Closed rob-heaney-itg closed 1 year ago
Hi @rob-heaney-itg ,
I might have thought a different opinion while revising the code. We can make a small change on that.
I might have thought a different opinion while revising the code. We can make a small change on that.
Thanks for replying @ozlerhakan , ok that's fair enough if you did. In that case, perhaps it's something which is best switchable on/off? Maybe some will prefer that it stop on the first error, and others not?
We can undo the change @rob-heaney-itg
Thanks very much @ozlerhakan 👍
Hi @ozlerhakan , bad news, I think the issue still applies, due to also being present on HSSFUnmarshaller.java#L102 as well 😞
Oh really, np we can fix it with 4.1.1
Oh really, np we can fix it with 4.1.1
Thank you @ozlerhakan 👍
Since v3.2.0/4.0.0 it seems as though the
PoijiMultiRowException
is never able to add more than onePoijiRowSpecificException
(in other words, it can't provide a full report of all errors in a document), because it stops processing after it hits the first issue.I think it might be because the
HSSFUnmarshaller
(line 200) is returning a newPoijiMultiRowException
as soon as the errors List is not empty, but the problem is that this is inside the for-loop, and not outside-of/after it, so it will never be able to continue the loop to potentially add more errors.This is different to the P.O.C that I submitted originally as part of issue https://github.com/ozlerhakan/poiji/issues/233 -- Was this change intentional? Seems like a defect to me.