Tests Using Locust and Faker

pom.xml

@@ -148,6 +148,15 @@
         <version>${scala.plugin.version}</version>
         <configuration>
           <launchers>
+            <launcher>
+              <id>clouseau</id>
+              <mainClass>com.cloudant.clouseau.Main</mainClass>
+              <jvmArgs>
+                <jvmArg>[email protected]</jvmArg>
+                <jvmArg>-Dclouseau.cookie=monster</jvmArg>
+                <jvmArg>-Dclouseau.dir=${basedir}/target/clouseau</jvmArg>
+              </jvmArgs>
+            </launcher>
             <launcher>
               <id>clouseau1</id>
               <mainClass>com.cloudant.clouseau.Main</mainClass>
@@ -346,6 +355,26 @@
           </execution>
         </executions>
       </plugin>
+      <!-- Code Coverage report generation -->
+      <plugin>
+        <groupId>org.jacoco</groupId>
+        <artifactId>jacoco-maven-plugin</artifactId>
+        <version>0.7.9</version>
+        <executions>
+          <execution>
+            <goals>
+              <goal>prepare-agent</goal>
+            </goals>
+          </execution>
+          <execution>
+            <id>generate-code-coverage-report</id>
+            <phase>test</phase>
+            <goals>
+              <goal>report</goal>
+            </goals>
+          </execution>
+        </executions>
+      </plugin>
     </plugins>
     <extensions>
       <extension>

test/README.md

@@ -0,0 +1,51 @@
+# Locust Test
+
+Test `Clouseau` using [Locust](https://github.com/locustio/locust) and [Faker](https://github.com/joke2k/faker).
+
+## Configuration options
+
+Locust configuration options.
+
+Command line | Description
+--- | ---
+--headless | Disable the web interface, and start the test
+--only-summary | Only print the summary stats
+--host | Host to load test
+-u | Peak number of concurrent Locust users
+-r | Rate to spawn users at (users per second)
+-t | Stop after the specified amount of time
+--docs-number | The number of generated documents (default: 10)
+
+```
+locust -f locustfile.py --headless --only-summary --docs-number 10 -u 1 -r 1 -t 10
+```
+
+## Basic Usage
+
+Run `CouchDB` and `Clouseau` in different terminals, and then run the locust test:
+
+```
+# Open 4 different terminals and run the command:
+./dev/run --admin=adm:pass
+mvn scala:run -Dlauncher=clouseau1
+mvn scala:run -Dlauncher=clouseau2
+mvn scala:run -Dlauncher=clouseau3
+```
+
+### Install dependencies:
+
+```
+./run install
+```
+
+### Run random_tree_generator tests:
+
+```
+./run locust
+```
+
+### Cleanup
+
+```
+./run clean
+```

test/data.py

@@ -0,0 +1,50 @@
+import math
+import json
+from datetime import date
+from random import choice
+from faker import Faker
+
+
+def write_to_files(data, filename, separator=0):
+    if separator:
+        chunks = [data[separator * i:separator * (i + 1)]
+                  for i in range(math.ceil(len(data) / separator))]
+        idx = filename.find('.json')
+        for i in range(len(chunks)):
+            with open(filename[:idx] + str(i) + filename[idx:], 'w') as outfile:
+                json.dump(chunks[i], outfile)
+    else:
+        with open(filename, 'w') as outfile:
+            json.dump(data, outfile)
+
+
+def gen_data(n=10, separator=10):
+    data = []
+    counter = {}
+    fake = Faker()
+    fields = ['married', 'ethnicity', 'gender']
+    counter['total_rows'] = n
+
+    for i in range(n):
+        data.append({'_id': str(i)})
+        data[i]['gender'] = choice(['M', 'F'])
+        data[i]['name'] = fake.name_male() if data[i]['gender'] == 'M' else fake.name_female()
+        data[i]['date_of_birth'] = fake.iso8601()
+        data[i]['age'] = date.today().year - int(data[i]['date_of_birth'][:4])
+        data[i]['married'] = 'False' if data[i]['age'] < 22 else choice(['True', 'False'])
+        data[i]['ethnicity'] = choice(['White', 'Black', 'Asian', 'Hispanic', 'non-Hispanic'])
+        data[i]['address'] = {'full_address': fake.address()}
+        data[i]['address']['city'] = data[i]['address']['full_address'][
+                                     data[i]['address']['full_address'].find('\n') + 1: -10]
+        data[i]['address']['area'] = data[i]['address']['full_address'][-8:-6]
+        data[i]['address']['zip'] = data[i]['address']['full_address'][-5:]
+        data[i]['lat'] = float(fake.latitude())
+        data[i]['long'] = float(fake.longitude())
+
+        for field in fields:
+            if field not in counter:
+                counter[field] = {}
+            counter[field].update({data[i][field]: counter[field].get(data[i][field], 0) + 1})
+
+    write_to_files(data, 'data.json', separator)
+    write_to_files(counter, 'analysis.json')

