add pie chart with detailed breakdown (#11)

.ruff.toml

@@ -1,2 +1,2 @@
 # Never enforce `E501` (line length violations).
-ignore = ["E501"]
+lint.ignore = ["E501"]

Makefile

@@ -8,14 +8,15 @@ test:
 
 .PHONY: lint
 lint:
-	ruff --format=github --select=E9,F63,F7,F82 --target-version=py37 .
-	ruff --format=github --target-version=py37 .
+	ruff check
 
 .PHONY: update-examples
 update-example:
 	pip3 install -r requirements.txt
 	python3 main.py -r "open-telemetry/opentelemetry-java-instrumentation" -l "java,groovy" -s "2022-11-15" -i 14 -o "./media/example_output.png"
 	python3 main.py -r "open-telemetry/opentelemetry-java-instrumentation" -l "groovy" -s "2022-11-15" -i 14 -o "./media/example_output2.png"
+	python3 count_by_instrumentation.py -r "open-telemetry/opentelemetry-java-instrumentation" -l "groovy" -o "./media/example_pie_output.png"
+
 
 
 .PHONY: all

count_by_instrumentation.py

@@ -0,0 +1,86 @@
+from datetime import datetime
+from typing import List
+import pandas as pd
+import seaborn as sns
+
+import matplotlib.pyplot as plt
+import argparse
+
+from data_filter import DataFilter
+from utilities import count_by_language_and_file_extension
+
+from github_client import GithubClient
+
+
+class App:
+    def __init__(self, languages: List[str], path_prefix: str, keyword: str):
+        self.client = GithubClient()
+        self.data_filter = DataFilter(languages=languages,
+                                      path_prefix=path_prefix, keyword=keyword)
+
+    def get_commit_by_date(self, repository, date):
+        return self.client.get_most_recent_commit(repository, date, "main")
+
+    def get_repository_by_commit(self, repository, commit):
+        repo_data = self.client.get_repository_at_commit(repository, commit)
+        repo_data = self.data_filter.parse_data(repo_data)
+
+        return repo_data
+
+
+def main(args):
+    app = App(
+        languages=[args.language],
+        path_prefix="instrumentation/",
+        keyword="test"
+    )
+
+    today = datetime.now().date().strftime("%Y-%m-%dT%H:%M:%SZ")
+
+    commit = app.get_commit_by_date(date=today, repository=args.repo)
+    repo_files = app.get_repository_by_commit(
+        repository=args.repo,
+        commit=commit
+    )
+    count = count_by_language_and_file_extension(files=repo_files["files"],
+                                                 languages=[args.language])
+
+    df = pd.DataFrame(list(count.items()), columns=['Key', 'Value'])
+    df = df.sort_values(by='Value', key=lambda col: col.astype(int), ascending=False)
+
+    sns.set_theme()
+    colors = sns.color_palette('pastel')[0:len(df)]
+
+    # Create a pie chart
+    explode = [0.05] * len(df)  # this will "explode" each slice from the pie
+    df.set_index('Key')['Value'].plot.pie(autopct='%1.0f%%', colors=colors,
+                                          explode=explode)
+
+    plt.title(f'Remaining {args.language} files by Instrumentation')
+    plt.ylabel('')
+
+    print(df.to_markdown(index=False))
+    print(f"| Total | {df['Value'].sum()} |")
+
+    if args.output is not None:
+        plt.savefig(args.output)
+    else:
+        plt.show()
+
+
+if __name__ == '__main__':
+    parser = argparse.ArgumentParser(
+        description='Show Pie chart of file count in test folders')
+    parser.add_argument("-r", "--repo",
+                        help="Repository name. "
+                             "ex: open-telemetry/opentelemetry-java-instrumentation",
+                        required=True)
+    parser.add_argument("-l", "--language",
+                        help="Language to analyze"
+                             "ex: groovy",
+                        required=True)
+    parser.add_argument("-o", "--output",
+                        help="File name to output graph to (leave blank and no file is generated)."
+                             "ex: pie-chart-counts.png")
+    arguments = parser.parse_args()
+    main(arguments)

