January 2005 - oXygen-user - www.oxygenxml.com

Project Files are have absolute file paths
by Peter Charles 11 Jan '05

11 Jan '05

Looking at the xml in the .xpr project files all file paths are absolute. This in unfortunate since it means that when you share projects via cvs, all users must have the same directory structure. Is it possible to set projects to use relative paths as a configuration option? Thanks, Peter

2 1

font.family Configuration
by Ray Miller 11 Jan '05

11 Jan '05

Utilizing oXygen 5.1 with the included Xalan processor. Can not get title.font.family body.font.family monospace.font.family all with values Helvetica, applied under under document | xml document | configure transformation scenario | edit | parameters to work. Is this a transformation engine or an oXygen limitation? Please advise. -- Ray Miller

2 3

huge memory usage
by lars 11 Jan '05

11 Jan '05

Hello, Oxygen seems very slow to restore when I've been using other applications. I understand that restoring from the page file can take time, but it seems really slow compared to other programs. Today I looked at the Task Manager during the Oxygen restore. The memory size of the Oxygen process was increasing at roughly .5MB/second, and was up near 200MB! And there was only one file open -- a small XSL stylesheet! (I had previously had about 6 or 7 stylesheets open -- none of them large.) It seems strange that Oxygen would need so much memory. I looked in the documentation for tips on performance, and it tells how to set the amount of memory available to the JVM, but doesn't indicate how to know what value to set it to. I guess I must have it on high since it was able to allocate over 200MB of space (oxygen.bat uses the -Xmx256m cmd-line option). Would it help to lower this setting? FWIW my laptop has 384MB RAM, and is running WinXP SP2. Windows has 600+MB of swap space on the drive. I'm running on power, not battery, so that's not an issue. I've just completely defragged my hard drive, so I know that's not the problem. These waits happen not only when I return to Oxygen from running other programs, but also when I return to a different buffer (file) in Oxygen that I haven't visited for a while, or use a function (like Find) for the first time in a while. After the wait, Oxygen runs pretty fast, but the waits are aggravating. Any suggestions? Other than more RAM (which I'm considering). Lars

2 1

[Fwd: Re: [oXygen-user] Validation and data entry]
by George Cristian Bina 11 Jan '05

11 Jan '05

This did not get through to the oxygen-user due to a config problem of my send mail server. Regards, George -------- Original Message -------- Subject: Re: [oXygen-user] Validation and data entry Date: Tue, 11 Jan 2005 00:07:46 +0200 From: George Cristian Bina <george(a)oxygenxml.com> To: António Mota <amsmota(a)gmail.com> CC: oXygen-user(a)oxygenxml.com References: <2fa55447050110075114e01532(a)mail.gmail.com> Hi Antonio, Oxygen will present in the content completion only the elements that keep the document valid. If you make the model view visible (you can find it in a tabbed pane together with the project view) you will be able to see also the content model for the current element as well as the information about the current element attributes. If you annotate your schemas oXygen will present the xs:annotation/xs:documentation next to the content completion proposals as the users edit the document, thus you can tell/explain the user the meaning of each element, attribute or value and that will be present exactly when the user needs it. You can see this in action if you edit an XML Schema or an XSLT stylesheet (they use annotated XML Schemas for content completion). Oxygen handles not only simple restriction types when it presents values in the content completion, it follows down the simple types until it gets the enumerations, thus we can handle also union types for instance. More it has support for list type values, after a value is inserted it will offer content completion for the following values. From 5.1 it has support for content completion also for elements with simple types, presenting the values as in case of attributes and it will handle also xs:IDREF and xs:IDREFS types presenting the xs:ID values collected in the last validation action. There is no real time validation support, but if the users will follow the content completion proposals and helped by the model view and by the schema annotations they should not really need that. Also in 5.1 oXygen added Schematron support. You can use that to create a standalone Schematron schema or to embed Schematron rules in your XML Schema and validate with both XML Schema and Schematron on a single action. This will let you apply very powerful checks on your instance documents as Schematron basically specifies XPath checks that are executed in specified contexts, so you have the full XPath power. Thus you can assert for instance that the sum of price elements from a list of products should be less than 500. Best Regards, George --------------------------------------------------------------------- George Cristian Bina <oXygen/> XML Editor, Schema Editor and XSLT Editor/Debugger http://www.oxygenxml.com António Mota wrote: > Hello: > > I'm new in here, i'm evaluating oXygen (among others) in order to buy > some licences to our company (nothing big, 4-5 licenses only). > > Until now i'm using 5.0, i like what i see so far, but i've a big > concern. The idea is to have the editor to our XML/XSL guy (me) do the > design/test/implementation, along with our Javascript guy (me) and the > CSS guy (me) cause the first project will be client-side tranforms, > and also with our Java guy (well, me) for other XML projects (based on > the XUI project). > > However after all these people :) do their work it will be the need > for some other people, probably with no or little understandig of XML, > to pick up the definitions and "write" a lot of XML files for > application use. > > Until now i'm testing Schema validation on data entry, but i can only > validate the data after their inclusion. Isn't a way to do some > "on-the-fly" real-time validation? > > Also, using 5.0, it seems the data introduction modes are very poor. > I've installed 5.1 just now. Is there some improvements on this? > > Excuse me also for my bad english writing... > > Cheers. > _______________________________________________ > oXygen-user mailing list > oXygen-user(a)oxygenxml.com > http://www.oxygenxml.com/mailman/listinfo/oxygen-user -- --------------------------------------------------------------------- George Cristian Bina <oXygen/> XML Editor, Schema Editor and XSLT Editor/Debugger http://www.oxygenxml.com

