Merge branch 'ft-1288-possibility-specify-variant-separator' into 'in…

…tegration'

FT 1288 Possibility to specify the default variant separator (, or custom)

See merge request pm4py/pm4py-core!504

pm4py/algo/clustering/trace_attribute_driven/variants/suc_dist_calc.py

@@ -51,7 +51,7 @@ def occu_var_suc(var_list, parameters=None):
 
     binarize = exec_utils.get_param_value(Parameters.BINARIZE, parameters, True)
 
-    comb_list = [var_list[i] + ',' + var_list[i + 1] for i in range(len(var_list) - 1)]
+    comb_list = [var_list[i] + constants.DEFAULT_VARIANT_SEP + var_list[i + 1] for i in range(len(var_list) - 1)]
     result = Counter(comb_list)  # count number of occurrence of each element
     df = pd.DataFrame.from_dict(dict(result), orient='index', columns=['freq'])
     df = df.reset_index().rename(columns={'index': 'direct_suc'})

pm4py/algo/enhancement/roles/common/algorithm.py

@@ -181,7 +181,7 @@ def aggregate_roles_iteration(roles, parameters=None):
         for j in range(i + 1, len(roles)):
             sim.append((i, j, roles[i][0], roles[j][0], -find_role_similarity(roles, i, j, parameters=parameters)))
 
-    sim = sorted(sim, key=lambda x: (x[-1], ",".join(x[-3]), ",".join(x[-2])))
+    sim = sorted(sim, key=lambda x: (x[-1], constants.DEFAULT_VARIANT_SEP.join(x[-3]), constants.DEFAULT_VARIANT_SEP.join(x[-2])))
 
     found_feasible = False
 
@@ -200,7 +200,7 @@ def aggregate_roles_iteration(roles, parameters=None):
 
             roles.append([total_set_act, total_set_res])
 
-            roles = sorted(roles, key=lambda x: ",".join(x[0]))
+            roles = sorted(roles, key=lambda x: constants.DEFAULT_VARIANT_SEP.join(x[0]))
 
             found_feasible = True
 
@@ -265,7 +265,7 @@ def get_initial_roles(res_act_couples, parameters=None):
     for act in roles0:
         roles.append([[act], roles0[act]])
 
-    roles = sorted(roles, key=lambda x: ",".join(x[0]))
+    roles = sorted(roles, key=lambda x: constants.DEFAULT_VARIANT_SEP.join(x[0]))
 
     roles = aggregate_roles_algorithm(roles, parameters=parameters)
 

pm4py/algo/evaluation/precision/utils.py

@@ -5,6 +5,8 @@
 from pm4py.objects.petri_net.utils.petri_utils import decorate_places_preset_trans, decorate_transitions_prepostset
 from pm4py.objects.petri_net.utils import align_utils as utils
 from pm4py.objects.petri_net.utils.incidence_matrix import construct
+from pm4py.util import constants
+
 
 def __search(sync_net, ini, fin, stop, cost_function, skip):
     decorate_transitions_prepostset(sync_net)
@@ -100,7 +102,7 @@ def get_log_prefixes(log, activity_key=xes_util.DEFAULT_NAME_KEY):
     for trace in log:
         for i in range(1, len(trace)):
             red_trace = trace[0:i]
-            prefix = ",".join([x[activity_key] for x in red_trace])
+            prefix = constants.DEFAULT_VARIANT_SEP.join([x[activity_key] for x in red_trace])
             next_activity = trace[i][activity_key]
             if prefix not in prefixes:
                 prefixes[prefix] = set()
@@ -123,7 +125,7 @@ def form_fake_log(prefixes_keys, activity_key=xes_util.DEFAULT_NAME_KEY):
     fake_log = EventLog()
     for prefix in prefixes_keys:
         trace = Trace()
-        prefix_activities = prefix.split(",")
+        prefix_activities = prefix.split(constants.DEFAULT_VARIANT_SEP)
         for activity in prefix_activities:
             event = Event()
             event[activity_key] = activity

