Home Information - Your home, your data, organized.
Python
import logging
import os
import xml.etree.ElementTree as ET
from hi.apps.common.file_utils import generate_unique_filename
from hi.apps.common.svg_models import SvgViewBox
logger = logging.getLogger(__name__)
ET.register_namespace('', 'http://www.w3.org/2000/svg')
DANGEROUS_TAGS = {
'script', 'foreignObject', 'iframe', 'object',
'animation', 'audio', 'video', 'style',
}
DANGEROUS_ATTRS = {
'onload', 'onclick', 'onmouseover',
}
HREF_ATTRS = {
'href', 'xlink:href',
'{http://www.w3.org/1999/xlink}href',
}
def process_svg_content( svg_content,
media_destination_directory,
source_filename,
remove_dangerous = True ):
"""
Process a full SVG document: validate, extract viewBox, strip the
outer <svg> wrapper, clean namespaces, scan for dangerous elements,
and generate a unique MEDIA_ROOT filename.
Args:
svg_content: Full SVG content string with outer <svg> element.
media_destination_directory: Relative directory in MEDIA_ROOT (e.g., 'location/svg').
source_filename: Original filename for generating unique destination name.
remove_dangerous: If True, silently remove dangerous elements.
If False, return counts but do not remove.
Returns:
dict with keys:
'svg_fragment_content': Inner SVG content string (outer <svg> stripped).
'svg_viewbox': SvgViewBox instance.
'svg_fragment_filename': Generated filename relative to MEDIA_ROOT.
'dangerous_tag_counts': dict of {tag_name: count} for dangerous tags found.
'dangerous_attr_counts': dict of {attr_name: count} for dangerous attrs found.
Raises:
ValueError: If content is not valid SVG or missing viewBox.
"""
root = ET.fromstring( svg_content )
if root.tag != '{http://www.w3.org/2000/svg}svg':
raise ValueError( 'Content is not a valid SVG.' )
view_box_str = root.attrib.get( 'viewBox' )
if not view_box_str:
raise ValueError( 'SVG must contain a viewBox attribute.' )
svg_viewbox = SvgViewBox.from_attribute_value( view_box_str )
dangerous_tag_counts = dict()
dangerous_attr_counts = dict()
for element in list( root.iter() ):
if element is root:
continue
if element.tag.startswith( '{http://www.w3.org/2000/svg}' ):
element.tag = element.tag.split( '}', 1 )[1]
tag_name = element.tag.split( '}' )[-1]
if tag_name in DANGEROUS_TAGS:
dangerous_tag_counts[tag_name] = dangerous_tag_counts.get( tag_name, 0 ) + 1
if remove_dangerous:
logger.debug( f'Removing dangerous SVG tag "{tag_name}"' )
root.remove( element )
continue
for attr_name in list( element.attrib ):
if attr_name in DANGEROUS_ATTRS:
dangerous_attr_counts[attr_name] = dangerous_attr_counts.get( attr_name, 0 ) + 1
if remove_dangerous:
logger.debug( f'Removing dangerous SVG attribute "{attr_name}"' )
del element.attrib[attr_name]
elif attr_name in HREF_ATTRS:
attr_value = element.attrib[attr_name].strip()
if not attr_value.startswith( '#' ):
dangerous_attr_counts[attr_name] = dangerous_attr_counts.get( attr_name, 0 ) + 1
if remove_dangerous:
logger.debug( f'Removing dangerous SVG href "{attr_name}={attr_value}"' )
del element.attrib[attr_name]
inner_content = ''.join(
ET.tostring( element, encoding='unicode' ) for element in root
)
svg_fragment_filename = os.path.join(
media_destination_directory,
generate_unique_filename( source_filename ),
)
return {
'svg_fragment_content': inner_content,
'svg_viewbox': svg_viewbox,
'svg_fragment_filename': svg_fragment_filename,
'dangerous_tag_counts': dangerous_tag_counts,
'dangerous_attr_counts': dangerous_attr_counts,
}