test/locustfile.py

@@ -0,0 +1,167 @@
+import math
+import json
+import time
+import requests
+from locust import events, HttpUser, constant, task, tag
+
+import data
+from logger import logger
+
+URL = "http://adm:pass@localhost:15984"
+DB = "http://adm:pass@localhost:15984/demo"
+SESSION = requests.session()
+timer = [time.perf_counter()]
+
+
+def create_database():
+    if SESSION.get(DB).status_code == 200:
+        SESSION.delete(DB)
+    SESSION.put(DB)
+
+
+def insert_docs(docs_number, separator):
+    for i in range(math.ceil(docs_number / separator)):
+        payload = {"docs": []}
+        with open("data" + str(i) + ".json") as json_file:
+            payload["docs"].extend(json.load(json_file))
+        SESSION.post(DB + "/_bulk_docs", json=payload, headers={"Content-Type": "application/json"})
+
+
+def create_indexes():
+    design_docs = {
+        "_id": "_design/search",
+        "indexes": {
+            "search_index": {
+                "index": "function(doc) {if(doc.gender) {index(\"gender\", doc.gender, {\"store\": true} );};"
+                         "if(doc.age) {index(\"age\", doc.age, {\"store\": true} );};"
+                         "if(doc.married) {index(\"married\", doc.married, {\"store\": true} );};"
+                         "if(doc.ethnicity) {index(\"ethnicity\", doc.ethnicity, {\"store\": true} );}}"
+            }
+        }
+    }
+    SESSION.put(f"{DB}/_design/search", data=json.dumps(design_docs))
+
+
+def get_result(condition, response, func_name):
+    response.success() if condition else response.failure(func_name + " FAILED.")
+
+
+@events.init_command_line_parser.add_listener
+def _(parser):
+    parser.add_argument("--docs-number", type=int, env_var="LOCUST_DOCS_NUMBER", default=100_000,
+                        help="How many documents do you want to generate")
+    parser.add_argument("--separator", type=int, env_var="LOCUST_SEPARATOR", default=5000,
+                        help="How many documents are stored in each JSON file")
+
+
+@events.test_start.add_listener
+def _(environment, **kw):
+    data.gen_data(environment.parsed_options.docs_number, environment.parsed_options.separator)
+    timer.append(time.perf_counter())
+    logger.critical(f"1. Generate documents ---- TIME: {timer[-1] - timer[-2]}")
+
+
+class LoadTest(HttpUser):
+    host = URL
+    wait_time = constant(1)
+
+    def on_start(self):
+        self.client.get("/", name=self.on_start.__name__)
+        create_database()
+        timer.append(time.perf_counter())
+        logger.debug(f"2. Create Database ---- TIME: {timer[-1] - timer[-2]}")
+        insert_docs(self.environment.parsed_options.docs_number, self.environment.parsed_options.separator)
+        timer.append(time.perf_counter())
+        logger.info(f"2. Insert docs ---- TIME: {timer[-1] - timer[-2]}")
+        create_indexes()
+        timer.append(time.perf_counter())
+        logger.info(f"2. Create design docs ---- TIME: {timer[-1] - timer[-2]}")
+        logger.critical("3. Start testing ... ")
+        with open("analysis.json") as json_file:
+            self.data = json.load(json_file)
+
+    @tag("get")
+    @task
+    def get_all_dbs(self):
+        with self.client.get("/demo/_all_dbs", catch_response=True, name="Get All DBs") as response:
+            get_result(
+                len(response.text) and response.elapsed.total_seconds() < 2.0,
+                response, self.get_all_dbs.__name__)
+
+    @tag("get")
+    @task
+    def get_all_docs(self):
+        with self.client.get("/demo/_all_docs", catch_response=True, name="Get All Docs") as response:
+            get_result(
+                len(response.text) and response.elapsed.total_seconds() < 2.0,
+                response, self.get_all_docs.__name__)
+
+    @tag("search")
+    @task
+    def search_all_docs(self):
+        with self.client.get("/demo/_design/search/_search/search_index?query=*:*",
+                             catch_response=True, name="Search All Docs") as response:
+            get_result(
+                response.status_code == 200 and response.json()["total_rows"] == self.data["total_rows"],
+                response, self.search_all_docs.__name__)
+
+    @tag("search")
+    @task
+    def search_gender_is_male(self):
+        with self.client.get("/demo/_design/search/_search/search_index?query=gender:m",
+                             catch_response=True, name="Search Gender is Male") as response:
+            get_result(
+                response.status_code == 200 and response.json()["total_rows"] == self.data["gender"]["M"],
+                response, self.search_gender_is_male.__name__)
+
+    @tag("search")
+    @task
+    def search_gender_is_male_with_limit_2(self):
+        with self.client.get("/demo/_design/search/_search/search_index?query=gender:m&limit=2",
+                             catch_response=True, name="Search Gender is Male with Limit 2") as response:
+            get_result(
+                response.status_code == 200 and len(response.json()["rows"]) == 2,
+                response, self.search_gender_is_male_with_limit_2.__name__)
+
+    @tag("search")
+    @task
+    def search_gender_is_female_and_sort_by_age(self):
+        with self.client.get("/demo/_design/search/_search/search_index?query=gender:f&sort=\"age\"",
+                             catch_response=True, name="Search Gender is Female AND Sort by age") as response:
+            result = response.json()
+            if self.data["gender"]["F"] >= 2:
+                conditions = result["total_rows"] == self.data["gender"]["F"] and \
+                             result["rows"][0]["order"][1] <= result["rows"][1]["order"][1]
+            else:
+                conditions = result["total_rows"] == self.data["gender"]["F"]
+            get_result(conditions, response, self.search_gender_is_female_and_sort_by_age.__name__)
+
+    @tag("search")
+    @task
+    def search_married_people_age_should_greater_than_21(self):
+        with self.client.get(
+                "/demo/_design/search/_search/search_index?query=married:true",
+                catch_response=True, name="Search married people age > 21") as response:
+            result = response.json()
+            for i in result["rows"]:
+                if i["fields"]["age"] <= 21:
+                    response.failure(self.search_married_people_age_should_greater_than_21.__name__)
+            response.success()
+
+    @tag("search")
+    @task
+    def search_ethnicity_White_OR_Asian(self):
+        with self.client.get(
+                f"/demo/_design/search/_search/search_index?query=ethnicity:White OR ethnicity:Asian",
+                catch_response=True, name="Search ethnicity White OR Asian") as response:
+            result = response.json()
+            get_result(
+                response.status_code == 200 and
+                result["total_rows"] == self.data["ethnicity"]["White"] + self.data["ethnicity"]["Asian"],
+                response, self.search_ethnicity_White_OR_Asian.__name__)
+
+    def on_stop(self):
+        self.client.get("/", name=self.on_stop.__name__)
+        timer.append(time.perf_counter())
+        logger.debug(f"4. Delete database, and shut down the locust ---- TIME: {timer[-1] - timer[-2]}")
+        SESSION.delete(DB)

