jaseg/gerbolyze
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Move from bs4 to etree

Browse source 
BeautifulSoup when using lxml in XML mode would mis-parse XML with very
long attributes. Specifically, a <polygon> with about 18MB in its points
attr would make lxml not return anything past that point in the file.

bs4 uses lxml, which uses libxml2. libxml2 has a config option for
parsing "huge" files that increases buffer sizes and avoids this error,
and this option is exposed in lxml, but AFAICT you can't tell bs4 to set
it, and bs4 just silently swallows the error from lxml.

Fixes one half of #46
This commit is contained in:
jaseg 2024-07-05 11:13:25 +02:00
parent 534c2e4ea3
commit bd2b3733ee
1 changed files with 14 additions and 7 deletions
Download patch file Download diff file Expand all files Collapse all files

21
gerbolyze/__init__.py
View file

@ -14,8 +14,8 @@ import warnings
import shutil
from zipfile import ZipFile, is_zipfile
from pathlib import Path
from xml.etree import ElementTree
from bs4 import BeautifulSoup
import numpy as np
import click
@ -23,6 +23,11 @@ import gerbonara as gn
__version__ = '3.1.8'
ET_NS= {'svg': 'http://www.w3.org/2000/svg',
'inkscape': 'http://www.inkscape.org/namespaces/inkscape'}
SVG_NS = f'{{{ET_NS["svg"]}}}'
INKSCAPE_NS = f'{{{ET_NS["inkscape"]}}}'
@click.group()
def cli():
pass
@ -71,16 +76,17 @@ def paste(input_gerbers, input_svg, output_gerbers, is_zip,
with tempfile.NamedTemporaryFile(suffix='.svg') as processed_svg:
run_cargo_command('usvg', *shlex.split(os.environ.get('USVG_OPTIONS', '')), input_svg, processed_svg.name)
with open(processed_svg.name) as f:
soup = BeautifulSoup(f.read(), features='xml')
et = ElementTree.parse(processed_svg)
for (side, use), layer in [
*stack.graphic_layers.items(),
(('drill', 'plated'), stack.drill_pth),
(('drill', 'nonplated'), stack.drill_npth)]:
logging.info(f'Layer {side} {use}')
if (soup_layer := soup.find('g', id=f'g-{side}-{use}')):
if not soup_layer.contents:
et_layer = et.find(f".//{SVG_NS}g[@id='g-{side}-{use}']")
if et_layer is not None:
if not len(et_layer):
logging.info(f' Corresponding overlay layer is empty. Skipping.')
else:
logging.info(f' Corresponding overlay layer not found. Skipping.')
@ -271,8 +277,9 @@ def convert(input_svg, output_gerbers, is_zip, dilate, curve_tolerance, subtract
with tempfile.NamedTemporaryFile(suffix='.svg') as processed_svg:
run_cargo_command('usvg', *shlex.split(os.environ.get('USVG_OPTIONS', '')), input_svg, processed_svg.name)
soup = BeautifulSoup(input_svg.read_text(), features='xml')
layers = {e.get('id'): e.get('inkscape:label') for e in soup.find_all('g', recursive=True)}
et = ElementTree.fromstring(input_svg.read_text())
layers = {node.get(f'id'): node.get(f'{INKSCAPE_NS}label')
for node in et.findall(f'{SVG_NS}g')}
stack = gn.LayerStack({}, None, None, [], board_name=input_svg.stem, original_path=input_svg)