Regina | Real-time MongoDB 's database server using Socket.IO | Socket library

Sort values by type and population:

• type by ascending order because 'C' is before 'P'
• population by descending order to have highest values at top

Then group by id and state columns and finally take the first value of each group:

It would probably be easiest to loop over the numbers and build the dictionaries, something like this:

Try to remove the limit on the number of displayed columns with:

As the name suggests, register_shutdown_function registers a handler to run every time PHP "shuts down" - that is, at the end of every request, regardless of whether it's ending "normally" or after an error.

It's therefore up to you, inside the handler function, to detect whether the shutdown was due to a fatal error or not. The common trick with this is to look at the last "error", which might better be termed "diagnostic", because it includes Warnings, Notices, and Deprecation messages. If there was a fatal error, we can expect it to be the last thing that happened before the shutdown handler is called, so if we immediately call error_get_last, we should get the information about it.

Let's trace through some key parts of your function:

The concrete error in your code was this line:

You could [index] a random city with Math.random(), using cities.length as your max value.

The complete error you are getting is probably this one:

Uncaught TypeError: Failed to execute 'drawImage' on 'CanvasRenderingContext2D': No function was found that matched the signature provided.

The problem here is that CanvasRenderingContext2D.drawImage requires an Image object as its first parameter.

Your findImage function currently returns the image path.

You could alter your function like this:

So add a check to see if you have any. And you really should use ternary to return a value, not as an if/else

You can add row_number to the partitions after windowing and filter based on this to limit records per window. You can control the maximum number of rows per window using max_number_of_rows_per_partition variable in the code below.

Since your question did not include the way you want stars and review_count ordered, I have assumed them to be descending.