1 0

XInclude Validation Errors
by Ray Miller 10 Jan '05

10 Jan '05

Utilizing oXygen 5.1 with Docbook 4.4b2 DTD and XSL 1.67.2 with the default oXygen Xalan transformer. A document validation check produces the following error: Element type "xi:include" must be declared along with a couple of related error messages. XInclude Processing is enabled utilizing Options | Preferences | XML Parser Options Docbook files compile correctly utilizing XInclude but the errors during the document validation are irritating. Please advise. -- Ray Miller

2 1

Validation and data entry
by António Mota 10 Jan '05

10 Jan '05

Hello: I'm new in here, i'm evaluating oXygen (among others) in order to buy some licences to our company (nothing big, 4-5 licenses only). Until now i'm using 5.0, i like what i see so far, but i've a big concern. The idea is to have the editor to our XML/XSL guy (me) do the design/test/implementation, along with our Javascript guy (me) and the CSS guy (me) cause the first project will be client-side tranforms, and also with our Java guy (well, me) for other XML projects (based on the XUI project). However after all these people :) do their work it will be the need for some other people, probably with no or little understandig of XML, to pick up the definitions and "write" a lot of XML files for application use. Until now i'm testing Schema validation on data entry, but i can only validate the data after their inclusion. Isn't a way to do some "on-the-fly" real-time validation? Also, using 5.0, it seems the data introduction modes are very poor. I've installed 5.1 just now. Is there some improvements on this? Excuse me also for my bad english writing... Cheers.

