MOSIP-44175:ARC - Run the ARC UI automation in French language

[damodarguru]

MOSIP-44175:ARC - Run the ARC UI automation in French language

Summary by CodeRabbit

• New Features

  • Centralized device orientation handling for consistent layouts
  • Conditional biometric capture flows driven by UI configuration
  • In-page scrolling helpers and new preview title checks
  • "Stay logged in" and improved auto‑logout detection controls

• Bug Fixes

  • Stabilized camera/photo and document upload flows with timing and orientation sync
  • Hardened UI element detection across locales

• Chores

  • Unified orientation initialization and test data addition
  • Enhanced logging and increased command timeout

[coderabbitai]

Note

Reviews paused

It looks like this branch is under active development. To avoid overwhelming you with review comments due to an influx of new commits, CodeRabbit has automatically paused this review. You can configure this behavior by changing the reviews.auto_review.auto_pause_after_reviewed_commits setting.

Use the following commands to manage reviews:

• @coderabbitai resume to resume automatic reviews.
• @coderabbitai review to trigger a single review.

Use the checkboxes below for quick actions:

• ▶️ Resume reviews
• 🔍 Trigger review

Walkthrough

Centralized device orientation handling added via BasePage.applyOrientation(); tests removed direct disableAutoRotation() calls. Numerous pages updated for orientation, scrolling, locator strategies, document camera flow timing, biometric conditional flows, OTP/email polling, and logging. Test data gains new "orientation" key.

Changes

Cohort / File(s)	Summary
Core orientation & infra ui-test/src/main/java/regclient/page/BasePage.java, ui-test/src/main/java/regclient/BaseTest/AndroidBaseTest.java, ui-test/src/main/java/regclient/BaseTest/BaseTest.java, ui-test/src/main/resources/testdata.json, ui-test/src/main/resources/DesiredCapabilities.json	Added applyOrientation() + ADB rotation helpers, logger, new wait/input helpers, instantiated BasePage in AndroidBaseTest and call to applyOrientation(), added orientation testdata key and increased Appium timeout. Review ADB execution, new behavior and logger usage.
Tests — rotation removals ui-test/src/main/java/regclient/androidTestCases/.../*.java	Removed many BasePage.disableAutoRotation() calls across test cases; a few minor assertion/comment toggles and duplicate logout removal. Check tests for unintended flakiness due to orientation changes.
Preview & scrolling helpers ui-test/src/main/java/regclient/pages/english/PreviewPageEnglish.java, ui-test/src/main/java/regclient/pages/french/PreviewPageFrench.java, ui-test/src/main/java/regclient/pages/arabic/PreviewPageArabic.java	Added scrollToText and replaced static element checks with dynamic scroll+check; added title helpers and try/catch-safe display checks. Verify scrolling reliability across locales.
Document upload & camera flows ui-test/src/main/java/regclient/pages/{english,french,arabic}/DocumentUploadPage*.java	Inserted waitTime(1) and applyOrientation() after camera actions, adjusted retake/scroll sequencing and added scrollToTop in update flows. Verify camera/retake flows on devices.
Biometrics & OTP/email flows ui-test/src/main/java/regclient/pages/{english,french,arabic}/BiometricDetailsPage*.java, ui-test/src/main/java/regclient/androidTestCases/NewRegistrationInfant.java, .../NewRegistrationAdult.java	Added conditional biometric scan branches based on FetchUiSpec, introduced auto-logout detection/stay handlers, replaced prints with logger, and implemented timed OTP polling (15 min timeout, 20s poll, 30s initial wait). Review polling, timeout handling and locale navigation after stay-logged-in.
Input helpers & demographic pages ui-test/src/main/java/regclient/pages/english/DemographicDetailsPageEnglish.java, .../DemographicDetailsPageFrench.java	Switched to clickAndsendKeysToTextBox3, added brief waits before inputs, adjusted appId textbox locators and non-foreigner option handling. Confirm input helper behavior and keyboard hiding.
Locator updates (locales) ui-test/src/main/java/regclient/pages/french/*, ui-test/src/main/java/regclient/pages/arabic/*, ui-test/src/main/java/regclient/pages/english/*	Replaced several accessibility/XPath locators with UiAutomator descriptionContains/clickable selectors and updated localized strings/locators. Review updated locators for target UIs.
Login / Keycloak / navigation tweaks ui-test/src/main/java/regclient/pages/english/LoginPageEnglish.java, ui-test/src/main/java/regclient/page/KeycloakPage.java, ui-test/src/main/java/regclient/pages/french/ApplicantBiometricsPageFrench.java	Call applyOrientation() after language select, switch Keycloak flow to openKeycloakWebView(), and adjust clickOnNextButton() to return locale-correct BiometricDetailsPage. Validate webview/context transitions.
Logging, waits, and cleanups ui-test/src/main/java/regclient/pages/*, ui-test/src/main/java/regclient/page/BiometricDetailsPage.java, assorted small files	Replaced System.out with logger, tightened wait timeouts and exception handling (OTP/polling), removed unused vars and minor formatting edits. Inspect logging and new wait durations for side effects.

