Skip to content

Commit

Permalink
Updated microservice tool study website with a spreadsheet on metrics…
Browse files Browse the repository at this point in the history 
… used in papers and our updated metrics results
  • Loading branch information
@dorawyy
dorawyy committed Sep 21, 2024
1 parent db12ffd commit 6b96ae2
Show file tree
Hide file tree
Showing 33 changed files with 1,546 additions and 364 deletions.
Binary file added artifacts/MicroserviceToolStudy/assets/data/excelsheets/paper_metrics.xlsx
View file
Binary file not shown.
164 changes: 107 additions & 57 deletions artifacts/MicroserviceToolStudy/assets/data/metrics/scripts/access/data_repository.py
View file
Original file line number Diff line number Diff line change
@@ -1,13 +1,14 @@
import json
import os
import pandas as pd

class DataRepository:
def __init__(self, decompositions_path):
application_src_base = os.path.abspath("../../../code/scripts/metrics/casestudies")
application_src_base = os.path.abspath("./casestudies")

self.data_base = os.path.abspath("../../../code/scripts/metrics/data")
self.data_base = os.path.abspath("./data")

self.base_file_path = os.path.abspath("../../../code/scripts/metrics/data")
self.base_file_path = os.path.abspath("./data")

self.application_src_paths = {
"demo": f"{application_src_base}/demo/src/main/java",
Expand All @@ -21,7 +22,7 @@ def __init__(self, decompositions_path):
self.entropy_output_path = f"{self.data_base}/metrics/entropy.csv"
self.statistics_output_path = f"{self.data_base}/metrics/statistics.csv"

self.config_path = "./config.json"
# self.config_path = "./config.json"

def get_method_list(self, application):
with open(
Expand All @@ -37,10 +38,10 @@ def get_git_log_path(self, application):
return f"{self.base_file_path}/relationship_graphs/{application}/method_level/history.log"

def get_commit_graph_path(self, application):
return f"{self.base_file_path}/relationship_graphs/{application}/method_level/commit_graph.csv"
return f"{self.base_file_path}/relationship_graphs/{application}/class_level/commits.csv"

def get_contributor_graph_path(self, application):
return f"{self.base_file_path}/relationship_graphs/{application}/method_level/contributor_graph.csv"
return f"{self.base_file_path}/relationship_graphs/{application}/class_level/contributors.csv"

def get_decompositions_manifest(self):
manifest_file = open(f"{self.decompositions_path}/manifest.json", "r")
Expand All @@ -57,76 +58,125 @@ def get_visualization_path(
return f"{folder_path}/{granularity.split('_')[0]}_visualization.json"

def write_metrics_to_csv(self, rows):
csv_data = (
"Application,Tool,Partition Count,Decomposition Type,Commits,Contributors,Class Names,"
"Static Structural,Granularity,SMQ,CMQ,CDP,TurboMQ_commits,TurbomMQ_contributors\n"
)
data = []
for row in rows:
if row["tool"] == "mem":
partition_count = row["partition"].split("_")[0] + "_partitions"
variant = row["partition"][2:].replace("_partitions", "")
csv_data += (
f"{row['application']},{row['tool']},{partition_count},{variant},"
f"X,X,X,"
f"{row['structural_static']},{row['granularity']},{row['structural_static-fosci']},X,{row['CDP']},{row['commits']},{row['contributors']}\n"
)
row_data = {
"Application": row["application"],
"Tool": row["tool"],
"Partition Count": partition_count,
"Decomposition Type": variant,
"Commits": "X",
"Contributors": "X",
"Class Names": "X",
"Static Structural": row["structural_static"],
"Granularity": row["granularity"],
"SMQ": row["structural_static-fosci"],
"CMQ": "X",
"CDP": row["CDP"],
"TurboMQ_commits": row["commits"],
"TurbomMQ_contributors": row["contributors"],
}
else:
csv_data += (
f"{row['application']},{row['tool']},{row['partition']},{row['decomposition_type']},"
f"X,X,X,"
f"{row['structural_static']},{row['granularity']},{row['structural_static-fosci']},X,{row['CDP']},{row['commits']},{row['contributors']}\n"
)

out_file = open(self.metrics_output_path, "w")
out_file.write(csv_data)
out_file.close()
row_data ={
"Application": row['application'],
"Tool": row['tool'],
"Partition Count": row['partition'],
"Decomposition Type": row['decomposition_type'],
"Commits": "X",
"Contributors": "X",
"Class Names": "X",
"Static Structural": row['structural_static'],
"Granularity": row['granularity'],
"SMQ": row['structural_static-fosci'],
"CMQ": "X",
"CDP": row['CDP'],
"TurboMQ_commits": row['commits'],
"TurbomMQ_contributors": row['contributors']
}
data.append(row_data)
df = pd.DataFrame(data)
df = df.sort_values(by=['Decomposition Type', 'Tool', 'Application'])
df.to_csv(self.metrics_output_path, index=False)

