Skip to content

User guide

Install

pip install "robotframework-marka[robot]"

Python users can install robotframework-marka without the Robot adapter dependencies.

Annotate your existing browser

*** Settings ***
Library    SeleniumLibrary
Library    Marka

*** Test Cases ***
Show Changes
    Highlight Element    id:email    color=coral    background=rgba(240,100,69,0.12)    group=profile
    Add Dot    id:email    text=1    position=left    group=profile
    Add Note    id:save    text=Save changes    position=right    group=profile
    Capture Annotated Screenshot    ${OUTPUT DIR}/changes.png

from marka import Annotator
from selenium.webdriver.common.by import By

# driver is your existing Selenium WebDriver.
marks = Annotator(driver)
email = driver.find_element(By.ID, "email")
marks.add(email, color="coral", background="rgba(240,100,69,0.12)")
marks.add(email, kind="dot", text="1", color="#FFBF47", position="left")
marks.capture("output/changes.png")  # clears marks by default

Behavior and limits

Open the browser with SeleniumLibrary first; Marka does not open another session. If SeleniumLibrary has an alias, use Library Marka selenium_library=Web. Marks are separate pointer-transparent overlays: original styles and layout are preserved. They follow scrolling, resizing and moving elements. Removed elements hide their marks. Highlight Elements marks every match; singular keywords use the first match. Dots accept text/number, CSS color, size and position. Notes have a pointer; labels are plain text. Border style supports solid, dashed, dotted and double; backgrounds accept CSS colors including RGBA. Use returned IDs with Remove Annotation, or Clear Annotations group=profile.

Window/frame scope: select the desired window or iframe before annotating. Cleanup affects that current document only. Select each annotated frame to clean it. Reload/navigation discards its marks. Robot performs best-effort cleanup in each driver's current frame at test end; use explicit cleanup for other frames.

Capture Annotated Screenshot captures the viewport, returns an absolute PNG path and clears by default. Set clear=${False} to retain marks. It writes a file without adding a duplicate image to the Robot log; SeleniumLibrary's capture keyword remains usable. Screenshots inside iframes should use the normal page capture after the marks are added. Very long notes should be shortened to fit the viewport. Full-page stitching, arrows and region cropping are future features.