Hello,
I think that the actual oversight is that the checkInitSteps should be
called at the beginning of processing initialization steps rather than
while processing -I, because currently other places modify the
initialization string (no-vacuum, foreign key) and thus are not checked.
As far as I read the code, runInitSteps() does the check. If the initialization
steps string contains unrecognized character, runInitSteps() emits an error.
Sure, but the previous step have been executed and committed, the point of
the check is to detect the issue before starting the execution.
* (We could just leave it to runInitSteps() to fail if there are wrong
* characters, but since initialization can take awhile, it seems friendlier
* to check during option parsing.)
The above comment in checkInitSteps() seems to explain why
checkInitSteps() is called at the beginning of processing initialization
steps.
Yep, the comment is right in the motivation, but not accurate anymore wrt
the submitted patch. V2 attached updates this comment.
--
Fabien.
