IRC Archive for channel #xwiki

Last modified by Vincent Massol on 2012/10/18 19:14

13:15 <mflorea> thanks
13:16 <jstoldt> i stand corrected: i want to but i cannot. ;)
13:17 <mflorea> you just said that the path inside the profile is not reliable. It can change and it seems to be different for LibreOffice so the best is to use an API but I couldn't find that API
13:17 <mflorea> ah, ok, let me do it
13:18 <jstoldt> true enough. i was just trying to suggest a dirty hack which would allow to change it at least for OO.o installations... and if you figure out what it is for LibO, also for that one. still only very dirty hacking...
13:18 <jstoldt> only other way i could imagine would be to provide a patch for OOo/LibO and hope they adapt it, given there really is no API at this time
13:19 <vmassol> sdumitriu1: seems like xpage=attachments isn't correct. For ex:
13:20 <sdumitriu1> That's right
13:20 <sdumitriu1> It should be viewer=attachments
13:21 <vmassol> ok...
13:21 <mflorea> jstoldt: maybe there is an API hidden somewhere. I'll asked them on the forum
13:22 <jstoldt> okay, i just joined #documentfoundation and asked them about the dir and where the information got moved
13:22 <jstoldt> will also inquire about the API
13:23 <jstoldt> just asked both questions. let's see if we get reply
13:25 <mflorea> let me join the channel too
13:29 <evalica> has quit
13:30 <florinciu> has joined #xwiki
13:34 <evalica> has joined #xwiki
13:49 <sabageru> has joined #xwiki
13:54 <+lucaa> guys, I have some trouble with tests
13:56 <+lucaa> the moment I add xwiki-core as a dep to my macro, my tests fail with this error:
13:56 <+lucaa>
13:56 <+lucaa> do you know what might cause it?
14:05 <+tmortagne> lucaa: what is your test trying to do when you get this issue ?
14:05 <+lucaa> wait, I will debug and let you know
14:05 <+lucaa> it's the dashboard test
14:06 <+lucaa> the one and only
14:06 <+tmortagne> lucaa: do I have a easy way to reproduce this ?
14:06 <+lucaa>
14:06 <+lucaa> I think just adding the xwiki-core as a dep to the project
14:06 <+lucaa> e.g. the dashboard
14:06 <+lucaa> dashboard macro
14:06 <+lucaa> and then try to run it
14:07 <+tmortagne> ok
14:07 <+tmortagne> will do
14:08 <+tmortagne> last time i had this issue was because I was playing with classloader but it's pretty specific to extension manager
14:12 <+tmortagne> lucaa: xwiki-core depends on dashboard macro already
14:12 <+lucaa> mmh...
14:12 <+lucaa> I see
14:12 <+tmortagne> maybe that's what makes maven classloader a bit lost
14:13 <+lucaa> but xwiki-core doesn't really depend on dashboard, it just needs to drag it in the distribution...
14:14 <+tmortagne> sure, xwiki-core is not the best example of clean maven project
14:14 <+tmortagne> you can move dashboard dependency from xwiki-core to web if you need to depends on core
14:14 <+lucaa> lemme try that, and see what happens
14:15 <+lucaa> it's gonna be odd because all other macros are in the core
14:16 <+tmortagne> sure but that's not the only solution, you could introduce a new component module for what needs to manipulate xwiki-core directly and use it in your macro like we usually do
14:17 <+lucaa> that's what I do
14:17 <+lucaa> but I get the same issue
14:17 <+lucaa> I got it when I put the dep directly
14:17 <+lucaa> and then I put my code that manipulates core in another module and made macro depend on that module
14:17 <+lucaa> and it does the same
14:19 <+tmortagne> you need a api module and an implementation module otherwise it's useless
14:19 <+tmortagne> macro depends on api module to not transitively depends on xwiki-core
14:20 <+lucaa> ah right
14:20 <+lucaa> but then it's a tiny little bit of overkill
14:22 <+tmortagne> depends of your use case
14:22 <+lucaa> I need skin extensions
14:22 <+tmortagne> it's just a matter of does this new module would make some sense for others
14:22 <+lucaa> it surely makes a lot of sense for other situations as well
14:23 <+lucaa> mmh. wait I have an idea...