pm4py/algo/filtering/log/paths/paths_filter.py

@@ -5,7 +5,7 @@
 from pm4py.objects.log.obj import EventLog, Trace
 from pm4py.util import exec_utils
 from pm4py.util import xes_constants as xes
-from pm4py.util.constants import PARAMETER_CONSTANT_ATTRIBUTE_KEY, PARAMETER_CONSTANT_TIMESTAMP_KEY
+from pm4py.util.constants import PARAMETER_CONSTANT_ATTRIBUTE_KEY, PARAMETER_CONSTANT_TIMESTAMP_KEY, DEFAULT_VARIANT_SEP
 import deprecation
 import sys
 
@@ -127,7 +127,7 @@ def get_paths_from_log(log, attribute_key="concept:name"):
     for trace in log:
         for i in range(0, len(trace) - 1):
             if attribute_key in trace[i] and attribute_key in trace[i + 1]:
-                path = trace[i][attribute_key] + "," + trace[i + 1][attribute_key]
+                path = trace[i][attribute_key] + DEFAULT_VARIANT_SEP + trace[i + 1][attribute_key]
                 if path not in paths:
                     paths[path] = 0
                 paths[path] = paths[path] + 1
@@ -209,7 +209,7 @@ def filter_log_by_paths(log, paths, variants, vc, threshold, attribute_key="conc
     fvft = variants[vc[0][0]][0]
     fvp = set()
     for i in range(0, len(fvft) - 1):
-        path = fvft[i][attribute_key] + "," + fvft[i + 1][attribute_key]
+        path = fvft[i][attribute_key] + DEFAULT_VARIANT_SEP + fvft[i + 1][attribute_key]
         fvp.add(path)
     for trace in log:
         new_trace = Trace()
@@ -221,7 +221,7 @@ def filter_log_by_paths(log, paths, variants, vc, threshold, attribute_key="conc
                 if j >= len(trace):
                     break
                 if attribute_key in trace[j] and attribute_key in trace[j + 1]:
-                    path = trace[j][attribute_key] + "," + trace[j + 1][attribute_key]
+                    path = trace[j][attribute_key] + DEFAULT_VARIANT_SEP + trace[j + 1][attribute_key]
                     if path in paths:
                         if path in fvp or paths[path] >= threshold:
                             new_trace.append(trace[j])

