Export to pdf with "plain text with embedded fonts" option does not work in all cases

Question

Hi

I have made a argument map using the "entity with attributes" node from the Entity Relationship palette. For my use, I have added one label in the top compartment of the node. In the bottom compartment, there's only the default label.

I exported it to pdf, with "plain text with embedded fonts" option.

In the pdf file, I can select/search text in both the labels in the top compartment.

But I cannot do that for the text in the bottom compartment label. [BUG]

Comments

I have tried, unsuccessfully, to recreate the problem you described.  Using yEd 3.17.2, and the "Entity with Attributes" node, the PDF output text of all labels can be selected and searched in the following cases:

•Node added directly from palette with default text
•Node with a label added to the top compartment
•Node with another node converted to a label in the top compartment
•Node with an additional top label and bottom label font changed

Can the label text be selected and searched in a PDF if you use the default node labels?  If that works, then there may be other settings that are causing the behaviour you observed.

---

I have uploaded an example here:
https://drive.google.com/open?id=1PSuYvGa_uGccBf7KffiH2k4Tq2--ExHz

Both ygml and pdf files are attached.

---

I have examined the files you posted and verified the symptom you observed.  There does seem to be a problem, but it's not specific to the Entity with Attributes node.

It appears that any label with the Configuration property set to "Cropping" cannot be selected when the diagram is exported to PDF with the Text Rendering Policy set to Plain Text or Plain Text with embedded Fonts.  If the label Configuration is set to "Standard", the PDF text selection works.

Also, be sure to remove any data from your nodes when posting examples.

---

I thought there is no data of any kind left in that.
The text is generic.

Please let me know if any fields still have potentially sensitive data in them. Thanks in advance!

BTW I intend to submit the whole original diagram as demo (how to create an argument map from a court verdict). So even if some data is inadvertently left in this sample, it's not a biggie! :)

Answer 1

This is still a bug two years later.  It's bad to expect end users to have to manually enter line returns in cropped boxes just to get all the text rendered when exporting to PDF.  Is there a plan to fix this?

Comments

This is not a bug. The text is rendered to the PDF document. It is, however, rendered in vectorized form (i.e. drawing primitives). This vectorization is not something that happens in yEd code, but in Java's rendering framework over which we have no control.