Merge pull request #513 from process-intelligence-solutions/dev

pm4py 2.7.13
process-intelligence-solutions · Jan 10, 2025 · fc13d12 · fc13d12
2 parents 91f8c26 + f1e42d1
commit fc13d12
Show file tree

Hide file tree

Showing 39 changed files with 923 additions and 136 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -1,6 +1,127 @@
 # Changelog of pm4py
 
 
+## pm4py 2.7.14 (2024.XX.YY)
+
+### Added
+
+### Changed
+
+### Deprecated
+
+### Fixed
+
+### Removed
+
+### Other
+
+
+---
+
+
+## pm4py 2.7.13 (2024.07.11)
+
+### Added
+
+### Changed
+* 75c893999e32b3a1a6081db854b9a3cd10eaaa44
+  * Dockerfile refactoring
+* 7b80dad1fcc6955730357f46754c4a511b83ee58
+  * updating requirements list
+* 52459f95d42e2fc39568b0d685bd794c319f205a
+  * refactoring dependencies generation script
+* 928ecf5f8192ff1ef7944c5b0a7be7c530814ad9
+  * refactoring OpenAI query interface
+* 5fee9b4ac61c227ac77cbd28b888a7321ba401bf
+  * refactor OCEL object repr
+* 6c1ea7a38b3ef18fe9d490164ecebd85c432915f
+  * possibility to set the title in pm4py visualizations
+
+### Deprecated
+
+### Fixed
+* 954c2fbdc258b90bcba85001592a5d2950a79724
+  * bug fix temporal profile conformance checking
+* cd8f9fb9f49a5955f0a26f547770aefa8ff222b8
+  * issue with closing in-memory files when parsing a XES from string
+* d51bd9e24aed442f011917d6b77793b118bbf89f
+  * minor fix OCEL interleavings computation
+* 39d769cad1327c47d5948b0c40be4a90e41fcce7
+  * removed indeterminism in TBR with duplicate transitions
+* 2fb8e6bffcac11b14fe13f152ac0b096cdc0d03a
+  * penwidth assignation in DFG edges
+* c753425ffd63da55fc0d3e71ab5b5fb62a2bf359
+  * OCEL exporters consistency check
+* e98d795c977348dcbe7495d62c8975fae0b9499c
+  * LTL filters documentation in the simplified interface
+
+### Removed
+
+### Other
+* ed708047252f96f3bf103ba81d6139b433c39b85
+  * printing Python versions in tests and examples
+* c48882b88326608dc5c6a4101b8123c71a852e18
+  * parameter to enable/disable visualizations
+* 607d2eb601299a23335f8c1976cc59f63bcae5f2
+  * added example for Inductive Miner LC
+
+---
+
+
+## pm4py 2.7.12.1 (2024.04.17)
+
+### Added
+
+### Changed
+* 5641f9f2c830b567c138524a0b5e705a3836102a
+  * minor change OCEL names_stripping
+* c48882b88326608dc5c6a4101b8123c71a852e18
+  * parameter to enable/disable visualizations' view
+
+### Deprecated
+
+### Fixed
+* 1c6887fda74d2b0a36f483351a6978738517e61e
+  * bug fix inductive miner infreuent
+* e7e8ebf6894d9bc85febc43c82613d185e16ee3b
+  * solving performance bottleneck in IM
+* 4811bbde07f6639deb7c87bdda61504de0889873
+  * resolved indeterminism in IM fallthroughs
+* df1db968adf8bb5f483410be0ceef2928c2b3e6c
+  * missing tree sort in IM
+* 4958e2b8407c01923e5b65c7e46a5f89d8f927dc
+  * missing fold of PT objects in IM
+* 8e404c6c2b53cf14687d1f82d53a4a0fc4ac99d6
+  * fixed folding of PTs
+* a871cbc4bdeaa81013fbfa7a6ec5054d494cb329
+  * fixed parsing of PTs
+* a72cd92410653bf2f0b2ff35ab5d06bf73a95305
+  eddae51259114caf4e1447beee4981f2e00bcb9c
+  69dcf7a836a5af3ab5da64701f43e3d8cf6b07aa
+  * fixed WF-net-to-PT conversion
+* 0980ac460d69beb9ede4403a3ed002c1cd30536e
+  * fixed docstring OCEL flattening
+* 6f9c67f48ad89e88d58c2481e85c6f3de9e6e1a5
+  * fix small issues with stochastic Petri net
+* 6781a90df2ba8d90155e7b5135520076843cac33
+  * bug fix DECLARE discovery and conformance checking
+* 7330d5100c8b8916eba9e4b0f31e8fa1f9157c98
+  * bug fix log skeleton discovery (always after, always before)
+* ec2f0b5b43d83a7b9fa154456241e3c685fcb721
+  * bug fix log-to-prefix-tree discovery (final nodes)
+* d580fedd4b770ed85f6864cd6d435db2477684f2
+  * bumping release number. forcing numpy<2 and pandas<3
+* 4f8bbf6d884e25c5474479fd86fe7331076ac2fe
+  * bug fixes for Pandas >= 3.0 future support
+
+### Removed
+
+### Other
+
+
+---
+
+
 ## pm4py 2.7.12 (2024.03.21)
 