14:23 <+lucaa> ah no it's not gonna work :(
14:24 <venkatna> has quit
14:24 <+lucaa> so, I still think it's a bit of overkill to have 2 different modules for skinx wrapper, but we could actually implement the skinx API in the core itself
14:26 <+tmortagne> yes
14:26 <+tmortagne> like we do for other things
14:26 <+tmortagne> or the plugin
14:26 <+tmortagne> since that's where this is really
14:27 <+lucaa> yes, I was thinking about the plugin actually
14:27 <+lucaa> I still need sergiu
14:27 <+lucaa> sergiu's agreement for that
14:27 <+lucaa> sdumitriu1: ?
14:28 <+tmortagne> it basically the same thing than starting to convert skinx from plugin to component concept
14:29 <+lucaa> yes, I had written this wrapper already, for this very purpose, to provide components interface for the skinx
14:30 <+lucaa> I sent it to sergiu and I am waiting for some response...
14:30 <vmassol> nice textual format:
14:31 <lucaa> has quit
14:33 <jstoldt> indeed, vmassol
14:36 <Menso> is there something like #foreach($page in $xwiki.spaces.spacename) ?
14:37 <vmassol> Menso: yes in ^xwiki
14:37 <sdumitriu1> is now known as <sdumitriu>
14:37 <vmassol> $xwiki
14:37 <vmassol> public List<String> getSpaceDocsName(String SpaceName) throws XWikiException
14:37 <vmassol> so $xwiki.getSpaceDocsName($space)
14:43 <arkub> has quit
14:43 <Menso> I'm using this now:
14:43 <Menso> #foreach($page in $xwiki.getSpaceDocsName(GemAppTest))
14:43 <Menso>  * [[$page]]
14:43 <Menso> #end
14:44 <vmassol> "GemAppTest"
14:44 <Menso> check
14:54 <lucaa> has joined #xwiki
15:01 <abusenius> has joined #xwiki
15:17 <+lucaa> tmortagne, guys, I just sent a vote on the list for the implementation of skinx component API, as we've discussed on this channel with tmortagne
15:17 <+lucaa> sdumitriu: please answer
15:18 <@sdumitriu> K
15:31 <abusenius> has quit
15:45 <lucaa> has quit
15:45 <lucaa> has joined #xwiki
16:01 <lucaa1> has joined #xwiki
16:01 <lucaa> has quit
16:09 <sabageru> has quit
16:10 <Menso> Mmm, I can't seem to find the code anywhere anymore, how do I use velocity again to check if a page is being viewed or editted?
16:12 <@sdumitriu> #if ($context.action == 'view')
16:12 <lucaa1> I will ask this again, is there a particular reason for which setParameter is not in the Block interface but in the AbstractBlock implementation?
16:12 <lucaa1> ( vmassol, tmortagne )
16:12 <Menso> thanks sdu
16:14 <vmassol> lucaa1: getParameter too apparently
16:14 <lucaa1> yes, both parameters APIs
16:14 <vmassol> everything related to parameters
16:15 <vmassol> I can't think of a good reason
16:16 <arkub> has joined #xwiki
16:17 <vmassol> I'd say it's an error
16:17 <+tmortagne> probably
16:17 <vmassol> well or rather we didn't have a use case for it yet
16:18 <vmassol> +1 to make it part of the Block interface from me
16:18 <+tmortagne> both, if we had use case we would have seen it ;)
16:18 <lucaa1> I think this is the third time I ask :D
16:18 <+tmortagne> lucaa1: don't remember that
16:19 <lucaa1> so, I open an issue, does anybody else than me want to fix it
16:19 <lucaa1> maybe I always picked christmas or some other holidays to ask about this
16:19 <lucaa1> :)
16:19 <lucaa1> because once was somewhere last week
16:20 <lucaa1> no, it was a 24 dec
16:20 <vmassol> nice timing
16:20 <Menso> When I do something like this: [[Software Producten.$doc.display("OpenDir_SoftProd")]] to make a link to a page it displays the pagename if "OpenDir_SoftProd" is empty
16:20 <lucaa1> which is a bad day to ask that, I agree
16:20 <Menso> is this a bug or am I doing something wrong?
