def drow_l_s(): # lista sasiedstwa
inciliststr = stext4.get('1.0', 'end-1c')
incidencelist = inciliststr.split('\n')
P = nx.parse_adjlist(incidencelist, nodetype=int)
nx.draw_networkx(P)
plt.show()
def overall_call(original_file, attributes_file, start_point_id, length_path,
set_progress_funk):
data = pd.read_csv(attributes_file)
list_of_adjacencies = data[['fid', 'fid_2']]
list_of_adjacencies = np.array(list_of_adjacencies)
lines = prepare(list_of_adjacencies)
G = nx.parse_adjlist(lines, nodetype=str)
l_dataframe = pd.read_csv(length_path)
l_dataframe = l_dataframe.astype({'id': 'str'})
last_vertex = distance_attr(G, str(start_point_id), l_dataframe)
rank_set(G, str(start_point_id), str(last_vertex), set_progress_funk)
iter_set_values(G, str(start_point_id))
dataframe = make_dataframe(G)
rivers = pd.read_csv(original_file)
rivers = rivers.astype({'fid': 'str'})
data_merged = pd.merge(rivers, dataframe, on='fid')
rows_count = data_merged.shape[0]
df_dict = {}
for i in range(rows_count):
df_dict[int(data_merged.iloc[i]['fid'])] = [
int(data_merged.iloc[i]['Rank']),
int(data_merged.iloc[i]['Value']),
int(data_merged.iloc[i]['Distance'])
]
return df_dict
def draw_graph_from_adjlist(adjlist, prog='dot', **kwargs):
lines = [' '.join(str(x) for x in line) for line in adjlist]
g = nx.parse_adjlist(lines, create_using=nx.DiGraph(), nodetype=int)
pos = nx.drawing.nx_pydot.graphviz_layout(g, prog=prog)
nx.draw(g, pos, with_labels=True, arrows=True, **kwargs)
#plt.show()
return pos
def create_graph_enzymes(file):
f = open(file, 'r')
lines = f.read().splitlines()
f.close()
# get the indices of the vertext, adj list and class
idx_vertex = lines.index("#v - vertex labels")
idx_adj_list = lines.index("#a - adjacency list")
idx_clss = lines.index("#c - Class")
# node label
vl = [int(ivl) for ivl in lines[idx_vertex+1:idx_adj_list]]
adj_list = lines[idx_adj_list+1:idx_clss]
sources = list(range(1,len(adj_list)+1))
for i in range(len(adj_list)):
if not adj_list[i]:
adj_list[i] = str(sources[i])
else:
adj_list[i] = str(sources[i])+","+adj_list[i]
g = nx.parse_adjlist(adj_list, nodetype=int, delimiter=",")
for i in range(1, g.number_of_nodes()+1):
g.node[i]['labels'] = np.array(vl[i-1])
c = int(lines[idx_clss+1])
return g, c
def from_adjlist(self, filename, delimiter=None, nodetype=int):
""" Loads a Cluster from its adjacency list stored in a file `filename`
"""
with open(filename, 'r') as f:
graph = nx.parse_adjlist(lines=f,
nodetype=nodetype,
delimiter=delimiter)
self.graph = graph
def MCL(cdr3, edgelist=None, mcl_hyper=[1.2, 2], outfile=None):
"""
Perform clustering on a network of CDR3 amino acid sequences with
a known hamming distance, using the Markov clustering (MCL) algorithm.
For more info about the inflation and expansion parameters,
visit: https://micans.org/mcl/
Parameters
----------
edgelist : set, optional
Tab-separated edgelist. The default is None.
mcl_hyper : list, optional
MCL hyperparameters: inflation and expansion.
The default is [1.2,2].
outfile : str, optional
Name of outfile. The default is None.
Returns
-------
clusters : pd.DataFrame
pd.DataFrame containing two columns: 'CDR3' and 'cluster'.
The first column contains CDR3 sequences, the second column
contains the corresponding cluster ids.
