channels in log_dict.values():
for day_content in day_content_all_channels:
day_log = day_content["log_data"]
today_conversation = [
[0] for i in range(config.MAX_EXPECTED_DIFF_NICKS)
]
for line in day_log:
flag_comma = 0
if (util.check_if_msg_line(line)):
parsed_nick = re.search(r"\<(.*?)\>", line)
corrected_nick = util.correctLastCharCR(
parsed_nick.group(0)[1:-1])
nick_sender = ""
nick_receiver = ""
nick_sender = util.get_nick_sen_rec(
config.MAX_EXPECTED_DIFF_NICKS, corrected_nick,
conn_comp_list, nick_sender)
for nick in nicks:
rec_list = [e.strip() for e in line.split(':')]
util.rec_list_splice(rec_list)
if not rec_list[1]:
break
rec_list = util.correct_last_char_list(rec_list)
msg_no_analysis_helper(rec_list, corrected_nick, nick,
conn_comp_list, conversations,
today_conversation)
if "," in rec_list[1]:
flag_comma = 1
rec_list_2 = [
e.strip() for e in rec_list[1].split(',')
]
for i in xrange(0, len(rec_list_2)):
if (rec_list_2[i]):
rec_list_2[i] = util.correctLastCharCR(
rec_list_2[i])
msg_no_analysis_helper(rec_list_2, corrected_nick,
nick, conn_comp_list,
conversations,
today_conversation)
if (flag_comma == 0):
rec = line[line.find(">") + 1:line.find(", ")]
rec = rec[1:]
rec = util.correctLastCharCR(rec)
if (rec == nick):
if (corrected_nick != nick):
nick_receiver = nick_receiver_from_conn_comp(
nick, conn_comp_list)
if DAY_BY_DAY_ANALYSIS:
today_message_number_graph = nx.DiGraph()
today_message_number_graph = message_no_add_egde(
today_message_number_graph, today_conversation)
year, month, day = util.get_year_month_day(day_content)
message_number_day_list.append([
today_message_number_graph, year + '-' + month + '-' + day
])
print "\nBuilding graph object with EDGE WEIGHT THRESHOLD:", config.THRESHOLD_MESSAGE_NUMBER_GRAPH
if not DAY_BY_DAY_ANALYSIS:
aggregate_message_number_graph = message_no_add_egde(
aggregate_message_number_graph, conversations)
if config.DEBUGGER:
print "========> 30 on " + str(len(conversations)) + " conversations"
print conversations[:30]
if DAY_BY_DAY_ANALYSIS:
return message_number_day_list
else:
return aggregate_message_number_graph