1 0

oXygen File-Open dialog wishes
by Lars Huttar 10 Jan '05

10 Jan '05

Dear oXygen folks, I have just purchased an academic license to oXygen, with 1-yr maintenance. Maybe this is a good time to express some of my wishes for Oxygen that have held me back from purchasing till now. I'm using version 5.0, so maybe some of this is fixed in 5.1. #1: Better directory-browsing functionality in File-Open dialog. This is a major reason why I still often use XMLSpy instead of oXygen. When you press Ctrl+O, the Open dialog that appears isn't the native Windows File-Open dialog, and doesn't meet expectations in some significant ways. I'm sure this is due to Oxygen being a cross-platform Java program. But I would hope that that doesn't mean the interface has to be inferior. a) Can't follow shortcuts. My code base is located several directories deep, so to get there, in XMLSpy I first open c:\mount.lnk, which is a shortcut to the actual codebase. When I select mount.lnk (e.g. double-click, or press Enter) the dialog stays open, but displays the folder that the shortcut points to. In Oxygen, it apparently attempts to open "mount.lnk" as a file, with no useful result. So I have to navigate the whole path by hand. b) File/folder name completion. As I start typing the name of the file, a native Windows dialog would suggest possible completions, and let me select using the up/down arrow keys. Oxygen does not. c) Type-to-find doesn't autoscroll. If you put focus in the pane of the Open dialog that lists all the files, and you type letters, the selection moves to the file/folder that begins with those letters; but if the newly selected file/folder happens to be one that is not showing on the screen (e.g. it's off to the right because there are too many files to show at once), the pane doesn't auto-scroll. Manual scrolling is required, which detracts from the ergonomic benefit of type-to-find. d) Pressing Enter after typing a folder name doesn't remove the folder name from the "File name" field. E.g. if you are in c:\Program Files and you type Apache Group and press Enter, the dialog shows the subfolders of c:\Program Files\Apache Group, but leaves "Apache Group" in the File name field. (Expectation: Apache Group would disappear, since it is not needed relative to the folder that is being shown, i.e. Apache Group.) So if you continue typing the name of a subfolder, you get "Apache Group\cocoon" in the File name field, and if you press Enter/OK, you will not see the contents of the cocoon subfolder. (Instead it will apparentlyl try to open c:\Program Files\Apache Group\Apache Group\cocoon.) e) If any of the above behavior leads to attempting to open a file that doesn't exist, Oxygen still closes the File-Open dialog and tries to open the file; thus I lose the File-Open dialog and have to hit Ctrl+O again. It would be nicer if the File-Open dialog would (beep or something and) stay open unless you click Cancel. f) (This is more icing on the cake...) In a native Windows file-Open dialog, you can perform other operations in the dialog, such as renaming or deleting files. This is often handy: e.g. if you want to save the file you're working on to a filename that already exists. In Oxygen you can't do this. That's probably enough for now. Lest I sound overly negative, let me say that the benefits of Oxygen have outweighed its limitations enough that I was willing to pay for it, vs. the free XMLSpy Home Edition. For example, the responsiveness to problems posted on this list has been a good selling point. I hope to see Oxygen improve to the point where I use it without having regretful thoughts about what I left behind. Lars

