Robot Framework — Advanced Playbook
§1 Project Setup & Configuration
Installation
pip install robotframework
pip install robotframework-seleniumlibrary # Web testing
pip install robotframework-requests # API testing
pip install robotframework-datadriver # Data-driven testing
pip install robotframework-pabot # Parallel execution
pip install robotframework-browser # Playwright-based (modern)
pip install robotframework-jsonlibrary # JSON handling
pip install robotframework-databaselibrary # Database testing
Project Structure
robot-tests/
├── tests/
│ ├── ui/
│ │ ├── login.robot
│ │ ├── dashboard.robot
│ │ └── checkout.robot
│ ├── api/
│ │ ├── users_api.robot
│ │ ├── products_api.robot
│ │ └── auth_api.robot
│ └── e2e/
│ └── purchase_flow.robot
├── resources/
│ ├── keywords/
│ │ ├── common.resource
│ │ ├── login_keywords.resource
│ │ └── api_keywords.resource
│ ├── variables/
│ │ ├── common.yaml
│ │ ├── staging.yaml
│ │ └── production.yaml
│ └── libraries/
│ ├── CustomLibrary.py
│ └── DatabaseHelper.py
├── data/
│ ├── test_users.csv
│ └── products.json
├── results/
└── robot.yaml # pabot / execution config
Variable Files (Environment Config)
# resources/variables/staging.yaml
BASE_URL: https://staging.example.com
API_URL: https://staging.example.com/api
BROWSER: headlesschrome
TIMEOUT: 15s
DB_HOST: staging-db.example.com
DB_NAME: testdb
# resources/variables/production.yaml
BASE_URL: https://www.example.com
API_URL: https://api.example.com
BROWSER: headlesschrome
TIMEOUT: 10s
Execution Commands
# Run all tests
robot --outputdir results tests/
# Run with variables file
robot --variablefile resources/variables/staging.yaml tests/
# Run by tag
robot --include smoke --outputdir results tests/
robot --exclude slow --outputdir results tests/
# Run specific suite
robot --suite tests.ui.login tests/
# Parallel execution with pabot
pabot --processes 4 --outputdir results tests/
# With Selenium Grid / LambdaTest
robot --variable REMOTE_URL:https://hub.lambdatest.com/wd/hub \
--variable BROWSER:chrome \
--variablefile resources/variables/staging.yaml \
tests/ui/
§2 Web UI Testing — SeleniumLibrary
Login Tests with Page Objects
*** Settings ***
Library SeleniumLibrary
Resource ../resources/keywords/login_keywords.resource
Variables ../resources/variables/staging.yaml
Suite Setup Open Browser To Login Page
Suite Teardown Close All Browsers
Test Teardown Go To ${BASE_URL}/login
*** Test Cases ***
Successful Login With Valid Credentials
[Documentation] Verify user can log in with valid email and password.
[Tags] smoke login critical
Login With Credentials user@test.com ValidPass123
Verify Dashboard Is Displayed
Page Should Contain Welcome back
Failed Login With Invalid Password
[Tags] login negative
Login With Credentials user@test.com WrongPassword
Verify Error Message Invalid email or password
Location Should Not Contain /dashboard
Failed Login With Empty Fields
[Tags] login validation
Click Button id:login-submit
Verify Error Message Email is required
Login Remembers Email After Failed Attempt
[Tags] login ux
Input Text id:email user@test.com
Input Password id:password wrong
Click Button id:login-submit
Textfield Value Should Be id:email user@test.com
Account Lockout After Multiple Failures
[Tags] login security
FOR ${i} IN RANGE 5
Login With Credentials user@test.com wrong-${i}
END
Verify Error Message Account locked
Login Keywords Resource
# resources/keywords/login_keywords.resource
*** Settings ***
Library SeleniumLibrary
*** Keywords ***
Open Browser To Login Page
[Documentation] Opens browser and navigates to login page.
