March 2009 - oXygen-user - www.oxygenxml.com

trying to create windows help html
by Bram Vogelaar 17 Mar '09

17 Mar '09

Dear all I am trying to create windows help files from one of our docbook files. All is well with the default settings while using the $ {frameworks}/docbook/xsl/htmlhelp/htmlhelp.xsl but the files end up in the /Applications/oxygen/ folder. (i use the mac stand alone version) When i try to change the basedir location, only the html files end up in the basedir not the .hh* files still end up in the /Applications/oxygen/ folder and the chm fails to render due to include issues. Does anyone know how to fix this? with regards Bram

2 2

DITA Maps Opens to Author Mode
by Paul Ryan 16 Mar '09

16 Mar '09

I think I've exhausted my search for how to open files from a DITA Map (in the DITA Maps editor) directly into the text page (text mode). I have all of my options set to text mode as my default such that if I open the individual .dita files regularly they open into text. From the DITA Map editor however open always goes to the author page (author mode) first. Any help in getting these files to open in the text page is appreciated. Thanks, --Paul

2 1

10.1 takes a step backward?
by Syd Bauman 13 Mar '09

13 Mar '09

First, if this has already been mentioned on the list or via your bug reporting system and I've missed it, my apologies. Second, let me state for the record that despite the following complaint, I am a big fan of oXygen. There is no other commercial proprietary software that I outright recommend people buy, except for Mac OS X. (And even that only for those not up to dealing with Ubuntu.) I have severe tendonitis to my wrists (bilaterally). One consequence of this is that I really try to avoid using the pointer (in my case a trackball or trackpad, as I cannot use a mouse at all), and try to type most commands. For those menu items that do not have keyboard shortcuts, I often use the operating system's "access the menu" shortcut (ctl-F2 on a Mac, alt-underlined_letter on my GNU/Linux system). I have not tried oXygen 10.1 on a GNU/Linux yet (I use oXygen there very rarely), but I use oXygen almost daily on Mac OS X 10.5 with "In windows and dialogs, press Tab to move the keyboard focus between" set to "All controls" in the Keyboard & Mouse pane of System Preferences. One action I perform reasonably frequently is Associate Schema, although it's usually via the New XML Document rather than explicitly associating a schema to an already existing XML document. Either way, in previous versions of oXygen I would perform this entirely from the keyboard. Something like: * cmd-N > new dialog * XML Document selected > defaulted from last time * enter > gets "Create an XML document" dialog * Use DTD or Schema checked > defaulted from last time * tab > moves me to set of schema language tabs * R-cursor R-cursor > selects Relax NG * tab tab > selected "browse local" folder icon [X] * space > gets open dialog box * [navigate to schema] > use normal Mac OS X keystrokes * return > syntax button automatically set from extension * return > done! But note the [X]. Here, in oXygen 10.0 (and earlier versions, but I don't know how far back), there was no visual indication that I had selected the button with the normal folder icon that means "Browse for local file" (the leftmost of the 2 or 3 folder icons for browsing local, remote, or archive), but it worked. Now, in 10.1, tabbing in this dialog box skips right over the browse icons! I've tried all sorts of key combinations (ctl-tab, opt-tab, cursor keys, cursor keys w/ modifiers, etc.), but cannot seem to select that button without using the pointer. Snif! Sob! PLEASE give me a way to create a new document with a schema associated without using the pointer! P.S. It turns out I can still create new files in oXygen without using the pointer so long as I don't associate a schema. Some time ago (9.2? 10.0? I forget), the "New" dialog box became a combined "New" and "New from template" box. As far as I could tell, there was no way to select one or the other of those tabs from the keyboard, though. However, I recently discovered that pressing ctl-tab 3 times or shift-ctl-tab once selected the tabs bar, so that either L or R cursor keys would select the other tab. Not exactly convenient, but I'm trying to teach my fingers to do that.