"""
if edgelist is None:
edgelist = create_edgelist(cdr3)
try:
G = nx.parse_adjlist(edgelist, nodetype=str)
m = nx.to_scipy_sparse_array(G)
# Run MCL
result = mcl.run_mcl(m, inflation=mcl_hyper[0], expansion=mcl_hyper[1])
mcl_output = mcl.get_clusters(result)
identifiers = list(G.nodes())
# Map cluster ids back to seqs
cluster_ids = dict()
for i in range(len(mcl_output)):
cluster_ids[i] = list(identifiers[i] for i in mcl_output[i])
# Generate nodelist
clusters = {"CDR3": [], "cluster": []}
for c in cluster_ids:
for seq in cluster_ids[c]:
clusters["CDR3"].append(seq)
clusters["cluster"].append(c)
clusters = pd.DataFrame(data=clusters)
# Write to file
if outfile is not None:
clusters.to_csv(outfile, sep="\t", index=False)
except nx.NetworkXError:
clusters = pd.DataFrame({"CDR3": [], "cluster": []})
return clusters
def fromDict(dict: Dict):
adj_list = dict["adj_list"]
graph = nx.parse_adjlist(adj_list, create_using=nx.DiGraph)
seed_id = dict["seed_id"]
params = dict["params"]
social_graph = SocialGraph(graph, seed_id, params)
return social_graph
def read_file(self, file):
graph_list = []
with open(file, 'r') as graph:
i = 0
for line in graph.readlines():
if i > 0:
graph_list.append(str(i) + ' ' + line)
i += 1
Adjlist = nx.parse_adjlist(graph_list, nodetype=int)
return Adjlist
def readFromFile():
with open("files/adjList.txt", 'r') as f:
tab = f.readlines()
ll = []
x = 0
for i in tab:
ii = str(x) + " " + i[1:-2].replace(',', '')
ll.append(ii)
x += 1
return nx.parse_adjlist(ll, nodetype = int)
def read_graph(self, filename):
G = nx.Graph()
lines = []
with open(filename, 'r') as graph:
num_data = graph.readline().split(' ')
num_node, num_edge = int(num_data[0]), int(num_data[1]) # not used
i = 0
for line in graph:
i = i + 1
lines.append(str(i) + ' ' + line[:-2])
G = nx.parse_adjlist(lines[:-1], nodetype=int)
return G
def readFromFile():
with open("files/adjList.txt", 'r') as f:
tab = f.readlines()
adjlist = []
x = 0
for i in tab:
i = str(x) + " " + i[1:-2].replace(',', '')
adjlist.append(i)
x += 1
return nx.parse_adjlist(adjlist)
def generate_adj_network_viz(networks):
'''
generates a networkx graph and from an adjacency list and saves as a pyplot drawing
:param networks: a list of lists where the inner lists are a list of characters that are a network, the first one
being the 'protagonist' or POV holder
:return: None, saves to png files
'''
# TODO remove the hardcoding of test indices
for index in range(1, 50, 5):
# this is janky. Because nx expects a string
curr_graph = nx.parse_adjlist([' '.join(networks[index])])
nx.draw_networkx(curr_graph)
plt.savefig('Chapter {} network.png'.format(index))
def get_fastg_digraph(fastg_name):
""" scans through fastg headers as an adjacency list
builds and returns a nx directed graph using adjacencies
note: no connections are created between each node and its
rc node - we need to take care to maintain these
"""
lines = []
fp = open(fastg_name, 'r')
for name,seq,qual in readfq(fp):
name = re.sub('[:,]'," ", name[:-1])
lines.append(name)
G = nx.DiGraph()
return nx.parse_adjlist(lines, create_using=G)
def get_fastg_digraph(fastg_name):
""" scans through fastg headers as an adjacency list
builds and returns a nx directed graph using adjacencies
note: no connections are created between each node and its
rc node - we need to take care to maintain these
"""
lines = []
fp = open(fastg_name, 'r')
for name, seq, qual in readfq(fp):
name = re.sub('[:,]', " ", name[:-1])
lines.append(name)
G = nx.DiGraph()
return nx.parse_adjlist(lines, create_using=G)
def louvain(cdr3, edgelist=None):
if edgelist is None:
edgelist = create_edgelist(cdr3)
try:
G = nx.parse_adjlist(edgelist, nodetype=str)
partition = community.best_partition(G)
except nx.NetworkXError:
partition = pd.DataFrame({"CDR3": [], "cluster": []})
return pd.DataFrame.from_dict(
partition, orient="index", columns=["cluster"]
).reset_index().rename(columns={'index': 'CDR3'})
def test_parse_adjlist(self):
lines = ["1 2 5", "2 3 4", "3 5", "4", "5"]
nx.parse_adjlist(lines, nodetype=int) # smoke test
with pytest.raises(TypeError):
nx.parse_adjlist(lines, nodetype="int")
lines = ["1 2 5", "2 b", "c"]
with pytest.raises(TypeError):
nx.parse_adjlist(lines, nodetype=int)
def calculate_cetrality(self, data):
logger.info("Starting calculating centrality")
adj_list = []
for user in data:
adj_list.append(str(user.user_oid_i) + " " + str(user.user_oid_j))
g = nx.parse_adjlist(adj_list, nodetype=int)
logger.info(str(len(adj_list)))
# calculo centralidad de grados
degree_centrality = nx.degree_centrality(g)
logger.info("Finish calculating centrality")
return degree_centrality
def calculate_cetrality(self, data):
logger.info("Starting calculating centrality")
adj_list = []
for user in data:
adj_list.append(str(user.user_oid_i) + " " + str(user.user_oid_j))
g = nx.parse_adjlist(adj_list, nodetype=int)
logger.info(str(len(adj_list)))
# calculo centralidad de grados
degree_centrality = nx.degree_centrality(g)
logger.info("Finish calculating centrality")
return degree_centrality
def test_graph_conversion_to_text(self):
"""
convert graph to text, as adjacency list.
:return:
"""
yt_script.convert_graph_to_text(self.MOCK_GRAPH, self.MOCK_FILE_OUTPUT)
with open(self.MOCK_FILE_OUTPUT) as f:
output = f.read()
output = output.split('\n')
result_graph = nx.parse_adjlist(output)
for node in self.MOCK_GRAPH.nodes():
self.assertIn(node, result_graph.nodes())
for edge in self.MOCK_GRAPH.edges():
try:
self.assertIn(edge, result_graph.edges())
except AssertionError:
edge = (edge[1], edge[0])
self.assertIn(edge, result_graph.edges())
continue
def test_graph_conversion_to_text(self):
"""
convert graph to text, as adjacency list.
:return:
"""
yt_script.convert_graph_to_text(self.MOCK_GRAPH, self.MOCK_FILE_OUTPUT)
with open(self.MOCK_FILE_OUTPUT) as f:
output = f.read()
output = output.split('\n')
result_graph = nx.parse_adjlist(output)
for node in self.MOCK_GRAPH.nodes():
self.assertIn(node, result_graph.nodes())
for edge in self.MOCK_GRAPH.edges():
try:
self.assertIn(edge, result_graph.edges())
except AssertionError:
edge = (edge[1], edge[0])
self.assertIn(edge, result_graph.edges())
continue
def main():
maximum_word_length = input("Enter max word length: ")
while type(maximum_word_length) != int and (int(maximum_word_length)<1 or int(maximum_word_length)>16 ):
maximum_word_length = input("please input a real integer value: ")
maximum_word_length = int(maximum_word_length)
board = random_board_2d()
display_2d_board(board)
board = unravel_board(board) #change this soon
#words have an n+1 length for an n length path
word_list = make_word_list('word_lists/words')
#make the graph
adj = build_adjacency_list(board)
G = nx.parse_adjlist(adj)
substrings = find_valid_paths(G,board,maximum_word_length)
valid_words = check_substrings_for_words(word_list,substrings)
print(len(valid_words))
for word in valid_words:
print(word)
def get_centrality_by_user(self, users):
# armo el "adjencency list" como string
adj_list = []
for user in users:
followers = ""
user_followers = [follower.user_id for follower in user.followers.all()]
if user.followers:
followers = " " + " ".join(str(x) for x in user_followers)
adj_list.append(str(user.user_id) + followers)
# armo el grafo a partir del string
g = nx.parse_adjlist(adj_list, nodetype=int)
