Batch translate all files & Custom Open-AI API support by JargeZ · Pull Request #85 · soberlemur/autopo

JargeZ · 2025-12-07T09:34:51Z

Hello my mates!

I proudly vibe-coded something 🤠 I can't verify myself from a Java perspective.
BUT
it works great and has really helped me use the local LMStudio to make translations for my micro-project.
That you can, btw, check out here bbq-me.org it's a pretty fun tool that helps me find places to hang out with friends on the weekends 🌭

I want to leave this in the pull request here to not abandon the work that I believe can help many other people who have found this program very useful, but I immediately got tired of clicking on each file manually and waiting to move on to the next one. And again after adding each new line for translation...

So

It kinda closes Add a menu entry to translate untranslated #45 Now you can just click one button and wait
Allows to connect any OpenAI compatible provider (qwen3-30b locally pretty good)

Copilot

Pull request overview

This PR adds batch translation functionality for all project files and support for custom OpenAI-compatible API endpoints (enabling use of local models like LMStudio and qwen3-30b).

Key Changes:

Adds a "Batch AI translation (All Files)" button that sequentially processes all translation files in a project
Introduces custom OpenAI base URL configuration and custom model name support for OpenAI-compatible providers
Enables editable model selection in UI to support arbitrary model names

Reviewed changes

Copilot reviewed 7 out of 7 changed files in this pull request and generated 8 comments.

Show a summary per file

File	Description
autopo-i18n/po/autopo.pot	Adds translation strings for the new "Batch AI translation (All Files)" feature
TranslationsTableToolbar.java	Implements TranslateAllFilesButton class with sequential file processing logic
OpenAiModelDescriptor.java	Adds HTTP client configuration, base URL support, and custom model name handling
OpenAISettings.java	Updates UI to support editable model names and custom base URL field
OpenAIPersistentProperty.java	Adds BASE_URL persistent property
module-info.java	Adds required module dependencies for HTTP client functionality
pom.xml	Adds langchain4j-http-client-jdk dependency

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot · 2026-01-03T16:47:12Z

+                if (n) {
+                    ofNullable(subscription[0]).ifPresent(Subscription::unsubscribe);
+                    // Move to next file after translation completes
+                    processNextFile(iterator, aiModelDescriptor, description);
+                }
+            });
+        }


The subscription array pattern creates a potential memory leak. If the translation never completes (e.g., due to an error or cancellation), the subscription will never be unsubscribed and the listener will remain active indefinitely. Consider handling error cases or adding a cleanup mechanism to ensure subscriptions are always released.

Suggested change

if (n) {

ofNullable(subscription[0]).ifPresent(Subscription::unsubscribe);

// Move to next file after translation completes

processNextFile(iterator, aiModelDescriptor, description);

}

});

}

// Always unsubscribe to avoid leaking the listener, regardless of outcome

ofNullable(subscription[0]).ifPresent(Subscription::unsubscribe);

// Move to next file after translation completes or in case of failure/cancellation

processNextFile(iterator, aiModelDescriptor, description);

});

}

}

Copilot · 2026-01-03T16:47:13Z

+            JdkHttpClientBuilder customHttpClientBuilder = JdkHttpClient.builder()
+                    .httpClientBuilder(HttpClient.newBuilder()
+                            .version(HttpClient.Version.HTTP_1_1) // Specify HTTP/1.1 for certain local models like vLLM
+                            .connectTimeout(Duration.ofSeconds(10))) // Set connection timeout
+                    .readTimeout(Duration.ofSeconds(60)); // Set read timeout


The code duplicates HTTP client configuration logic between translationModel() and validationModel() methods. Consider extracting this configuration into a private helper method to improve maintainability and ensure consistency.

Copilot · 2026-01-03T16:47:13Z

+            var baseUrl = repo.getString(OpenAIPersistentProperty.BASE_URL.key(), (String) null);
+            if (isNotBlank(baseUrl)) {
+                builder.baseUrl(baseUrl);
+            } else {
+                builder.baseUrl(DEFAULT_OPENAI_BASE_URL);
+            }


The base URL configuration logic is duplicated between translationModel() and validationModel() methods. Consider extracting this logic into a private helper method to improve maintainability and reduce duplication.

Copilot · 2026-01-03T16:47:13Z

+            var modelName = repo.getString(OpenAIPersistentProperty.MODEL_NAME.key(), GPT_4.name());
+            try {
+                builder.modelName(OpenAiChatModelName.valueOf(modelName));
+            } catch (IllegalArgumentException e) {
+                // Custom model name not in enum, use as string
+                builder.modelName(modelName);
+            }


The model name configuration logic with exception handling is duplicated between translationModel() and validationModel() methods. Consider extracting this logic into a private helper method to improve maintainability and ensure consistent error handling.

