Release Notes - Version 0.4.3-beta

Changes:

• Fixed Most CSS Conflicts: Resolved conflicts with other plugins that were causing issues with the styling of our application. The UI should now display correctly without any interference from other plugins.

• Startup Template Loading: Fixed an issue where templates were not loading correctly at startup. Users will now experience a smoother and more reliable application launch.

• Enhanced Key Encryption: We have improved the encryption for keys, enhancing the security of your data. Additionally, we've added an option that allows users to disable encryption if needed, providing more flexibility in managing their data security and compatibility with mobile devices

• Section Resizing: Fixed a bug where sections would have a weird maxHeight value when resizing the application window. Now, sections should adapt properly to the window size changes, providing a better user experience.

Full Changelog: https://github.com/nhaouari/obsidian-textgenerator-plugin/compare/0.4.2-beta...0.4.3-beta

Release Notes - Version 0.4.2-beta

New Features:

• Added support for Obsidian Properties in templates.

Bug Fixes:

• Addressed issues causing unexpected behaviors in stream generation.
• Fixed the auto-suggest issue when Vim is in insert mode. Thanks to @chrisgrieser for the fix.

We appreciate your continued feedback as we move towards a stable release. Please report any issues or suggestions to our GitHub Issues page.

Thank you for being a part of our community!

Release Notes: Version 0.4.1-beta

Changes:

• Error Handling: Now displays an error message for invalid inputs instead of showing "NAN".

• UI Update: The commands pallet now displays the template ID, replacing the previous path.

Bugfixes:

• Input Behavior: Resolved an issue where Temperature and other number-based settings inputs behaved unexpectedly.

• Display Issue: Fixed a problem where the range input did not display its value.

Thank you for updating to the latest version. We always value feedback from our users to help us improve. If you encounter any issues or have suggestions, please let us know!

Version 0.4.0 Release Notes

Major Updates:

• Revamped Settings Page: Introducing a fresh design with added search functionality.
• Expanded LLm Provider Support: Beyond OpenAI, we now support various LLm providers. Even self-hosted solutions are compatible!
• Langchain Integration: We've integrated with Langchain, paving the way for numerous future use cases.
• Code Overhaul: We've completed a comprehensive code cleaning and refactoring. Error-free code makes contributions a breeze.
• Streaming Stability: Streaming has been stabilized and is now activated by default.

Minor Updates:

• Enhanced Extractors: Added extractors.all to templates, allowing for extraction of all content without the need for double looping.

  {{#each extractions.all}}
     {{this}}
  {{/each}}
  

• Whisper's Custom Basepath: Users can now set a custom basepath to whisper.
• React Settings Page: We've rebuilt the settings page using React, enhancing customizability.
• Bug Fixes: Addressed numerous bugs to ensure a smoother experience.
• Streaming Mode: Just a reminder, streaming is now the default mode!

Version 0.3.20

• Add "gpt-3.5-turbo-16k-0613"

New OpenAI Models are available in Text Generator

• "gpt-3.5-turbo-16k"

• "gpt-3.5-turbo-0613"

• "gpt-4-0613" (waitlist)

• "gpt-4-32k-0613" (waitlist)

V0.3.18

We are delighted to announce the official rollout of version v0.3.18 for Text Generator. This particular update includes an exciting new feature.

Innovative Feature Addition

• We have implemented clipboard functionality within the context. The clipboard's content can now be accessed via the variable {{{clipboard}}}.

• This enhancement enables seamless utilization of templates on past text. For instance, you can use a template that simplifies and summarizes copied text.

We highly value the opinions of our user community and encourage all users to share their feedback and report any potential issues on our Discord server.

We sincerely appreciate your continued support and use of Text Generator!

Release Notes - Text Generator v0.3.17

Fixes

• Fixed extractions in templates, ensuring that they are now properly extracted and displayed.

We always welcome your feedback, so please don't hesitate to join our Discord server at Discord if you have any questions or suggestions for future releases.

Release Notes - Version 0.3.16-beta

We are excited to announce the release of version 0.3.16-beta of Text Generator!

New Feature

• Streaming option for Text Generator ( generate text and generate text with metadata) is now available! This feature allows users to generate text in real time using OpenAI models.

If you have any feedback or questions about this release, please join our Discord server at https://discord.com/invite/BRYqetyjag.

Thank you for using Text Generator!

Release Notes - Text Generator v0.3.15: Enhanced Auto-Suggest & Bug Fixes

We are thrilled to introduce Text Generator version 0.3.15, which brings a variety of enhancements to our auto-suggest feature and resolves several bugs. Get ready for a smoother and more efficient text generation experience!