Sequence Diagram(s)

sequenceDiagram
    participant Runner as Test Runner
    participant AndroidSetup as AndroidBaseTest.setup
    participant BasePage as BasePage.applyOrientation
    participant Appium as Appium Driver
    participant ADB as adb (device)

    Runner->>AndroidSetup: start test setup
    AndroidSetup->>BasePage: new BasePage(driver) & applyOrientation()
    BasePage->>BasePage: read "orientation" from TestDataReader
    BasePage->>Appium: query/apply driver orientation
    BasePage->>ADB: execute adb command to set/lock rotation
    ADB-->>BasePage: adb result
    BasePage->>Appium: ensure driver orientation applied
    BasePage-->>AndroidSetup: orientation applied
    AndroidSetup-->>Runner: continue with test execution

Loading

Estimated code review effort

🎯 4 (Complex) | ⏱️ ~60 minutes

Possibly related PRs

• MOSIP-44176:ARC - Run the ARC UI automation in Arab language #679: Overlapping edits to BasePage and context/rotation methods; likely directly related.

Suggested reviewers

• mohanachandran-s
• Prafulrakhade

Poem

🐰 A hop, a tweak, a gentle nudge,

orientation set without a grudge.
Cameras wait and scrolls align,
OTPs polled in patient time.
The rabbit claps — tests pass in kind!

🚥 Pre-merge checks | ✅ 3 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 0.00% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (3 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title clearly summarizes the main change: enabling ARC UI automation to run in French language, which aligns with the substantial updates across French localization files (PreviewPageFrench, DocumentUploadPageFrench, BiometricDetailsPageFrench, DemographicDetailsPageFrench, etc.) and orientation handling infrastructure.
Merge Conflict Detection	✅ Passed	✅ No merge conflicts detected when merging into develop

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[coderabbitai]

Actionable comments posted: 7

Caution

Some comments are outside the diff and can’t be posted inline due to platform limitations.

⚠️ Outside diff range comments (2)

ui-test/src/main/java/regclient/page/KeycloakPage.java (1)

63-80: ⚠️ Potential issue | 🟡 Minor

Remove unused openKeycloakWebView() method.

The method has no callers anywhere in the codebase and should be deleted to reduce clutter.

ui-test/src/main/java/regclient/pages/english/SupervisorBiometricVerificationpageEnglish.java (1)

289-292: ⚠️ Potential issue | 🟡 Minor

isZoomButtonDisplayed() is hardcoded to return true, bypassing the actual check.

This silently skips zoom button verification in all tests that rely on it. If the zoom button locator is unreliable, consider fixing the locator rather than disabling the check. At minimum, add a TODO comment explaining why this was disabled.

Suggested fix if the check should be restored

 public boolean isZoomButtonDisplayed() {
-    return true;
-//    return isElementDisplayed(zoomButton);
+    return isElementDisplayed(zoomButton);
 }

🤖 Fix all issues with AI agents

In `@ui-test/src/main/java/regclient/androidTestCases/LoginTest.java`:
- Around line 228-231: The test currently calls
profilePage.clickOnLogoutButton() twice in a row which looks accidental; either
remove the duplicate call or make the intent explicit: if the first click opens
a confirmation modal and the second confirms, replace the second call with a
clearly named action (e.g., profilePage.confirmLogout() or
profilePage.clickConfirmLogout()) or add a brief comment above the two calls
explaining that pattern, and add an assertion that the confirmation modal is
present between the two calls (use methods on ProfilePage such as
isLogoutConfirmationVisible() or create one) to make the behavior explicit and
robust.

In `@ui-test/src/main/java/regclient/androidTestCases/Settings.java`:
- Around line 191-198: Add a tracked issue and annotate the disabled validation
so it doesn't remain silently skipped: create a ticket describing the regression
caused by the added submit button and include its ID in the test file, then
replace the inline comment above the commented-out assert with a short
TODO/FIXME that references that ticket ID and the affected method names
(validateJobCardFields, isJobDisplayed, clickOnSyncButton, isToastVisible); when
the bug is fixed, re-enable the check by uncommenting the
assertTrue(settingsPage.validateJobCardFields("Master Data Sync")) (or update
validateJobCardFields to accommodate the submit-button flow) and restore the
isToastVisible assertion, linking the ticket to the PR that re-enables it.

In `@ui-test/src/main/java/regclient/page/BasePage.java`:
- Line 402: The inline comment after the Pause action in BasePage is stale (it
says "Pause for 200ms" while the code uses Duration.ofMillis(600)); update the
comment next to addAction(new Pause(finger, Duration.ofMillis(600))) to reflect
600ms (or change the Duration to 200ms if the intent was 200ms) so the comment
and code match, ensuring the comment text is correct and consistent with the
Pause duration.
- Around line 172-180: The sendKeys method leaves the on-screen keyboard open
which can block later interactions; update sendKeys (in class BasePage) to
dismiss the keyboard after el.sendKeys(text) by either calling
driver.navigate().back() or invoking ((HidesKeyboard) driver).hideKeyboard(),
matching the behavior of clickAndsendKeysToTextBox, clickAndsendKeysToTextBox2,
and sendKeysToTextBox; place the dismissal immediately after sendKeys so
subsequent waits/clicks are not obstructed.

In `@ui-test/src/main/java/regclient/page/CameraPage.java`:
- Around line 27-28: The device-specific locator okBtn (AndroidFindBy id
"com.hihonor.camera:id/done_button") causes TimeoutException on non-Honor
devices; update the clickOkButton() method to try the Honor-specific okBtn first
and if it's not present or clickable within a short/conditional check, fall back
to the generic okButton xpath locator and call the existing click helper (e.g.,
clickOnElement or clickOnElement2) on okButton instead. Ensure you reference
okBtn, okButton, and clickOnElement2/clickOnElement in the implementation so the
Honor-specific path is attempted but the generic xpath is used as a graceful
fallback.

