Contact support. That should not happen.
-- Mike
On 1 Sep 2011, at 14:49 , Will Thompson wrote:
> I recently added several fragment roots to my local MLS, and based on all
> previous experience reindexing, this operation was taking an inordinate
> amount of time. I left it running overnight, a
I recently added several fragment roots to my local MLS, and based on all
previous experience reindexing, this operation was taking an inordinate amount
of time. I left it running overnight, and it was still going this morning at
the same progress level that I left it. So I removed the new fragm
Sorry, missed the 'total disk' part.. ;-P
-Oorspronkelijk bericht-
Van: general-boun...@developer.marklogic.com
[mailto:general-boun...@developer.marklogic.com] Namens Geert Josten
Verzonden: donderdag 1 september 2011 20:28
Aan: General MarkLogic Developer Discussion
Onderwerp: Re: [Mark
Keith,
You are right. That was not the intent, but more of a typo in the code - I
was calling collection() twice, when I thought the first time I was just
defining the collection name! So I was passing the documents returned by the
first collection() into the second one. This IS embarrassing, I
Neil,
It would appear that you are passing the contents of the document to
fn:collection() instead of the URIs of the documents.
- Keith
From: general-boun...@developer.marklogic.com
[mailto:general-boun...@developer.marklogic.com] On Behalf Of Neil
Sent: Thursday, September 01, 2011 1:08 PM
T
Your call to fn:doc returns the entire document which is then passed as the
parameter to fn:collection. I don't think that is what you want. Fn:collection
is looking for a uri not a document and so it complains that the document is
not a valid uri.
Sent from my iPhone
On Sep 1, 2011, at 15:04,
MarkLogic will run fine on top of OS, filesystem or hard drive level
encryption. As we leave control of the filesystem to the OS the encryption will
be transparent to us.
Sent from my iPhone
On Sep 1, 2011, at 13:54, "bek" wrote:
> If my data on disk needs to be encrypted (not hashed), does
Hi,
I am getting a strange error when I try to access some document via a
particular collection.
This works OK, and returns the document I want:
fn:doc("/DOCS/0001012-002-064--YYY-ZZ02_2010-05-17_17-43-14-109-
2.6.xml")
So the problem is not with the URI of that docume
Hi Bek,
Data files themselves should be already encrypted, but it won't make much sense
encrypting the data within the databases itself if it needs to be searchable.
There are still some options though. You could encrypt sensitive character data
leaving xml structure intact, and you could add a
If my data on disk needs to be encrypted (not hashed), does MLS
support/thrive/tolerate a product like PGP Total Disk Encryption? I don't see
any references outside of different comments on hashing data from xdmp
functions.
Cheers
bek
___
General mai
Thanks, Mike
On 08/31/2011 11:20 PM, Michael Blakeley wrote:
> I think it's better to order by empty-sequence:
>
>for $x in cts:search(...)
>order by ()
>return $x
>
> Using the profiler in cq, I can measure a small but consistent difference
> between this and 'order by cts:score($x)'
Once again I am bailed out with everyone's help in this mail list.
Thanks!
gary
-Original Message-
From: general-boun...@developer.marklogic.com
[mailto:general-boun...@developer.marklogic.com] On Behalf Of Michael
Blakeley
Sent: Wednesday, August 31, 2011 5:34 PM
To: General MarkLogic De
Michael Blakeley wrote:
Hi,
> The final XPath step '.../dateline' reorders the results into
> node order, as specified by the W3C. A sequence of constructed
> nodes has indeterminate node order, so you will see
> unpredictable results.
More precisely, nodes from different trees have impleme
13 matches
Mail list logo