# calculo centralidad de grados
degree_centrality = nx.degree_centrality(g)
# closeness_centrality = nx.closeness_centrality(g)
# betweenness_centrality = nx.betweenness_centrality(g)
# eigenvector_centrality = nx.eigenvector_centrality(g)
return degree_centrality
def get(path):
with open(path, 'r') as f:
tab = f.readlines()
list = []
for i in tab:
list.append(i[2:-2])
list[len(list)-1] = list[len(list)-1].replace(' ' , '')
adj_list = []
for k, v in enumerate(list):
adj_list.append([])
for key, value in enumerate(v):
if int(value) == 1:
adj_list[k].append(key)
list = []
for k, v in enumerate(adj_list):
list.append(str(k))
for i in v:
list[k] += " " + str(i)
G = nx.parse_adjlist(list)
return G
def draw_graph_from_adjacency_list(self, adjacency_list):
self.graph = nx.parse_adjlist(adjacency_list, nodetype=int)
self.draw_graph()
def bnb(inst, cutoff, seed):
graph = nx.Graph()
adj_list = []
v = 0
e = 0
with open(inst, 'r') as readGraph:
i = 0
for line in readGraph.readlines():
if i == 0:
tmp = line.split()
v, e = int(tmp[0]), int(tmp[1])
i += 1
else:
adj_list.append(str(i) + ' ' + line)
i += 1
graph = nx.parse_adjlist(adj_list, nodetype=int)
graphDup = graph.copy()
vDup = maxDeg(graphDup)
f = []
f.append((vDup[0], 0, (-1, -1)))
f.append((vDup[0], 1, (-1, -1)))
best = []
current = []
sol = []
ub = graph.number_of_nodes()
s = time.time()
end = s
t = end - s
while f != [] and t < cutoff:
(vx, state, parent) = f.pop()
bt = False
if state == 0:
for i in list(graphDup.neighbors(vx)):
current.append((i, 1))
graphDup.remove_node(i)
elif state == 1:
graphDup.remove_node(vx)
else:
pass
current.append((vx, state))
currentSz = vcsz(current)
if graphDup.number_of_edges() == 0:
if currentSz < ub:
best = current.copy()
ub = currentSz
sol.append([time.time() - s, currentSz])
bt = True
else:
currlb = currentSz + lb(graphDup)
if currlb < ub:
vy = maxDeg(graphDup)
f.append((vy[0], 0, (vx, state)))
f.append((vy[0], 1, (vx, state)))
else:
bt = True
if bt:
if f:
#nnp = f[-1][2]
if f[-1][2] in current:
i = current.index(f[-1][2]) + 1
while i < len(current):
currN, currState = current.pop()
graphDup.add_node(currN)
currvcn = list(map(lambda t: t[0], current))
for n in graph.neighbors(currN):
if n in graphDup.nodes() and n not in currvcn:
graphDup.add_edge(n, currN)
elif f[-1][2] == (-1, -1):
current.clear()
graphDup = graph.copy()
end = time.time()
t = end - s
for v in best:
if v[1] == 0:
best.remove(v)
vc = []
for v, d in best:
vc.append(v)
output_file(inst, cutoff, len(vc), vc, sol)
def create_graph(from_seconds=30):
D = nx.DiGraph()
return nx.parse_adjlist(extract_nodes(from_seconds), create_using=D)
for record in tweet:
users_topico.append(record['user']['id'])
users_filter = db.user.find({"id": {"$in": users_topico}})
# armo el "adjencency list" como string
adj_list = []
for user in users_filter:
followers = ""
if user['followers']:
followers = " " + " ".join(str(x) for x in user['followers'])
adj_list.append(str(user['id']) + followers)
# armo el grafo a partir del string
g = nx.parse_adjlist(adj_list, nodetype = int)
# calculo centralidad de grado
degree_centrality = nx.degree_centrality(g)
# closeness_centrality = nx.closeness_centrality(g)
# betweenness_centrality = nx.betweenness_centrality(g)
# eigenvector_centrality = nx.eigenvector_centrality(g)
# obtengo el usuario con mayor centralidad
max = max(degree_centrality.iterkeys(), key = lambda k: degree_centrality[k])
user = db.user.find_one({u'id': max})
if __name__ == '__main__':
if len(sys.argv) != 2:
print "Missing topic"
def draw_graph_from_adjlist(adjlist):
lines = [' '.join(str(x) for x in line) for line in adjlist]