3 5

Open Big Files in read only with validation capabilities
by mozer 12 Mar '09

12 Mar '09

Hi, Recently I felt the need for being able to : A) Open a big Files in Oxygen B) Do some validation and see where the problem happen to be (and perhaps launch an XSLT Transformation) C) Do some indenting But I don't want to edit the file I get around 1) Open Big files : it is indeed read only, but we don't have access to validation and transformation 2) Open Tree Editor : it is indeed fast to load, but it takes ages before the validation can finish 3) Open normal editor : it takes 1,5 Gb and at least 10 minutes to stabilise for 100 Mb files I checked the performance options to drop undo information on save Is there anything else I need to know ? Xmlizer

2 2

RelaxNG and Schematron
by Christian Kosmowski 11 Mar '09

11 Mar '09

Hi there! Due to some restrictions of XML Schema we are currently changing to RelaxNG. I now tried to take some Schematron rules of the old schema and put it in the new RelaxNG Schema to test if Schematron will work. This is my "root" File which includes other RelaxNG Schemas: <?xml version="1.0" encoding="UTF-8"?> <grammar xmlns="http://relaxng.org/ns/structure/1.0" xmlns:a="http://relaxng.org/ns/compatibility/annotations/1.0" xmlns:sch="http://www.ascc.net/xml/schematron" datatypeLibrary="http://www.w3.org/2001/XMLSchema-datatypes" ns="http://www.docscape.de/layoutV4.2"> <include href="layout_definitions.rng"/> <sch:pattern> <sch:rule context="//Regel[@datensatz and @name]"> <sch:assert test="count(//Regel[@name=current()/@name and @datensatz=current()/@datensatz]) le 1">Die Regel »<sch:value-of select="@name"/>« zu dem Datensatz »<sch:value-of select="@datensatz"/>« ist mehrfach vorhanden.</sch:assert> </sch:rule> <sch:rule context="//Regel[@datensatz and not(@name)]"> <sch:assert test="count(//Regel[@datensatz=current()/@datensatz and not(@name)]) le 1">Die Regel zum Datensatz »<sch:value-of select="@datensatz"/>« ist mehrfach vorhanden.</sch:assert> </sch:rule> <sch:rule context="//Regel[@name and not(@datensatz)]"> <sch:assert test="count(//Regel[@name=current()/@name and not(@datensatz)]) le 1">Die Regel »<sch:value-of select="@name"/>« ist mehrfach vorhanden.</sch:assert> </sch:rule> <sch:rule context="//Zuweisung"> <sch:assert test="(//DefVariable[@name=current()/@name]) or exists(Name)">Variable »<sch:value-of select="@name"/>« nicht definiert.</sch:assert> </sch:rule> </sch:pattern> <start> <choice> <element name="Layout"> <optional> <attribute name="version"> <text/> </attribute> </optional> <optional><attribute name="datensatz"><text/></attribute></optional> <zeroOrMore> <interleave> <optional><ref name="g_Deklarationen"/></optional> <optional><ref name="g_Documentation"/></optional> <optional><ref name="x_Layoutinclude"/></optional> </interleave> </zeroOrMore> </element> <ref name="x_Layoutinclude"/> </choice> </start> </grammar> The're is just no effect to the validation. The rest of the Schema works fine but the schematron rules have no effect. They seme to be ignored. Any suggestions? Thanks, Christian -- QuinScape GmbH Christian.Kosmowski(a)QuinScape.de http://www.QuinScape.de http://www.docscape.de

2 1

String normalized in the result of XQuery?
by Florent Georges 06 Mar '09

06 Mar '09

Hi, When I evaluate the following XQuery: 'a&_#10;b' using a scenario configured with the MarkLogic "transformer," I get the following result: 'a b' (a file with those three characters, note this is a space instead of a linefeed.) If I use the MarkLogic CQ, I get the expected result, with a linefeed instead of a space. Sounds like a bug to me. I am not sure what other info I could provide you with. Regards, -- Florent Georges http://www.fgeorges.org/

