libxml2-text-input-testcase.py

#!/usr/bin/env python3

import io
import libxml2
import xml.sax.xmlreader
from drv_libxml2 import LibXml2Reader

def parse_binary():
  source = xml.sax.xmlreader.InputSource()
  source.setByteStream(io.BytesIO(b'<?xml version="1.0" encoding="UTF-8"?>\n<root />\n'))
  return LibXml2Reader().parse(source)

def parse_text():
  source = xml.sax.xmlreader.InputSource()
  source.setCharacterStream(io.StringIO('<?xml version="1.0"?>\n<root />\n'))
  return LibXml2Reader().parse(source)

# Succeeds (prints "None")
print(parse_binary())

# Fails, and prints:
#
#   xmlPythonFileRead: result is not a String
#   <unknown>:1:-1: Unknown IO error
#
#   None
#
print(parse_text())

# Root cause is that LibXml2Reader.parse() does [1]:
#
#  input = libxml2.inputBuffer(source.getByteStream())
#
# but since source.getByteStraem() returns None, this creates an ioWrapper that
# only ever returns -1 [2], triggering the "result is not a String" errror [3].
#
# 1. https://github.com/GNOME/libxml2/blob/master/python/drv_libxml2.py#L149
# 2. https://github.com/GNOME/libxml2/blob/67ff748c3eba93745a09157e11759d09f864492f/python/libxml.py#L73-L74
# 3. https://github.com/GNOME/libxml2/blob/67ff748c3eba93745a09157e11759d09f864492f/python/libxml.c#L333