g = nx.parse_adjlist(lines, create_using=nx.DiGraph(), nodetype=int)
pos = nx.drawing.nx_pydot.graphviz_layout(g, prog='dot')
nx.draw(g, pos, with_labels=True, arrows=True)
plt.show()
def read_txt(f): graph = nx.parse_adjlist(f) return process_graph(graph)
#!/usr/bin/python
import sys
#import csv
import networkx as nx
# for line in sys.stdin:
# print line
# edge_list = [node[0].replace("\t",",") for node in line]
# G = nx.parse_adjlist(edge_list, delimiter=",", nodetype=int)
# print G
input = sys.argv[0]
with open(input, "rb") as connect:
connections = csv.reader(connect, delimiter="\n")
edge_list = [node[0].replace("\t",",") for node in connections]
G = nx.parse_adjlist(edge_list, delimiter=",", nodetype=int)
# print edge_list
for node in nx.nodes(G):
for non_con in nx.non_neighbors(G, node):
for result in nx.common_neighbors(G, node, non_con):
print '{}\t{}\t{}'.format(node, non_con, 1)
def load_network(input_data):
# Create country_decade feature
dt_collect = input_data[['eventid', 'country_txt', 'iyear', 'gname']]
dt_collect["decade"] = (dt_collect['iyear'] // 10) * 10
dt_collect['country_decade'] = dt_collect[
'country_txt'] + '_' + dt_collect['decade'].map(str) + 's'
dt_collect = dt_collect[dt_collect.gname != 'Unknown']
# Create a country_decade edgelist
gnames_country_decade = dt_collect.groupby(['gname',
'country_decade']).agg({
'eventid':
'count'
}).reset_index()
gnames_country_decade_edgelist = pd.merge(gnames_country_decade,
gnames_country_decade,
on='country_decade',
how='left')
gnames_country_decade_edgelist.drop(['eventid_x', 'eventid_y'],
axis=1,
inplace=True)
gnames_country_decade_edgelist.columns = [
'source', 'country_decade', 'target'
]
gnames_country_decade_edgelist = gnames_country_decade_edgelist[
gnames_country_decade_edgelist.source !=
gnames_country_decade_edgelist.target]
G_country_decade = nx.from_pandas_edgelist(gnames_country_decade_edgelist,
source='source',
target='target')
# Create edgelist from the related column
dt_collect = input_data['related']
dt_collect.dropna(inplace=True)
gname_event_mapping = input_data[['eventid', 'gname']].drop_duplicates()
gname_event_mapping.eventid = gname_event_mapping.eventid.astype(str)
G_related = nx.parse_adjlist(dt_collect.values,
delimiter=', ') # attacks that are related
df_related = nx.to_pandas_edgelist(G_related)
df_related.replace(' ', '', regex=True, inplace=True)
df_source_gname = pd.merge(df_related,
gname_event_mapping,
how='left',
left_on='source',
right_on='eventid')
df_source_gname.rename(columns={'gname': 'gname_source'}, inplace=True)
df_target_gname = pd.merge(df_source_gname,
gname_event_mapping,
how='left',
left_on='target',
right_on='eventid')
df_target_gname.rename(columns={'gname': 'gname_target'}, inplace=True)
# Filtering and cleaning
gnames_relations_edgelist = df_target_gname[
df_target_gname.gname_source != df_target_gname.gname_target]
gnames_relations_edgelist = gnames_relations_edgelist[
gnames_relations_edgelist.gname_source != 'Unknown']
gnames_relations_edgelist = gnames_relations_edgelist[
gnames_relations_edgelist.gname_target != 'Unknown']
gnames_relations_edgelist = gnames_relations_edgelist[[
'gname_source', 'gname_target'
]]
gnames_relations_edgelist.dropna(inplace=True)
G_rel = nx.from_pandas_edgelist(gnames_relations_edgelist,
source='gname_source',
target='gname_target')
# Merging two graphs
G = nx.compose(G_country_decade, G_rel)
return G
def get_user_to_friends_graph(self, user):
query = self.user_friends_graph_collection.find_one({'user': user})
graph = nx.parse_adjlist(query['adj_list'])
return graph
def load_network(input_data):