def write_entropy_to_csv(self, rows):
csv_data = "Application,Tool,Partition Count,Variant,Granularity,Database Entropy,Use Case Entropy, M2M Use Case Entropy, Sarah BCP\n"
data = []
for row in rows:
if row["tool"] == "mem":
partition_count = row["long_partition"].split("_")[0]
variant = row["long_partition"][2:].replace("_partitions", "")
csv_data += (
f"{row['application']},{row['tool']},{partition_count},{variant},"
f"{row['granularity']},"
f"{row['database_entropy']},{row['use_case_entropy']},{row['m2m_use_case_entropy']},{row['sarah_bcp']}\n"
)
row_data ={
"Application": row['application'],
"Tool": row['tool'],
"Partition Count": partition_count,
"Variant": variant,
"Granularity": row['granularity'],
"Database Entropy": row['database_entropy'],
"Use Case Entropy": row['use_case_entropy'],
"M2M Use Case Entropy": row['m2m_use_case_entropy'],
"Sarah BCP": row['sarah_bcp']
}
else:
csv_data += (
f"{row['application']},{row['tool']},{row['partition_count']},{row['variant']},"
f"{row['granularity']},"
f"{row['database_entropy']},{row['use_case_entropy']},{row['m2m_use_case_entropy']},{row['sarah_bcp']}\n"
)

out_file = open(self.entropy_output_path, "w")
out_file.write(csv_data)
out_file.close()
row_data ={
"Application": row['application'],
"Tool": row['tool'],
"Partition Count": row['partition_count'],
"Variant": row['variant'],
"Granularity": row['granularity'],
"Database Entropy": row['database_entropy'],
"Use Case Entropy": row['use_case_entropy'],
"M2M Use Case Entropy": row['m2m_use_case_entropy'],
"Sarah BCP": row['sarah_bcp']
}
data.append(row_data)
df = pd.DataFrame(data)
df = df.sort_values(by=['Variant', 'Tool', 'Application'])
df.to_csv(self.entropy_output_path, index=False)

def write_statistics_to_csv(self, rows):
csv_data = "Application,Tool,Partition Count,Variant,Granularity,NED,Completeness,Actual Partition Count,Min,Max,Mean,Stdev,Median\n"
data = []
for row in rows:
if row["tool"] == "mem":
partition_count = row["long_partition"].split("_")[0]
variant = row["long_partition"][2:].replace("_partitions", "")
csv_data += (
f"{row['application']},{row['tool']},{partition_count},{variant},"
f"{row['granularity']},{row['ned']},"
f"{row['completeness']},{row['actual_partition_count']},"
f"{row['min']},{row['max']},{row['mean']},{row['stdev']},{row['median']}\n"
)
row_data = {
"Application": row['application'],
"Tool": row['tool'],
"Partition Count": partition_count,
"Variant": variant,
"Granularity": row['granularity'],
"NED": row['ned'],
"Completeness": row['completeness'],
"Actual Partition Count": row['actual_partition_count'],
"Min": row['min'],
"Max": row['max'],
"Mean": row['mean'],
"Stdev": row['stdev'],
"Median": row['median']
}
else:
csv_data += (
f"{row['application']},{row['tool']},{row['partition_count']},{row['variant']},"
f"{row['granularity']},{row['ned']},"
f"{row['completeness']},{row['actual_partition_count']},"
f"{row['min']},{row['max']},{row['mean']},{row['stdev']},{row['median']}\n"
)

out_file = open(self.statistics_output_path, "w")
out_file.write(csv_data)
out_file.close()

row_data = {
"Application": row['application'],
"Tool": row['tool'],
"Partition Count": row['partition_count'],
"Variant": row['variant'],
"Granularity": row['granularity'],
"NED": row['ned'],
"Completeness": row['completeness'],
"Actual Partition Count": row['actual_partition_count'],
"Min": row['min'],
"Max": row['max'],
"Mean": row['mean'],
"Stdev": row['stdev'],
"Median": row['median']
}
data.append(row_data)
df = pd.DataFrame(data)
df = df.sort_values(by=['Variant', 'Tool', 'Application'])
df.to_csv(self.statistics_output_path, index=False)

def get_dependency_path(self, application):
return f"{self.data_base}/applications/{application}/dependencies.xml"

