headers
subs | 11 Jan 2006 19:30

Why can't node IDs start with a number?


Hi,

According to Node.php, node IDs must match /^[a-zA-Z_][a-zA-Z0-9_]*$/.
Is there a particular reason why they cannot start with a number? I
can't see anywhere that the IDs need to conform to PHP function/method
naming rules, unless I've missed something obvious.

Cheers,
Ben

_______________________________________________
dev mailing list
dev@...
http://lists.binarycloud.com/mailman/listinfo/dev
Permalink | Reply | 1 comment |
headers
Wim Niemans | 19 Dec 2005 16:19

Re: an outsiders view [2] (resent)

>The Conf Class is awesome! Why isn't this incorporated more into
>the Node
>class? Instead of having setParam and getParam why not have a 
>node
>automagically create its own conf namespace so all nodes have an 
>easy
>place to reference;  a uniform interface. In addition with node 
>creation
>the Conf class could be used to autocreate a tree manager so you 
>could
>crawl the tree and pre and post process nodes in this fashion.

Nathan, Node creation and conf are different holons.
A NodeFactory may create Nodes, but doesn't use conf for that.
Crawling a tree in pre- or postOrder mode, or Depth-or BreadthFirst 
mode (I prefer these terms), should be *outside* Node, indeed. I agree 
totally.

>I also think the Event class is great. I just think its a shame 
>that it
>isn't being used more, Nodes should fire events when entering 
>the
>initialization/process/render steps and fire events that other 
>nodes can
>listen for so they can respond without having to know the 
>parent/child
>relationship. This would make Nodes insanely powerful.

I can' agree more. Though the Event system lacks true Object 
registration at the moment.
(Continue reading)

Permalink | Reply | 0 comments |
headers
Wim Niemans | 19 Dec 2005 16:19

Re: an outsiders view

I am repeating my message here, since devlist persists in 'shrubbing' my emails as 
'attachments'.

>one could do: 
>$related_node = $this->getRelatedNode('../someid'); 

Wim, do you have any input here? 

Well, yes. I could have something to say. 

As we discussed, Bas, I've got a version of Node where all  redundant code is removed: 
Caching, Navigating, Creating. 

-: I deleted Caching since it is a bare copy of Smarty Caching. 

-: I've developed a NodeFactory to encapsulate the tree creating  process. Nodes are 
created using a NodeCreator Object. 

-: Nodes are equipped now with a 'pivot', meaning that all  referencing to an UpperNode 
or a Sibling goes through it's Pivot. 
This frees Nodes from traversing, since the pivot is 'visiting' a  Node on preOrder and 
postOrder traversal. Yes, it works clean and  efficient. 

-: To locate a Node in the tree, I don't think one should use the  naming ID's, since BC 
creates a Top node everywhere with the same  name. Alternatives are position (in the 
tree, horizontally) and  level (in the tree, vertically). However the need to do something  
like that is merely a symptom of bad design. 

-: I came to the conclusion that a Node belongs to a Tree and  points to its Top (I'm 
naming that a 'trunk'), and has some  important anchestor (like form and its inputs). It's
(Continue reading)

Permalink | Reply | 7 comments |
headers
Wim Niemans | 19 Dec 2005 12:42

Re: an outsiders view[2]

>The Conf Class is awesome! Why isn't this incorporated more into >the Node
>class? Instead of having setParam and getParam why not have a >node
>automagically create its own conf namespace so all nodes have an >easy
>place to reference;  a uniform interface. In addition with node >creation
>the Conf class could be used to autocreate a tree manager so you >could
>crawl the tree and pre and post process nodes in this fashion.

Nathan, Node creation and conf are different holons.
A NodeFactory may create Nodes, but doesn't use conf for that.
Crawling a tree in pre- or postOrder mode, or Depth- or BreadthFirst mode (I prefer these terms), should be *outside* Node, indeed. I agree totally.

>I also think the Event class is great. I just think its a shame >that it
>isn't being used more, Nodes should fire events when entering >the
>initialization/process/render steps and fire events that other >nodes can
>listen for so they can respond without having to know the >parent/child
>relationship. This would make Nodes insanely powerful.

I can' agree less. Though the Event system lacks true Object registration at the moment.
Otherwise, IMO, a Node should not know it's parent at all.
A Node may controll it's children, obsoleting the need of it's children to give feedback. Well, it's IMO.

>Having Smarty/Templates integrated so closely to a node sucks in >my
>option, if you want to output to a template create a template >node, or
>have a "/template" namespace in Conf. Render should have >settings like
>HTML,SVG,PDF,JPG etc The public interface to a node is way too >big.

Most BC people will tend that Smarty is tightly integrated. Let me tell you from experience: it is *not*. The Smarty API is copied into the Node. Eliminating that copied-API is easy without breaking anything.
The idea of resource-types for Nodes is interesting. Do you have more ideas on that, ie. how to realize such, or examples?

>Speaking of NDFs, I hate XML config files, they are great for >porting
>configs from one environment to another but to work with by hand >is

I do see ndf as a concept, for the moment coded with XML.
I still do use Phing to generate native php arrays.
MyConf has several 'packers'. These support several fileformats,
line ini, txt, php_script(!), sql and on. But the concept remains conf and/or ndf. The 'packers' are just convenience.

>painful. I have patched my Binarycloud install so it runs native >php
>config files, I find it much easier to deal with. My though with >this is
>to create a simple web base interface for handling the >editing/creation of
>config files. In this fashion we could then create resources and >the
>config manager could automagically manage them and validate that >all
>required parameters are filled and valid.

Agree on that. It is a definitive need. Though some will say that it is discussed already and before. Herewith ignoring that the need-to-have-it is existent.

>There is allready an RenderManager but i  am
>not sure you can just add a renderer.

Bas, it would be wonderfull if you may point to RenderManager in some more extend. I'm afraight that the term Render is not clear to vereybody in tha same way. At least to me, that is.....

> But that might change if you need to hook node up in  a
> different way than a simple tree. Not talking about the sepeartion of
> Node and Rendering.

>I would be more then willing to take on this task or help out.

Nodes form a simple Tree. This Tree can be further parameterized, if there are needs. The next step will be a Graph. I think if you go that way, the application of Nodes will be bent as well.

Example: Entities in a database could form a graph with data and relations. Searching the Graph is an application, like: find me the fastest travel connection fromA to B; or: find me the cheapiest supplier for this DVD recorder; or such.

IMO: there is a fundamental difference between an Entity and a Node. That's why Nodes form a Tree and not a Graph. IMO, that is.

Cheers,
wim niemans
_______________________________________________
dev mailing list
dev@...
http://lists.binarycloud.com/mailman/listinfo/dev
Permalink | Reply | 0 comments |
headers
B. Kamer | 15 Dec 2005 13:13
Picon

house keeping

Hi all,

Builds of the r3 and r4 branches are now available from

http://www.binarycloud.com/index.php/Releases/NightlyTarballs

Also the nightly build of http://testsite.binarycloud.com has been  
restored. It is php 4 though (r3) until we install php 5.
_______________________________________________
dev mailing list
dev@...
http://lists.binarycloud.com/mailman/listinfo/dev
Permalink | Reply | 1 comment |
headers
Bas Kamer | 13 Dec 2005 14:07
Picon
Gravatar

new Roadmap

Everybody,

The trac roadmap and timeline should be redefined, especially since  
there is an branch that holds an binarycloud version that runs only  
on php 5. This is a short proposal for such a change. This is -not-  
decided so please comment on it... I feel main development should  
focus on the php5 version now, but it is important that the r3 branch  
is 'closed' in an orderly fashion. You might not agree here, please  
shooot

- We release the current r3 branch as 3.0.0 Final.
- A milestone for 3.0.0 bug fixes should be created (3.0.1?) and any  
open tickets that concern bugs in 3.0.0 Final should go there.
- Any other tickets would then reflect changes that should be applied  
to the r4 branch and new Milestones should be created. Some tickets  
will be closed. This would then make a new Roadmap.

4.0.0RC1 could be released sometime time soon. The idea here is that  
it would be functionality wise mostly the same as the 3.0.0Final  
release. Mainly php4->php5 conversions.

4.0.0RC2 current 3.0.0RC4 tickets

4.0.0RC3 current 3.0.0RC5 tickets

4.0.0 all other tickets

4.1.0 anyone?

_______________________________________________
dev mailing list
dev@...
http://lists.binarycloud.com/mailman/listinfo/dev
Permalink | Reply | 0 comments |
headers
alex black | 12 Dec 2005 22:14

Re: [Fwd: New role requested in the binarycloud project]

Hey all, re the tigris site:

> would be good to close it -- or simply place a redirection for  
> current site.

we should close it, eh?

_a

--

-- 
alex black, founder
the turing studio, inc.

510.666.0074
root@...
http://www.turingstudio.com

2600 10th street, suite 635
berkeley, ca 94710

_______________________________________________
dev mailing list
dev@...
http://lists.binarycloud.com/mailman/listinfo/dev
Permalink | Reply | 4 comments |
headers
alex black | 8 Dec 2005 18:21

test


--

-- 
alex black, founder
the turing studio, inc.

510.666.0074
root@...
http://www.turingstudio.com

2600 10th street, suite 635
berkeley, ca 94710

_______________________________________________
dev mailing list
dev@...
http://lists.binarycloud.com/mailman/listinfo/dev
Permalink | Reply | 1 comment |
headers
Bas Kamer | 3 Dec 2005 13:48
Picon
Gravatar

help! svn r3 & r4

help!

I made a mistake and am not sure how to undo this.

revisions 8635 & 8636 were done to the r3 branch and not to what i  
indended the r4 brach created in 8635.

is there an easy way to delete / remove 8635 & 8636?

or even better is there an easy way to undo these changes in r3 and  
apply them to r4? since 8635 is a copy of r3 that might be possible...

bas
_______________________________________________
dev mailing list
dev@...
http://lists.binarycloud.com/mailman/listinfo/dev
Permalink | Reply | 2 comments |
headers
alex black | 2 Dec 2005 20:14

new wildcard certificate

hi all,

I have just installed a new wildcard certificate on binarycloud.com -  
the renewal was kindly donated by GoDaddy. Thanks GoDaddy!

All seems to be running well, let me know if you have any problems  
with SVN etc.

best,

_alex

--

-- 
alex black, founder
the turing studio, inc.

510.666.0074
root@...
http://www.turingstudio.com

2600 10th street, suite 635
berkeley, ca 94710

_______________________________________________
dev mailing list
dev@...
http://lists.binarycloud.com/mailman/listinfo/dev
Permalink | Reply | 1 comment |
headers
alex black | 2 Dec 2005 00:47

Administration on the binarycloud.com box

hey all,

I'll be futzing with the wildcard cert over the next day or two, also  
I will be updating the box for security reasons which may (or may  
not) break stuff. I'll send notifications to the list.

best,

_alex

--

-- 
alex black, founder
the turing studio, inc.

510.666.0074
root@...
http://www.turingstudio.com

2600 10th street, suite 635
berkeley, ca 94710

_______________________________________________
dev mailing list
dev@...
http://lists.binarycloud.com/mailman/listinfo/dev
Permalink | Reply | 0 comments |

Gmane