# Create country_decade feature
dt_collect = input_data[["eventid", "country_txt", "iyear", "gname"]]
dt_collect["decade"] = (dt_collect["iyear"] // 10) * 10
dt_collect["country_decade"] = (dt_collect["country_txt"] + "_" +
dt_collect["decade"].map(str) + "s")
dt_collect = dt_collect[dt_collect.gname != "Unknown"]
# Create a country_decade edgelist
gnames_country_decade = (dt_collect.groupby(["gname",
"country_decade"]).agg({
"eventid":
"count"
}).reset_index())
gnames_country_decade_edgelist = pd.merge(gnames_country_decade,
gnames_country_decade,
on="country_decade",
how="left")
gnames_country_decade_edgelist.drop(["eventid_x", "eventid_y"],
axis=1,
inplace=True)
gnames_country_decade_edgelist.columns = [
"source", "country_decade", "target"
]
gnames_country_decade_edgelist = gnames_country_decade_edgelist[
gnames_country_decade_edgelist.source !=
gnames_country_decade_edgelist.target]
G_country_decade = nx.from_pandas_edgelist(gnames_country_decade_edgelist,
source="source",
target="target")
# Create edgelist from the related column
dt_collect = input_data["related"]
dt_collect.dropna(inplace=True)
gname_event_mapping = input_data[["eventid", "gname"]].drop_duplicates()
gname_event_mapping.eventid = gname_event_mapping.eventid.astype(str)
G_related = nx.parse_adjlist(dt_collect.values,
delimiter=", ") # attacks that are related
df_related = nx.to_pandas_edgelist(G_related)
df_related.replace(" ", "", regex=True, inplace=True)
df_source_gname = pd.merge(
df_related,
gname_event_mapping,
how="left",
left_on="source",
right_on="eventid",
)
df_source_gname.rename(columns={"gname": "gname_source"}, inplace=True)
df_target_gname = pd.merge(
df_source_gname,
gname_event_mapping,
how="left",
left_on="target",
right_on="eventid",
)
df_target_gname.rename(columns={"gname": "gname_target"}, inplace=True)
# Filtering and cleaning
gnames_relations_edgelist = df_target_gname[
df_target_gname.gname_source != df_target_gname.gname_target]
gnames_relations_edgelist = gnames_relations_edgelist[
gnames_relations_edgelist.gname_source != "Unknown"]
gnames_relations_edgelist = gnames_relations_edgelist[
gnames_relations_edgelist.gname_target != "Unknown"]
gnames_relations_edgelist = gnames_relations_edgelist[[
"gname_source", "gname_target"
]]
gnames_relations_edgelist.dropna(inplace=True)
G_rel = nx.from_pandas_edgelist(gnames_relations_edgelist,
source="gname_source",
target="gname_target")
# Merging two graphs
G = nx.compose(G_country_decade, G_rel)
return G
def __init__(self, adjlist, root=int()):
self.__root = root
self.__graph = nx.parse_adjlist(adjlist,
create_using=nx.DiGraph,
nodetype=int)
for (node, status, time_step) in new_infected: state_dict[node] = (status, time_step) if len(new_infected) == 0: break for key, (a,b) in state_dict.iteritems(): if a == 1: print "%s %s %s" % (iteration, key, b) urls = ["http://people.seas.harvard.edu/~yaron/cs284rFall2013/data/epinions.txt"] # ["http://people.seas.harvard.edu/~yaron/cs284rFall2013/data/enron.txt"] "http://people.seas.harvard.edu/~yaron/cs284rFall2013/data/livejournal.txt"] for url in urls : line_array = download_graphs(url) graph = nx.parse_adjlist(line_array,nodetype=int) voter_model(graph, 1000) title = url #make_graphs(graph, title)
def test_clustering(self):
adjlist = self.test_case[0].strip().split('\n')
g = networkx.parse_adjlist(adjlist, create_using=PrecedenceGraph())
for a, b in zip(self.test_case[1], g.clustering()):
self.assertSetEqual(set(a), set(b))
def load(self, filename=None):
'''
Load adjacency list and data from file. This will remove the graph
created by the editor in the current session.