Open Browser ${BASE_URL}/login ${BROWSER}
... options=add_argument("--no-sandbox");add_argument("--disable-dev-shm-usage")
Set Selenium Implicit Wait ${TIMEOUT}
Set Selenium Speed 0.1s
Login With Credentials
[Arguments] ${email} ${password}
Wait Until Element Is Visible id:email timeout=${TIMEOUT}
Clear Element Text id:email
Input Text id:email ${email}
Input Password id:password ${password}
Click Button id:login-submit
Sleep 1s # Wait for response
Verify Dashboard Is Displayed
Wait Until Location Contains /dashboard timeout=${TIMEOUT}
Wait Until Element Is Visible id:dashboard-content timeout=${TIMEOUT}
Verify Error Message
[Arguments] ${expected_message}
Wait Until Element Is Visible css:.error-message timeout=5s
Element Should Contain css:.error-message ${expected_message}
Dynamic Content & Waits
*** Test Cases ***
Product Search Returns Results
[Tags] ui search
Go To ${BASE_URL}/products
Input Text id:search-input laptop
Click Button id:search-button
Wait Until Element Is Visible css:.product-card timeout=10s
${count}= Get Element Count css:.product-card
Should Be True ${count} > 0 No search results found
Pagination Works Correctly
[Tags] ui pagination
Go To ${BASE_URL}/products
Wait Until Element Is Visible css:.product-card timeout=10s
${page1_first}= Get Text css:.product-card:first-child .product-name
Click Element css:.pagination .next
Wait Until Element Is Not Visible css:.loading-spinner timeout=10s
${page2_first}= Get Text css:.product-card:first-child .product-name
Should Not Be Equal ${page1_first} ${page2_first}
Modal Dialog Interaction
[Tags] ui modal
Click Button id:delete-account
Wait Until Element Is Visible id:confirm-modal timeout=5s
Element Should Contain id:confirm-modal Are you sure?
Click Button id:modal-cancel
Wait Until Element Is Not Visible id:confirm-modal timeout=5s
§3 API Testing — RequestsLibrary
CRUD API Tests
*** Settings ***
Library RequestsLibrary
Library Collections
Library JSONLibrary
Variables ../resources/variables/staging.yaml
Suite Setup Authenticate API User
Suite Teardown Delete All Sessions
*** Variables ***
${AUTH_TOKEN} ${EMPTY}
*** Test Cases ***
Create User Via API
[Tags] api crud smoke
${body}= Create Dictionary
... name=Test User
... email=newuser-${TIMESTAMP}@test.com
... role=viewer
${response}= POST On Session api /users json=${body}
Should Be Equal As Integers ${response.status_code} 201
Dictionary Should Contain Key ${response.json()} id
Set Suite Variable ${CREATED_USER_ID} ${response.json()}[id]
Get User Returns Created User
[Tags] api crud
${response}= GET On Session api /users/${CREATED_USER_ID}
Should Be Equal As Integers ${response.status_code} 200
Should Be Equal ${response.json()}[name] Test User
Should Be Equal ${response.json()}[role] viewer
Update User Via API
[Tags] api crud
${body}= Create Dictionary name=Updated User role=editor
${response}= PUT On Session api /users/${CREATED_USER_ID} json=${body}
Should Be Equal As Integers ${response.status_code} 200
Should Be Equal ${response.json()}[name] Updated User
List Users With Pagination
[Tags] api pagination
${params}= Create Dictionary page=1 limit=10 sort=name
${response}= GET On Session api /users params=${params}
Should Be Equal As Integers ${response.status_code} 200
${users}= Get Value From Json ${response.json()} $.data
${count}= Get Length ${users[0]}
Should Be True ${count} <= 10
Delete User Via API
[Tags] api crud
${response}= DELETE On Session api /users/${CREATED_USER_ID}
Should Be Equal As Integers ${response.status_code} 204
Verify Deleted User Returns 404
[Tags] api crud negative
${response}= GET On Session api /users/${CREATED_USER_ID}
... expected_status=404
*** Keywords ***
Authenticate API User
${timestamp}= Evaluate int(time.time()) time
Set Suite Variable ${TIMESTAMP} ${timestamp}
Create Session api ${API_URL} verify=${TRUE}
${body}= Create Dictionary email=apitest@test.com password=TestPass123
${response}= POST On Session api /auth/login json=${body}
${token}= Set Variable ${response.json()}[token]
${headers}= Create Dictionary
... Authorization=Bearer ${token}
... Content-Type=application/json
Create Session api ${API_URL} headers=${headers} verify=${TRUE}
Error Handling & Validation
*** Test Cases ***
Create User Missing Required Fields Returns 422
[Tags] api validation
${body}= Create Dictionary name=Only Name
${response}= POST On Session api /users json=${body}
