Thanks for the hint regarding parsers.
After spending a few hours trying to understand what special tricks I needed to
put in a resolver, I realized that there were none. The resolver juts needs to
fetch the data (I would have expected the Parser to do this itself)
schema4=io.StringIO('''<schema xmlns="http://www.w3.org/2001/XMLSchema";
xmlns:patch="urn:paulhiggs:my-patch" targetNamespace="urn:paulhiggs:my-patch"
elementFormDefault="qualified" attributeFormDefault="unqualified">
<include
schemaLocation="https://www.iana.org/assignments/xml-registry/schema/patch-ops.xsd"/>
<element name="Patch" type="patch:PatchType"/>
<complexType name="PatchType">
<choice minOccurs="1" maxOccurs="unbounded">
<element name="add" type="patch:add"/>
<element name="remove" type="patch:remove"/>
<element name="replace" type="patch:replace"/>
</choice>
<attribute name="paulsAttrib" type="string" use="required"/>
</complexType>
</schema>''')
import requests
class PrefixResolver(etree.Resolver):
# https://lxml.de/resolvers.html
def __init__(self, prefix):
self.prefix = prefix.lower()
def resolve(self, url, pubid, context):
if url.lower().startswith(self.prefix):
res=requests.get(url, allow_redirects=True)
return self.resolve_string(res.text, context)
parser=etree.XMLParser(load_dtd=True, no_network=False, huge_tree=True,
resolve_entities=True)
parser.resolvers.add( PrefixResolver("https") )
parser.resolvers.add( PrefixResolver("http") )
my_schema=etree.XMLSchema(etree.parse(schema4, parser))
This now works OK for an include of HTTP and HTTPS!
I need to look into the workings of libxml2 to see if loading for INCLUDE and
IMPORT are somehow handled differently – I have never had a problem with an
HTTP or HTTPS IMPORT
Paul
-----Original Message-----
From: Paul Higgs <paul_hi...@hotmail.com>
Sent: 25 June 2021 14:51
To: holger.jo...@lbbw.de; l...@lxml.de
Subject: [lxml] Re: Cannot <include> from a network location
Thanks Holger
First, I am not behind any proxy - CURL to http: and https: both give the schema
Second, Your test_schema.py using http: also works for me, however is
https:/www.loc.gov... is used then I get an error
>python test_schema.py
Traceback (most recent call last):
File "G:\lxml-test\test_schema.py", line 14, in <module>
schema = etree.XMLSchema(tree)
File "src\lxml\xmlschema.pxi", line 88, in lxml.etree.XMLSchema.__init__
lxml.etree.XMLSchemaParseError: Element
'{http://www.w3.org/2001/XMLSchema}include': Failed to load the document
'https://www.loc.gov/standards/xlink/xlink.xsd' for inclusion., line 8
Third, 'https://www.loc.gov/standards/xlink/xlink.xsd' does exist - CURL
retrieves it OK
>curl --head https://www.loc.gov/standards/xlink/xlink.xsd
HTTP/2 200
date: Fri, 25 Jun 2021 13:44:27 GMT
content-type: text/xml
content-length: 3180
last-modified: Thu, 23 Aug 2007 19:02:01 GMT
etag: "119c982-c6c-867afc40"
accept-ranges: bytes
cf-cache-status: DYNAMIC
cf-request-id: 0ae5033c37000054ab1ab16000000001
expect-ct: max-age=604800,
report-uri="https://report-uri.cloudflare.com/cdn-cgi/beacon/expect-ct";
server: cloudflare
cf-ray: 664ea1738c7e54ab-MAN
>curl --head http://www.loc.gov/standards/xlink/xlink.xsd
HTTP/1.1 200 OK
Date: Fri, 25 Jun 2021 13:45:16 GMT
Content-Type: text/xml
Content-Length: 3180
Connection: keep-alive
Last-Modified: Thu, 23 Aug 2007 19:02:01 GMT
ETag: "119c982-c6c-43862867afc40"
Accept-Ranges: bytes
X-Frame-Options: deny
Set-Cookie: HttpOnly
CF-Cache-Status: DYNAMIC
cf-request-id: 0ae503f9cd0000000a56ab5000000001
Server: cloudflare
CF-RAY: 664ea2a2ed37000a-MAN
Paul
-----Original Message-----
From: holger.jo...@lbbw.de<mailto:holger.jo...@lbbw.de>
<holger.jo...@lbbw.de<mailto:holger.jo...@lbbw.de>>
Sent: 25 June 2021 13:31
To: l...@lxml.de<mailto:l...@lxml.de>
Subject: [lxml] Re: Cannot <include> from a network location
> Just a thought: Might this be proxy- or https-related? Does it work if you
> locally serve the xs:included schema with http?
