xwiki-devs

xwiki-devs@xwiki.org

9400 discussions

[xwiki-devs] [Office Import] Mid Term Review and todos

by Vincent Massol

Hi Wang, I haven't run your code yet but I've looked at it and I wanted to send you a quick review before the end of the midterm. * General comment: good work! This is promising. Now we need to work on making it generate xwiki syntax and in making the XHTML parser work fine. I'd expect your help on this now that you've done the basic work for the rest. * Please use the following issue to notify about your progress from now on: http://jira.xwiki.org/jira/browse/XSANDBOX-32. Also whenever you do a commit you should now refer to this issue in your commit comment. * Some comments on OfficeImporter interface public interface OfficeImporter { /** * convert the inputStream which contain the office document to a outputStream which contain the result html code * * @param inputStream * @param inputfileExtension the extension of source content * @param outputStream * @param outputfileExtension the extension of target content */ void convert(InputStream inputStream, String inputfileExtension, OutputStream outputStream, String outputfileExtension); a) The javadoc should be improved. I had to look at the source code to understand the meaning of inputfileExtension... which doesn't look like a file extension at all :) I think it's a file type rather than an extension. Thus I suggest to create an OfficeDocumentType enum class to list all supported file types and use that enum for input and output file types. b) Also I think the output stream should be returned instead of being passed as a parameter. It would also be more homogeneous with the byte[] convert() method which returns the results as an array of bytes. c) Since it's an importer the method should be named import(). However I think it's a converter more than an importer so I'd rather you rename it to OfficeConverter. WDYT? * OfficeImporterFactory. What is this required for? * We need to discuss on the list about how to best integrate your work in XE. This includes discussing the needs for an application or a macro (I'm not sure they are required and in any case should be left for later IMO). * ImporterServerConnectionFactory. IMO should be renamed to OfficeServerConnection and should be implemented as a Plexus component (singleton). * OfficeImporterImp should also be implemented as a Plexus component and should extend AbstractLogEnabled. * OfficeImporterImp: what is this method used for: convert2HTML? It doesn't seem to be used. Same for convert2PDF. They are also not in the interface so I'm not sure what they are for. * Need better exception handling. You need to create a OfficeConverterException exception IMO. * We're using dom4j and not jdom in xwiki so I think it would be better if you could use it instead of jdom. WDYT? (unless there's a good reason of course :)) * I think it would be better to transform your Util class into a set of Filters. Basically the idea would be to implement a Filter chain that would be executed against the generated XHTML and before passing the XHTML to the XWiki XHTML parser. You would implement them as components and have one Filter component per filtering needed (one for removing PinLi, one for transforming the images elements, etc). * BTW I need to check but it's possible that you don't need to transform the img elements into {image} macros since this might be done by the XHTML parser automatically. Same for removing the HTML, HEAD or BODY elements. * Note that xwiki has some util method for creating temporary directories so you shouldn't create them yourself with: // create the new temp folder String oriTempDir = System.getProperty("java.io.tmpdir"); File tempFile = new File(new File(oriTempDir), "xwikiOfficeImportTemp"); if (!tempFile.exists()) { tempFile.mkdirs(); } System.setProperty("java.io.tmpdir", tempFile.getAbsolutePath()); * The build doesn't work right now. It's missing some artifacts in maven remote repositories. We need to work on making it even more simpler to use. * The tests should be improved (better tests, no try catch, etc). Could you send an email on the list asking for feedback from xwiki committers for defining the integration points for this office converter feature? Thanks -Vincent

16 years, 7 months

Re: [xwiki-devs] syntax coloring problem Fixed

by malaka ekanayake

Hi Fabio/Venkatesh, I have committed my work to the svn.I have fixed the problem of syntax colouring. At the present code there are only 5 partition types including the default.At present both List and text enhancements(bold,italic ...) are taken as default partition type.But I think text enhancements and list should be taken as different partitions.Also Headings should belong to another partition. Shall I do the above changes next ? WDUT On Fri, Jul 11, 2008 at 9:46 AM, malaka ekanayake < malaka.ekanayake(a)gmail.com> wrote: > Hi Fabio/Venkatesh, > > I have fixed the syntax coloring problem in the org.xwiki.eclipse.ui > plugin :D. > As far as I can understand was the partitioning of the document.As I > suspected the class which suppose to do the partitionning in the > org.xwiki.eclipse.ui plugin won't execute .In the standalone plugin written > by venketesh works fine because in that,those code are executed through a > extention point.So I move that code inside to the PageDocumentProvider's > createDocument which provide the document. > > I have also remove some classes which are not used in the new code. > > But My ADSL connection is still down.So I'm sending the modified code as a > attachment.Peace look at it and let me know that my fix is correct or not.I > will commit the code ASAP > > cheers > -- Malaka Ekanayake > CSE UOM -- Malaka Ekanayake CSE UOM

16 years, 7 months

[xwiki-devs] Tests broken?

by Tiago Rinck Caveden

Hello, I'm trying to build XWiki but some tests are not passing. They are related to rendering, apparently an expected line feed is not being printed. Take a look at one of the errors: junit.framework.ComparisonFailure: expected:<beginDocument[ onMacro: [testsimplemacro] [param1=value1|param2=value2] [ content ] endDocument ] > but was:<beginDocument[ onMacro: [testsimplemacro] [param1=value1|param2=value2] [ content ] endDocument] > This has been happening since yesterday, so I'd like to know if it's really happening to everybody or if I did something stupid here. I am indeed trying to change one of XWiki's macros, the GraphViz one, I'm adding some functionalities to it - allowing it to use more features that GraphViz provides. But I really don't see how the code I did would break this rendering mechanism, I don't render anything there except the GraphViz stuff. -- Tiago Rinck Caveden http://caveden.multiply.com

16 years, 7 months

[xwiki-devs] Username And Password session variables

by MikhaelSantos

Hi guys! Here is my problem: I have a function that xwiki.myplugin.showdata(command, visualtype) but I need the username and password in the function to show the data. The easy way would be to make it possible to insert username and password "xwiki.myplugin.showdata(command, visualtype, username, password)" but that wouldn't be safe because if anyone edited the page they would see the users password. What I would like to do is be able to in the function showdata() access the information I need: username and password. Can anyone help out? -- View this message in context: http://www.nabble.com/Username-And-Password-session-variables-tp18405471p18… Sent from the XWiki- Dev mailing list archive at Nabble.com.

16 years, 7 months

[xwiki-devs] PhotoAlbumClassSheet Template

by Ambika Corent

I have created the PhotoAlbumClassTemplate in new space called "My Space", so in template file i have include the form as #includeForm("UPM Subscriber Demo.PhotoAlbumClassSheet") - no issue till now. Now, if i try to add new Photo album then below url get submitted * URL : *//<myserver>/xwiki/bin/inline/This Space/NewAlbum?parent=Photos.WebHome&template=This+Space.PhotoAlbumClassTemplate&sheet=1&webname=This+Space&name=NewAlbum Due to empty space ("My Space" - between My and Space ), Add photo album page get loaded without inline form Thanks in advance

16 years, 7 months

[xwiki-devs] [GSOC] Distributed Search status and problem committing the source code

by saikrishna＠research.iiit.ac.in

Hi David and Ludovic, I have finished implementing distributed search for XWiki using the existing plugin "lucene". In short what the new code does. (on each individual installation) 1. As soon is XWiki is started, lucene plugin is initiated, during the process of initialization, the machine makes is index remote available. 2. A Parallel searcher is created over all the remotely available indexes, during the initialization of lucene plugin. These 2 steps will achieve a simple distributed search. But Index updates would not be reflected. 3. As and when there is an index update, index reader which was made remotely available earlier in step 1 is re-opened and is made remotely available. With this Index updates on remote machines are reflected in search process. But what if one the machines go down during the process ?? 4. When ever there is an error while performing distributed search, the parallel searcher in step 2 is redefined, i.e instead of searching 'n' machines (earlier) it will search only 'n-m' machines. At some point the machines went down will be started again. How to search on these machines again? 5. A thread keeps on checking if searcher is set to search over all the slaves defined in xwiki.cfg or not . If it is not searching on all the slaves, It will try to redefine the searcher in step 2 as did in step 4. 6. I have modified existing XWiki.Results to make compatible to both distributed/normal search. 7. Modified xwiki.cfg.vm, to have fews options for distributed search. I have tested this on a cluster of 5 machines and it is working fine. In addition to distributed search I have also attempted to fix 2 bugs/issues in earlier implementation of lucene. a. Fix Lucene plugin scoring algorithm -- *XPLUCENE-8<http://jira.xwiki.org/jira/browse/XPLUCENE-8> *b. Remove duplicates from Lucene search results -- *XPLUCENE-5<http://jira.xwiki.org/jira/browse/XPLUCENE-5> *All these changes exist in the modified version of lucene plugin. *Cannot commit the source code : *I was trying to commit the source code to https://svn.xwiki.org/svnroot/sandbox/plugins/lucene, but I couldn't This is what I did, 1. svn co https://pvsaikrishna@svn.xwiki.org/svnroot/sandbox/plugins/~/xwiki-sandbox-… 2. svn export lucene ~/xwiki-sandbox-plugins 3. cd ~/xwiki-sandbox-plugins 4. svn add lucene 5. svn commit --username pvsaikrishna -m 'distributed search using lucene' *I got the following error* svn: Commit failed (details follow): svn: MKACTIVITY of '/svnroot/sandbox/!svn/act/e5cbc51c-1643-4805-9019-0ce92d036d2b': 500 Internal Server Error (https://svn.xwiki.org) --Sai Krishna

