Ok since our last vote wasn't successful I'm restarting it (since
we've had lots of reasons given the opinion of each of us may have
changed). Please note that we also need to be quick since 1.6 release
is approaching fast and this is going to take some time to implement.
Proposal:
* Replace 1[.1]* syntax with =[=]* syntax
* Force closing = with any number of equal signs (including none).
** Rationale 1: this is what Creole suggest. See http://www.wikicreole.org/wiki/Creole1.0#section-Creole1.0-Headings
** Rationale 2: makes it easier for lazy people
* Allow wiki syntax in headers (we need that to auto generate them
using velocity syntax for example). Creole considers this optional so
we're good.
* White space allowed before the left =
* Any characters after the closing = signs are considered to be part
of the paragraph after the header. Note: I don't know what Creole
suggests for this. Note also that I'll have to check with the wiki
model parser what is easier to do. The other alternative is to simply
ignore chars there.
* Default (in wysiwyg editor for ex) is to generate closed headers
(with matching = signs on the right)
Here's my +1 to all.
Thanks
-Vincent
Hi, devs.
I committed first version of XWQL (or XQL) language to our sandbox
(xwiki-query/xwql) for review.
I worked ~2 last weeks for it.
install info is in xwiki-query/README.
Currently it works according my proposal[1], but it is quite easy to modify.
It is not perfect, but it works and will be improved.
You can test it via QueryManager. (for example in velocity:
$xwiki.xWiki.store.queryManager.createQuery("select doc.fullName from
Document as doc, doc.XWiki.ArticleClass as a where a.title like
'%Launch'", "xwql").execute() )
plan for xwql (for fix in next week):
* use xwiki-bridge instead of xwiki-core in xwql.
(some methods in bridge are needed)
* create query velocity service
accessible like $services.query.xwql('...')
(according Vincent's velocity services proposal [2])
* move xwiki-query maven module from sandbox to platform/components/
* move QueryManager component from xwiki-core to xwiki-query
I would like to integrate xwql for hql in 1.6M2 (8 Sept) and improve it
in RC if needed.
JCRStore:
old JCRStore with some modifications is partly working (save/load
document is ok) on recent Jackrabbit, but missing some new stores and
need redesign.
plan:
* create xwiki-store/jcr maven module in platform/components (at
sandbox first)
it have to depend on xwiki-core (but not reverse). Is it a problem
for platform/components?
* split JCRStore into components. (RepositoryFacory, SessionFactory,
SessionPool, Stores, DAOs)
* jcrstore will contains DefaultStores which will contain only
business logic on data model, and use DAOs components for store/load data.
DefaultStores will be reused in future HibernateStore to remove
duplicating of code.
* get rid of patched graffito-jcr-mapping (~ORM for JCR). it is old and
unsupported.
candidates:
jcrom [3],
jackrabbit-ocm (former graffito),
simple hand-made annotation based mapping.
I'll look at jcrom first.
I already have some code and plan to work on it in sandbox, after finish
with xwql-hql.
I plan to create working version of JCRStore at the end of September.
xwql on JCRStore will work only on JCRv2 implementation.
[1] http://xwiki.markmail.org/message/uar73uglivj6fbbn
[2] http://markmail.org/message/jwcjomrck3yo2kjq
[3] http://code.google.com/p/jcrom/
--
Artem Melentyev
Hello everybody,
Discussing we Vincent we agreed that currently XWiki Core components on
JIRA are very scattered and do not make very much sense. We propose to
reorganize the project and have one component per module in platform
core and web, which would be :
* action
* cache
* component
* containers
* context
* xwiki-core
* observation
* patchservice
* plexus
* rendering
* url
* velocity
* xml
* xmlrpc
* web-standard
* web-exo
* web-gwt
* web-wysiwyg
+1 for that.
A question now I have is : since it will be quite a big refactoring, do
we move all issues ? or only open issues ?