Expand Down
3 changes: 2 additions & 1 deletion artifacts/MicroserviceToolStudy/assets/data/metrics/scripts/access/decompositions.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -67,10 +67,11 @@ def get_decomposition_folder(self, tool, application, partition_count, variant):

def write_decomposition(self, tool, application, partition_count, variant, granularity, decomposition, filtered=True):
file_path = self.get_decomposition_path(tool, application, partition_count, granularity, variant, filtered)
with open(file_path, "w") as filtered_decomposition_file:
with open(file_path, "w", newline='\n') as filtered_decomposition_file:
for partition_name in decomposition.keys():
old_partition = decomposition[partition_name]
new_partition = list(map(lambda node: {"id": node}, old_partition))
new_partition.sort(key=lambda x: x['id'])
decomposition[partition_name] = new_partition

filtered_decomposition_file.write(json.dumps({"tool": {"decomposition": decomposition}}, indent=4))
2 changes: 1 addition & 1 deletion artifacts/MicroserviceToolStudy/assets/data/metrics/scripts/access/manifest.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -38,7 +38,7 @@ def get_decomposition_variants(self, tool):
def get_partition_count_options(self, application, tool):
if tool == "mem":
return get_subfolder_names(
f"../../../code/scripts/metrics/data/decompositions/{application}/mem"
f"./data/decompositions/{application}/mem"
)
return list(
filter(
Expand Down
2 changes: 1 addition & 1 deletion artifacts/MicroserviceToolStudy/assets/data/metrics/scripts/calculator/clean_mem.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -70,7 +70,7 @@ def get_file_names(folder_path):

with open(
f"{data_path}/{data_app_name}/mem/{variant}{variant_suffix}/class_decomposition.json",
"w",
"w", newline='\n'
) as decomp_file:
decomp_file.write(
json.dumps({"MEM": {"decomposition": decomp}}, indent=4)
Expand Down
18 changes: 11 additions & 7 deletions artifacts/MicroserviceToolStudy/assets/data/metrics/scripts/calculator/evaluate.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -5,6 +5,7 @@
from access.manifest import Manifest
from utils.utils import Utils
import numpy as np
import os


class Evaluator:
Expand Down Expand Up @@ -56,7 +57,7 @@ def calculate_cycles(self, edges, decomposition, partition_assignments):

return (interdependent_pairs / total_pairs) * 100

def calculte_ratio_icf_ecf(
def calculate_commit_mq(
self, decomposition, partition_assignments, app_name, granularity
):
internal_changes = {}
Expand All @@ -66,9 +67,11 @@ def calculte_ratio_icf_ecf(
internal_changes[partition_name] = 0
external_changes[partition_name] = 0

filename = os.path.abspath(f"./data/relationship_graphs/{app_name}/class_level/commits.csv")
with open(
f"./data/relationship_graphs/{app_name}/class_level/evolutionary_commits.csv"
filename
) as edge_file:

for line in edge_file.readlines():
row = line.strip().split(",")
source = Utils().shorten_name(row[1], granularity, True)
Expand Down Expand Up @@ -96,7 +99,7 @@ def calculte_ratio_icf_ecf(
cf += (2 * icf) / ((2 * icf) + ecf)

if len(decomposition) == 0:
print(f"WARNING: Decomposition is empty")
print(f"WARNING: Decomposition is empty for file {filename} [calculate_commit_mq]")
return 0
return (cf / len(decomposition)) * 100

Expand All @@ -110,8 +113,9 @@ def calculte_contrib_mq(
internal_changes[partition_name] = 0
external_changes[partition_name] = 0

filename = os.path.abspath(f"./data/relationship_graphs/{app_name}/class_level/contributors.csv")
with open(
f"./data/relationship_graphs/{app_name}/class_level/evolutionary_contributors.csv"
filename
) as edge_file:
for line in edge_file.readlines():
row = line.strip().split(",")
Expand Down Expand Up @@ -140,7 +144,7 @@ def calculte_contrib_mq(
cf += (2 * icf) / ((2 * icf) + ecf)

if len(decomposition) == 0:
print(f"WARNING: Decomposition is empty")
print(f"WARNING: Decomposition is empty for file {filename}[calculte_contrib_mq]")
return 0
return (cf / len(decomposition)) * 100

Expand Down Expand Up @@ -169,7 +173,7 @@ def calculate_normalized_turbomq(
cf += (2 * internal_edges) / ((2 * internal_edges) + external_edges)

if len(decomposition) == 0:
print(f"WARNING: Decomposition is empty")
print(f"WARNING: Decomposition is empty [calculate_normalized_turbomq]")
return 0
return (cf / len(decomposition)) * 100

Expand Down Expand Up @@ -287,7 +291,7 @@ def calculate_turbomq(
fosci_mq = self.calculate_fosci_mq(
edges, decomposition, partition_assignments
)
rei = self.calculte_ratio_icf_ecf(
rei = self.calculate_commit_mq(
decomposition, partition_assignments, application, granularity
)
contrib_mq = self.calculte_contrib_mq(
Expand Down
Loading

0 comments on commit 6b96ae2

Please sign in to comment.