pm4py/algo/filtering/pandas/paths/paths_filter.py

@@ -6,6 +6,7 @@
 from pm4py.util.constants import PARAMETER_CONSTANT_ATTRIBUTE_KEY
 from pm4py.util.constants import PARAMETER_CONSTANT_CASEID_KEY
 from pm4py.util.constants import PARAMETER_CONSTANT_TIMESTAMP_KEY
+from pm4py.util.constants import DEFAULT_VARIANT_SEP
 from enum import Enum
 from pm4py.util import exec_utils
 from copy import copy
@@ -52,14 +53,14 @@ def apply(df: pd.DataFrame, paths: List[Tuple[str, str]], parameters: Optional[D
     attribute_key = exec_utils.get_param_value(Parameters.ATTRIBUTE_KEY, parameters, DEFAULT_NAME_KEY)
     timestamp_key = exec_utils.get_param_value(Parameters.TIMESTAMP_KEY, parameters, DEFAULT_TIMESTAMP_KEY)
     positive = exec_utils.get_param_value(Parameters.POSITIVE, parameters, True)
-    paths = [path[0] + "," + path[1] for path in paths]
+    paths = [path[0] + DEFAULT_VARIANT_SEP + path[1] for path in paths]
     df = df.sort_values([case_id_glue, timestamp_key])
     filt_df = df[[case_id_glue, attribute_key]]
     filt_dif_shifted = filt_df.shift(-1)
     filt_dif_shifted.columns = [str(col) + '_2' for col in filt_dif_shifted.columns]
     stacked_df = pd.concat([filt_df, filt_dif_shifted], axis=1)
     stacked_df = stacked_df[stacked_df[case_id_glue] == stacked_df[case_id_glue + '_2']]
-    stacked_df["@@path"] = stacked_df[attribute_key] + "," + stacked_df[attribute_key + "_2"]
+    stacked_df["@@path"] = stacked_df[attribute_key] + DEFAULT_VARIANT_SEP + stacked_df[attribute_key + "_2"]
     stacked_df = stacked_df[stacked_df["@@path"].isin(paths)]
     i1 = df.set_index(case_id_glue).index
     i2 = stacked_df.set_index(case_id_glue).index
@@ -104,15 +105,15 @@ def apply_performance(df: pd.DataFrame, provided_path: Tuple[str, str], paramete
     attribute_key = exec_utils.get_param_value(Parameters.ATTRIBUTE_KEY, parameters, DEFAULT_NAME_KEY)
     timestamp_key = exec_utils.get_param_value(Parameters.TIMESTAMP_KEY, parameters, DEFAULT_TIMESTAMP_KEY)
     positive = exec_utils.get_param_value(Parameters.POSITIVE, parameters, True)
-    provided_path = provided_path[0] + "," + provided_path[1]
+    provided_path = provided_path[0] + DEFAULT_VARIANT_SEP + provided_path[1]
     min_performance = exec_utils.get_param_value(Parameters.MIN_PERFORMANCE, parameters, 0)
     max_performance = exec_utils.get_param_value(Parameters.MAX_PERFORMANCE, parameters, sys.maxsize)
     df = df.sort_values([case_id_glue, timestamp_key])
     filt_df = df[[case_id_glue, attribute_key, timestamp_key]]
     filt_dif_shifted = filt_df.shift(-1)
     filt_dif_shifted.columns = [str(col) + '_2' for col in filt_dif_shifted.columns]
     stacked_df = pd.concat([filt_df, filt_dif_shifted], axis=1)
-    stacked_df["@@path"] = stacked_df[attribute_key] + "," + stacked_df[attribute_key + "_2"]
+    stacked_df["@@path"] = stacked_df[attribute_key] + DEFAULT_VARIANT_SEP + stacked_df[attribute_key + "_2"]
     stacked_df = stacked_df[stacked_df["@@path"] == provided_path]
     stacked_df["@@timedelta"] = (stacked_df[timestamp_key + "_2"] - stacked_df[timestamp_key]).astype('timedelta64[s]')
     stacked_df = stacked_df[stacked_df["@@timedelta"] >= min_performance]

pm4py/algo/organizational_mining/roles/common/algorithm.py

@@ -181,7 +181,7 @@ def aggregate_roles_iteration(roles, parameters=None):
         for j in range(i + 1, len(roles)):
             sim.append((i, j, roles[i][0], roles[j][0], -find_role_similarity(roles, i, j, parameters=parameters)))
 
-    sim = sorted(sim, key=lambda x: (x[-1], ",".join(x[-3]), ",".join(x[-2])))
+    sim = sorted(sim, key=lambda x: (x[-1], constants.DEFAULT_VARIANT_SEP.join(x[-3]), constants.DEFAULT_VARIANT_SEP.join(x[-2])))
 
     found_feasible = False
 
@@ -200,7 +200,7 @@ def aggregate_roles_iteration(roles, parameters=None):
 
             roles.append([total_set_act, total_set_res])
 
-            roles = sorted(roles, key=lambda x: ",".join(x[0]))
+            roles = sorted(roles, key=lambda x: constants.DEFAULT_VARIANT_SEP.join(x[0]))
 
             found_feasible = True
 
@@ -265,7 +265,7 @@ def get_initial_roles(res_act_couples, parameters=None):
     for act in roles0:
         roles.append([[act], roles0[act]])
 
