1. Gary Eden
  2. PowerBuilder
  3. Wednesday, 27 March 2019 14:57 PM UTC

Hi,

I thought I would mention an issue that I seem to have picked up between development and live environment.

I have a 100 page report and I perform the SaveAs PDF using native pdf, this works fine in development and produces a 100 page pdf report. In the live environment the report can be any size and occasionally manages to be the correct size. Checking the return code for the incomplete pdf's I get a -1 which indicates a failure however I don't get these failures within the development environment. The live environment happens to be cloud based and the client remote desktops onto the server to run the application.

The worse part of it is that the user is sitting with a partial PDF file that we can inform them is incomplete but would definitely be better if the SaveAs code cleaned up when it detected a failure, also the partial pdf file is viewable within Adobe.

It would also be nicer if the user is not presented with the actual pdf file and the emf as it is creating the pdf, if this was done in memory would it not also speed the process up? maybe part of the problem is the disk access within a cloud environment hence the high failure rate?

 

Thanks

Gary Eden

 

Chris Pollach @Appeon Accepted Answer Pending Moderation
  1. Wednesday, 27 March 2019 15:21 PM UTC
  2. PowerBuilder
  3. # 1

Hi Gary;

   Are you on the latest version & build of Appeon PB (V2017R3 MR01 - build 1880)? There have been many fixes and enhancements to the NativePDF feature in R2/R3.

Regards ... Chris

Comment
  1. Chris Pollach @Appeon
  2. Thursday, 28 March 2019 13:57 PM UTC
Hi Gary;

First of all ... definitely not a recommended practice to mix & match DLL's from different revisions or even random one offs. This can certainly lead to instability issues!

The answer to your R2 & R3 on the same machine ... no (sorry).

Regards ... Chris
  1. Helpful
  1. Gary Eden
  2. Friday, 29 March 2019 17:10 PM UTC
Hi Chris,



I have upgraded to PB2017R3(1880) and all seems a lot better. We do not notice the secondary file anymore although the speed is still somewhat slow.

We do end up with the full datawindow in the pdf and the saveas completing successfully.

I do have a concern regarding the different sizes of the distribution dll's from R2 to R3 and how much regression testing may be required to ensure another issue does not arise.



Thanks

Gary
  1. Helpful
  1. Chris Pollach @Appeon
  2. Friday, 29 March 2019 18:31 PM UTC
Hi Gary;

Most of the changes from R2 => R3 were new features like: Encoding, Encryption, oAuth2.0, and minor improvements to: JSON, PBC, REST etc and NativePDF (of course). Please use that as an indicator as to how much regression testing you need to do. There are more details in the PB 2017R3 MR01 Help File attached to the build 1880 IDE.

Regards ... Chris
  1. Helpful
There are no comments made yet.
  • Page :
  • 1


There are no replies made for this question yet.
However, you are not allowed to reply to this question.