main.py

@@ -122,7 +122,7 @@ def main(args):
 
     plt.xlabel('Date', fontsize=14)
     plt.ylabel('Count', fontsize=14)
-    plt.title('Test File Count by Language in Instrumentation Directory', fontsize=16)
+    plt.title('Test File Count in Instrumentation Directory', fontsize=16)
     plt.xticks(rotation=45)
 
     plt.legend()

readme.md

@@ -84,6 +84,59 @@ Output:
 
 ![Example](./media/benchmark_output.png)
 
+## Detail Count by Instrumentation module
+
+### Arguments
+
+| Argument   | Command        | Description                                    | Example                                                    |
+|------------|----------------|------------------------------------------------|------------------------------------------------------------|
+| Repository | -r, --repo     | Repository name.                               | --repo "open-telemetry/opentelemetry-java-instrumentation" |
+| Language   | -l, --language | Language to get count for                      | --language "groovy"                                        |
+| Output     | -o, --output   | (Optional) Location where file should be saved | --output "./media/example.png"                             |
+
+### Example Usage:
+
+In the `open-telemetry/opentelemetry-java-instrumentation` repository, analyze the files in test directories in the 
+`instrumentation` directory and outputting counts by module.
+
+`python count_by_instrumentation.py -r "open-telemetry/opentelemetry-java-instrumentation" -l "groovy"`
+
+Output: 
+
+| Key               | Value |
+|:------------------|------:|
+| spring            |    52 |
+| jaxrs             |    37 |
+| servlet           |    23 |
+| restlet           |    22 |
+| couchbase         |    18 |
+| aws-sdk           |    17 |
+| ratpack           |    16 |
+| elasticsearch     |    15 |
+| play              |    15 |
+| jaxws             |    15 |
+| vertx             |    14 |
+| mongo             |    10 |
+| jdbc              |     8 |
+| apache-dubbo-2.7  |     7 |
+| jaxrs-client      |     5 |
+| netty             |     5 |
+| apache-httpclient |     3 |
+| opentelemetry-api |     3 |
+| grizzly-2.3       |     3 |
+| grails-3.0        |     3 |
+| undertow-1.4      |     3 |
+| kafka             |     3 |
+| internal          |     2 |
+| dropwizard        |     2 |
+| hibernate         |     1 |
+| rediscala-1.8     |     1 |
+| spymemcached-2.12 |     1 |
+| twilio-6.6        |     1 |
+| Total             |   305 |
+
+![Example](./media/example_pie_output.png)
+
 ## Approach
 
 - Query Github for point in time snapshots based on commits around times spanning a timeframe

requirements.txt

@@ -3,6 +3,7 @@ matplotlib
 argparse
 pytest
 pytest-cov
-ruff
+ruff==0.5.7
 seaborn
-pandas
+pandas
+tabulate

utilities.py

@@ -1,6 +1,6 @@
 from datetime import datetime, timedelta
 from collections import defaultdict
-from typing import List
+from typing import List, Dict
 
 
 def get_dates_between(start_date_str, end_date, interval):
@@ -37,6 +37,18 @@ def count_by_file_extension(files: List[str], languages: List[str]) -> dict:
     return file_counts
 
 
+def count_by_language_and_file_extension(files: List[str], languages: List[str]) -> Dict[str, Dict[str, int]]:
+    counts = defaultdict(int)
+    for file in files:
+        file_parts = file.split('/')
+        if len(file_parts) < 3:
+            continue
+        instrumentation = file_parts[1]
+        extension = file_parts[-1].split('.')[-1]
+        if extension in languages:
+            counts[instrumentation] += 1
+    return counts
+
 def convert_to_plot(input_dict: dict, items):
     result = {}
     dates = []