[TYPO3-Solr] Multilanguage Indexing, one domain per lang & realurl

hard.one Software GmbH - Martin Peter Hanke mphanke at hardone-it.ch
Mon Jan 13 10:05:15 CET 2014


yes, exactly...
But I thing you hinted me to the catch 22...

With the realurl domain switch as described in the link, you should not 
have domain records.
I will investigate in this direction.

Thanks a lot for pointing this out!

Best

Martin

On 13.01.2014 09:50, Hans Höchtl wrote:
> So you have something like this:
>
> Single Tree
> | - PageXYZ
>      | - L=0 (English)   example.com
>      | - L=1 (German)  example.de
>
> Do you have Domainrecords for every domain?
>
> Best
> Hans
>
>
>
> 2014/1/13 hard.one Software GmbH - Martin Peter Hanke <mphanke at hardone-it.ch
>>
>
>> Hi Hans
>>
>> we have only one root.
>> To prevent misunderstandings:
>> All we do is switching the domain as described here by Dimitry:
>> http://www.dmitry-dulepov.com/2009/02/realurl-separate-
>> language-domains-in.html
>>
>> Might this be the challenge for the indexer?
>>
>> Best
>>
>> Martin
>>
>>
>> On 13.01.2014 09:41, Hans Höchtl wrote:
>>
>>> Hi Martin,
>>>
>>> you can install devlog and turn on
>>> plugins.tx_solr.logging.indexing = 1
>>> plugins.tx_solr.logging.indexing.indexQueueInitialization = 1
>>>
>>> than you can see what values are being indexed. Take a look at the field
>>> siteHash.
>>> Also take a look inside the table tx_solr_indexqueue_item if all records
>>> are listed correctly with the correct root. Otherwise they would not be
>>> indexed.
>>>
>>> Regards,
>>> Hans
>>>
>>>
>>> 2014/1/13 hard.one Software GmbH - Martin Peter Hanke <
>>> mphanke at hardone-it.ch
>>>
>>>>
>>>>
>>>   Hi Hans
>>>>
>>>> can you hint me to some place I can can check why the indexer behaves in
>>>> this odd way?
>>>> Before updating to the latest version and switching to realurl domain per
>>>> lang things were running.
>>>>
>>>> Best
>>>>
>>>> Martin
>>>>
>>>>
>>>> On 13.01.2014 09:29, Hans Höchtl wrote:
>>>>
>>>>   Hi,
>>>>>
>>>>> no that's not necessary. The Indexer generates a domain hash which
>>>>> automatically filters the output. So this would work out of the box. We
>>>>> have a multitree instance with 28 trees up and running and it works.
>>>>>
>>>>> Best regards,
>>>>> Hans
>>>>>
>>>>>
>>>>> 2014/1/13 hard.one Software GmbH - Martin Peter Hanke <
>>>>> mphanke at hardone-it.ch
>>>>>
>>>>>
>>>>>>
>>>>>>     Hi all
>>>>>
>>>>>>
>>>>>> we are experiencing some odd indexing behaviour on your website.
>>>>>>
>>>>>> We have a domain based langugage switch and two solr 4.6 cores along
>>>>>> with
>>>>>> the git-version of the extension.
>>>>>>
>>>>>> Some pages are not indexed at all whereas others are indexed and it
>>>>>> seems
>>>>>> randomly inserted into the cores.
>>>>>>
>>>>>> Do we have to implement an own page indexer for this scenario? If so,
>>>>>> could you give us advice on how to do this in the right manner?
>>>>>>
>>>>>> Best
>>>>>>
>>>>>> Martin
>>>>>> _______________________________________________
>>>>>> TYPO3-project-solr mailing list
>>>>>> TYPO3-project-solr at lists.typo3.org
>>>>>> http://lists.typo3.org/cgi-bin/mailman/listinfo/typo3-project-solr
>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>>    _______________________________________________
>>>>>
>>>> TYPO3-project-solr mailing list
>>>> TYPO3-project-solr at lists.typo3.org
>>>> http://lists.typo3.org/cgi-bin/mailman/listinfo/typo3-project-solr
>>>>
>>>>
>>>
>>>
>>>   _______________________________________________
>> TYPO3-project-solr mailing list
>> TYPO3-project-solr at lists.typo3.org
>> http://lists.typo3.org/cgi-bin/mailman/listinfo/typo3-project-solr
>>
>
>
>


More information about the TYPO3-project-solr mailing list