16:21 <vmassol> Menso: don't use display it generates HTML
16:21 <vmassol> $doc.getValue("...")
16:22 <lucaa1> anyone else handling the Block thing, or I should do it?
16:22 <vmassol> lucaa1: go ahead
16:23 <vmassol> (I have too many local changes in the rendering module to handle it right now)
16:23 <Menso> vmassol: issue remains
16:23 <vmassol> Menso: you should test what you get
16:23 <vmassol> print it
16:23 <lucaa1> vmassol: ok
16:26 <lucaa1> getParameters, getParameter, setParameters, setParameter, right?
16:26 <vmassol> yep
16:26 <Menso> vmassol: the problem seems to be when I put it in a link
16:26 <vmassol> lucaa1: make sure you refactor the javadoc too
16:27 <Menso> for example, see
16:27 <vmassol> to use {inheritdoc} and @see
16:27 <lucaa1> yes, I'll handle that
16:27 <vmassol> and change the @since tags
16:27 <vmassol> ok thanks
16:27 <lucaa1> is there any documentation i need to update also? :P
16:27 <lucaa1> I need to change since tags? for what? for the interface?
16:27 <Menso> getValue and Display below 'Functies:' use $doc.display and $doc.getValue and both show nothing correctly, below that I put [[Space.$doc.getValue("somevalue")]] and it shows the title of the page in the link instead of being empty
16:28 <vmassol> lucaa1: you're lucky this time, no, althoug there are always stuff to improve on ;)
16:28 <vmassol> @since for the interface
16:28 <lucaa1> hmm... that's how since works?
16:28 <lucaa1> we change it when we change a function?
16:28 <vmassol> no
16:28 <vmassol> you're creating new apis
16:28 <lucaa1> yes, yes
16:29 <lucaa1> but we change @since tag when I do that?
16:29 <lucaa1> what does since mean?
16:29 <vmassol> you can leave existing @since in AbstractBlock
16:29 <vmassol> s/can/should/
16:31 <lucaa1> wait, we have since tags for each function
16:31 <lucaa1> that's the since tag you mean, right? not the class since tag
16:32 <vmassol> yes method
16:32 <vmassol> I see      * @since 1.7M2 for setParameters
16:32 <lucaa1> yes, I will leave that in the impl
16:33 <vmassol> brb
16:37 <abusenius> has joined #xwiki
16:42 <jvelo> has joined #xwiki
16:42 <jvelo> sdumitriu, Pong
16:43 <@sdumitriu> Ping?
16:43 <Menso> anyone any idea on my earlier question regarding links to empty values?
16:47 <lucaa1> Menso: you make a link to
16:47 <lucaa1> MySpace.$doc.display("something")
16:47 <lucaa1> so when $doc.display("something") is empty, your link goes to
16:47 <lucaa1> MySpace.
16:47 <lucaa1> which means, resolved as a reference, the current page
16:47 <lucaa1> what is the result that you expect?
16:50 <Menso> That makes sense, I was hoping it would not show any link
16:50 <Menso> now I need to check all the values if empty and if not display and otherwise not I suppose? :)
16:50 <@sdumitriu> You do write [[
16:50 <@sdumitriu> You get a link when you do that
16:51 <Menso> yes, but I don't understand why it shows the page title of the current page
16:51 <lucaa1> I think even [[]] displays a link, to current page as well
16:51 <Menso> instead of showing nothing
16:51 <lucaa1> Menso: you cannot have a link with an empty label
16:51 <@sdumitriu> The wiki syntax doesn't know XWiki internals, so it can't decide on its own "ok, this is not a proper link, let's just forget this and go home"
16:51 <@sdumitriu> You link to a document
16:51 <lucaa1> because that's not a link anymore
16:51 <@sdumitriu> The document is always valid
16:51 <lucaa1> there are some defaults
16:52 <Menso> right, I was expecting it to be translated into <a href="foo"></a>
16:52 <@sdumitriu> You can't have a "not a document" document
16:52 <lucaa1> Menso: that does not make sense
16:52 <lucaa1> actually you might be able to force it if you put the label explicitely
16:52 <lucaa1> because a link with only a reference and no label will have an automatically generated label, which is the name of the page it links to
16:52 <lucaa1> so
16:53 <Menso> lucaa1: see everywhere it says 'Haarlem' now I want it to be empty since the value is null and only become a link if the value is not null
16:54 <lucaa1> MySpace.