 ### Added
@@ -34,8 +155,10 @@
 
 ### Other
 
+
 ---
 
+
 ## pm4py 2.7.11 (2024.03.01)
 
 ### Added

diff --git a/pm4py/meta.py b/pm4py/meta.py
@@ -21,7 +21,7 @@
 '''
 
 __name__ = 'pm4py'
-VERSION = '2.7.12.4'
+VERSION = '2.7.13'
 __version__ = VERSION
 __doc__ = 'Process mining for Python'
 __author__ = 'Process Intelligence Solutions (PIS)'

diff --git a/pm4py/objects/bpmn/importer/variants/lxml.py b/pm4py/objects/bpmn/importer/variants/lxml.py
@@ -305,15 +305,13 @@ def parse_element(bpmn_graph, counts, curr_el, parents, incoming_dict, outgoing_
                     incoming_dict[seq_flow_id][3], incoming_dict[seq_flow_id][4])
         for seq_flow_id in outgoing_dict:
             if outgoing_dict[seq_flow_id][0] in nodes_dict:
-                outgoing_dict[seq_flow_id] = (
-                nodes_dict[outgoing_dict[seq_flow_id][0]], outgoing_dict[seq_flow_id][1], outgoing_dict[seq_flow_id][2],
-                outgoing_dict[seq_flow_id][3], outgoing_dict[seq_flow_id][4])
+                outgoing_dict[seq_flow_id] = (nodes_dict[outgoing_dict[seq_flow_id][0]], outgoing_dict[seq_flow_id][1], outgoing_dict[seq_flow_id][2], outgoing_dict[seq_flow_id][3], outgoing_dict[seq_flow_id][4])
 
         # also supports flows without waypoints
         flows_without_waypoints = set(flow_info).union(set(outgoing_dict).intersection(set(incoming_dict)))
         for flow_id in flows_without_waypoints:
             flow_info[flow_id] = []
-
+        
         for flow_id in flow_info:
             if flow_id in outgoing_dict and flow_id in incoming_dict:
                 flow = None

diff --git a/pm4py/objects/ocel/obj.py b/pm4py/objects/ocel/obj.py
@@ -25,6 +25,7 @@
 from pm4py.util import exec_utils, pandas_utils
 import pandas as pd
 import numpy as np
+from collections import Counter
 from copy import copy, deepcopy
 
 
@@ -39,7 +40,8 @@ class Parameters(Enum):
 
 
 class OCEL(object):
-    def __init__(self, events=None, objects=None, relations=None, globals=None, parameters=None, o2o=None, e2e=None, object_changes=None):
+    def __init__(self, events=None, objects=None, relations=None, globals=None, parameters=None, o2o=None, e2e=None,
+                 object_changes=None):
         if parameters is None:
             parameters = {}
 
