fix: update Twitter URL and improve functionality/fix: resolved various data and calculation errors

[sergeypanin1994]

fix: update Twitter URL and improve functionality

• Updated Twitter URL to reflect rebranding to x.com format
• Enhanced social card generation and MDX file processing
  • Improved sanitizeSlug function to support Unicode and clean special characters
  • Fixed splitTitleIntoLines logic to prevent text loss
  • Added template image validation before processing in generateSocialCard
  • Improved error handling and logging for better debugging
• Optimized MDX file processing for efficient changes
• Improved directory handling with recursive subdirectory processing

fix: resolved various data and calculation errors

• Fixed calculateDailyIncrease time range calculation
• Added check for values.length < 2 in daily increase computation
• Corrected totalSize handling to avoid empty data errors
• Improved axios request error handling and added API response validation
• Fixed getWeekNumber logic for accurate week number calculation
• Improved JSON file handling to prevent data corruption
• Added additional logging for better diagnostics

Node size data is now processed and saved correctly.

[vercel]

@sergeypanin1994 is attempting to deploy a commit to the Consensys Team on Vercel.

A member of the Team first needs to authorize it.

[jlwllmr]

Thanks @sergeypanin1994. Just the one thing to catch — TIL how week numbering works 😅

I'll tag in @sammysango to take a look at the social card function.

[jlwllmr]

We need to retain line 31 in this function otherwise there's a risk that week numbering will be wrong. The ISO system uses Thursday as an anchor day (hence the +4, where 0 is Monday, the week start) to ensure that the week 1 is actually the first week of the year that has at least 4 days in the calendar year. See here.

[sammysango]

@sergeypanin1994 Card script improvements looking great! One question, for this point:

Fixed splitTitleIntoLines logic to prevent text loss

As far as I can tell, this removes the logic that handles long titles by truncating them with .... With this change, the title will end part-way through if it hits the max, but without a ... indication that the text has been cut short. So either way, text loss occurs, but the previous method deals with this more elegantly imo. Please correct me if I'm wrong/missing something though!

I don't think this actually affects any of the docs/cards at their current title length, but it'd be preferable to retain as a safeguard against future lengthy titles, if they occur.