16:54 <lucaa1> reference is resolved as a ref to current page, because that's how resolver works
16:54 <lucaa1> and then the link label generator will generate the current page name as the label of that link
16:54 <lucaa1> Menso: that's not correct: if you don't want links, there should be no links. having <a href="foo"></a> when in fact you don't want a link is against the principles of HTML
16:54 <lucaa1> which should also be _semantic_ not only visual
16:55 <lucaa1> the information is in the markup as well, not only in what the user can read
16:55 <lucaa1> so, if you don't want links when the value is empty, you'll have to do an if
16:55 <lucaa1> #if(myValueIsNotEmpty)
16:55 <lucaa1>  [[ ... ]]
16:55 <lucaa1> #end
16:55 <Menso> 22 times :D
16:56 <Menso> I'll see if I can maybe make a function in velocity to do it
16:56 <lucaa1> yes, you can write a velocity macro
17:03 <lucaa1> am trimis vot numai ca am vb cu tmortagne pe lista despre asta
17:04 <lucaa1> si vbisem si cu sergiu, si ma gandeam ca na, sa zica acolo toata lumea parerea pentru posteritate
17:04 <lucaa1> insa...
17:04 <lucaa1> cu tmortagne pe IRC, nu pe lista
17:04 <lucaa1> sorry guys very wrong chat
17:04 <sburjan`> :)
17:05 <lucaa1> (it was bound to happen)
17:05 <sburjan`> sooner or later :)
17:09 <jstoldt> lol, i was actually wondering just now X-D
17:09 <jstoldt> anyway, good bye
17:09 <jstoldt> have a nice day
17:09 <jstoldt> has quit
17:14 <npm_> has joined #xwiki
17:18 <npm> has quit
17:18 <npm_> is now known as <npm>
17:19 <jvelo> has quit
17:25 <Menso> using a velocimacro creates <span> stuff in html output which seems to have a rather disruptive effect on my table
17:33 <Menso> e.g. code in the page:
17:33 <Menso> |=CMS|$doc.getValue("CMS_Status")|#createLink("Xwiki" "SoftProd")|#createLink($doc.getValue("CMS_Supplier") "Supplier")|$doc.getValue("CMS_Desc")
17:34 <Menso> output in HTML:
17:34 <Menso> <tr><th scope="row">CMS</th><td>ge&#239;mplementeerd</td><td><span class="wikicreatelink"><a href="view-source:"><span class="wikigeneratedlinkcontent">Xwiki</span></a></span></td></tr><tr><td><span class="wikicreatelink"><a href="view-source:"><span class="
17:35 <lucaa1> what's the problematic span?