test/logger.py

@@ -0,0 +1,16 @@
+import logging
+
+from rich.logging import RichHandler
+
+logger = logging.getLogger(__name__)
+
+shell_handler = RichHandler()
+
+logger.setLevel(logging.DEBUG)
+shell_handler.setLevel(logging.DEBUG)
+
+fmt_shell = '%(message)s'
+shell_formatter = logging.Formatter(fmt_shell)
+shell_handler.setFormatter(shell_formatter)
+
+logger.addHandler(shell_handler)

test/run

@@ -0,0 +1,29 @@
+#!/bin/bash
+case $1 in
+h | help)
+  echo "Common options:
+  h, help       Show this help message
+  i, install    Install dependencies
+  l, locust     Run locust tests
+                (Specify the docs number to be tested, run $./run l {docs_number}})
+  c, cleanup    Cleanup the directory"
+  ;;
+i | install)
+  echo Install dependencies
+  python3 -m pip install Faker locust rich
+  ;;
+l | locust)
+  if [ ! -z $2 ] && [ $2 -ge 1 ]; then
+    locust -f locustfile.py --headless --only-summary --docs-number $2 --separator 5000 -u 1 -r 1 -t 300
+  else
+    locust -f locustfile.py --headless --only-summary --docs-number 100 --separator 50 -u 1 -r 1 -t 10
+  fi
+  ;;
+c | clean)
+  rm -rf *.json __pycache__/
+  echo cleanup DONE
+  ;;
+*)
+  echo don\'t know
+  ;;
+esac