In
`@ui-test/src/main/java/regclient/pages/arabic/BiometricDetailsPageArabic.java`:
- Around line 161-165: The method isAdditionalInfoRequestIdTextboxDisplayed in
BiometricDetailsPageArabic currently declares a local By
additionalInfoRequestIdTextbox which shadows the `@AndroidFindBy` WebElement field
of the same name; remove the local By and call the isElementDisplayed overload
that accepts the existing WebElement field additionalInfoRequestIdTextbox (or
alternatively, if a By was intended, rename the By to avoid the shadow and
update usages), ensuring the `@AndroidFindBy` field is used for the display check
rather than the locally created selector.

In
`@ui-test/src/main/java/regclient/pages/arabic/UpdateOperatorBiometricspageArabic.java`:
- Around line 125-126: The zoomButton field in
UpdateOperatorBiometricspageArabic uses an overly-generic UiSelector (className
+ clickable) which can target the wrong view; update the locator for WebElement
zoomButton in class UpdateOperatorBiometricspageArabic to be specific (e.g., add
.instance(N) to target the correct occurrence or add a descriptionContains/text
qualifier to anchor it to the zoom control), aligning its approach with other
Arabic pages like IntroducerBiometricPageArabic and
ApplicantBiometricsPageArabic that use positional/XPath-style selectors; ensure
the new selector uniquely identifies the intended zoom control without affecting
other locators.

🧹 Nitpick comments (24)

ui-test/src/main/resources/DesiredCapabilities.json (1)

15-15: Consider making newCommandTimeout environment-specific.

Bumping to 3600s can hide stalled sessions and slow failure feedback. If this is needed for French runs only, consider parameterizing by profile/env to avoid lengthening all suites.

ui-test/src/main/java/regclient/androidTestCases/NewRegistrationInfant.java (1)

545-549: Duplicate clickServerStatusDropdown() calls — is this intentional?

Calling clickServerStatusDropdown() twice in succession (lines 545 and 547) will toggle the dropdown open and then closed (or vice versa), which is effectively a no-op unless there's a side effect being relied upon (e.g., dismissing a popup or forcing a UI refresh). The previous code apparently called clickDismissButton() here instead.

If this is a deliberate workaround, please add a brief comment explaining why the double-click is needed. If the dismiss button was removed by mistake, this should be reverted.

ui-test/src/main/java/regclient/pages/english/ManageApplicationsPageEnglish.java (1)

87-90: backButton field is declared but never used.

clickOnBackButton() (Line 234) now uses driver.navigate().back() instead of interacting with the backButton element. The field at Lines 87–90 is dead code and should be removed to avoid confusion.

Proposed fix

-	`@AndroidFindBy`(xpath =
-			  "(//android.widget.TextView[`@content-desc`='Manage Applications']" +
-			  "/ancestor::android.view.View)[1]//android.widget.Button")
-			private WebElement backButton;

Also applies to: 233-235

ui-test/src/main/java/regclient/page/KeycloakPage.java (2)

176-188: Consider using WebDriverWait instead of manual polling loop.

waitForLoginPage() manually reimplements a polling wait with Thread.sleep in a loop. The class already imports and uses WebDriverWait with ExpectedConditions (e.g., line 90). Using the existing utility would be more idiomatic and consistent with the rest of this file.

♻️ Proposed refactor

 public void waitForLoginPage() {
-    for (int i = 0; i < 15; i++) {
-        try {
-            driver.findElement(By.id("username"));
-            return;
-        } catch (Exception e) {
-            try {
-                Thread.sleep(1000);
-            } catch (InterruptedException ignored) {}
-        }
-    }
-    throw new RuntimeException("Login page not loaded");
+    new WebDriverWait(driver, Duration.ofSeconds(15))
+        .until(ExpectedConditions.presenceOfElementLocated(By.id("username")));
 }

---

123-131: Field naming and placement are inconsistent with the rest of the class.

usernameTextBox1 and passwordTextBox1 are By locators declared inline between methods, shadowing the existing @FindBy WebElement fields usernameTextBox and passwordTextBox. This creates two parallel locator strategies for the same elements.

If the By-based approach is now preferred for these fields, consider removing the unused @FindBy WebElement counterparts (lines 33-37) and giving the By fields clearer names (e.g., usernameLocator, passwordLocator), grouped with the other field declarations at the top of the class.

ui-test/src/main/java/regclient/androidTestCases/ResetPassword.java (1)

47-108: Consider extracting language-based page factory into a shared utility.

The if/else language dispatch is repeated ~8 times across the two methods (and likely across other test classes). A simple factory method like createLoginPage(language, driver) would eliminate this duplication and make adding new languages a one-line change.

Also applies to: 70-108, 94-108, 124-138, 232-246, 254-268, 275-289, 305-319

ui-test/src/main/java/regclient/pages/arabic/ApplicantBiometricsPageArabic.java (1)

85-86: zoomButton field is no longer used after switching to coordinate-based click.

The zoomButton WebElement (Line 85) is declared but clickOnZoomButton() now uses clickAtCoordinates(1035, 1077) instead of clicking the element. The field and its locator change (from View[3] to View[2]) are dead code. This applies to the French variant as well.

