header images share id-space with document

[eswarthammana]

python: 3.10.4
python-docx: 0.8.11
MS Office: 2016

Added two images to the word document body and an image as a header using MS word 2016(Manually added)
Document.xml file

Header1.xml

The below details after add the word document using python-docx
Document.xml

Header1.xml

It's clearly the id issue while during merge both id's are conflicting and overridden the other.
As a temp fix replaced the updated the image id's in the document.xml with the max(id) + 1.

My request is - how i can read the header1.xml image id. So that i will check for duplicate id presence and update the same id with max(id) + 1

[icegreentea]

The <w:drawing> element isn't implemented, so you'll have to lxml you're way down to the <wp:docPr> element (which is mapped as CT_NonVisualDrawingProps fortunately).

Everything below is assuming a single section.

For the body:

doc = docx.Document("blah.docx")
for para in doc.paragraphs:
	for drawing_props in para._element.xpath(".//wp:docPr"):
		# drawing_props will be instances of CT_NonVisualDrawingProps
		print(drawing_props.id)
		# you can set the id to whatever you want here
		drawing_props.id = 10

For the header:

doc = docx.Document("blah.docx")
for section in doc.sections:
	for drawing_props in section.header._element.xpath(".//wp:docPr"):
		print(drawing_props.id)

[eswarthammana]

Thanks @icegreentea

i have updated the below way, while final save of the document.

    `doc_lst = doc._part._element.xpath('.//wp:docPr')
    doc_ids = [int(doc_pr.get('id')) for doc_pr in doc_lst]
    header_lst = doc.sections[0].header._element.xpath(".//wp:docPr")
    header_ids = [int(doc_pr.get('id')) for doc_pr in header_lst]
    match_found = [i for i in header_ids if i in doc_ids]
    if len(match_found) > 0:
        for doc_pr in header_lst:
            doc_pr.set('id', str(int(doc_pr.get('id'))+max(doc_ids)))`

[scanny]

This is a good catch. It looks like the image "id-space" is shared across the document so we'll need a different get_next_id() method for that case that considers all images in the document.