@@ -54,10 +56,12 @@ def __init__(self, events=None, objects=None, relations=None, globals=None, para
         self.event_timestamp = exec_utils.get_param_value(Parameters.EVENT_TIMESTAMP, parameters,
                                                           constants.DEFAULT_EVENT_TIMESTAMP)
         self.qualifier = exec_utils.get_param_value(Parameters.QUALIFIER, parameters, constants.DEFAULT_QUALIFIER)
-        self.changed_field = exec_utils.get_param_value(Parameters.CHANGED_FIELD, parameters, constants.DEFAULT_CHNGD_FIELD)
+        self.changed_field = exec_utils.get_param_value(Parameters.CHANGED_FIELD, parameters,
+                                                        constants.DEFAULT_CHNGD_FIELD)
 
         if events is None:
-            events = pandas_utils.instantiate_dataframe({self.event_id_column: [], self.event_activity: [], self.event_timestamp: []})
+            events = pandas_utils.instantiate_dataframe(
+                {self.event_id_column: [], self.event_activity: [], self.event_timestamp: []})
         if objects is None:
             objects = pandas_utils.instantiate_dataframe({self.object_id_column: [], self.object_type_column: []})
         if relations is None:
@@ -67,11 +71,15 @@ def __init__(self, events=None, objects=None, relations=None, globals=None, para
         if globals is None:
             globals = {}
         if o2o is None:
-            o2o = pandas_utils.instantiate_dataframe({self.object_id_column: [], self.object_id_column+"_2": [], self.qualifier: []})
+            o2o = pandas_utils.instantiate_dataframe(
+                {self.object_id_column: [], self.object_id_column + "_2": [], self.qualifier: []})
         if e2e is None:
-            e2e = pandas_utils.instantiate_dataframe({self.event_id_column: [], self.event_id_column+"_2": [], self.qualifier: []})
+            e2e = pandas_utils.instantiate_dataframe(
+                {self.event_id_column: [], self.event_id_column + "_2": [], self.qualifier: []})
         if object_changes is None:
-            object_changes = pandas_utils.instantiate_dataframe({self.object_id_column: [], self.object_type_column: [], self.event_timestamp: [], self.changed_field: []})
+            object_changes = pandas_utils.instantiate_dataframe(
+                {self.object_id_column: [], self.object_type_column: [], self.event_timestamp: [],
+                 self.changed_field: []})
         if self.qualifier not in relations:
             relations[self.qualifier] = [None] * len(relations)
 
@@ -111,10 +119,13 @@ def get_summary(self) -> str:
         ret.append(", number of object types: %d" % (self.objects[self.object_type_column].nunique()))
         ret.append(", events-objects relationships: %d)" % (len(self.relations)))
         ret.append("\n")
-        ret.append("Activities occurrences: " + str(self.events[self.event_activity].value_counts().to_dict()))
+        ret.append("Activities occurrences: " + str(Counter(self.events[self.event_activity].value_counts().to_dict())))
         ret.append("\n")
         ret.append("Object types occurrences (number of objects): " + str(
-            self.objects[self.object_type_column].value_counts().to_dict()))
+            Counter(self.objects[self.object_type_column].value_counts().to_dict())))
+        ret.append("\n")
+        ret.append("Unique activities per object type: " + str(
+            Counter(self.relations.groupby(self.object_type_column)[self.event_activity].nunique().to_dict())))
         ret.append("\n")
         ret.append(
             "Please use <THIS>.get_extended_table() to get a dataframe representation of the events related to the objects.")
@@ -123,7 +134,8 @@ def get_summary(self) -> str:
     def is_ocel20(self):
         unique_qualifiers = []
         if self.qualifier in self.relations.columns:
-            unique_qualifiers = [x for x in pandas_utils.format_unique(self.relations[self.qualifier].unique()) if not self.__check_is_nan(x)]
+            unique_qualifiers = [x for x in pandas_utils.format_unique(self.relations[self.qualifier].unique()) if
+                                 not self.__check_is_nan(x)]
 
         return len(self.o2o) > 0 or len(self.object_changes) > 0 or len(unique_qualifiers) > 0
 