17:35 <Menso> for some reason it's throwing in extra </td></tr> stuff
17:35 <lucaa1> ah
17:35 <jvelo> has joined #xwiki
17:35 <lucaa1> you have a newline in your #createLink macro
17:35 <lucaa1> very probably
17:36 <Menso> this is the macro:
17:36 <Menso> #macro ( createLink $xValue $xProp )
17:36 <Menso> #if ($xValue != "")
17:36 <Menso> #if ($xProp == "SoftProd")
17:36 <Menso> [[Software Producten.$xValue]]
17:36 <Menso> #elseif ($xProp == "Supplier")
17:36 <Menso> [[Leveranciers.$xValue]]
17:36 <Menso> #end
17:36 <Menso> #end
17:36 <Menso> #end
17:36 <lucaa1> when macro is executed, it adds a new line in the table
17:36 <lucaa1> yes, after each link you have a newline
17:37 <lucaa1> #if ($xValue != "") #if ($xProp == "SoftProd") [[Software Producten.$xValue]] #elseif ($xProp == "Supplier") [[Leveranciers.$xValue]] #end #end
17:37 <lucaa1> try this
17:37 <lucaa1> it will do the job ;)
17:37 <lucaa1> when you have a newline, it will mean end of table line
17:37 <lucaa1> that's why the </td></tr>
17:38 <Menso> so much for readable code :)
17:38 <Menso> but you are right, it fixed it :D
17:38 <lucaa1> there are ways to make it more readable than that
17:38 <lucaa1> but I took the fast way
17:39 <vmassol> Menso: if you want readable, you can use a filter, see
17:39 <vmassol>
17:40 <Menso> ah nice, thanks vmassol and lucaa1 :-)
17:40 <Menso> I just need to get used to stuff like this breaking code
17:41 <Menso> or rather, newlines introducing </td></tr> in the output
17:41 <vmassol> Menso: more precisely you can use the html filter and use $sp and $nl to specify spaces and newlines
17:41 <vmassol> {{velocity filter="html"}}....
17:48 <sdumitriu> has quit
17:48 <Menso> contains only works for lists?
17:53 <sburjan`> are we sure we want such a page on .org ?
17:53 <lucaa1> sburjan`: there's a mail about it on the users list
17:53 <lucaa1> I think we sorta agreed :)
17:54 <vmassol> sburjan`: I see value for the community to know strenghts and weaknesses of xwiki vs other solutions
17:54 <vmassol> don't you think so?
17:55 <Menso> fixed typo
17:56 <sburjan`> vmassol: I agree to that, it's just that IMO competitors is a little harsh as a word to use.
17:56 <vmassol> yes we could replace with another word
17:57 <vmassol> XWiki Alternatives, etc
17:57 <vmassol> you can propose one
17:58 <sburjan`> yes, Alternative is the best I can think of. Competitors leaves the impression we want to crush those products and we see them as a threat. This is how I interpret it
18:02 <Menso> It might turn potential xwiki customers to your competitors or 'alternatives'
18:03 <sburjan`> yes, also true
18:03 <vmassol> yes but we shouldn't be afraid of this
18:03 <vmassol> on the contrary
18:04 <sburjan`> well this depends on personal perception how safe to we feel on the market
18:05 <vmassol> if we're bad in a domain, hiding it is not going to help, instead we'll just need to improve on that domain, and making it visible will help us address it
18:05 <Menso> I think it's very important to explain the difference.. the application I'm making for NOiV now I could never have made in Confluence for example
18:06 <vmassol> yes good explanation is key
18:06 <Menso> But if you're not interested in that, Confluence might look a lot easier to use and people might move to that instead
18:06 <vmassol> we want xwiki to be easier to use than confluence so we also need to know what we're lacking in this domain if that's the case
18:07 <evalica> has quit
18:08 <Menso> I've got XWiki and Confluence both deployed at customers, I think the big difference is in quality control atm
18:08 <sburjan`> Menso: what do you mean ?
