PDF parser: add required attributes from the document context to section

[buchen]

No description provided.

[buchen]

As discussed, this change allows to "require" document context attributes.

It really does not reduce the number of lines of code in the extractor - but it fails with better error messages if a property such as currency is actually missing.

What do you think?
Merge? Or not worth it?

[Nirus2000]

Hello @buchen
Is it still possible that documentContext could also be optional?
Sometimes it is possible that we need to search and compare multiple attributes.

Regards
Alex

[buchen]

Is it still possible that documentContext could also be optional?

If it is required, we could add a

.optionalDocumentContext("currency")

I was trying to avoid it, because it makes the code harder to understand and maintain.

Why? The attributes of the sections are always required. The flag "optional" is an all-or-nothing flag on the whole section. Now we would add an optional flag no the attributes level - but only on some attributes. From a semantic point of view it becomes kraut and rüben.

Plus: the added value of defining an optional document attributes is very small. For required document attributes we fail if they do not exist - but obviously we do not check this for optional attributes. So it boils down to writing v.get("X") instead of type.getCurrentContext().get("X") which is a difference of only 23 characters.

[Nirus2000]

Okay, i understand.
Let's start...