Also applies to: 165-167

ui-test/src/main/java/regclient/page/BiometricDetailsPage.java (1)

42-43: Remove commented-out code instead of leaving it in the abstract class.

The commented-out handleBiometricDetails() method adds noise. If it's not needed, delete it; if it's planned, track it in an issue.

Proposed fix

 	public abstract void enterAdditionalInfoUsingEmail(String emailId);
-	
-//	public abstract void handleBiometricDetails();
 }

ui-test/src/main/java/regclient/pages/english/DemographicDetailsPageEnglish.java (1)

238-240: Good fix: explicit "Non-Foreigner" selection instead of generic first-view click.

Selecting a specific dropdown option is more reliable than clicking the first android.view.View. However, remove the commented-out line to keep the code clean.

Clean up commented-out code

 					if (!isElementDisplayed(dropdownElement)) {
-//						clickOnElement(findElement(By.className("android.view.View")));
 						clickOnElement(nonForeignerOption);

ui-test/src/main/java/regclient/androidTestCases/LostUin.java (2)

301-302: Remove commented-out code or add a TODO explaining why it's disabled.

The commented call //biometricDetailsPage.handleBiometricDetails(); looks like a debug artifact. If this is intentionally disabled, add a brief comment explaining why; otherwise, remove it to keep the test clean.

Proposed fix

-				
-//				biometricDetailsPage.handleBiometricDetails();

---

134-134: Unused variable documentuploadPage.

documentuploadPage is declared but never assigned or used in this test method (the Documents screen case was removed). The DocumentUploadPage import (line 20) and related language-specific imports (lines 35, 50, 65, 80, 95, 110) are also unused now.

ui-test/src/main/java/regclient/pages/arabic/PreviewPageArabic.java (1)

160-164: New scrollToText helper — consider moving to BasePage.

This utility is likely reusable across other language pages. If similar scroll-to-text patterns emerge in other page objects, consider promoting this to BasePage to avoid duplication. Low priority given this is a regional language page.

ui-test/src/main/java/regclient/page/BasePage.java (1)

91-94: Consider a more descriptive method name than clickOnElement2.

The 2 suffix doesn't convey the semantic difference (waits for clickable vs. visible). Something like clickOnClickableElement or clickWhenClickable would make the intent clear at call sites.

♻️ Suggested rename

-	protected void clickOnElement2(WebElement element) {
+	protected void clickOnClickableElement(WebElement element) {
 		waitForElementToBeClickable(element);
 		element.click();
 	}

ui-test/src/main/java/regclient/pages/english/DocumentuploadPageEnglish.java (1)

169-169: Commented-out assertion may hide test failures.

The isRetakeButtonDisplayed() check is active in the other document upload branches (lines 108 and 137) but commented out here in the proofOfRelationship block. If this was intentionally disabled to work around a flaky condition, consider adding a brief comment explaining why. Otherwise, re-enable or remove it to maintain consistency.

ui-test/src/main/java/regclient/pages/english/SettingsPageEnglish.java (2)

267-276: Scroll target may not reach the desired job.

scrollIntoView scrolls to the first android.widget.EditText found in the scrollable container, which may not be the one matching jobName if it's further down the list. Consider scrolling directly to the element matching the job name for reliability.

Suggested improvement

 public boolean isJobDisplayed(String jobName) {
-    // Scroll to jobs list (safe for long lists)
-    driver.findElement(MobileBy.AndroidUIAutomator("new UiScrollable(new UiSelector().scrollable(true))"
-            + ".scrollIntoView(new UiSelector().className(\"android.widget.EditText\"))"));
-
-    By job = By.xpath("//android.widget.EditText[contains(`@hint`,'" + jobName + "')]");
-
-    return isElementDisplayed(job);
+    // Scroll directly to the target job element
+    try {
+        driver.findElement(MobileBy.AndroidUIAutomator(
+            "new UiScrollable(new UiSelector().scrollable(true))"
+            + ".scrollIntoView(new UiSelector().className(\"android.widget.EditText\").textContains(\"" + jobName + "\"))"));
+    } catch (Exception e) {
+        // Element may already be visible
+    }
+    By job = By.xpath("//android.widget.EditText[contains(`@hint`,'" + jobName + "')]");
+    return isElementDisplayed(job);
 }

---

1-56: Duplicate imports should be cleaned up.

There are several duplicate imports in this file:

• java.io.IOException (lines 3 and 28)
• io.appium.java_client.android.Activity (lines 17 and 43)
• regclient.api.FetchUiSpec (lines 46 and 56)

These won't cause compilation errors but are unnecessary clutter.

ui-test/src/main/java/regclient/api/FetchUiSpec.java (2)

514-601: Remove commented-out code; track future work in an issue instead.

~90 lines of commented-out methods (getBioAttributesUsingId, getBiometricLabel) add significant noise. If these are needed later, create a tracked issue or keep them in a feature branch rather than checked into develop as dead code.

Would you like me to open an issue to track implementing these methods when needed?

---

19-28: Newly added imports are unused.

org.springframework.test.context.TestContext (line 19) and regclient.BaseTest.BaseTest (line 28) are imported but not referenced anywhere in this file. These appear to be leftover from the commented-out code or an incomplete refactoring.

ui-test/src/main/java/regclient/pages/arabic/OperationalTaskPageArabic.java (1)

36-41: Local variable updateOperatorBiometricsButton shadows the class field of the same name (line 14).

