[Koha-bugs] [Bug 11064] batch import ignoring framework choice

[bugzilla-daemon at bugs.koha-community.org]

http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=11064

--- Comment #5 from Jacek Ablewicz <abl at biblos.pk.edu.pl> ---
(In reply to David Roberts from comment #4)
> Is this related to
> http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=10906?

In a way I guess (hard to say for sure, because original bug report is a bit
vaque, and over 1.5 years old). It is somehow complicated.. There are 2
possible cases when importing biblio records:

1) some of the biblios in the given import batch do match some existing local
record[s], and our choice is to replace such local records contents with
imported ones if they match

2) some of the biblios in the import batch do not match any existing record[s]
in the local catalogue, or we wish to import them 'as new' even if there are
some possible matches found for them

Before Bug 10906 fix, for cases #1, framework choice in import batch commit
form was being ignored no matter what (but note that this particular setting is
worded as 'Add _new_ bibliographic records into this framework', so it should
not apply for records whitch are not imported 'as new' - ?) and also the
previousy set framework in the existing recods being replaced by imported ones
was getting lost in the process

After Bug 10906, for cases #1, previously set framework in the existing records
is not getting lost any longer, but framework choosen in the import batch
commit form (if any) is still being ignored for the records which are not
imported "as new". But I guess it probably should behave like this = this is an
expected behaviour for such cases?

Now, for cases #2 (biblios imported "as new" for whatever reason - there are no
matches, or we don't wish to replace any existing records even if there is a
match), framework choosen in the import batch commit form _should_ apply.
But it seems to me that it _does_ apply and is not being ignored (at least not
for records imported "as new") - ? Note: I tested it only in 3.18 and in ~2
months old master (and it appears to work fine / as expected), but not in the
current clean master - perhaps it got somehow broken in more recent versions.

-- 
You are receiving this mail because:
You are watching all bug changes.