18:09 <Menso> With new XWiki releases there's often also some new bugs in the system
18:09 <Menso> A lot of the time those are quickly fixed in a new version but for the customer that means they need to upgrade to this new version again, etc
18:10 <sburjan`> I understand
18:10 <Menso> With confluence, we never have this, you turn it on and can walk out the door
18:10 <vmassol> Menso: one thing we're working on is to ease upgrades so at least this will mitigate the issue
18:11 <Menso> Then again, with confluence, we can't build applications or customize it for customers, etc :)
18:11 <vmassol> note that it's much easier for confliuence
18:11 <vmassol> since you can't modify base pages
18:11 <vmassol> so there's no merge to be done
18:11 <Menso> vmassol: true, confluence upgrade is extract to /usr/local/confluence<version/, turn off old one, turn on new one, done
18:12 <Menso> ah well, it's hard to compare, they are two different tools for different purposes I suppose
18:13 <vmassol> indeed
18:14 <vmassol> also confluence is mostly for dev teams (even though it can be used by non dev teams)
18:17 <sburjan`> vmassol: I'm proposing "Other Alternatives to XWiki" instead of competitors or rivals
18:18 <vmassol> sburjan`: you could propose it to gregory
18:18 <vmassol> or propose it on that page
18:18 <sburjan`> or, proposing to G
18:20 <sburjan`> sorry for missing the Thread on userlist and opening the discussion again
18:20 <vmassol> or on the list indeed
18:20 <vmassol> best place
18:22 <sburjan`> problem is that I got until 5 minutes ago the userlist on Digest mode... so I don't see the votes. I only see the mail Gregory send and the one you replied. No +1
18:24 <sburjan`> but I will talk to G
18:27 <vmassol> mflorea: ping
18:27 <vmassol> we'll need to talk about how you'd like to handle transformations in the wysiwyg editor
18:28 <vmassol> (knowing that a transformation can modify any block, and add blocks too)
18:29 <vmassol> so for example you might need to have a changeset id to know that several modifications come from the same transformation
18:29 <mflorea> well, same as with macros. I need to detect them and make the output readonly
18:29 <vmassol> well you also need to offer the ability to change the original text
18:30 <mflorea> you mean edit a transformation that isn't a macro?
18:30 <vmassol> for example imagine you have emoticon support
18:30 <vmassol> by using a drop down list of emoticons
18:30 <vmassol> when the user clicks on an emoticon image it outputs for ex : )
18:30 <vmassol> this is transformed
18:30 <vmassol> but the user needs to be able to edit it to change the emoticon
18:31 <vmassol> (to : ( for ex)
18:31 <vmassol> macros or not macro, we need a generic support for all transformations I think
18:31 <vmassol> and special handling for well-known transformations
18:31 <mflorea> the difference is that macros have parameters
18:32 <vmassol> with the changes I' m mkaing all tx are the same
18:32 <vmassol> the origianl blocks are output and the HTML renderer will use the xml serialized to serailize them
18:32 <mflorea> in this case you needs to make the macro markers usable for all transformations
18:32 <vmassol> in a generic way
18:33 <vmassol> there are no longer macro markers
18:33 <mflorea> how do you mark a transformation in the HTML?
18:33 <vmassol> but I think we still need a transformation id
18:33 <vmassol> the changeset id I was referring to
18:33 <Menso> vmassol: confluence mostly for dev teams? I have succes with it mostly with companies without any technical people whatsoever :-)
18:33 <+tmortagne> mflorea: you find it the same way that current macro markers
18:33 <+tmortagne> but in a more generic format
18:34 <mflorea> tmortagne: (07:33:06 PM) vmassol: there are no longer macro markers
18:34 <+tmortagne> that doe snot mean there is not marker
18:34 <+tmortagne> that means the markers are more generic
18:34 <mflorea> what do you mean by "changeset id"?