Lines at the beginning starting with # are ignored.
The adjacency list for the matrix is first.
The line following DATA that contains a list of all the nodes with
attributes must be on a single line (no line breaks).
It should be the output of graph.nodes(data=True).
Lines after this line are ignored.
Example file:
#weighted_graph.py
# GMT Fri Aug 16 22:24:35 2013
#
N0 N1 N2 N3
N1
N2
N3
DATA
[('N0', {'weight': -3, 'parent': -1}), ('N1', {'weight': -3, 'parent':
0}), ('N2', {'weight': -3, 'parent': 0}), ('N3', {'weight': -3,
'parent': 0})]
'''
if not filename:
# open file options
options = {}
options['defaultextension'] = '.txt'
options['filetypes'] = [('all files', '.*'), ('text files', '.txt')]
filename = tkFileDialog.askopenfilename(**options)
if filename == '': # canceled
return
# attempt to open file
try:
adjfile = open(filename, 'r')
self.info = filename
except:
if self.gui:
tkMessageBox.showwarning('Open file',
'Cannot open file %s.' %filename \
+'Aborting operation; please try again.')
raise IOError('failed to open file')
# get the adjacency list
adjlist = [] # list of strings
try:
while 1:
line = adjfile.readline()
if not line: # 'DATA' line should come before EOF
raise IOError('failed to load - possibly no graph data')
if line[:4] == 'DATA' or line[:4] == 'data':
break
if line[0] == '#' or line == '' or line == '\n': # skip
continue
adjlist.append(line[:-1]) # don't want \n char
self.graph = nx.parse_adjlist(adjlist) # got graph
self.num_nodes = len(self.graph.nodes(data=False))
if not is_forest(self.graph):
raise ValueError('not a forest (disjoint union of trees)')
missing_nodes = self.missing_nodes()
if missing_nodes != set([]):
raise ValueError('missing node(s)')
# get the node data- must be preceded by a line starting with 'DATA'
while 1:
node_data = adjfile.readline()
if not node_data: # EOF
raise IOError('no data')
if node_data != '' and node_data != '\n':
break
# use eval convert node_data (string) to list; make sure is a list
if node_data.strip()[0] != '[' or node_data.strip()[-1] != ']':
raise ValueError('cannot parse data')
# add node attributes (weight, parent)
for node in eval(node_data):
for attr in node[1].keys():
self.graph.node[node[0]][attr] = node[1][attr]
# node[1] is a dict of attr
# update self.nodes (keep in order)
self.nodes = []
for index in range(self.num_nodes):
self.nodes.append('N%i' % index)
except Exception as error:
if self.gui:
tkMessageBox.showwarning('Loading', 'Loading failed - %s%s\n%s'\
%(type(error),filename, traceback.format_exc()))
print traceback.print_exc()
return
# update graph control options
if self.gui:
self.n_parent_opt = range(-1, self.num_nodes)
self.n_parentmenu = OptionMenu(self.frame, self.n_parent_var,
*self.n_parent_opt)
self.n_parentmenu.grid(row=1, column=1)
self.e_node_opt = range(self.num_nodes)
self.e_nodemenu = OptionMenu(self.frame, self.e_node_var,
*self.e_node_opt)
self.e_nodemenu.grid(row=4, column=1)
self.e_parent_opt = ['same']
self.e_parent_opt.extend(range(-1, self.num_nodes))
self.e_parentmenu = OptionMenu(self.frame, self.e_parent_var,
*self.e_parent_opt)
self.e_parentmenu.grid(row=4, column=2)
self.del_node.destroy()
self.del_node = Label(self.frame,text='Node #%i'%(self.num_nodes-1))
self.del_node.grid(row=6, column=1)
self.update_graph()
print 'Graph data successfully loaded from %s' % filename
def convert_networkx(self):
'''Saves as a networkx graph object'''
adj_list = [' '.join(map(str, i)) for i in self.graph]
self.G = nx.parse_adjlist(adj_list)
