On Feb 16, 2010, at 11:33 PM, Will Scheidegger wrote:

> 
> Well, it's surely faster... but I don't know about "cleaner". I personally 
> don't like keeping data in a redundant form unless I'm really forced to by 
> performance issues for instance.

Understandable. If a|b|c work for you and your projects, no reason to dig 
further indeed :) I just think (correct me if i'm wrong) that (d) would be a 
more general/generic solution (i.e independent of the property name(s) used to 
store the text, for instance) - that's what I meant with "cleaner". Data is 
redundant, but could also be cleaned or re-constructed easily. (with my good 
old friend the visitor out)

(additionally, this is the kind of thing I'd store in a namespaced property, to 
"mark" it as a "system" or "generated" property, as opposed to your regular 
content properties such as text, i.e mgnl:links, or something of the sort)

-g


> 
> -will
> 
> On 16.02.2010, at 23:23, Grégory Joseph wrote:
> 
>> 
>> d) post process properties on saving, store found links in a specific, 
>> multi-value, property (at page level maybe, instead of paragraph level), and 
>> query that instead. (cleaner, faster, and you don't have to know where/how 
>> links are encoded in your texts)
>> 
>> Not that it's trivial to implement at the moment, but it's something I'd 
>> like to see happening at some point ... :)
>> 
>> Cheers,
>> 
>> -g
>> 
>> 
>> On Feb 16, 2010, at 8:55 PM, Will Scheidegger wrote:
>> 
>>> 
>>> Sure. 
>>> 
>>> How is that link stored
>>> a) in a nodeData by itself using a "link" control
>>> b) in a nodeData by itself using a "uuidLink" control
>>> c) somewhere in the content created by a "fckEdit" control
>>> 
>>> What you do is you let the target page query the whole website tree for 
>>> nodes pointing to itself. 
>>> 
>>> For case a) this could look like so:
>>> - target page at /path/to/targetpage
>>> - nodeData containing the link called "link"
>>> - query: "SELECT * FROM nt:base WHERE link = '/path/to/targetpage'
>>> 
>>> For case b) the query could look like this:
>>> - target page uuid is abcd-1234-efgh-5678-ij90
>>> - nodeData containing the uuid of the linked page called "linkedPageUUID"
>>> - query: "SELECT * FROM nt:base WHERE linkedPageUUID = 
>>> 'abcd-1234-efgh-5678-ij90'
>>> 
>>> For case c) fckEdit stored both the handle (path) and the uuid in the text, 
>>> e.g. like this:
>>> <a 
>>> href="${link:{uuid:{f0ac0fe9-62cd-4892-ada2-33d9cd43fa5a},repository:{website},handle:{/path/to/targetpage},nodeData:{},extension:{html}}}">Click
>>>  here</a>
>>> If you decide you should want to search for the uuid (what I would do), 
>>> things could work like this:
>>> - target page uuid is abcd-1234-efgh-5678-ij90
>>> - nodeData containing the html code which somewhere contains the uuid of 
>>> the linked page called "text"
>>> - query: "SELECT * FROM nt:base WHERE CONTAINS(text, 
>>> '%abcd-1234-efgh-5678-ij90%')
>>> 
>>> Cheers,
>>> -will
>>> 
>>> On 16.02.2010, at 19:31, [email protected] wrote:
>>> 
>>>> 
>>>> Assuming i have a paragraph with an internal link to a page, is there any 
>>>> way from the other page to collect all the links pointing to?
>>>> 
>>>> Tx fof any advice
>>>> 
>>>> Luca
>>>> 
>>>> ----------------------------------------------------------------
>>>> For list details see
>>>> http://www.magnolia-cms.com/home/community/mailing-lists.html
>>>> To unsubscribe, E-mail to: <[email protected]>
>>>> ----------------------------------------------------------------
>>>> 
>>> 
>>> 
>>> ----------------------------------------------------------------
>>> For list details see
>>> http://www.magnolia-cms.com/home/community/mailing-lists.html
>>> To unsubscribe, E-mail to: <[email protected]>
>>> ----------------------------------------------------------------
>> 
>> 
>> ----------------------------------------------------------------
>> For list details see
>> http://www.magnolia-cms.com/home/community/mailing-lists.html
>> To unsubscribe, E-mail to: <[email protected]>
>> ----------------------------------------------------------------
>> 
> 
> 
> ----------------------------------------------------------------
> For list details see
> http://www.magnolia-cms.com/home/community/mailing-lists.html
> To unsubscribe, E-mail to: <[email protected]>
> ----------------------------------------------------------------


----------------------------------------------------------------
For list details see
http://www.magnolia-cms.com/home/community/mailing-lists.html
To unsubscribe, E-mail to: <[email protected]>
----------------------------------------------------------------

Reply via email to