2 1

oXygen 5.1 beta
by George Cristian Bina 07 Jan '05

07 Jan '05

Hi everybody! First of all I would like to wish you all a Happy New Year! We have almost ready a new version of oXygen and we made available a couple of distributions of a beta version of oXygen 5.1. Windows without a JVM: http://www.oxygenxml.com/v5.1beta/InstData/Windows/NoVM/oxygen.exe All platforms: http://www.oxygenxml.com/v5.1beta/InstData/All/oxygen.tar.gz Eclipse site URL: http://www.oxygenxml.com/v5.1beta/InstData/Eclipse/site.xml I would like to invite you to try it out and let us know if you find any problem with it. Although it is a minor release it contains a lot of new additions and improvements. I will mention the most important features below: * Folding support oXygen has a light folding implementation and offers more that collapse/expand actions. The most useful action is IMO collapse child folds (CTRL + Numpad-) which will collapse the child fold elements of the current fold (that is it will present the children elements of the current element collapsed). Using this starting from the root element and going down to the element of interest will have as result the document presented with all the elements collapsed except the element of interest. Also the folding structure is preserved after a format and indent action and even if you close the file and reopen it. * Really improved Relax NG support (I think this will make a lot of people happy :) ) The content completion support for Relax NG was rewritten from scratch to offer context sensitive proposals and it is not almost at the same level as the support for XML Schema and DTD. Also we have now an implementation of the Model View based on the Relax NG schema associated with the instance document that shows the element model and the possible attributes. The most important missing part that we plan to address in a future release is the annotation support, that is to present the annotations in the content completion documentation window and in the Model View. * Schematron support oXygen will offer full support for Schematron 1.5. This includes editing and validating Schematron schemas, validating XML documents with Schematron schemas, with Relax NG schemas with embedded Schematron rules and with XML Schemas with embedded Schematron rules. * Support for XSLTProc, MSXML 3.0, MSXML 4.0 and MSXML.Net as XSLT transformers. * Support for ID values on content completion oXygen will present the ID values collected in the last validate action on the content completion proposals where an IDREF or an IDREFS type is specified. This works not only with documents that have a DTD associated but also with documents that have an XML Schema or a Relax NG schema associated. * Code templates The content completion support can be enhanced with code templates. oXygen already ships with some code templates for XSLT (thanks to M. David Peterson) and XSD. One example is typing cwto and pressing CTRL+Space will get into an XSLT document something like: <xsl:choose> <xsl:when test="|"> </xsl:when> <xsl:otherwise> </xsl:otherwise> </xsl:choose> where | represents the caret position. The code templates are triggered with CTRL+Space and they are presented together with eventual content completion proposals. * Automatic detection of formatting settings There are two options that control this feature. The first (true by default) controls the detection of indent size and indent characters. The second (false by default) controls the detection of line width. * Hard line wrap as you edit This is a time saving option. When editing content one needs to use almost at every line the format and indent action in order to get the content wrapped. Now the wrapping can be triggered automatically (if the option is enabled) and one can issue a format and indent action eventually when he finishes the editing of that content. There are more features, I will just mention their short description below: * Content completion support also on element values (for elements with XML Schema simple type content) * Saving of WSDL SOAP Requests * Open Web Service Call Result into the Editor * Extended Go to Options * Optional Content in new XML Document Creation * Scenario based XSLT debugging session * Save the Debug Trace * Extended Batch Validation * Saxon 8.SA can be used as XQuery Processor * Annotation Tooltips * Sortable Tables * Option for saving all files before validation or transformation * Project improvements * Fixed the bidirectional editing support * Better undo support * Updated the TEI framework * File properties page Best Regards, George --------------------------------------------------------------------- George Cristian Bina <oXygen/> XML Editor, Schema Editor and XSLT Editor/Debugger http://www.oxygenxml.com

1 1

Associating and validating with xsd
by Mark Peyer 05 Jan '05

05 Jan '05

Hi, I am exporting data out of FileMaker Pro's XML function, converting them with XSL to comply with an .xsd-file. I can associate the .xsd, and Oxygen 5 will insert xsi:schemaLocation="http://www.xmlData.ch/xmlInvoice/XSD filepath/schemafile.xsd" Then I can validate the .xml-file. The receiver of my .xml-files wants the schemaLocation in the form xsi:schemaLocation="http://www.xmlData.ch/xmlInvoice/XSD schemafile.xsd" If I associate the .xsd-file without the filepath, I cannot validate it. In Project mode, I can validate one or all .xml-files using "Validate selection with...". Here validation functions with or without filepath, but the line xsi:schemaLocation="http://www.xmlData.ch/xmlInvoice/XSD schemafile.xsd" Is not inserted. As far as I can see, there is no possibility to batch associate a schema file. Are there any workarounds? Am I missing something? Thanks for your help. -- Mark A. Peyer

2 1

DocBook Table Problems
by Ray Miller 04 Jan '05

04 Jan '05

I believe this question has been submitted before, I am unable to locate a suitable resolution. Current version of oXygen on JDK15; Fedora2 platform. Docbook DTD is 4.4b2 XSL is version supplied with oXygen Transformer is Saxon 6.5.3 <table width="1224" border="0" cellpadding="0" cellspacing="0"> <title>Test table</title> <tgroup cols="3"> <colspec colwidth="796" colnum="1" colname="c1"/> <colspec colwidth="74" colnum="2" colname="c2"/> <colspec colwidth="354" colnum="3" colname="c3"/> <spanspec namest="c1" nameend="c2" spanname="s1"/> <tbody> <row valign="top"> <entry spanname="s1"><para>cols 1 and 2</para></entry> <entry><para>col 3</para></entry> </row> </tbody> </tgroup> </table> The colwidth attribute does not function; cannot create a DocBook rendering of an OpenOffice text document. Help please. Ray

2 1