... expected_status=422
Should Contain ${response.json()}[errors][0][field] email
Invalid JSON Returns 400
[Tags] api negative
${headers}= Create Dictionary Content-Type=application/json
${response}= POST On Session api /users
... data=not-json
... headers=${headers}
... expected_status=400
Unauthorized Request Returns 401
[Tags] api security
Create Session noauth ${API_URL} verify=${TRUE}
${response}= GET On Session noauth /users/me expected_status=401
§4 Data-Driven Testing
DataDriver with CSV
# tests/data-driven/login_data_driven.robot
*** Settings ***
Library SeleniumLibrary
Library DataDriver file=data/login_tests.csv dialect=unix
Resource ../resources/keywords/login_keywords.resource
Suite Setup Open Browser To Login Page
Suite Teardown Close All Browsers
Test Template Login Should ${expected_result}
*** Test Cases ***
Login with ${email} and ${password} should ${expected_result} [Tags] data-driven
*** Keywords ***
Login Should succeed
[Arguments] ${email} ${password} ${expected_result}
Login With Credentials ${email} ${password}
Verify Dashboard Is Displayed
Login Should fail
[Arguments] ${email} ${password} ${expected_result}
Login With Credentials ${email} ${password}
Verify Error Message Invalid
# data/login_tests.csv
email,password,expected_result
valid@test.com,ValidPass123,succeed
invalid@test.com,WrongPass,fail
,password,fail
user@test.com,,fail
FOR Loops for Data Iteration
*** Test Cases ***
Verify All Product Categories Load
[Tags] api data-driven
@{categories}= Create List electronics clothing books sports home
FOR ${category} IN @{categories}
${response}= GET On Session api /products params=category=${category}
Should Be Equal As Integers ${response.status_code} 200
${count}= Get Length ${response.json()}[data]
Should Be True ${count} > 0 No products in category: ${category}
END
Verify Bulk User Creation
[Tags] api bulk
@{users}= Create List
FOR ${i} IN RANGE 5
${body}= Create Dictionary name=User ${i} email=bulk${i}@test.com
${response}= POST On Session api /users json=${body}
Should Be Equal As Integers ${response.status_code} 201
Append To List ${users} ${response.json()}[id]
END
# Cleanup
FOR ${user_id} IN @{users}
DELETE On Session api /users/${user_id}
END
§5 Custom Python Libraries
Custom Library with @keyword Decorator
# resources/libraries/CustomLibrary.py
from robot.api.deco import keyword, library
from robot.api import logger
import json
import hashlib
import time
import random
import string
@library(scope='GLOBAL')
class CustomLibrary:
"""Custom keywords for application-specific testing."""
def __init__(self):
self._created_resources = []
@keyword("Generate Unique Email")
def generate_unique_email(self, prefix="test"):
"""Generate a unique email address for testing."""
timestamp = int(time.time() * 1000)
random_str = ''.join(random.choices(string.ascii_lowercase, k=4))
email = f"{prefix}-{timestamp}-{random_str}@test.com"
logger.info(f"Generated email: {email}")
return email
@keyword("Generate Test User Data")
def generate_test_user_data(self, **overrides):
"""Generate a complete test user with realistic data."""
user = {
"name": f"Test User {random.randint(1000, 9999)}",
"email": self.generate_unique_email(),
"password": "TestPass123!",
"role": "viewer",
}
user.update(overrides)
return user
@keyword("Calculate MD5 Hash")
def calculate_md5_hash(self, content):
"""Calculate MD5 hash of content for file integrity checks."""
return hashlib.md5(content.encode()).hexdigest()
@keyword("Parse JSON Response Field")
def parse_json_response_field(self, response_body, json_path):
"""Extract a nested field from JSON using dot notation."""
data = json.loads(response_body) if isinstance(response_body, str) else response_body
for key in json_path.split("."):
if key.isdigit():
data = data[int(key)]
else:
data = data[key]
return data
@keyword("Track Created Resource")
def track_created_resource(self, resource_type, resource_id):
"""Track resources for cleanup in teardown."""
self._created_resources.append((resource_type, resource_id))
logger.info(f"Tracking {resource_type}/{resource_id} for cleanup")
@keyword("Get Created Resources")
def get_created_resources(self):
"""Return list of tracked resources."""
return list(self._created_resources)
@keyword("Clear Tracked Resources")
def clear_tracked_resources(self):
"""Clear tracked resources after cleanup."""
