I found a potential blocker for 0.4-incubator, CONNECTORS-342. I
think it is severe enough to require respin of the candidate. The bad
HTML causes the "Save" button to vanish and makes the File Connector
only barely usable.
I've also found UI problems with the CMIS and Alfresco connectors; the
way hiddens are managed is incorrect so that you can get ${xxx} values
stuffed in the fields on the initial post/repost. In addition they
both suffer from lack of proper escaping of values in the UI. The
OpenSearchServer connector also suffers from the latter. These bugs
are pretty severe in my opinion and I've captured them in
CONNECTORS-343.
I've checked in a fix to the branch for CONNECTORS-342. The fix for
CONNECTORS-343 is a bit more extensive but is certainly doable within
the 0.4-incubating release. Piergiorgio, do you agree that we should
try to fix this for this release?
Karl
On Tue, Dec 20, 2011 at 8:40 AM, Karl Wright <[email protected]> wrote:
> As long as your changes are committed to trunk, I can push them to the
> site immediately. They just won't be in the documentation that ships
> with ManifoldCF 0.4-incubating, is all.
>
> So it sounds like we have a vote to release, as soon as we wait the
> full 72 hours. ;-)
> Karl
>
> On Tue, Dec 20, 2011 at 8:33 AM, Piergiorgio Lucidi
> <[email protected]> wrote:
>> 2011/12/20 Karl Wright <[email protected]>:
>>> Piergiorgio wants to include a section in the "how to build and
>>> deploy" documentation for the Alfresco connector. To me this is not
>>> really an important enough change to spin another RC, but since we're
>>> still waiting for Tommaso to review our releases there may be time to
>>> do it.
>>>
>>> But, right now, I vote +1 for RC0. We'll see if Piergiorgio withdraws his
>>> vote.
>>
>> It's not a problem for me to release this RC0, but I would like to
>> update at least the website documentation about how to build the
>> Alfresco Connector. I hope this night to work on this new section and
>> then we can push an update of the website.
>>
>> Piergiorgio
>>
>>>
>>> Karl
>>>
>>> On Mon, Dec 19, 2011 at 7:04 PM, Shinichiro Abe
>>> <[email protected]> wrote:
>>>> Thank you. I understand.
>>>>
>>>> +1 for me.
>>>>
>>>> Shinichiro Abe
>>>>
>>>> On 2011/12/19, at 16:25, Karl Wright wrote:
>>>>
>>>>> The hsqldb configuration used by multiprocess-example is an external
>>>>> hsqldb database. It does not require the
>>>>> org.apache.manifoldcf.hsqldbdatabasepath property. If you want to try
>>>>> out the multiprocess example, take the following steps:
>>>>>
>>>>> (1) In one shell prompt, start the database: "start-database.bat/.sh"
>>>>> (2) In another shell prompt, initialize the database: "initialize.bat/.sh"
>>>>> (3) When that is done, start the agents proces: "start-agents.bat/.sh"
>>>>>
>>>>> Thanks,
>>>>> Karl
>>>>>
>>>>>
>>>>> On Mon, Dec 19, 2011 at 2:05 AM, Shinichiro Abe
>>>>> <[email protected]> wrote:
>>>>>> Hi.
>>>>>>
>>>>>> I checked 0.4 by running ant test, okay but one problem.
>>>>>>
>>>>>> In multiprocess-example, at default properties.xml, HSQLDB database
>>>>>> requires
>>>>>> 'org.apache.manifoldcf.hsqldbdatabasepath' property, containing a
>>>>>> relative path:
>>>>>>
>>>>>> <property name="org.apache.manifoldcf.hsqldbdatabasepath" value="."/>
>>>>>>
>>>>>> Now exception is thrown, so we need to add this property.
>>>>>> Should I send a patch? Next?
>>>>>>
>>>>>> Regards,
>>>>>> Shinichiro Abe
>>>>>>
>>>>>> On 2011/12/18, at 2:27, Karl Wright wrote:
>>>>>>
>>>>>>> I've again uploaded an apache-manifoldcf-0.4-incubating package which
>>>>>>> you
>>>>>>> can download from http://people.apache.org/~kwright. If you agree
>>>>>>> this should be released please vote +1.
>>>>>>>
>>>>>>> [Note that the release of this artifact is dependent on the release of
>>>>>>> the three plugin packages upon which it depends. If we need to change
>>>>>>> those packages we'll need to spin a new RC of this one. Also please
>>>>>>> note that I've include SHA1 signatures for the first time, as per
>>>>>>> Jukka's suggestion.]
>>>>>>>
>>>>>>> I apologize again for the earlier hiccup. This one is the real thing,
>>>>>>> though, and passes all tests.
>>>>>>>
>>>>>>> Thanks!
>>>>>>> Karl
>>>>>>
>>>>
>>
>>
>>
>> --
>> Piergiorgio Lucidi
>> http://www.open4dev.com