-    roles = sorted(roles, key=lambda x: ",".join(x[0]))
+    roles = sorted(roles, key=lambda x: constants.DEFAULT_VARIANT_SEP.join(x[0]))
 
     roles = aggregate_roles_algorithm(roles, parameters=parameters)
 

pm4py/algo/simulation/playout/dfg/variants/classic.py

@@ -227,7 +227,7 @@ def apply(dfg: Dict[Tuple[str, str], int], start_activities: Dict[str, int], end
         # returns the variants instead of the log
         variants = []
         for p, tr in final_traces:
-            variants.append({"variant": ",".join(tr), "count": math.ceil(-p * max_no_variants)})
+            variants.append({"variant": constants.DEFAULT_VARIANT_SEP.join(tr), "count": math.ceil(-p * max_no_variants)})
         return variants
     else:
         event_log = EventLog()

pm4py/algo/transformation/log_to_trie/algorithm.py

@@ -5,7 +5,7 @@
 from pm4py.objects.trie.obj import Trie
 from pm4py.statistics.variants.log import get as get_variants
 from typing import Optional, Dict, Any, Union, Tuple
-from pm4py.util import variants_util
+from pm4py.util import variants_util, constants
 
 
 class Parameters(Enum):
@@ -17,7 +17,7 @@ def apply(log: EventLog, parameters: Optional[Dict[Union[str, Parameters], Any]]
     root = Trie()
     variants = get_variants.get_variants(log, parameters=parameters)
     if variants_util.VARIANT_SPECIFICATION == variants_util.VariantsSpecifications.STRING:
-        variants = list(map(lambda v: v.split(','), variants))
+        variants = list(map(lambda v: v.split(constants.DEFAULT_VARIANT_SEP), variants))
     else:
         variants = list(variants)
 

pm4py/evaluation/precision/utils.py

@@ -1,6 +1,6 @@
 from collections import Counter
 from pm4py.objects.log.obj import EventLog, Event, Trace
-from pm4py.util import xes_constants as xes_util
+from pm4py.util import xes_constants as xes_util, constants
 import heapq
 from pm4py.objects.petri_net.utils.petri_utils import decorate_places_preset_trans, decorate_transitions_prepostset
 from pm4py.objects.petri_net.utils import align_utils as utils
@@ -100,7 +100,7 @@ def get_log_prefixes(log, activity_key=xes_util.DEFAULT_NAME_KEY):
     for trace in log:
         for i in range(1, len(trace)):
             red_trace = trace[0:i]
-            prefix = ",".join([x[activity_key] for x in red_trace])
+            prefix = constants.DEFAULT_VARIANT_SEP.join([x[activity_key] for x in red_trace])
             next_activity = trace[i][activity_key]
             if prefix not in prefixes:
                 prefixes[prefix] = set()
@@ -123,7 +123,7 @@ def form_fake_log(prefixes_keys, activity_key=xes_util.DEFAULT_NAME_KEY):
     fake_log = EventLog()
     for prefix in prefixes_keys:
         trace = Trace()
-        prefix_activities = prefix.split(",")
+        prefix_activities = prefix.split(constants.DEFAULT_VARIANT_SEP)
         for activity in prefix_activities:
             event = Event()
             event[activity_key] = activity

pm4py/filtering.py

@@ -202,7 +202,7 @@ def filter_variants(log: Union[EventLog, pd.DataFrame], variants:  Union[Set[str
     from pm4py.util import variants_util
     parameters = get_properties(log)
     if variants_util.VARIANT_SPECIFICATION == variants_util.VariantsSpecifications.STRING:
-        variants = [",".join(v) for v in variants]
+        variants = [constants.DEFAULT_VARIANT_SEP.join(v) for v in variants]
     if check_is_pandas_dataframe(log):
         check_pandas_dataframe_columns(log)
         from pm4py.algo.filtering.pandas.variants import variants_filter

pm4py/objects/conversion/log/variants/df_to_event_log_1v.py

@@ -29,7 +29,7 @@ def apply(df, parameters=None):
         pm4_constants.PARAMETER_CONSTANT_ACTIVITY_KEY] if pm4_constants.PARAMETER_CONSTANT_ACTIVITY_KEY in parameters else xes.DEFAULT_NAME_KEY
     log = EventLog()
     for vd in variant_stats:
-        variant = vd['variant'].split(",")
+        variant = vd['variant'].split(pm4_constants.DEFAULT_VARIANT_SEP)
         trace = Trace()
         for activity in variant:
             event = Event()

