BigText.ADDTEXT loop is slow in RTC

I had a code where I was combining several XML documents into one and used a BigText variable for the job.  At design time everything looked fine but when I executed the code in the Role Tailored Client everything was slow.  I sent this issue to Microsoft and the product team has responded.

It is true that AddText used in a loop is slower in RTC than in CC. The reason is the underlying .Net string type. As mentioned in the immutability section in: http://msdn.microsoft.com/en-us/library/system.string.aspx#Immutability the creation of many immutable strings can be very time consuming. Since the backing field in NAV is a string we have the same problem – and it would be vastly inefficient in all other scenarios to create a StringBuilder for each BigText variable – and you only save instantiating strings if you do more than one concatenation.

This leaves us the following options:
–          If partner is adding texts – consider using .Net Interop and use the StringBuilder class – and document the changed behavior of BigText
–          Refactor code to avoid hundreds / thousands of small additions (e.g. by building the total big text in chunks of e.g. 1000 – but will have to look at perf of that)
–          Create a new feature for a StringBuilder class in A/L (would possibly have to work for the other text types as well?)

  • I doubt that this would be prioritized in Sicily

I know that this is an unwanted side effect of using .Net strings – but that is also what buys us Unicode etc., so I would prefer that we document this new behavior.

So I rewrote this bit and am now using OutStream instead.  Just wanted you to know.

One Reply to “BigText.ADDTEXT loop is slow in RTC”

  1. Hi Gunnar

    Thank you for reply.
    So our product team agreed to do global solution:
    – Increase msdn documentation about BigText.ADDTEXT(‘’) usage risk in performance.
    – Make own function in next version to resolve current problem.

    So it looks in next version things go better.
    Thank you for reporting this issue, your findings make NAV better.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.