batch processing - Would the performance improve significantly if SC and snapshot only have to be loaded only once?

[aisbergde]

My main goal is still to export a diagram for each object by batch with the SC. Interactively with the shell this works quite well:

connect --server=offline --database=dhw_self.ser
grep --grep-tables=.*\.repo\.RepoObject
filter --parents=1 --children=1
load --info-level=standard
execute --command=schema --output-file=[repo].[Repo=bject].svg

I can do this also in one non-interactive command, which I can generate (outside SC, for example in a database) and execute in batch for each table:

schemacrawler --server=offline --database=dhw_self.ser --grep-tables=.*\.repo\.RepoObject --parents=1 --children=1 --info-level=standard --command=schema --output-file=[repo].[repoobject].svg

So I have a solution.

But I don't know how performant this is if SC and the offline snapshot are reloaded every time. And I wonder: can this also be done directly with SC?

• load SC only once
• load offline snapshot only once
• loop through all objects in offline snapshot
  • grep, filter, load, command

I am already sure that it can be done. There are many examples for different programming languages, and also for the extensive API.

But is it worth the effort? Would the performance improve significantly because SC and snapshot only have to be loaded once?