18:34 <+tmortagne> instead of macro information you find transformation informations
18:34 <vmassol> this changeset id is just an idea I've had (not discussed with tmortagne yet)
18:35 <vmassol> but I think you might need it
18:35 <vmassol> imagine a tx that replaces some blocks in some location and adds some others in other places in the XDOM
18:35 <mflorea> but why not using xml comments, just make the macro syntax more general
18:35 <vmassol> you'll get markers for each change
18:35 <vmassol> but you need to know that it's part of the same tx
18:36 <mflorea> sure, but that can be part of the marker syntax
18:36 <vmassol> s/need/might need/
18:36 <+tmortagne> mflorea: that's the plan
18:36 <vmassol> yes the changeset id is going to be output in the marker syntax
18:36 <+tmortagne> again you will have a transformation marker
18:36 <+tmortagne> i dn't know about this changeset id
18:37 <+tmortagne> mflorea: for now what i see is that you will get a block marker instead of a macroblock marker basically
18:37 <+tmortagne> just a different syntax from which you can unserialize a Block
18:37 <mflorea> so, as long as I can detect the transformation by looking at the HTML and I have the transformation source in the HTML then I can add generic support in the wysiwyg editor
18:38 <+tmortagne> like now you unserailize a macro block basically
18:38 <vmassol> mflorea: it's more complex
18:38 <vmassol> I think
18:38 <vmassol> again imagine a tx that does 2 things
18:38 <vmassol> replace some blocks + adds some other at the end of the doc
18:39 <vmassol> when you double click on the added blocks
18:39 <vmassol> you need to find the original content
18:39 <vmassol> which is split
18:39 <+tmortagne> vmassol: not really
18:39 <vmassol> unless we repeat it for each marker
18:40 <+tmortagne> it depends of the transformation
18:41 <mflorea> vmassol: a transformation must have an "insertion point" i.e. the place where its source appears in the wiki syntax. This "insertion point" can be marked in the HTML and the rest of the "sub"-transformation markers must refer to this place using an id as you suggested
18:41 <+tmortagne> the important part is that the WYSIWYG can know what content is generated
18:42 <+tmortagne> then there is kind of transformations like macro it know how to handler is a specific way
18:42 <+tmortagne> but for some other it will just be readonly content
18:42 <+tmortagne> it's a case by case basis i think
18:43 <vmassol> tmortagne: well the user should always be able to change the original content
18:43 <vmassol> so when you click on "sub" marker (as mflorea calls them) you still need to be able to find the original content to edit
18:43 <+tmortagne> vmassol: you can't do that in a generic way
18:44 <vmassol> or each marker repeats the whole original content
18:44 <+tmortagne> every transformations are differents
18:44 <mflorea> sure, I'll find the marker for the insertion point and take the source from it
18:44 <vmassol> tmortagne: for me it's just about having a changeset id
18:44 <vmassol> you could use it this way in XDOM: block.startChangeset(String id), endChangeset(id)
18:45 <vmassol> then when you add, replace, etc it saves the "history" with the changeset id
18:45 <vmassol> and when we output XHTML the id is used in the markers
18:45 <vmassol> (haven't thought it through yet though still need some thinking)
18:46 <+tmortagne> i don't see how the WYSIWYF is supposed to do with your id
18:46 <+tmortagne> s/how/what/
18:46 <florinciu> has quit
18:47 <vmassol> for ex,show all parts modified by the same tx in the same color, allow to edit the same original text when you click on any marker with the same changeset it
18:48 <mflorea> when the editor find a transformation marker it can use the id to lookup the "insertion point" and take the source of the transformation. Basically the id is useful only for editing the transformation
18:48 <mflorea> well, yes, for styling too
18:51 <mflorea> guys, I have to go
18:51 <vmassol> mflorea: yep me too, let's continue tomorrow
18:54 <mflorea> has quit
18:59 <sburjan`> vmassol: email sent to users
18:59 <sburjan`> reply
19:18 <Menso> is there an easy way to 'clone' properties from a class, in case a user wants to set two different values to it?
19:19 <Menso> for example property cms_product should be 'joomla' and cms_status 'phased out' while another cms_product should be 'xwiki' and cms_status be 'implemented'
19:25 <tmortagne> has quit
19:26 <Menso> I could use multiple select to solve issue of selecting multiple options but then the status would be unclear for each of the selected options
19:45 <sdumitriu> has joined #xwiki
19:55 <lucaa1> guys, I want to commit XE-787
19:55 <lucaa1> should I commit it on 2.7 as well?
19:57 <@sdumitriu> +1
19:59 <sburjan`> sdumitriu: could you vote on the userlist regarding : [xwiki-users] [request] Comparing XWiki to other solutions
19:59 <@sdumitriu> It's not a vote
20:00 <sburjan`> but I proposed something on that thread ...
20:00 <sburjan`> didn't wanted to start a new thread
20:00 <@sdumitriu> K
20:03 <jvelo> has quit
20:04 <lucaa1> ah wait
20:04 <lucaa1> it's not XE
20:04 <lucaa1> it's XAADMINISTRATIO
20:04 <lucaa1> N
20:10 <sburjan`> what's the diff between XE/Administration and XAADMINISTRATION ?
