Providing support for all E. Crane Computing products, including POWERGEN, PBSERVICE, HARPB, VERSIONEDIT, PBLRESCUE, and PBLEXPLORER.
You are not logged in.
Pages: 1
Hi ,
i-m doing a bootstrap import including "regeneration after import" and the logfile shows around 900 errors during regen. when building the just imported application through powergen afterwards (again including regeneration) no errors occur and the generation of exe/pbds runs without further problems.
To get around this i changed my batchscript to split the generation in 2 parts: first bootstrap with /NBR and then a second run with just /a. Works so far but increases the time needed for a complete build process.
When looking at the errors during the regeneration phase of the bootstrap import it is obvious that all calls to objects that are not yet regenerated (ex: calls to a popup-menu in a user object) lead to an error.
Is there a chance to get rid of these errors durcing bootstrap?
Thanks
Dirk
Hi Dirk,
Something is not right here, even though you are getting the final result that you want.
You shouldn'e be getting the errors and splitting the operation shouldn't make a difference.
Can you send your PowerGen project file and an output log that shows the errors you are getting.
Regards,
Phil
Hi Phil,
please find enclosed
- project file (ecrane.gen)
- logfile bootstrap (bootstrap_powergen.log)
- logfile generate (generate_powergen.log)
Thanks for your help
Dirk
Hi Dirk,
For some reason the attachments didn't make it. Try again or just e-mail me at philw@ecrane.com.
Regards,
Phil
Hi Phil,
hope you got the files with my mail.
Regards
Dirk
Hi Dirk,
No I didn't get them. Can you try again: philw@ecrane.com?
Sorry about the delay. I'll get back to you as soon as I can see the files.
Regards,
Phil
Hi Phil,
okay, second mail is on its way.
Regards
Dirk
Hi Dirk,
Sorry for the delayed response.
These are "interesting" results and I can't explain them yet. But some observations and questions.
First, the reason that splitting the process gives you a error build (in the end), is that you are actually performing a second full regeneration when you do this. That would also explain why it takes longer then the single operation.
Second the initial errors you're getting occur during the Regeneration of Descendants of Imported Objects phase, even though the objects had earlier in the process imported with no errors. I've never seen that before. What happens if you choose NOT to regenerate the descendants?
Sometimes it's helpful to look at a particular error and try to determine why it's occurring. For example the first error I see is:
Undefined variable: am_dw
Where is that variable defined?
Are you deleting the PBLs before you start? If not try that.
This may a take a couple iterations, but I'm sure there's a good explanation to this. We have no reports to applications that won't build using this process.
Regards,
Phil
Pages: 1