Thanks,
Jerome.
Hi Devs,
>From today onwards I'll be working on integrating the xwiki-webdav plugin in
the xwiki ui. And in the process i will resurrect the foxwiki firefox
extension as well.
By integrating xwiki-webdav on the xwiki ui, we mean to provide the users
the ability to open / edit / save various types of attachments with their
favourite editor.
For an example, we'll provide a "open with word" link next to a ".doc"
attachment in the attachments section of the UI. When the user clicks on
this link it will fire up word and let the user edit the content and save it
back as he/she wishes.
Currently this functionality is available natively on IE (phew) and there
isn't much to be done on that. But for this to work on firefox we're going
to have to resurrect the foxwiki extension. So the tentative workplan would
be as bellow:
* Make foxwiki work with firefox 3 - (Estimated : 2 days - Already started
working on this)
* Add the functionality to load the corresponding application (to edit the
attachment) via foxwiki - (Estimated : 3 days ?)
* Provide a common javascript API (javascript.vm ?) that can be used to
launch an application from pages - (Estimated : ?)
* Modify attachments.vm to automatically include the links for editing
attachments.
These are the goals I currently have in my mind and i'm still bit shaky
about some topics myself. If any of you have sugesstions / comments /
thoughts please let me know.
Thanks.
- Asiri
Hi XWikiers,
Those last months we've discussed a lot about precise elements of the
new XWiki 2.0 syntax :
- usage of 2 chars for **bold** and [[links]]
- {{macro/}} syntax
- Linebreaks on carriage returns
- Linking to attachments/Insert link to image
- |table| syntax
- styling syntax
- emoticons
So far the idea was to keep the xwiki/2.0 syntax as close as possible
from xwiki/1.0.
IMHO we've reached a point where xwiki/2.0 is so different from
xwiki/1.0 that it could be the occasion to go further and replace our
few radeox leftovers by their widespread equivalent.
That's the first thing I'd like to discuss, do we all agree to make
other substantial modifications to xwiki/2.0 ?
Here's my +1.
Being completely in line with Creole [1] or WikiSyntaxCommon [2] is
not an option since we've already made some incompatible choices (like
linebreaks on carriage returns) and it would imply a big amount of
work now that the new rendering is approaching a stable state. That's
the reason why this proposal is about 2 basic (and important) wiki
syntax elements, those 2 modifications would make xwiki syntax
familiar to wikiPowerUsers.
1) Usage of // instead of ~~ for italic
Creole rationale : A slash (/) looks like slanted italics, so it is
intuitive and thus easier to remember.
See http://www.wikicreole.org/wiki/BoldAndItalicsReasoning
Note : on all the keyboard mappings I know slashes are more easily
reachable than ~.
Here's my +1.
2) Usage of equal signs instead of 1./1.1/etc for Headings
Creole rationale : Using equal signs (=) is the most popular wiki
header markup. Since there are more equal signs for smaller headers,
subheaders will become more indented making it easier to get a visual
overview from the markup alone.
See http://www.wikicreole.org/wiki/HeadingsReasoning
Note : We've experienced problems in the past with the 1/1.1 syntax
(bad document section computation, problems with table cells starting
with 1, etc) but those problems are addressed in the new rendering.
Anyway I think users are more likely to unintentionaly start a line
with a 1 than a = (like in "1 is more than 2" in the middle of a
paragraph).
What I personnaly like with this syntax is that it makes sections very
visible when editing a page in wiki mode.
About unclosed titles :
=== Also level 3
=== Also level 3 =
=== Also level 3 ==
a) Don't interpret them
+0.
b) Interpret them but make closed title our default (generated by wysiwyg)
+1, as I wrote above I think that having clear sections in a page
helps to navigate in edit mode.
c) Interpret them and make it our default
-0.
[1] : http://www.wikicreole.org/wiki/Creole1.0
[2] : http://code.google.com/p/wikimodel/wiki/WikiSyntaxCommon
Thanks,
JV.
