GPT support on CLI

Introducing Spheron GPT, a CLI tool to enhance code generation capabilities using ChatGPT. It will be a valuable addition to developers workflows. Now, developers can perform various tasks such as creating, updating, bug finding, bug fixing, optimizing, creating test cases, and transpiling code conveniently from their terminals.

The new commands include:

• spheron gpt --prompt <prompt> [--filepath <filepath>]

  • prompt: The code you wish to generate.
  • filepath (optional): Path for the file based on which you want to generate new code.

• spheron gpt update --prompt <prompt> --filepath <filepath>

  • prompt: The code you wish to add in the file.
  • filepath: Path for the file you want to update.

• spheron gpt findbugs --filepath <filepath>

  • filepath: Path for the file you want to debug

• spheron gpt improve --filepath <filepath>

  • filepath: Path for the file you want to optimise

• spheron gpt transpile --filepath <filepath> --language <language>

  • filepath: Path for the file you want to transpile
  • language: Language you want to transpile your code to.

• spheron gpt ctc --filepath <filepath>

  • filepath: Path for the file you want to create test cases for.

SDK New Methods

Added a new method getCdnRecords on Storage, Site and Compute. You can use this method to get the values that should be put into your record so that the domain will start pointing to your data.

Compute Revamp

We are excited to announce the redesigned compute deployment page, aimed at providing an enhanced user experience. Here are some highlighted features that we have introduced:

• Customizable Plans: Users can now easily tailor their compute plans by selecting desired CPU and RAM options from a convenient dropdown menu.
• Flexible Storage Options: We have introduced support for both ephemeral and persistent storage, allowing users to choose from a range of predefined plans or create custom storage configurations.
• Streamlined Deployment Process: Users can now specify the number of replicas in the advanced configuration, enabling simultaneous deployment of multiple instances.
• Real-Time Pricing: We have integrated real-time pricing information into the deployment page, providing users with an estimated cost for their compute resources.

Improvements & Fixes

• Time-out Uploads: Resolved the issue of Upload deployments timing out prematurely.
• Ignore files and folders: Added logic to ignore certain files and folders by default and prevent them from being uploaded during deployment for security and performance reasons. Blacklisted files and folders are:

.hg
.git
.gitmodules
.svn
.cache
.next
.now
.vercel
.npmignore
.dockerignore
.gitignore
._.swp
.DS_Store
.wafpicke-_
.lock-wscript
.env.local
.env.*.local
.venv
npm-debug.log
config.gypi
node_modules
**pycache**
venv
CVS

• Top navigation bar UI improvement: Updated the UI of the top navigation bar, allowing easy switching between the web app and compute platforms, and optimized the menu item in the navbar.
• CLI login page: Fixed the CLI login page, making it accessible even when logged out.
• Time-out Deployments: Addressed the problem of timed-out deployment status not updating and being stuck in the loading state.
• Missing Logs: Fixed the issue preventing deployment logs from displaying on the UI.
• Loading Issue: Resolved the loading issue with the "Load More Instances" button on the cluster main page.
• Password Protection Bonus: Limited the password protection bonus amount to 1.