The By local variable has the same name as the WebElement field declared at line 14–15. While not a bug (the field is unused in this method), it reduces readability. Consider renaming the local variable to avoid confusion.

This is a minor note — the runtime locator approach itself is fine.

ui-test/src/main/java/regclient/pages/arabic/RegistrationTasksPageArabic.java (1)

37-41: Remove the commented-out old locator.

Lines 37–38 are dead code. Since the replacement locator on line 40 is already in place, the old code can be removed.

ui-test/src/main/java/regclient/pages/arabic/AcknowledgementPageArabic.java (1)

46-50: Local goToHomeButton shadows the class field on line 28 — same pattern noted in other Arabic pages.

The field goToHomeButton (line 28) is now unused in this method since a local By with the same name takes over. Consider either removing the field or renaming the local variable for clarity.

ui-test/src/main/java/regclient/pages/english/PreviewPageEnglish.java (1)

52-53: Unused field biometricCorrectionTitle.

The @AndroidFindBy field biometricCorrectionTitle (line 52-53) is no longer referenced — isBiometricCorrectionTitleDisplayed() now uses dynamic lookup via FetchUiSpec.getTitleUsingId("BIO"). Consider removing the dead field.

🧹 Remove unused field

-	`@AndroidFindBy`(accessibility = "Biometric correction")
-	private WebElement biometricCorrectionTitle;

ui-test/src/main/java/regclient/androidTestCases/BiometricCorrection.java (1)

816-827: Retry mechanism for authentication page is a good resilience improvement.

The 3-attempt retry loop handles cases where the continue button click doesn't immediately navigate to the authentication page. Consider using BasePage.waitTime() instead of Thread.sleep(2000) for consistency with other retry patterns in this file (e.g., line 830 uses BasePage.waitTime(1)).

♻️ Use BasePage.waitTime for consistency

 		for (int i = 0; i < 3; i++) {
 			previewPage.clickOnContinueButton();
-			Thread.sleep(2000);
+			BasePage.waitTime(2);
 			if (authenticationPage.isAuthenticationPageDisplayed()) {
 				isAuthenticationPageDisplayed = true;
 				break;
 			}
 		}

ui-test/src/main/java/regclient/pages/arabic/BiometricDetailsPageArabic.java (1)

3-3: Unused JUnit import — this project uses TestNG.

import static org.junit.Assert.assertTrue is imported but not used within this file. Mixing JUnit and TestNG assertion imports can cause confusion. Remove if unused.

[coderabbitai]

Actionable comments posted: 8

Caution

Some comments are outside the diff and can’t be posted inline due to platform limitations.

⚠️ Outside diff range comments (2)

ui-test/src/main/java/regclient/androidTestCases/NewRegistrationInfant.java (1)

337-392: ⚠️ Potential issue | 🟡 Minor

Introducer biometric flag reset and conditional flow looks correct.

Resetting all flags to "no" before re-fetching via getBiometricDetails("introducerBiometrics") ensures clean state for the introducer-specific checks. The conditional blocks for each biometric modality are well-structured.

One inconsistency: line 391 reassigns biometricDetailsPage from clickOnNextButton() only in the face block. The other blocks (iris at 351, rightHand at 361, leftHand at 371, thumb at 381) discard the return value. If the last enabled introducer biometric is not face (e.g., only thumb is "yes"), the return from clickOnNextButton() on line 381 is silently dropped. This works today because the page object wraps the shared driver, but it's fragile — if the last enabled biometric's clickOnNextButton() ever returns a different page type, biometricDetailsPage.clickOnContinueButton() at line 394 would operate on a stale reference.

Consider capturing the return value from whichever block executes last, or assigning it consistently in every block.

ui-test/src/main/java/regclient/pages/arabic/BiometricDetailsPageArabic.java (1)

233-241: ⚠️ Potential issue | 🟠 Major

Same 10-minute wait issue as BiometricDetailsPageEnglish — isAutoLogoutPopupDisplayed() blocks too long inside a polling loop.

This method waits 10 minutes for the popup, but it's called on every poll iteration within enterAdditionalInfoUsingEmail. If the popup isn't present, the poll cycle becomes ~10 minutes. Consider reducing to a short timeout (e.g., seconds) to match the French implementation which uses Duration.ofSeconds(3).

🤖 Fix all issues with AI agents

In `@ui-test/src/main/java/regclient/page/BasePage.java`:
- Around line 887-901: applyOrientation() currently calls
ScreenOrientation.valueOf(...) with the raw string from
TestDataReader.readData("orientation"), which throws IllegalArgumentException
for empty or invalid values; guard against this by first checking if the
returned orientation string is null or blank and only then attempt to parse, and
additionally catch IllegalArgumentException around ScreenOrientation.valueOf to
handle unknown values gracefully (e.g., log a warning and skip setting
desiredOrientation). Update references in the method (applyOrientation,
TestDataReader.readData, desiredOrientation, lockSystemRotation, and the
((SupportsRotation) driver).rotate call) so rotation is only performed when a
valid ScreenOrientation is obtained. Ensure logging provides the invalid input
and that the method returns without throwing for empty/invalid orientation.
- Around line 903-920: The lockSystemRotation method currently calls
Runtime.getRuntime().exec(...) twice without capturing or handling the returned
Process, which leaks handles and doesn't ensure the adb commands complete;
update lockSystemRotation to run the commands using ProcessBuilder (or
Runtime.exec(String[])) capturing each Process from the two commands,
consume/close their input, output and error streams, call waitFor() on each
Process and handle InterruptedException/IOExceptions, and ensure streams are
closed in finally or try-with-resources so the adb commands reliably finish
before proceeding.