count = len(self._created_resources)
self._created_resources.clear()
return count
Using Custom Library in Tests
*** Settings ***
Library ../resources/libraries/CustomLibrary.py
Library RequestsLibrary
*** Test Cases ***
Create User With Generated Data
[Tags] api custom-lib
${user_data}= Generate Test User Data role=admin
${response}= POST On Session api /users json=${user_data}
Should Be Equal As Integers ${response.status_code} 201
Track Created Resource user ${response.json()}[id]
Test Teardown Cleans Up Resources
[Teardown] Cleanup All Tracked Resources
*** Keywords ***
Cleanup All Tracked Resources
${resources}= Get Created Resources
FOR ${resource} IN @{resources}
${type}= Set Variable ${resource}[0]
${id}= Set Variable ${resource}[1]
Run Keyword If '${type}' == 'user'
... DELETE On Session api /users/${id} expected_status=any
END
Clear Tracked Resources
§6 Browser Library (Playwright-Based)
Modern Web Testing with Browser Library
*** Settings ***
Library Browser
Suite Setup New Browser chromium headless=${TRUE}
Suite Teardown Close Browser
*** Test Cases ***
Fill and Submit Form
[Tags] browser form
New Page ${BASE_URL}/register
Fill Text id=name Test User
Fill Text id=email register@test.com
Fill Text id=password SecurePass123!
Click id=submit-btn
Wait For Elements State id=success-message visible timeout=10s
Get Text id=success-message == Registration successful!
Handle Network Requests
[Tags] browser network
${promise}= Promise To Wait For Response **/api/products
New Page ${BASE_URL}/products
${response}= Wait For ${promise}
Should Be Equal As Numbers ${response}[status] 200
Take Screenshot on Failure
[Tags] browser screenshot
[Teardown] Take Screenshot EMBED
New Page ${BASE_URL}/missing-page
Get Title == 404 Not Found
Test Responsive Layout
[Tags] browser responsive
New Page ${BASE_URL}
Set Viewport Size 375 812 # iPhone X
Get Style id=sidebar display == none
Set Viewport Size 1920 1080 # Desktop
Get Style id=sidebar display == flex
§7 LambdaTest Cloud Integration
Remote Browser Configuration
*** Settings ***
Library SeleniumLibrary
*** Variables ***
${LT_USERNAME} %{LT_USERNAME}
${LT_ACCESS_KEY} %{LT_ACCESS_KEY}
${REMOTE_URL} https://${LT_USERNAME}:${LT_ACCESS_KEY}@hub.lambdatest.com/wd/hub
*** Keywords ***
Open LambdaTest Browser
[Arguments] ${browser}=chrome ${test_name}=Robot Test
${lt_options}= Create Dictionary
... name=${test_name}
... build=Robot-${BUILD_NUMBER}
... project=MyProject
... platformName=Windows 11
... selenium_version=4.0
... w3c=${TRUE}
... visual=${TRUE}
... network=${TRUE}
... console=${TRUE}
${options}= Evaluate
... selenium.webdriver.ChromeOptions()
... modules=selenium.webdriver
Call Method ${options} set_capability LT:Options ${lt_options}
Create Webdriver Remote
... command_executor=${REMOTE_URL}
... options=${options}
Set Selenium Implicit Wait 15s
Mark LambdaTest Test Status
[Arguments] ${status} ${reason}=${EMPTY}
Execute Javascript
... lambda_status = "${status}";
... lambda_reason = "${reason}";
... window.lambdatest_action = {"action": "setTestStatus", "arguments": {"status": lambda_status, "remark": lambda_reason}};
Cross-Browser Test Suite
*** Settings ***
Library SeleniumLibrary
Resource ../resources/keywords/common.resource
*** Test Cases ***
Cross Browser Login Test - Chrome
[Tags] cross-browser chrome
Open LambdaTest Browser chrome Login-Chrome
Run Login Test Flow
[Teardown] Run Keywords Mark LambdaTest Test Status passed AND Close Browser
Cross Browser Login Test - Firefox
[Tags] cross-browser firefox
Open LambdaTest Browser firefox Login-Firefox
Run Login Test Flow
[Teardown] Run Keywords Mark LambdaTest Test Status passed AND Close Browser
*** Keywords ***
Run Login Test Flow
Go To ${BASE_URL}/login
Input Text id:email user@test.com
Input Password id:password TestPass123
Click Button id:login-submit
Wait Until Location Contains /dashboard timeout=15s
Page Should Contain Welcome
§8 CI/CD Integration
GitHub Actions
name: Robot Framework Tests
on:
push:
branches: [main, develop]
pull_request:
branches: [main]
jobs:
robot-tests:
runs-on: ubuntu-latest
strategy:
matrix:
suite: [ui, api, e2e]
steps:
- uses: actions/checkout@v4
- uses: actions/setup-python@v5
with:
python-version: '3.12'
cache: 'pip'
- name: Install dependencies
run: |
pip install robotframework \
robotframework-seleniumlibrary \
robotframework-requests \
