Cannot show optional content group: AttributeError: module 'fitz.mupdf' has no attribute 'pdf_array_push_drop'

[stefaneidelloth]

Description of the bug

I tried to set the visibility of an optional content group with

doc.set_layer(config=-1, on=[first_id])

and got

AttributeError: module 'fitz.mupdf' has no attribute 'pdf_array_push_drop'

=> How to correctly show/hide optional content groups?

How to reproduce the bug

Full example:

import fitz  # PyMuPDF

doc = fitz.open()
page = doc.new_page()

# Define the items for the combo box
combo_items = ['first', 'second', 'third']

# Create a combo box field
combo_box = fitz.Widget()  # create a new widget
combo_box.field_type = fitz.PDF_WIDGET_TYPE_COMBOBOX
combo_box.field_name = "myComboBox"
combo_box.field_value = combo_items[0]
combo_box.choice_values = combo_items
combo_box.rect = fitz.Rect(50, 50, 200, 75)  # position of the combo box
combo_box.script_change = """
var value = event.value;
app.alert('You selected: ' + value);

//todo: show selected image/optional content group/layer

"""

# Insert the combo box into the page
# https://pymupdf.readthedocs.io/en/latest/page.html#Page.add_widget
page.add_widget(combo_box)

# Create optional content groups
# https://github.com/pymupdf/PyMuPDF-Utilities/blob/master/jupyter-notebooks/optional-content.ipynb


# Load images and create OCGs for each
optional_content_group_ids = {}
for i, item in enumerate(combo_items):
    optional_content_group_id = doc.add_ocg(item, on=False)
    optional_content_group_ids[item] = optional_content_group_id
    rect = fitz.Rect(50, 100, 250, 300)
    image_file_name = f'{item}.png'
    # xref = page.insert_image(
    #    rect,
    #    filename=image_file_name,
    #    oc=optional_content_group_id,
    # )


first_id = optional_content_group_ids['first']
second_id = optional_content_group_ids['second']
third_id = optional_content_group_ids['third']

# https://pymupdf.readthedocs.io/en/latest/document.html#Document.set_layer


doc.set_layer(-1, basestate="OFF")
layers = doc.get_layer()
doc.set_layer(config=-1, on=[first_id])

# https://pymupdf.readthedocs.io/en/latest/document.html#Document.set_layer_ui_config
# configs = doc.layer_ui_configs()
# doc.set_layer_ui_config(0, fitz.PDF_OC_ON)
# doc.set_layer_ui_config('third', action=2)

# Save the PDF
doc.save("output.pdf")
doc.close()

Example images first.png, second.png, third.png

PyMuPDF version

1.23.23

Operating system

Windows

Python version

3.11

[julian-smith-artifex-com]

Thanks for reporting this. I have a fix in my tree.

[stefaneidelloth]

Thank you. I think this does not make the ocm visible, yet.

I would expect something like

doc.show_ocgs([first_id])

or

doc.set_ocg(first_id, on=True)

But there are no such methods.

=> How to programmatically show a previously hidden ocg (or layer)?

[JorjMcKie]

@stefaneidelloth Please have a look at layer_ui_configs() and set_layer_ui_config(number, action=0).
They get / set the temporary visibility - like a supporting PDF viewer would via its UI.

Close by in that part of the documentation, you will also find ways to make permanent changes.

[stefaneidelloth]

Thanks. As you see in the comments /end of the example, I already tried that. What would be a modified version of the example that works? (I also attached example image files just now.)

[julian-smith-artifex-com]

The pdf_array_push_drop bug is fixed in 1.23.24. I'll leave this issue open though until the other questions are understood/answered.