performance: use Array instead of Collection

[Nick-vanGemeren]

What performance improvements do you expect for small and very large JSON arrays? Note that every time you call ReDim Preserve the entire array is copied to a new array.

There are some backward-compatibility issues:

• Collections have first element (1). Arrays have first element (0). This could be fixed in the ReDim statement.
• Existing code may be using the Count/Add/Remove methods which are not available for Arrays.
• Existing code may be using Typename/Vartype methods to navigate (and verify) the data structure.

So your ParseArray should be available as a non-default compilation option.

Note that no maintenance is being done on this repo. Don't hold your breath for the PR.

[WilianZilv]

What performance improvements do you expect for small and very large JSON arrays? Note that every time you call ReDim Preserve the entire array is copied to a new array.

There are some backward-compatibility issues:

• Collections have first element (1). Arrays have first element (0). This could be fixed in the ReDim statement.
• Existing code may be using the Count/Add/Remove methods which are not available for Arrays.
• Existing code may be using Typename/Vartype methods to navigate (and verify) the data structure.

So your ParseArray should be available as a non-default compilation option.

Note that no maintenance is being done on this repo. Don't hold your breath for the PR.

Actually, i forgot to explain what i mean by performance. I dont think the change by itself benefits the parser.

I did this change because i'm using json to write data in the sheet, to do that i need to convert the Collection to an array, which is too slow (i was testing with 100k values), with Collections it can take 1+ minutes, with arrays the conversion to the structure i need is much less expensive, resulting in less than a second to do so.