Hi Radu, yes, we used the MS Azure OpenAI connector of Positron Assistant Enterprise and liteLLM as a proxy to OpenAI API to an Anthropic Claude LLM. The confusion was that I first tried to configure the Anthropic connector which gave the error. Things finally worked using the MS Azure OpenAI connec...

Solved, sorry to bother, it was a configuration problem of the LLM connector.

Hi, I am trying to connect the Positron AI Assistant (which is really a great tool btw) to a LLM using Bedrock service from AWS but I always get the error message baseUrl must end in /: The line in the log file looks like: 62353 DEBUG [ AWT-EventQueue-0 ] com.oxygenxml.positron.core.ExternalAIComple...

Hi Julien,
ah okay that was easy
thank you
Regards
Jörn

Hi, I have a PDF2 transformation with a property file, in the DITA OT call there is a parameter \bin\dita.bat ... --propertyfile=<name> When I set up a transformation scenario in Oxygen and add this to the parameter section, it generates a dita call with -Dpropertyfile=<name> It puts a D in front of...

Hello, In Oxygen 25 and 26 we encountered a strange behavior when evaluating an xpath expression with the XPath/XQuery builder. We want to select all attributes in the entire document except the first attribute of each element with this expression: //@*[position()>1] The result is confusing. It sele...

Hi Julien,

we have customized DITA and removing the <p> within the note leads to "unexpected character data" error.
However your second solution does what I wanted. Thanks again!

Regards
Jörn

Thanks Julien, yes this works much better. :D Although I still have the line break after the note, even when adding !important to the span.note__title rule: image.png Here's the rendered HTML <span class="note__title">Note:</span> <p class="- topic/p p">The supply voltage is conn...

Hi I want to customize the output of notes in the PDF transformation with css. There should be no image and no linebreak after the bold "Note: " Sounds like a simple task fulfilled by: *[class ~= "topic/note"]:lang(en):before{ font-weight: bold; content: "Note: "; } and...

Hi I am learning about subject scheme maps and ran into a problem. I built the following example: The ditamap file: <?xml version="1.0" encoding="UTF-8"?> <!DOCTYPE map PUBLIC "-//OASIS//DTD DITA Map//EN" "map.dtd"> <map> <title>Example</title> <mapref href=&q...

This works well, thank you!
Sometimes it is hard to figure out these cryptic notations

We have to use the relative paths because the XML data is in a repository and will be downloaded to different locations.

Regards
Joern

Hello, I need some help again. I use the XSLT and CSS to PDF Transformation Scenario and have specified a background image for the title page @page front-page{ background-image: url("../img/title.png"); } It works fine as long as we need the same title page all the time. But to produce dif...

Hello Julien,

thank you, that was really easy.
I have put it together with current-date() and format-date() functions now.

Regards,
Joern

Hi, we want to put the current date in the footer of a PDF document using Oxygen with the "DITA Map PDF based on HTML & CSS" Transformation. It is done by customizing the p-pages-and-headers.css like this : @page :left, :right{ ... /* footer */ @bottom-left{ ... content: "16.12.20...