In
`@ui-test/src/main/java/regclient/pages/english/BiometricDetailsPageEnglish.java`:
- Around line 227-232: The isAutoLogoutPopupDisplayed() call inside
BiometricDetailsPageEnglish's polling loop blocks up to 10 minutes per
invocation and should be shortened; modify the call or the implementation used
by BiometricDetailsPageEnglish so it uses a short timeout (e.g.,
Duration.ofSeconds(3)) like BiometricDetailsPageFrench, or add an
overload/parameter to isAutoLogoutPopupDisplayed(Duration) and call that
short-duration variant before clickOnStayLoggedInButton(); ensure the polling
loop calls the non-blocking short-wait check so each 20s poll iteration does not
wait for 10 minutes.

In
`@ui-test/src/main/java/regclient/pages/english/UpdateOperatorBiometricspageEnglish.java`:
- Around line 125-126: The locator for zoomButton uses an overly broad
UiSelector; replace the `@AndroidFindBy`(uiAutomator = "new
UiSelector().className(\"android.view.View\").clickable(true)") on the
zoomButton field with a specific XPath-based locator consistent with other
English pages (for example the same pattern used in
SupervisorBiometricVerificationpageEnglish.java such as
"//android.widget.ScrollView/android.view.View[3]" or the exact hierarchical
XPath used across the codebase) by changing the annotation to
`@AndroidFindBy`(xpath = "…") so the zoomButton field targets the same specific
element as other implementations.

In `@ui-test/src/main/java/regclient/pages/french/AutoLogoutPageFrench.java`:
- Around line 23-24: The accessibility selectors are swapped/incorrect: change
the field annotated for stayLoggedInButton to use the French text for "Stay
Logged In" (e.g., "RESTER CONNECTÉ") and change logoutButton to use the French
"DÉCONNEXION"; update the annotations on the WebElement fields
stayLoggedInButton and logoutButton accordingly so clickOnStayLoggedInButton()
targets the correct element.

In
`@ui-test/src/main/java/regclient/pages/french/BiometricDetailsPageFrench.java`:
- Around line 60-61: The stayLoggedInButton is incorrectly mapped to the logout
accessibility label and the click method returns the Arabic page; update the
`@AndroidFindBy` on stayLoggedInButton to use the correct "stay logged in"
accessibility value used elsewhere in the app (e.g., "RESTER CONNECTÉ" or the
exact app's French string) instead of "DÉCONNEXION", and change
clickOnStayLoggedInButton() to return new RegistrationTasksPageFrench(driver)
rather than new RegistrationTasksPageArabic(driver) so the click maps to the
stay-logged-in action and navigates to the French page.

In
`@ui-test/src/main/java/regclient/pages/french/DemographicDetailsPageFrench.java`:
- Around line 420-429: The locator for appIdTextbox in fetchPreregApplicationId
is fragile because it uses a positional XPath; replace it with a hint- or
content-desc-based locator (like the English page’s hint-based XPath) to
reliably target the Application ID field (update the appIdTextbox By to use an
XPath such as one that contains the French hint text or content-desc for
"Application ID"), keep the existing appIdLabel check and continue using
applicationIdTextBox with
clickAndsendKeysToTextBox3(AdminTestUtil.getPreRegistrationFlow(age)) and
clickOnElement(fetchDataButton) so only the locator changes.

In
`@ui-test/src/main/java/regclient/pages/french/UpdateOperatorBiometricspageFrench.java`:
- Around line 101-102: The zoomButton locator in
UpdateOperatorBiometricspageFrench is overly broad (matches any clickable
android.view.View); update the `@AndroidFindBy` on the zoomButton field to use the
same, more specific selector used in the English page (e.g., narrow by
resource-id, content-desc, text, or a more specific UiSelector chain) so it
targets the actual zoom control; locate the field named zoomButton in class
UpdateOperatorBiometricspageFrench and replace the generic UiSelector with the
specific attribute-based selector used in the English counterpart.

🧹 Nitpick comments (7)

ui-test/src/main/java/regclient/androidTestCases/NewRegistrationInfant.java (1)

155-169: Extract language-based page creation into a factory or helper method.

There are ~11 nearly identical if-else-if chains that map a language code to a language-specific page object. This is a significant DRY violation that makes the test hard to maintain — adding a new language requires edits in every chain.

