NumesSanguis NumesSanguis - 1 month ago 11
Python Question

Python lxml parsing svg file

I'm trying to parse .svg files from http://kanjivg.tagaini.net/ , but I can't successfully extract the information inside.

Edit 1:(full file) http://www.filedropper.com/0f9ab

A part of

0f9ab.svg
looks like this:

<svg xmlns="http://www.w3.org/2000/svg" width="109" height="109" viewBox="0 0 109 109">
<g id="kvg:StrokePaths_0f9ab" style="fill:none;stroke:#000000;stroke-width:3;stroke-linecap:round;stroke-linejoin:round;">
<g id="kvg:0f9ab" kvg:element="嶺">
<g id="kvg:0f9ab-g1" kvg:element="山" kvg:position="top" kvg:radical="general">
<path id="kvg:0f9ab-s1" kvg:type="㇑a" d="M53.26,9.38c0.99,0.99,1.12,2.09,1.12,3.12c0,0.67,0.06,8.38,0.06,13.01"/>
<path id="kvg:0f9ab-s2" kvg:type="㇄a"
</g>
</g>
</g>


My .py file:

import lxml.etree as ET

svg = ET.parse('0f9ab.svg')
print(svg) # <lxml.etree._ElementTree object at 0x7f3a2f659ec8>

# AttributeError: 'lxml.etree._ElementTree' object has no attribute 'tag'
print(svg.tag)

# TypeError: 'lxml.etree._ElementTree' object is not subscriptable
print(svg[0])

# TypeError: 'lxml.etree._ElementTree' object is not iterable
for child in svg:
print(child)

# None
print(svg.find("./svg"))

# []
print(svg.findall("//g"))

# []
print(svg.xpath("//g"))


Purpose



I tried all kinds of operations I could think of, but nothing gets me any data from the .svg file.
I want to extract the kanji (Japanese character) in
kvg:element="kanji"
(which are at different depth levels).

Question




  1. Is using
    lxml
    the wrong package for this?

  2. If not, how do I extract information from my parsed .svg file?



Other solution




  • I could of course I could just read the file as a string and search
    for
    kvg:element="
    , but I would like to proper way of extracting xml
    / svg.

  • I used
    xmltodict
    before, but my code became really messy extracting
    kvg:element
    , because they were at different depth levels.


Answer

.parse() returns an ElementTree, which represents the tree as a whole. To query individual nodes, you need an Element, most likely the root element of the tree.

Replace part of your code with this:

xml = ET.parse('0f9ab.svg')
svg = xml.getroot()
print(svg)  # <lxml.etree._ElementTree object at 0x7f3a2f659ec8>

and I think you'll have some success.

Note also that .findall() requires a relative path:

print(svg.findall(".//g"))