Hello,
I would like to write a document containing some javascript functions... For
example:
in "MySpace.MyJSDoc":
{pre}
print('This is my javascript');
{/pre}
Then, instead of using Velocity #includeMacros, I would like to use it with
the classical external javascript link "<script src=""/> and take advantage
of usual JS caching features of web browsers in order not to load this JS
everytime... Thus, keeping a way of editing my javascript online...
In the middle of a page, I would like to put something like:
... Some wiki content to render...
<script type="text/javascript"
src="$xwiki.getURL('MySpace.MyJSDoc')?xpage=plain"/>
... Some wiki content to render...
(I use "xpage=plain" to retrieve only the JS content...)
But it appears that web browsers don't wait until the whole page is received
and make a nested call to the XWiki engine immediately after reading the
<script src=""> and the receiving of the remaining part of the page is
simply stopped at this stage (XWiki seems to trash the remaining part of the
page).
So my questions:
is my idea totally stupid?
if not, do you have an idea to make it possible or at least, at first sight,
do you see what happens exactly to give me some clues before going deeper
with debuggers?
best regards
Pascal
Hello,
I've tried building platform and XE from trunk under Eclipse 3.4(Ganymede)
with last M2Eclipse (Maven embedded 2.1.0 with m2eclipse 0.9.5) which brings
lots of new features but also some new problems...
I've checked out full platform trunk and tried to use the "nested modules
management" of m2eclipse in order to have something clean... so check out
using Subclipse and then activated Maven dependency management, Workspace
resolution and Nested Modules...
Due to some new features of both maven and m2eclipse and due to new nested
modules, there are some problems building everything from scratch...
Here are my short remarks (I'm only maven and eclipse user so this is what I
can seem with an external point of view):
- the famous out-of-memory still exists when building aspects so still needs
to creates a specific Run target with the now famous "-XmxLOTS_OF_MEMORY"
- now in Maven, compiled classes are separated: main classes goes in
target/classes and test classes goes in target/test-classes (which is not so
bad)
- In XWiki platform, we have now up to 3 levels of nested modules and
apparently the maven inheritance is not well managed by m2eclipse. By
default, maven tests are not run using parent compiled classes and I had to
explicitely trigger "resolve Workspace artifact" in my specific Maven Run
target not to have "no class def found exception for ...XWikiContext" when
running maven:test.
- the "resolve Workspace artifact" is sometimes not good: for web/standard,
it tries to copy the file xwiki-core/target/classes but as it is a
directory, it fails to copy it :)...
- Moreover, m2eclipse seems to run maven tests using target/classes from
parent or locally dependent maven modules but not target/test-classes. And
in XWiki, xwiki-rendering/xwiki-rendering-tests, for example, uses classes
from xwiki-core tests classes. I haven't found a clever solution for this.
- I can also see some code errors in Eclipse in web/gwt because Eclipse trie
to resolve some deprecated functions from xwiki-core trunk... not a problem
when compiling with maven
One other question about something I remark:
when compiling, I see this tens of times:
url = http://maven.xwiki.org/externals
Downloading:
http://maven.xwiki.org/externals/groovy/groovy-all-1.0-jsr/06/groovy-all-1.…
url = http://maven.xwiki.org/releases
Downloading:
http://maven.xwiki.org/releases/groovy/groovy-all-1.0-jsr/06/groovy-all-1.0…
url = http://repo1.maven.org/maven2
Downloading:
http://repo1.maven.org/maven2/groovy/groovy-all-1.0-jsr/06/groovy-all-1.0-j…
Apparently, it doesn't see it has already downloaded it... Do you know why?
So finally, compiling module per module triggering some options is OK but
compiling from scratch implies skipping tests...
Pascal
Summary :
1) Usage of // instead of ~~ for italic
Three +1, two 0, no -1.