A factory or registry approach would consolidate this into one place:

♻️ Sketch of a PageFactory approach

// New utility class (e.g., regclient/utils/PageFactory.java)
public class PageFactory {
    public static LoginPage createLoginPage(AppiumDriver driver, String language) {
        switch (language.toLowerCase()) {
            case "eng": return new LoginPageEnglish(driver);
            case "hin": return new LoginPageHindi(driver);
            case "fra": return new LoginPageFrench(driver);
            case "kan": return new LoginPageKannada(driver);
            case "tam": return new LoginPageTamil(driver);
            case "ara": return new LoginPageArabic(driver);
            default: throw new IllegalStateException("Unsupported language: " + language);
        }
    }
    // Similar methods for RegistrationTasksPage, ConsentPage, etc.
}

Then in the test:

-       if ("eng".equalsIgnoreCase(language)) {
-           loginPage = new LoginPageEnglish(driver);
-       } else if ("hin".equalsIgnoreCase(language)) {
-           loginPage = new LoginPageHindi(driver);
-       } else if ("fra".equalsIgnoreCase(language)) {
-           loginPage = new LoginPageFrench(driver);
-       } else if ("kan".equalsIgnoreCase(language)) {
-           loginPage = new LoginPageKannada(driver);
-       } else if ("tam".equalsIgnoreCase(language)) {
-           loginPage = new LoginPageTamil(driver);
-       } else if ("ara".equalsIgnoreCase(language)) {
-           loginPage = new LoginPageArabic(driver);
-       } else {
-           throw new IllegalStateException("Unsupported language in testdata.json: " + language);
-       }
+       loginPage = PageFactory.createLoginPage(driver, language);

This would benefit all test classes (not just this one) that repeat the same pattern.

Also applies to: 177-191, 196-210, 220-234, 237-251, 256-270, 275-289, 292-306, 322-336, 398-412, 415-429, 447-461, 465-479, 482-496, 525-539, 564-578

ui-test/src/main/java/regclient/pages/english/BiometricDetailsPageEnglish.java (1)

192-197: Inconsistent logging: System.out.println on line 193 while the rest of the method uses logger.info.

This is the only System.out.println remaining in the method — all other log statements were migrated to logger.info.

Proposed fix

 	    try {
-	        System.out.println("Waiting 30 seconds for email delivery...");
+	        logger.info("Waiting 30 seconds for email delivery...");
 	        Thread.sleep(30000);

ui-test/src/main/java/regclient/BaseTest/BaseTest.java (1)

5-5: BasePage import appears unused in this class.

BasePage is not referenced anywhere in BaseTest.java. If it's only needed in AndroidBaseTest (the subclass), the import should live there instead.

ui-test/src/main/java/regclient/androidTestCases/PreRegFetchingPacket.java (1)

154-168: Consider extracting a page-factory helper to eliminate the repeated language-dispatch blocks.

Each page object instantiation (LoginPage, RegistrationTasksPage, SelectLanguagePage, ConsentPage, etc.) repeats the exact same 12-line if/else-if chain ~15 times per test method across three nearly identical methods. A single helper like PageFactory.create(Class<T> pageType, String language, AppiumDriver driver) using a Map<String, Supplier<T>> or a simple registry would collapse all of these into one-liners, dramatically reducing the ~1500-line file and making it far easier to add new languages.

This is pre-existing debt but heavily amplified by this PR's French language addition.

Also applies to: 177-191, 197-211, 220-234

ui-test/src/main/java/regclient/BaseTest/AndroidBaseTest.java (1)

24-25: RuntimeException thrown without a message or cause.

When driver initialization fails, the original exception e is discarded, making debugging difficult.

Proposed fix

-		} catch (Exception e) {
-			throw new RuntimeException();
+		} catch (Exception e) {
+			throw new RuntimeException("Failed to start Appium server or get driver", e);
		}

ui-test/src/main/java/regclient/pages/english/PreviewPageEnglish.java (1)

65-73: Redundant scroll + element re-lookup for Demographic Information.

scrollToText("Demographic Information") scrolls to the element, then line 68-69 performs a second find using new UiSelector().text(...). The class field demographicInformationInPreviewPage (line 23-24) already has a UiScrollable.scrollIntoView locator that scrolls and finds in one step. In contrast, isDocumentsInformationInPreviewPageDisplayed (line 77-79) correctly reuses the existing field after calling scrollToText.

Consider reusing the existing field here too:

Proposed fix

 public boolean isDemographicInformationInPreviewPageDisplayed() {
 	try {
 		scrollToText("Demographic Information");
-		WebElement demographicInformationInPreviewPage = driver
-				.findElement(MobileBy.AndroidUIAutomator("new UiSelector().text(\"Demographic Information\")"));
 		return isElementDisplayed(demographicInformationInPreviewPage);
 	} catch (Exception e) {
 		return false;
 	}
 }

ui-test/src/main/java/regclient/page/BasePage.java (1)

155-164: clickAndsendKeysToTextBox3 is nearly identical to clickAndsendKeysToTextBox.

The only difference is waitForElementToBeClickable vs waitForElementToBeVisible. Consider parameterizing the wait strategy or renaming to convey the distinction (e.g., clickAndsendKeysToTextBoxWaitClickable). The current naming convention (3 suffix) makes it hard to know which variant to use.

[coderabbitai]

Actionable comments posted: 1

🤖 Fix all issues with AI agents

In
`@ui-test/src/main/java/regclient/pages/english/BiometricDetailsPageEnglish.java`:
- Line 194: In BiometricDetailsPageEnglish (method where "Waiting 30 seconds for
email delivery..." is printed), replace the System.out.println call with the
existing logger.info usage to make logging consistent; locate the println
statement in BiometricDetailsPageEnglish and change it to use
logger.info("Waiting 30 seconds for email delivery...") so it matches the other
log statements in that method.

🧹 Nitpick comments (5)

ui-test/src/main/java/regclient/page/BasePage.java (3)

891-891: desiredOrientation field is declared but never assigned or read.