pm4py/objects/conversion/log/variants/df_to_event_log_nv.py

@@ -39,7 +39,7 @@ def apply(df, parameters=None):
     log = EventLog()
     all_variants_log = {}
     for vd in variant_stats:
-        variant = vd['variant'].split(",")
+        variant = vd['variant'].split(pm4_constants.DEFAULT_VARIANT_SEP)
         variant_count = vd[case_glue]
         trace = Trace()
         for activity in variant:

pm4py/objects/log/util/pandas_numpy_variants.py

@@ -59,7 +59,7 @@ def apply(dataframe: pd.DataFrame, parameters=None):
         variants[acts] += 1
 
     if variants_util.VARIANT_SPECIFICATION == variants_util.VariantsSpecifications.STRING:
-        variants = {",".join(x): y for x, y in variants.items()}
+        variants = {constants.DEFAULT_VARIANT_SEP.join(x): y for x, y in variants.items()}
     else:
         variants = {x: y for x, y in variants.items()}
 

pm4py/objects/log/util/prefix_matrix.py

@@ -188,7 +188,7 @@ def get_prefix_matrix_from_var_str(var_str, activities, parameters=None):
     skip_last = parameters[SKIP_LAST] if SKIP_LAST in parameters else False
     prefix_mat = []
     this_prefix_repr = [0] * len(activities)
-    variant = var_str.split(",")
+    variant = var_str.split(constants.DEFAULT_VARIANT_SEP)
     for index, act in enumerate(variant):
         if skip_last and index == len(variant) - 1:
             break

pm4py/statistics/traces/generic/pandas/case_statistics.py

@@ -179,7 +179,7 @@ def get_variants_df(df, parameters=None):
     activity_key = exec_utils.get_param_value(Parameters.ACTIVITY_KEY, parameters, xes.DEFAULT_NAME_KEY)
 
     if variants_util.VARIANT_SPECIFICATION == variants_util.VariantsSpecifications.STRING:
-        new_df = df.groupby(case_id_glue)[activity_key].agg(lambda col: ",".join(pd.Series.to_list(col))).to_frame()
+        new_df = df.groupby(case_id_glue)[activity_key].agg(lambda col: constants.DEFAULT_VARIANT_SEP.join(pd.Series.to_list(col))).to_frame()
     elif variants_util.VARIANT_SPECIFICATION == variants_util.VariantsSpecifications.LIST:
         new_df = df.groupby(case_id_glue)[activity_key].agg(lambda col: tuple(pd.Series.to_list(col))).to_frame()
 
@@ -219,7 +219,7 @@ def get_variants_df_with_case_duration(df, parameters=None):
 
     df1 = None
     if variants_util.VARIANT_SPECIFICATION == variants_util.VariantsSpecifications.STRING:
-        df1 = grouped_df[activity_key].agg(lambda col: ",".join(pd.Series.to_list(col))).to_frame()
+        df1 = grouped_df[activity_key].agg(lambda col: constants.DEFAULT_VARIANT_SEP.join(pd.Series.to_list(col))).to_frame()
     elif variants_util.VARIANT_SPECIFICATION == variants_util.VariantsSpecifications.LIST:
         df1 = grouped_df[activity_key].agg(lambda col: tuple(pd.Series.to_list(col))).to_frame()
     new_cols = list(df1.columns)

pm4py/util/variants_util.py

@@ -60,6 +60,6 @@ def get_variant_from_trace(trace, parameters=None):
     activity_key = exec_utils.get_param_value(Parameters.ACTIVITY_KEY, parameters, xes_constants.DEFAULT_NAME_KEY)
 
     if VARIANT_SPECIFICATION == VariantsSpecifications.STRING:
-        return ",".join([x[activity_key] for x in trace])
+        return constants.DEFAULT_VARIANT_SEP.join([x[activity_key] for x in trace])
     elif VARIANT_SPECIFICATION == VariantsSpecifications.LIST:
         return tuple([x[activity_key] for x in trace])