16 years, 7 months

[xwiki-devs] OpenID support

by Markus Lanthaler

Hi! I'm currently adding OpenID authentication support to XWiki but I'm experiencing some problems with the current architecture. First of all the current architecture doesn't support multiple authenticators simultaneously (at least I didn't find how). I wanted to add OpenID as an additional (optional) authentication method as shown in my proposal at http://dev.xwiki.org/xwiki/bin/view/Design/SingleSignOn. This is not possible with the current design - at least not in an elegant manner. Should I add OpenID so that it is always shown if the MyFormAuthenticator is used? Or should I think about how we could make XWiki more flexible? I can't implement the whole XWikiAuthService interface because OpenID doesn't use passwords. This could cause problems. Should we change that interface or create a new one for authentication methods that rely only on XWikiContext? I need a mechanism to handle the response of the OpenID provider (at a known URL). What's the best approach for that? Using an action, a servlet, ...? I need to parse and evaluate the response and depending on return to the login form showing an error message or logging in the user and forwarding it to the requested resource. Thanks, Markus

16 years, 7 months

[xwiki-devs] [VOTE] Add .xml extension to exported documents files names

by Thomas Mortagne

Hi xwikiers, I'm reopening the old thread at http://markmail.org/message/qnfqvdaxyt46hdll. Now that the problem related at http://markmail.org/message/5hola4gk5mvcggq3 is "fixed" I propose to revote for it. So to summarize the goal is to provide an more clear extension to exported wiki pages. To be completed, this means : - modify the core packager plugin to add extension to pages (I think we should make it configurable in xwiki.cfg and maybe in XWikiPreferences) - rename the files in subversion server Here is my +1 (and +1 for making the extension configurable). WDYT ? -- Thomas Mortagne

16 years, 7 months

[xwiki-devs] [gsoc] Presentation Application

by Arpit Jain

Hi, I have developed the basic prototype for Wysiwyg editor for creating Presentation Applications in Xwiki. You can test the editor at http://www.arpitjain.com/xwiki/editor.html The features that are available right now are: 1.) Basic text editing features : Bold/Italic underline 2.) Indentation 3.) Bullets 4.) Support for colors/images/links is in progress and not complete. (You might see a popup for insering colors/images/link, but it is not working fine) 5.) Tables (TODO) *Right now, there are two type of slide templates : * 1.) Title or Cover slides 2.) Regular bulleted list slides Each template has fields that you can click on. When the field is undefined it shows an explanatory text, e.g. click to add title. This disappears when you click on it and restores if you move elsewhere without having typed, or if you delete what you wrote before. This is working fine in IE but not in Firefox. (Event handlers are not working on dom node elements in Firefox, any help ??) *In a bulleted list item(in Regular slides), first Enter starts a paragraph. The second Enter starts the next bullet. To indent or outdent a list item as a whole, you click on the appropriate button or press tab or shift + tab. * You can read more about it and other keypress events at http://arpitjain.com/xwiki/design-notes.html You can insert new slides, but transition between slides is not available right now and I am working on it. Once this is complete, I would work on saving the slides by AJAX and then creating layouts/themes for them. I have tested the features in Firefox(2 and 3) and IE. It should work fine in Safari also as standard workarounds have been taken care of, though I am not sure. @Jerome: It is now working fine for Firefox 2 and Flock in Debian. Please let me know, if you find some issues in other browsers. Thanks -- Arpit Jain Senior Undergraduate Student Department of Computer Science & Engineering Indian Institute of Technology, Kharagpur, India. Web: http://www.arpitjain.com

16 years, 7 months

[xwiki-devs] Come join me on PickensPlan...

by Michael "Ollie" Oliver

PickensPlan: Michael "Ollie" Oliver has invited you to join PickensPlan -------------------- Energy Plan. Don't you think it is at least worth a look? Check out PickensPlan: http://push.pickensplan.com/?xgi=8PYl4lu If your email program doesn't recognize the web address above as an active link, please copy and paste it into your web browser -------------------- Members already on PickensPlan Bill Broadhead, Rob Meyer, Chris Bushnell, Bob, Heather -------------------- About PickensPlan... 19127 members 401 photos 35 videos 743 discussions 12 events 90 blog posts -------------------- To control which emails you receive on the corner, or to opt-out, go to: http://push.pickensplan.com/profiles/profile/emailSettings

16 years, 7 months

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

xwiki-devs