8 Aug
2016
8 Aug
'16
12:07 p.m.
XWikiCacheStore#loadXWikiDoc is actually a bit tricker than that right
now for retro compatibility and laziness (really need to rewrite this
whole store API) reasons and before applying what you propose the way
it deals with document uid should be modified a bit.
Hibernate store is always loading the document from the wiki indicated
in the context completely ignoring the wiki in the XWikiDocument
object. That means that doc.getKey() may be different than the key of
the document that will actually come from the hibernate store. So
before adding DocumentLoader jingling a first thing to do is generate
the initial cache key with a mix of the wiki id located in the context
and the XWikiDocument reference.
Then what you propose sounds good.
On Mon, Aug 8, 2016 at 8:53 AM, Fabian Pichler
<fabian.pichler(a)synventis.com> wrote:
Hi devs
http://jira.xwiki.org/browse/XWIKI-13623
I would like to discuss the suggested solution, before I start implementing
it:
*I suggest the following solution:*
Introducing a nested DocumentLoader class which loads the document calling
the basestore loadXWikiDoc in a synchronized getDocOrLoad method. The
DocumentLoader objects gets instantiated in the cacheStore on a cache miss.
It then gets stored in a ConcurrentHashMap using the same key which is used
for the document cache (DocumentReference including language). Using
putIfAbsent to insert the DocumentLoader and calling the synchronized load
method on the object in the map solves the memory visibility issue and
moreover ensures that the same XWikiDocument is not loaded multiple times
concurrently from the database. It is still possible to load different
documents in parallel. A DocumentLoader will load the XWikiDocument only
once and the returns the identical document object on any following call of
the synchronized getDocOrLoad method.
WDYT?
Kind Regards,
Fabian
_______________________________________________
devs mailing list
devs(a)xwiki.org
http://lists.xwiki.org/mailman/listinfo/devs
--
Thomas Mortagne