20:11 <lucaa1> I have no clue... I would say the the XE/Administration is a left over
20:11 <lucaa1> from the time XAADMINISTRATION was not a standalone app, but part of the XE
20:12 <sburjan`> but now it's part of XE, right ?
20:12 <lucaa1> no, it's an independent application, which is bundled in XE
20:12 <lucaa1> but for example, on a clean platform (the installed war) with no pages, you can decide to install only the administration application
20:13 <sburjan`> this was my next question. But I guess installing Admin App doesnt install the actual appz, right ?
20:13 <lucaa1> what appz?
20:13 <lucaa1> the sources are here: it has its own release cycle, etc
20:13 <sburjan`> example : invitation, Import/Export, OpenOffice Server, etc
20:14 <lucaa1> Import/Export is part of the administration application
20:14 <lucaa1> openoffice server is not an application, is just a configuration section
20:14 <lucaa1> which I don't remember if it comes by def with admin app or not
20:15 <lucaa1> no, it seems it's not
20:16 <sburjan`> so XE is pretty complex as a XAR. a lot of bundled stuff within it
20:16 <lucaa1> yes, there are very little things which are "pure XE" if you wanna call it like that. It bundles a lot of applications, though (panels, scheduler, watchlist iirc, etc)
20:16 <lucaa1> annotations as well
20:17 <sburjan`> yeah ...
20:37 <arkub> has quit
20:41 <lucaa1> sdumitriu: is XAADMINISTRATION-193  incompatible with 2.7?
20:41 <@sdumitriu> Yes
20:41 <DrLou> has joined #xwiki
20:41 <lucaa1> (because it's also fixed on XAADMINISTRATION 1.45 and I was wondering, if I want to commit XAADMINISTRATION-194 for 2.7, I need to branch, right?)
20:41 <lucaa1> this is gonna be fuuuun
20:42 <lucaa1> to branch 1.44
20:42 <@sdumitriu> Yes
20:42 <lucaa1> wow fun
20:44 <DrLou> Hey XWiki friends: can anyone recommend anXWiki app/module for email campaigns/CRMs?
21:12 <lucaa1> how do I commit the change of dep for administration application in 2.7?
21:12 <lucaa1> I mean, I make an issue? what commit log I put?
21:12 <lucaa1> do I put the XAADMINISTRATION application issues?
21:13 <lucaa1> s/issues/issue
21:13 <@sdumitriu> You edit enterprise/branches/x-e-2.7/pom.xml
21:13 <lucaa1> in the commit log I mean
21:13 <@sdumitriu> Update the version
21:13 <lucaa1> yes, i know but what do I commit?
21:13 <@sdumitriu> And commit with:
21:13 <@sdumitriu> [misc] Updating dependency versions
21:13 <lucaa1> I don't say what for, nothing?
21:13 <lucaa1> ok
21:31 <@sdumitriu> lucaa1: Did you manually copy the branch?
21:31 <lucaa1> I did svn copy
21:31 <@sdumitriu> You should have used the maven release plugin for that
21:32 <lucaa1> :(
21:32 <lucaa1> it's the first time I do this...
21:32 <@sdumitriu> Don't worry, it's no big deal, just more work for you
21:32 <lucaa1> more work?
21:32 <@sdumitriu> The end result is the same
21:32 <lucaa1> to update the current dev version, the scm tags and? what else?
21:33 <@sdumitriu> That's all
21:33 <lucaa1> ok, I did those
21:33 <lucaa1> I hope I did it right
21:33 <@sdumitriu> It's good
21:40 <lucaa1> done
21:40 <lucaa1> it should be done now
21:41 <lucaa1> how do I close the issue now? I create XAADMINISTRATION-1.44.1 on jira?
21:54 <@sdumitriu> Yes
22:02 <lucaa1> ok done done now
22:09 <jvdrean> has quit
22:31 <jvelo> has joined #xwiki
22:39 <jvelo> has quit
22:47 <lucaa1> has quit
23:24 <Enygma`> has quit
23:53 <vmassol> has quit

Get Connected