Difference in decoding of OCGs names between doc.get_ocgs() and page.get_drawings()

[Rodrigodd]

Description of the bug

The list returned by doc.get_ocgs() contains the expected names of the OCGs, but the ones in the 'layer' field of the elements in page.get_drawings() are misdecoded.

How to reproduce the bug

Running the following code:

import fitz
import sys
from pprint import pprint

filename = sys.argv[1]
doc = fitz.open(filename)

pprint(doc.get_ocgs())

for page in doc:
    print('# Page:', page.number)
    print('## bboxlog layers')
    for _, _, l in page.get_bboxlog(layers=True):
        print(l)
    print('## drawing layers')
    for obj in page.get_drawings():
        print(obj.get('layer'))

With the following file:

layers-test2.pdf

Gives the following output:

PS > python .\show_layers.py .\layers-test2.pdf
{1: {'intent': [], 'name': '0', 'on': True, 'usage': None},
 2: {'intent': [], 'name': 'braço', 'on': True, 'usage': None},
 3: {'intent': [], 'name': 'braço1', 'on': True, 'usage': None},
 4: {'intent': [], 'name': 'ênvelo', 'on': True, 'usage': None},
 5: {'intent': [], 'name': 'camada', 'on': True, 'usage': None}}
# Page: 0
## bboxlog layers
0
braço
braço
braço1
ênvelo
camada
## drawing layers
0
braço
braço
braço1
ênvelo
camada

PyMuPDF version

1.23.21

Operating system

Windows

Python version

3.12

[JorjMcKie]

Thanks for submitting this.
We will look at it now.

[julian-smith-artifex-com]

Fixed in 1.23.22.