>
> I *think* libxml2 respects http_proxy but I don’t know anything about https
> support.
Just found that for an example adapted from
https://bugs.launchpad.net/lxml/+bug/1234114/comments/3
(xs:include instead of xs:import):
##############
# test_schema.py
XSD = b"""<?xml version="1.0" encoding="UTF-8"?> <xs:schema
xmlns:xs="http://www.w3.org/2001/XMLSchema";
xmlns:xlink="http://www.w3.org/1999/xlink";
xmlns="lxmltest"
targetNamespace="http://www.w3.org/1999/xlink";
elementFormDefault="qualified"
attributeFormDefault="unqualified">
<xs:include schemaLocation="http://www.loc.gov/standards/xlink/xlink.xsd";
/> </xs:schema> """
from lxml import etree
parser = etree.XMLParser(
load_dtd=True, no_network=True, huge_tree=True, resolve_entities=True) tree
= etree.fromstring(XSD, parser=parser) schema = etree.XMLSchema(tree)
print(schema)
##############
I can successfully run this if the xs:include location is http but not https
(the xlink.xsd is available both with https and http URLs).
If I change it to https I get
Traceback (most recent call last):
File "test_schema.py", line 16, in <module>
schema = etree.XMLSchema(tree)
File "src/lxml/xmlschema.pxi", line 88, in lxml.etree.XMLSchema.__init__
lxml.etree.XMLSchemaParseError: Element
'{http://www.w3.org/2001/XMLSchema}include': Failed to load the document
'https://www.loc.gov/standards/xlink/xlink.xsd' for inclusion., line 9
As I'm behind a proxy:
I also just found out that while curl happily accepts
http_proxy=my.proxy.address.net:8080 lxml (libxml2) only works if this is set
to http_proxy=http://my.proxy.address.net:8080 i.e. with an explicit <scheme>://
Cheers, H.
Landesbank Baden-Wuerttemberg
Anstalt des oeffentlichen Rechts
Hauptsitze: Stuttgart, Karlsruhe, Mannheim, Mainz HRA 12704 Amtsgericht
Stuttgart HRA 4356, HRA 104 440 Amtsgericht Mannheim HRA 40687 Amtsgericht Mainz
Die LBBW verarbeitet gemaess Erfordernissen der DSGVO Ihre personenbezogenen
Daten.
Informationen finden Sie unter https://www.lbbw.de/datenschutz.
_______________________________________________
lxml - The Python XML Toolkit mailing list --
lxml@python.org<mailto:lxml@python.org> To unsubscribe send an email to
lxml-le...@python.org<mailto:lxml-le...@python.org>
https://mail.python.org/mailman3/lists/lxml.python.org/
Member address: paul_hi...@hotmail.com<mailto:paul_hi...@hotmail.com>
_______________________________________________
lxml - The Python XML Toolkit mailing list --
lxml@python.org<mailto:lxml@python.org> To unsubscribe send an email to
lxml-le...@python.org<mailto:lxml-le...@python.org>
https://mail.python.org/mailman3/lists/lxml.python.org/
Member address: paul_hi...@hotmail.com<mailto:paul_hi...@hotmail.com>
_______________________________________________
lxml - The Python XML Toolkit mailing list -- lxml@python.org
To unsubscribe send an email to lxml-le...@python.org
https://mail.python.org/mailman3/lists/lxml.python.org/
Member address: arch...@mail-archive.com