2) Usage of equal signs instead of 1./1.1/etc for Headings
Three +1, two 0, no -1 for this change.
Three +1, two 0, no -1 for choice a) : consider a line as a heading
only if there is the same number of equal signs at the beginning and
at the end of the line.
Thanks,
JV.
On Fri, Sep 12, 2008 at 5:29 PM, Jean-Vincent Drean <jv(a)xwiki.com> wrote:
> Hi XWikiers,
>
> Those last months we've discussed a lot about precise elements of the
> new XWiki 2.0 syntax :
> - usage of 2 chars for **bold** and [[links]]
> - {{macro/}} syntax
> - Linebreaks on carriage returns
> - Linking to attachments/Insert link to image
> - |table| syntax
> - styling syntax
> - emoticons
>
> So far the idea was to keep the xwiki/2.0 syntax as close as possible
> from xwiki/1.0.
> IMHO we've reached a point where xwiki/2.0 is so different from
> xwiki/1.0 that it could be the occasion to go further and replace our
> few radeox leftovers by their widespread equivalent.
> That's the first thing I'd like to discuss, do we all agree to make
> other substantial modifications to xwiki/2.0 ?
>
> Here's my +1.
>
> Being completely in line with Creole [1] or WikiSyntaxCommon [2] is
> not an option since we've already made some incompatible choices (like
> linebreaks on carriage returns) and it would imply a big amount of
> work now that the new rendering is approaching a stable state. That's
> the reason why this proposal is about 2 basic (and important) wiki
> syntax elements, those 2 modifications would make xwiki syntax
> familiar to wikiPowerUsers.
>
> 1) Usage of // instead of ~~ for italic
>
> Creole rationale : A slash (/) looks like slanted italics, so it is
> intuitive and thus easier to remember.
> See http://www.wikicreole.org/wiki/BoldAndItalicsReasoning
> Note : on all the keyboard mappings I know slashes are more easily
> reachable than ~.
>
> Here's my +1.
>
> 2) Usage of equal signs instead of 1./1.1/etc for Headings
>
> Creole rationale : Using equal signs (=) is the most popular wiki
> header markup. Since there are more equal signs for smaller headers,
> subheaders will become more indented making it easier to get a visual
> overview from the markup alone.
> See http://www.wikicreole.org/wiki/HeadingsReasoning
>
> Note : We've experienced problems in the past with the 1/1.1 syntax
> (bad document section computation, problems with table cells starting
> with 1, etc) but those problems are addressed in the new rendering.
> Anyway I think users are more likely to unintentionaly start a line
> with a 1 than a = (like in "1 is more than 2" in the middle of a
> paragraph).
> What I personnaly like with this syntax is that it makes sections very
> visible when editing a page in wiki mode.
>
> About unclosed titles :
>
> === Also level 3
> === Also level 3 =
> === Also level 3 ==
>
> a) Don't interpret them
>
> +0.
>
> b) Interpret them but make closed title our default (generated by wysiwyg)
>
> +1, as I wrote above I think that having clear sections in a page
> helps to navigate in edit mode.
>
> c) Interpret them and make it our default
>
> -0.
>
> [1] : http://www.wikicreole.org/wiki/Creole1.0
> [2] : http://code.google.com/p/wikimodel/wiki/WikiSyntaxCommon
>
> Thanks,
> JV.
>
--
Jean-Vincent Drean
Hi,
I have a problem in the new rendering. We cannot differentiate between
a URI and a multiwiki link. For example: [[prefix:value]]
This can be considered as a URI or as a space/page (another ambiguity)
pointing to the "prefix" subwiki...
Right now what I've done is to only support mailto: URI and URI with
the format scheme://value.
Any other prefix:value is considered a multiwiki link.
Ideally we should have a better format so that there's no ambiguity.
WDYT? (I don't have a good proposal right now).
Thanks
-Vincent