This field appears to be leftover from a prior iteration. applyOrientation() uses the local variable finalOrientation instead. Consider removing it to avoid confusion.

Proposed fix

-	private ScreenOrientation desiredOrientation;
-

---

927-950: Inconsistent logging in executeAdbCommand.

Lines 944 and 948 use System.out.println and e.printStackTrace() while the rest of the file has been migrated to logger. Consider using logger.warn(...) and logger.error(...) for consistency.

Proposed fix

 			if (exitCode != 0) {
-				System.out.println("ADB command failed with exit code: " + exitCode);
+				logger.warn("ADB command failed with exit code: {}", exitCode);
 			}
 
 		} catch (Exception e) {
-			e.printStackTrace();
+			logger.error("Error executing ADB command: {}", Arrays.toString(command), e);
 		}

---

162-171: clickAndsendKeysToTextBox3 naming is inconsistent with clickAndSendKeysToTextBox (line 863).

One uses camelCase And while the other uses lowercase And + different casing of send. Consider aligning the naming convention across these helper methods for discoverability. This is a nit — no functional impact.

ui-test/src/main/java/regclient/pages/english/BiometricDetailsPageEnglish.java (2)

302-302: Move logger declaration to the top of the class.

Convention is to declare static final loggers as the first field in the class, not at the bottom. This improves discoverability.

Proposed fix

 public class BiometricDetailsPageEnglish extends BiometricDetailsPage {
+	private static final Logger logger = LoggerFactory.getLogger(BiometricDetailsPageEnglish.class);
+
 	`@AndroidFindBy`(...)
 	private WebElement irisScanIcon;

And remove line 302.

---

228-235: Empty catch block silently swallows failures from clickOnStayLoggedInButton().

If the stay-logged-in click itself fails (e.g., element goes stale), you'll have no indication. Consider at least logging the exception.

Proposed fix

 		} catch (Exception ignored) {
+			logger.info("Auto-logout handling failed: " + ignored.getMessage());
 		}

[coderabbitai]

Actionable comments posted: 1

🤖 Fix all issues with AI agents

In
`@ui-test/src/main/java/regclient/pages/english/BiometricDetailsPageEnglish.java`:
- Around line 209-211: The endsWith check is using the original untrimmed id
which can cause a duplicated SUFFIX when id has trailing whitespace; in
BiometricDetailsPageEnglish (method using id, SUFFIX and finalId) trim the id
first into a variable (e.g., trimmedId = id.trim()), then perform
trimmedId.endsWith(SUFFIX) and build finalId from trimmedId, and update the
logger.info call to reference the trimmed value and the computed finalId so the
check and concatenation use the same trimmed string.

🧹 Nitpick comments (1)

ui-test/src/main/java/regclient/pages/english/BiometricDetailsPageEnglish.java (1)

302-302: Nit: logger declaration is conventionally placed at the top of the class, near other field declarations.

[coderabbitai]

Actionable comments posted: 1

🤖 Fix all issues with AI agents

In
`@ui-test/src/main/java/regclient/pages/english/BiometricDetailsPageEnglish.java`:
- Around line 295-302: The fallback loop in the method that reads element text
(iterating over attributes "text", "hint", "content-desc", "name" on variable
el) risks returning placeholder/hint content instead of the actual entered
value; remove "hint" from the fallback attribute list so the loop only checks
"text", "content-desc", and "name" (or otherwise ensure getAttribute("hint") is
not used for verification) to avoid false positives/mismatches in typeAndVerify;
update the attribute array in the method (where el.getAttribute(...) is called)
accordingly and keep the try/catch behavior intact.

🧹 Nitpick comments (3)

ui-test/src/main/java/regclient/pages/english/BiometricDetailsPageEnglish.java (3)

194-199: Inconsistent interrupt handling between initial sleep and polling sleep.

When InterruptedException is caught during the initial 30-second sleep (line 197-198), the interrupt flag is restored but execution continues into the polling loop. In contrast, the polling sleep (line 251-253) restores the flag and throws AssertionError. This means an interrupt during the initial wait is partially swallowed — the method will continue polling until the next sleep iteration detects the flag.

Consider making the behavior consistent:

Proposed fix

 		try {
 			logger.info("Waiting 30 seconds for email delivery...");
 			Thread.sleep(30000);
 		} catch (InterruptedException e) {
 			Thread.currentThread().interrupt();
+			throw new AssertionError("Interrupted while waiting for email delivery", e);
 		}

---

307-307: Logger declaration should be near the top of the class.

By Java convention, private static final fields (especially loggers) are declared at the top of the class, before constructors and methods. Moving this to line ~34 (after the class declaration) improves discoverability.

Proposed fix

 public class BiometricDetailsPageEnglish extends BiometricDetailsPage {
+	private static final Logger logger = LoggerFactory.getLogger(BiometricDetailsPageEnglish.class);
+
 	`@AndroidFindBy`(uiAutomator = "new UiScrollable(...)

And remove line 307.

---

221-230: AssertionError from line 226 bypasses the catch (Exception e) block — intended but worth a note.

AssertionError extends Error, not Exception, so the throw on line 226 will never be caught by the catch (Exception e) on line 228. Both paths ultimately propagate an AssertionError, so the runtime behavior is correct. However, the error from line 226 will lack a cause chain while the one from line 229 preserves it. If this distinction matters for debugging, consider restructuring to a single throw point.