Plutext

HI all,

I used docx4 to copy some docm templates. There are embedded excel charts in the templates. I use their names to identify them later and fill them with data.
But docxj4 do not actually transfer their names.

It transfers their descriptions however, and I could get around this problem.
But it is still better than one uses chart names to identify them and not chart decsriptions.

docx4j exposes what is in the docm, nothing more and generally speaking, nothing less.

Could you please pinpoint exactly where the name you mention appears in the docm?

In the main document part document.xml, the reference to the chart is by relId, for example:



though I see there also <wp:docPr id="1" name="Chart 1"/>. Does that correspond to your name?

You'd think that a name could appear in the chart part itself, but I didn't notice it in the example I looked at.

I am sorry, it is not the name, it is the title

in the original docm

<wp:docPr id="53" name="Diagramm 53" descr="LoadShareOverTotalOperatingHours" title="LoadShareOverTotalOperatingHours" />

It is in the docPr but as it seems, it is not parsed.

This is the copied through docxj

<wp:docPr id="53" name="Diagramm 53" descr="LoadShareOverTotalOperatingHours" />

The title is missing.

I am using docx4j-2.7.1

It is just funny that the description is transferred and not the title.

So , the problem is actually with parsing the document.xml

There were also not methods for the title. It is just an omission, but still could be corrected.

It looks like there may be a problem with the schema.

We have



where wp is "http://schemas.openxmlformats.org/drawingml/2006/wordprocessingDrawing"

The schema https://github.com/plutext/docx4j/blob/ ... rawing.xsd contains:



There is a discrepancy between namespaces here which needs to be investigated. I've created an issue for this on GitHub, see https://github.com/plutext/docx4j/issues/1

Interestingly, roundtripping a file, IN:



becomes OUT:



so the element is correctly written to the correct namespace, but @title is lost. org.docx4j.dml.wordprocessingDrawing.Inline contains field org.docx4j.dml.CTNonVisualDrawingProps docPr, which doesn't contain a title field, which explains that.

But it is not clear to me how JAXB manages to get the namespace right? I though xsd:import kept the element in its original namespace (ie drawingml/2006/main) as opposed to moving it to wp:. If that's correct, how does JAXB know what to do when it encounters this element in the wp namespace? Looks like it is just happily assuming that namespace, since the reference is from there, and there is no field annotation which tells it to use a different namespace. (So, did xjc do the right thing in writing this element without a field annotation, or not?)

Yes, I would like to investigate that too.