robotframework-datadriver \
robotframework-pabot \
robotframework-jsonlibrary
- name: Install Chrome
if: matrix.suite == 'ui' || matrix.suite == 'e2e'
uses: browser-actions/setup-chrome@v1
- name: Install ChromeDriver
if: matrix.suite == 'ui' || matrix.suite == 'e2e'
uses: nanasess/setup-chromedriver@v2
- name: Run Robot Tests
run: |
robot \
--outputdir results/${{ matrix.suite }} \
--variablefile resources/variables/staging.yaml \
--variable BROWSER:headlesschrome \
--loglevel INFO \
--include ${{ matrix.suite }} \
--exitonfailure \
tests/${{ matrix.suite }}/
env:
BASE_URL: ${{ secrets.STAGING_URL }}
LT_USERNAME: ${{ secrets.LT_USERNAME }}
LT_ACCESS_KEY: ${{ secrets.LT_ACCESS_KEY }}
- name: Generate combined report
if: always()
run: |
rebot \
--outputdir results/combined \
--merge \
results/${{ matrix.suite }}/output.xml
- uses: actions/upload-artifact@v4
if: always()
with:
name: robot-results-${{ matrix.suite }}
path: results/
- name: Publish Robot Results
uses: joonvena/robotframework-reporter-action@v2.4
if: always()
with:
gh_access_token: ${{ secrets.GITHUB_TOKEN }}
report_path: results/${{ matrix.suite }}
Parallel Execution with Pabot
# Run tests in parallel across 4 processes
pabot --processes 4 \
--outputdir results \
--variablefile resources/variables/staging.yaml \
--variable BROWSER:headlesschrome \
tests/
# Parallel with suite-level granularity
pabot --processes 4 \
--testlevelsplit \
--outputdir results \
tests/
§9 Debugging Table
| # | Problem | Cause | Fix |
|---|
| 1 | No keyword with name 'X' found | Library not imported or keyword name mismatch | Check *** Settings *** imports; keyword names are case-insensitive but space-sensitive |
| 2 | WebDriverException: Message: unknown error | ChromeDriver version mismatch | Match ChromeDriver to Chrome version; use webdriver-manager |
| 3 | ElementNotInteractableException | Element hidden, overlapped, or not ready | Use Wait Until Element Is Visible before interaction; scroll into view |
| 4 | Variable '${VAR}' not found | Variable not defined or wrong scope | Check variable file path; use Set Suite Variable for cross-test sharing |
| 5 | Session 'api' already exists | Session created in setup and test | Use Create Session in Suite Setup only; or use unique session names |
| 6 | Tests pass locally, fail in CI | Browser not installed or different env | Install headless Chrome; use --variable BROWSER:headlesschrome |
| 7 | DataDriver finds 0 test cases | CSV path wrong or format mismatch | Use relative path from test file; check dialect matches CSV format |
| 8 | Custom library not found | Wrong import path | Use absolute path or add to PYTHONPATH; check __init__.py exists |
| 9 | Pabot tests interfere with each other | Shared state (database, session) | Use unique test data per process; isolate with --pabotlib for shared resources |
| 10 | HTTPError: 401 in RequestsLibrary | Token expired between tests | Re-authenticate in Suite Setup; use shorter-lived fixtures |
| 11 | Screenshots not captured on failure | Missing Test Teardown | Add [Teardown] Capture Page Screenshot or use listener |
| 12 | FOR loop syntax error | Wrong Robot Framework version | RF 5+ uses FOR/END; RF 4 uses \ continuation; upgrade to RF 5+ |
§10 Best Practices Checklist
- Use resource files for reusable keywords — keep test files focused on scenarios
- Use variable files for environment config — switch staging/production with
--variablefile
- Use
[Tags] on every test — enable --include/--exclude for selective execution
- Use
[Documentation] on tests and keywords — improve report readability
- Use custom Python libraries for complex logic — keep
.robot files readable
- Use
Wait Until keywords — never use Sleep for synchronization
- Use
expected_status in RequestsLibrary — avoid manual status code checks
- Use DataDriver for parameterized tests — separate test data from test logic
- Use pabot for parallel execution — reduce CI pipeline duration
- Use
Set Suite Variable carefully — prefer test-level isolation over shared state
- Use
[Teardown] for cleanup — ensure resources are released even on failure
- Use meaningful keyword names — tests should read like specifications
- Use Browser Library for modern apps — Playwright-based auto-waits are more reliable
- Use
rebot for report merging — combine parallel results into single report