@@ -143,7 +155,8 @@ def __repr__(self):
         return str(self.get_summary())
 
     def __copy__(self):
-        return OCEL(self.events, self.objects, self.relations, copy(self.globals), copy(self.parameters), copy(self.o2o), copy(self.e2e), copy(self.object_changes))
+        return OCEL(self.events, self.objects, self.relations, copy(self.globals), copy(self.parameters),
+                    copy(self.o2o), copy(self.e2e), copy(self.object_changes))
 
     def __deepcopy__(self, memo):
         return OCEL(self.events.copy(), self.objects.copy(), self.relations.copy(), deepcopy(self.globals),

diff --git a/pm4py/objects/ocel/util/ocel_consistency.py b/pm4py/objects/ocel/util/ocel_consistency.py
@@ -22,6 +22,7 @@
 
 from pm4py.objects.ocel.obj import OCEL
 from typing import Optional, Dict, Any
+import warnings
 
 
 def apply(ocel: OCEL, parameters: Optional[Dict[Any, Any]] = None) -> OCEL:
@@ -45,12 +46,12 @@ def apply(ocel: OCEL, parameters: Optional[Dict[Any, Any]] = None) -> OCEL:
         parameters = {}
 
     fields = {
-        "events": ["ocel:eid", "ocel:activity"],
-        "objects": ["ocel:oid", "ocel:type"],
-        "relations": ["ocel:eid", "ocel:oid", "ocel:activity", "ocel:type"],
-        "o2o": ["ocel:oid", "ocel:oid_2"],
-        "e2e": ["ocel:eid", "ocel:eid_2"],
-        "object_changes": ["ocel:oid"]
+        "events": [ocel.event_id_column, ocel.event_activity],
+        "objects": [ocel.object_id_column, ocel.object_type_column],
+        "relations": [ocel.event_id_column, ocel.object_id_column, ocel.event_activity, ocel.object_type_column],
+        "o2o": [ocel.object_id_column, ocel.object_id_column+"_2"],
+        "e2e": [ocel.event_id_column, ocel.event_id_column+"_2"],
+        "object_changes": [ocel.object_id_column]
     }
 
     for tab in fields:
@@ -62,4 +63,14 @@ def apply(ocel: OCEL, parameters: Optional[Dict[Any, Any]] = None) -> OCEL:
             df = df[df[fie].str.len() > 0]
             setattr(ocel, tab, df)
 
+    # check if the event IDs or object IDs are unique
+    num_ev_ids = ocel.events[ocel.event_id_column].nunique()
+    num_obj_ids = ocel.objects[ocel.object_id_column].nunique()
+
+    if num_ev_ids < len(ocel.events):
+        warnings.warn("The event identifiers in the OCEL are not unique!")
+
+    if num_obj_ids < len(ocel.objects):
+        warnings.warn("The object identifiers in the OCEL are not unique!")
+
     return ocel
diff --git a/pm4py/objects/trie/obj.py b/pm4py/objects/trie/obj.py
@@ -63,3 +63,22 @@ def _get_depth(self):
     label = property(_get_label, _set_label)
     final = property(_get_final, _set_final)
     depth = property(_get_depth, _set_depth)
+
+    def repr_trie(self, indent_level=0):
+        stri = []
+
+        if self.label:
+            stri.append("\t"*indent_level + self.label)
+            indent_level += 1
+        for child in self.children:
+            stri.append(child.repr_trie(indent_level=indent_level))
+        if self.final:
+            stri.append("\t"*indent_level + "-- END --")
+
+        return "\n".join(stri)
+
+    def __repr__(self):
+        return self.repr_trie()
+
+    def __str__(self):
+        return self.repr_trie()
diff --git a/pm4py/util/constants.py b/pm4py/util/constants.py
@@ -149,6 +149,7 @@ def get_default_is_aware_enabled():
 OPENAI_EXEC_RESULT = True if get_param_from_env("PM4PY_OPENAI_EXEC_RESULT", "False").lower() == "true" else False
 DEFAULT_GVIZ_VIEW = get_param_from_env("PM4PY_DEFAULT_GVIZ_VIEW", None)
 DEFAULT_ENABLE_VISUALIZATIONS_VIEW = get_param_from_env("PM4PY_DEFAULT_ENABLE_VISUALIZATIONS_VIEW", True)
+DEFAULT_ENABLE_GRAPH_TITLES = get_param_from_env("PM4PY_DEFAULT_ENABLE_GRAPH_TITLES", False)
 
 JQUERY_LINK = "https://code.jquery.com/jquery-3.6.3.min.js"
 GRAPHVIZJS_LINK = "https://github.com/mdaines/viz-js/releases/download/v1.8.2/viz.js"