2 2

XQuery using Oxygen to MarkLogic server
by Lee, David 06 Mar '09

06 Mar '09

I've tried the forum but so far am not successful. Maybe someone on this list has an idea I could try. I'm trying to run an XQuery against a marklogic server. I've setup the connection, can browse the DB with DatabaseExplorer (WebDav), I stored a trivial xquery file and tried to execute it. No matter what I do I get errors from the marklogic server. I think I need to put something magic in the "XQuery URL" field but I have no idea what. If I use ${currentFileURL} I get the error from the server: SystemID: http://DLEE@localhost:8004/xquery/dxtest1.xquery Engine name: MarkLogic Severity: error Description: Query evaluation request rejected (404, Not Found). Is this an XDBC server? So I think that MarkLogic doesnt want to use the WebDav connection for its URL to the query. I've tried explictly usin the "browse remote" path or what I think is the path http://DLEE:******@localhost:8004/xquery/dxtest1.xquery Same error I changed the path to be port 8002 which the XDBC driver runs on SystemID: http://DLEE@localhost:8004/xquery/dxtest1.xquery Engine name: MarkLogic Severity: error Description: 404 Not Found for: http://localhost:8002/xquery/dxtest1.xquery ---------------------------------------- David A. Lee Senior Principal Software Engineer Epocrates, Inc. dlee(a)epocrates.com <mailto:dlee@epocrates.com> 812-482-5224

2 6

Create a new file on MarkLogic through WebDAV
by Florent Georges 06 Mar '09

06 Mar '09

Hi, I have MarkLogic in a virtual box for developing purposes (on Mac OS X) and tried to create a new file from within oXygen, by using the "Open URL" action. I filled in the info and pressed "Browse" to be sure the info are ok, and everything goes fine. But when I press "OK", I receive a 404. Is it possible to create a new file with this action? If not, is it possible by any other means? Regards, -- Florent Georges http://www.fgeorges.org/

3 4

Out of memory fixed
by Mark Wilson 05 Mar '09

05 Mar '09

Thanks Barton (and Eliot), Worked like a charm, and without the quotes. Yes, I know that I have a lot of wasted memory, but at the time I installed, it was the cheapest way to get the 3.5 I do get to use. In the good old days, running OS-9 on a Tandy Color Computer, I had to hard-wire bank switching on a 64K memory chip for a 6009 processor that could only address 32K. Things were so simple then. Although I am highly skilled with C++, I am totally at sea with java and know absolutely nothing at all about Linux - my bad. >From the Arizona desert, Thanks Mark

1 0

Out of memory problem
by Mark Wilson 05 Mar '09

05 Mar '09

This is likely not the right list for this question, but please help if you can. Using Oxygen's scenario mechanism, I am able to run the built-in FOP and produce my PDF file with no problems. However, when I try to do the same thing from a BAT using my stand-alone FOP 0.95, I get an "out of java heap space memory". I Googled and found lots of information for people who know java or Linux (I know neither). I know I have to change the fop.bat file but do not understand exactly how to do that. Concerning java, the fop.bat has the lines: set JAVAOPTS=-Denv.windir=%WINDIR% if "%JAVA_HOME%" == "" goto noJavaHome if not exist "%JAVA_HOME%\bin\java.exe" goto noJavaHome if "%JAVACMD%" == "" set JAVACMD=%JAVA_HOME%\bin\java goto runFop :noJavaHome if "%JAVACMD%" == "" set JAVACMD=java :runFop rem ECHO "%JAVACMD%" "%JAVACMD%" %JAVAOPTS% %LOGCHOICE% %LOGLEVEL% -cp "%LOCALCLASSPATH%" org.apache.fop.cli.Main %FOP_CMD_LINE_ARGS% What do I add or change? I have 4 gig of memory on a Vista machine. What do I add or change so that my stand-alone FOP will have the same heap size as the Oxygen built-in FOP? Thanks for any consideration you can give me, Mark

3 2