Copilot · 2026-01-03T16:47:13Z

+        add(new Label(i18n().tr("Base URL:")), 0, 4);
+        var urlField = new TextField();
+        urlField.setPromptText("https://api.openai.com/v1/");
+        ofNullable(repo.getString(OpenAIPersistentProperty.BASE_URL.key(), (String) null)).ifPresent(urlField::setText);
+        urlField.textProperty().subscribe((o, n) -> repo.saveString(OpenAIPersistentProperty.BASE_URL.key(), n));
+        setFillWidth(urlField, true);
+        add(urlField, 1, 4, 2, 1);


The Base URL field lacks input validation. Users could enter malformed URLs or non-HTTPS endpoints which could lead to runtime errors or security issues. Consider adding URL format validation and potentially restricting to HTTPS for security purposes.

Copilot · 2026-01-03T16:47:14Z

+            try {
+                builder.modelName(OpenAiChatModelName.valueOf(modelName));
+            } catch (IllegalArgumentException e) {
+                // Custom model name not in enum, use as string
+                builder.modelName(modelName);
+            }


The IllegalArgumentException catch block silently falls back to using the model name as a string without logging or notifying the user. This could make it difficult to debug issues when a user enters an invalid model name. Consider logging the exception or providing user feedback when falling back to custom model names.

Copilot · 2026-01-03T16:47:14Z

+        add(new Label(i18n().tr("Base URL:")), 0, 4);
+        var urlField = new TextField();
+        urlField.setPromptText("https://api.openai.com/v1/");
+        ofNullable(repo.getString(OpenAIPersistentProperty.BASE_URL.key(), (String) null)).ifPresent(urlField::setText);
+        urlField.textProperty().subscribe((o, n) -> repo.saveString(OpenAIPersistentProperty.BASE_URL.key(), n));
+        setFillWidth(urlField, true);
+        add(urlField, 1, 4, 2, 1);
+
        Button clearButton = new Button(i18n().tr("Clear"));
        clearButton.setTooltip(new Tooltip(i18n().tr("Clear OpenAI settings")));
        clearButton.setGraphic(FontIcon.of(FluentUiFilledAL.ERASER_24));
        clearButton.setOnAction(e -> {
            repo.clean();
            apiField.setText("");
            modelCombo.getSelectionModel().clearSelection();
+            urlField.setText("");
        });
        add(clearButton, 0, 3);


The Base URL field is added at row 4, but the Clear button is at row 3, causing the UI layout to be inconsistent with a gap between the Temperature field (row 2) and the Clear button (row 3). The Base URL should be at row 3, and the Clear button should be moved to row 4 to maintain a logical top-to-bottom flow without gaps.

Copilot · 2026-01-03T16:47:14Z

+                final Subscription[] subscription = new Subscription[1];
+                subscription[0] = poFile.status().subscribe(status -> {
+                    if (status == LoadingStatus.LOADED) {
+                        ofNullable(subscription[0]).ifPresent(Subscription::unsubscribe);
+                        translateCurrentFile(aiModelDescriptor, description, iterator);
+                    } else if (status == LoadingStatus.ERROR) {
+                        ofNullable(subscription[0]).ifPresent(Subscription::unsubscribe);
+                        // Skip this file and continue with next
+                        processNextFile(iterator, aiModelDescriptor, description);
+                    }
+                });


The subscription array pattern used here creates a potential memory leak. The subscription is stored in an array to allow self-reference, but if the status never becomes LOADED or ERROR, the subscription will never be unsubscribed and the listener will remain active indefinitely. Consider adding a timeout or cleanup mechanism to ensure the subscription is always eventually unsubscribed.

JargeZ added 2 commits December 7, 2025 20:21

Create button to translate all files in project

4946793

Support custom model and OpenAI-compatible API-endponts

4b0aef3

JargeZ changed the title ~~Batch ranslate all files & Custom Open-AI API support~~ Batch translate all files & Custom Open-AI API support Dec 7, 2025

torakiki requested a review from Copilot January 3, 2026 16:42

Copilot started reviewing on behalf of torakiki January 3, 2026 16:43 View session

Copilot AI reviewed Jan 3, 2026

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Batch translate all files & Custom Open-AI API support - #85

Batch translate all files & Custom Open-AI API support#85
JargeZ wants to merge 2 commits into
soberlemur:masterfrom
JargeZ:features

JargeZ commented Dec 7, 2025

Uh oh!

Copilot AI left a comment

Uh oh!

Copilot AI Jan 3, 2026

Uh oh!

Copilot AI Jan 3, 2026

Uh oh!

Copilot AI Jan 3, 2026

Uh oh!

Copilot AI Jan 3, 2026

Uh oh!

Copilot AI Jan 3, 2026

Uh oh!

Copilot AI Jan 3, 2026

Uh oh!

Copilot AI Jan 3, 2026

Uh oh!

Copilot AI Jan 3, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Uh oh!

Conversation

JargeZ commented Dec 7, 2025

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Copilot AI Jan 3, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Jan 3, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Jan 3, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Jan 3, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Jan 3, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Jan 3, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Jan 3, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Jan 3, 2026

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants