Hai...
I would like to apologize that I haven't communicate well during last few
weeks.
I need some feedback regarding the document type manager demo
http://gsoc.myxwiki.org/xwiki/bin/view/DocTypeManagerDemo/ its still in the
initial stage.Its just provide the basic sketch of the UI. Currently I'm
creating the "Description" pannel. I'm trying to load the details of the
documents of the left side in to the description form as a start and to
create a class using the description pannel. But I'm still not clear about
how to list the document types in the left side tree. Currently it loads the
list of classes relevant to specific spaces.Is it ok to continue wiith the
current UI. Hoping to catch up my work in next few weeks.
Thank you.
Hi,
We need to start working on the new module to install wiki pages,
components, UI extensions, etc.
I propose to call it Extension Manager instead of the Application
Manager name we had used till now.
If ok I'll rename the existing page at
http://dev.xwiki.org/xwiki/bin/view/Design/ApplicationManager
Thanks
-Vincent
Hi devs,
In velocity scripting and groovy we have the binding "context"
containing com.xpn.xwiki.api.Context (which should change at some
point when we will have the new model, anyway the point is that it
contains a XWiki custom context).
The problem is that "context" binding is reserved in JSR 223 (which
sounds very restrictive to me since there is no reason for that
binding to be technically reserved) so we can't have our own context
binding in all JSR223 based macros which mean all the macro executing
script (Velocity macro is still using old api but at some point it
will use standard ScriptContext). I managed to do a ugly hack for
groovy macro to have context being com.xpn.xwiki.api.Context but we
need to find a generic rule about that.
I can see several general rules:
1) change "context" binding for something like "xcontext": that's the
cleaner easier way from JSR223 POV but it means breaking all existing
script using "context" binding in 2.0 syntax. This is not an issue
jruby, jython etc... which are not yet used but it makes the
ocnversion between 1.0 syntax and 2.0 syntax harder for Velocity and
Groovy scripting (note that it's not difficult for the 1.0->2.0
converter to change $context to $xcontext)
2) make com.xpn.api.Context implements ScriptContext interface: this
means we need to make the ScriptContext a XWiki context behind the
scene for it to be used in both needs.
3) we reimplement our own JSR223 ScriptEngine and don't use the
standard to be able to put anything we want in the "context" binding:
this could mean lot's of work and no one could just put groovy (or any
other engine embedding JSR223 bridge) jar in lib/ folder and have
groovy scripting in XWiki anymore which is the goal of the generic
{{script}} macro
I'm -1 for 3), it would make using JSR223 useless.
Now between 1) and 2) i'm not sure, I'm 1) for future languages
support and even for Groovy but it's a very big move for Velocity.
There is also the
4) we do 1) by default and we introduce the "xcontext" binding in
Velocity and deprecate the "context" binding
If think the best is 4) for now.
WDYT ?
--
Thomas Mortagne
Hi,
I think the new search results page (the main search one not the
lucene one) has some limitations that could be improved:
- it doesn't scale with large number of documents (if you do a search
that returns a lot of documents)
- it doesn't allow refining your search easily
- the information presented take up more vertical space than needed
(currently 3 lines instead of one) and thus less results can be
presented
- it could be reused by other pages which require listing documents
such as the SpaceIndex which lists all pages in a given space. It's
actually currently reused but as a consequence the SpaceIndex page
isn't nice at all and doesn't scale.
Would it make sense to:
- have a livetable
- display all docs by default when you arrive on the search page (ie
no filter set)
- make the current search field a field bound to the livetable and
that filters that table when text is typed inside
- remove the need for the spaces combo box since it would be a filter
field in the livetable already
- same for the wiki field
Pros:
- all the limitations listed above are resolved
- the search becomes live (that's a + compared to other solutions
since I haven't seen other apps do this)
- it unifies the way we use livetable across XE
- if we do this we could also modify the AllDocs page to have the
global search field. Basically wherever we display documents we could
also include this search field.
Main Question:
- would the live search be costly? With Lucene search probably not,
with the DB search, it could (for ex if document content isn't indexed).
WDYT?
Thanks
-Vincent
Hi,
Need 1: Be able to prevent some macros from execution in some
contexts. For example in the comments field prevent usage of script
and html macros
Proposed solution: Use a Rendering Transformation with a high priority
that removes (or that generate an error) those macros (the list of
macros can be retrieved from a config file or from some execution
context)
Need 2: Ensure that renderers generate valid content (eg valid XHTML
for the WYSIWYG editor). For example if the user uses the HTML macro
and puts invalid HTML the WYSIWYG editor should still work
Proposed solution: Use a Rendering Transformation with a low priority
that traverses the XDOM to validate it (for example remove inline
elements that are placed in a standalone context).
WDYT?
Thanks
-Vincent
Hi,
We've already discussed this a long time ago but just want to raise
the topic again and document it. Do we agree that:
* info is used for logs that should be printed by default (default
logging configuration)
* logs that are informational but that shouldn't be printed by default
should be debug logs
* warnings are not critical problems but raised so that the users sees
them and remedies the problem, shown by defaut too
* error are critical problems that cause the application to misbehave
and should most of the time have an associated stack trace
As a consequence I'd like to modify the default log4.properties file
so that info level logs are printed by default (at least for all
org.xwiki.* code).
Here's my +1
Thanks
-Vincent