Key updates in this release include:

• Auto-suggest functionality now compatible with GPT-4 model
• Automatic plugin loading when options are modified

Important: Please be aware that utilizing the GPT-4 model for auto-suggest may incur higher costs. We advise reviewing the pricing details on the OpenAI website prior to using this model for auto-suggest.

We value your input and encourage you to share your feedback and ideas with us. Join our vibrant Text Generator community on our Discord server to share your thoughts and engage with fellow users.

Thank you for your ongoing support!

Version 0.3.13

• fix tg codeblock

Version 0.3.12

Release Notes - Version 0.3.10-beta

We are excited to announce the release of Text Generator version 0.3.10-beta! This latest version includes a Text Extractor Tool that helps to extract text from various sources such as webpages, YouTube videos, PDF files, audio files, and images.

We have worked on enhancing the Web page and YouTube video extractors for improved functionality and usability.

Note: to use the new Text Extractor Tool, you will need to enable it from the plugin settings and reload it.

As always, we welcome any feedback or suggestions on how we can continue to improve Text Generator. Join our Discord community at Discord to provide your feedback and engage with other users.

Thank you for your support!

Version 0.3.9-beta

We are excited to announce the release of Text Generator version 0.3.9-beta! This version includes several enhancements and bug fixes that improve the performance and functionality of our open source project.

This release includes:

• Better variables extractions in templates: We have improved the way variables are extracted from templates, making it more efficient and accurate.

• getContext in templates is faster: We have optimized getContext in templates, resulting in faster execution times with slash commands.

• Templates modals shows the values of variables from actional context: The templates modals now display the values of variables using the context of the active note making it easier to debug and troubleshoot issues.

As always, we welcome your feedback on this release. Please join our Discord server at https://discord.com/invite/BRYqetyjag to share your thoughts and suggestions with us.

Thank you for your support, and happy generating!

Release Notes - Text Generator v0.3.8-beta

We are pleased to announce the release of Text Generator version 0.3.8-beta! This release includes several new features, bug fixes, and improvements to enhance your experience with the plugin.

New Features

• Modal-suggest option: A new modal-suggest option has been added that allows you to show templates as modals that can appear in the list of suggestions after writing slash at the start of a new line after enabling the feature model-suggest from the settings of the plugin and reloading it.
• Custom OpenAI API access endpoints: We have introduced custom OpenAI API access endpoints to facilitate changing the OpenAI access endpoint such as Azure users thanks to dtcui.
• Tailwind framework: We have integrated Tailwind framework which will facilitate styling.

Bug Fixes and Improvements

• Refactored auto-suggest: The auto-suggest function has been refactored for better performance and accuracy.
• Fixed typos: Several typos have been fixed in this release, improving overall readability.
• Improved token estimation: The token estimation is now more accurate and provides results similar to OpenAI results. It is calculated using the function published in the following example https://github.com/openai/openai-cookbook/blob/main/examples/How_to_count_tokens_with_tiktoken.ipynb

We encourage all users to update their Text Generator plugin to take advantage of these new features and improvements. As always, if you encounter any issues or have any feedback, please join our Discord server here where we will be happy to assist you.

Thank you for using Text Generator!

https://user-images.githubusercontent.com/9850722/233844874-145f2d42-07f5-490e-bfab-0ede093e857b.mp4

0.3.7-beta

Text Generator v0.3.6-beta Release Notes

We are excited to announce the release of Text Generator v0.3.6-beta. In this update, we have focused on enhancing the user experience by improving various aspects of the application. Below is a summary of the changes introduced in this release:

Enhancements:

1. Web-Page-Extractor Improvements for PC: We have improved the web page extraction on PCs, making it more efficient and effective in capturing the relevant text. Additionally, we have implemented a mechanism to escape media URLs (images, audio, and URLs) in the web-page-extractor, ensuring a cleaner extraction of text data.

2. Insert, Replace, and Rename Modes: Users can now specify the desired mode for completions in the config file. The available modes are:

• insert: Inserts the completion at the cursor's position.
• replace: Replaces the selected text with the completion.
• rename: Renames the title of the active file with the completion.

Example of the config file with mode selection:

config:
 mode: insert

3. System and Messages in Config: You can now add system and messages in the config file to format your interactions with the Text Generator. For example, configuring the system to act as an English translator, spelling corrector, and improver that responds in corrected and improved English. Additionally, you can define messages for various scenarios.

Example of config file with system and messages configuration:

4. Addition of EQ Helper Function: We have added the eq function to the available helper functions. This allows users to modify template behavior based on frontmatter. This can be useful in conditional situations, for instance:

config:
 mode: insert
 system: <Your system description>
 messages:
 - <user Message>
 - <assistant Message>
 - <user Message>
 - <assistant Message>

We hope that these improvements make your experience with Text Generator v0.3.6-beta even more enjoyable and productive. As always, we welcome your feedback and suggestions to help us make the Text Generator better with each release.