What is the XSL version supported by Xalan 1.8 library ?
The XSLT version that is supported by all versions of Xalan is XSLT 1.0.
What is the XSL version supported by Xalan 1.8 library ?
1.0
Just in case it changes in the future: generally, you can ask any XSLT processor to report on the version it supports with:
<xsl:value-of select="system-property('xsl:version')" />
Related
Apache Arrow in their documentation list that each release has two versions, a Library Version and a Format Version: https://arrow.apache.org/docs/format/Versioning.html
It appears that over the last year there have been 4 Library Versions, but it's hard to tell if the format version has changed in any of these Library Versions. Is there a way to tell what the Format Version is for a given Library Version?
Currently, I do not see a way to determine the format version programmatically. This seems like a good idea. I have opened ARROW-12594.
The latest format version is also listed on the specification page.
Since the latest version is 1.0.0 you can safely assume that all Arrow versions >= 1.0.0 and <= 6.0.0 are compiled against format version 1.0.0
I am new to xslt and we are using currently version 1.0, now we have to upgrade to version 3.0
what are the steps i have to follow? Is there any processor i need to replace or just all the xsl files will have new version like
<xsl:stylesheet version="3.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
Is ML.NET support on ARM64 yet?
If so, can someone provide a link or perhaps some steps to test this on ARM64, please?
thank you!
Yes, but I need to install .NET Core 2.1 or later.
While ML.NET packages work with both .NET Core 2.1 or later and .NET Framework 4.6.1, only .NET Core 2.1 (starting from the release 2.1.609) supports ARMx64 binaries.
References
https://github.com/dotnet/machinelearning
https://dotnet.microsoft.com/download/dotnet-core/2.1
Background
We have dependencies to several Xalan and Xerces versions in our rather large C++ system. This has accumulated over the years, and is not optimal for several reasons.
Now we're migrating from Solaris to Linux, and we think it makes sense to consolidate third party dependencies to a minimum.
We've determined that we still need Xerces 2.8 and Xerces 3.x. And Xalan 1.x
Target platform is RHEL6.
Problem
From RHEL repo we can get Xerces 3.0 and Xalan 1.10 (build against Xerces 3.0)
We've installed Xerces 2.8, under /usr/local...
We can't figure out how Xalan is supposed to work with Xerces (link wise). That is:
Do we need two Xalan libraries, one linked against 3.x and one against 2.8?
Can we have one Xalan version, let's say 1.11, that (by some magic) works with both the xerces versions?
We can't be the only one with this problem, or are we just idiots? (which would be the best scenario... :))
From what I've just read at xerces.apache.org , Xerces 2.X and Xerces 3.x have different API's and therefore you'll need to link with one OR the other.
And as you can see from xalan.apache.org:
"The Xalan-C/C++ 1.11 library requires the Xerces-C/C++ XML Parser 3.0 or newer, version 3.1.1 is preferred.", while Xalan 1.10 requires Xerces 2.7.0 (will probably work with 2.8.0).
So yes, you will need two different versions of Xalan libraries, unless you can resolve your dependencies to Xalan 1.11.0 only, in which case you can have only one installed Xalan and one Xerces.
Probably most of all problems of xerces + xalan are because of xerces definition of XML character "XMLCh" in Xerces_autoconf_config.hpp (generated header). I have been trying to build on windows xerces 3.2.2 and xalan 1.11, and faced types missmaching...
E.g., DirectoryEnumerator.hpp, struct FindFileStruct, has function:
const XalanDOMChar* getName() const
{
return name;
}
name is "wchar_t name[260];" from corecrt_wio.h, struct _wfinddata64i32_t (it is Windows part...).
XalanDOMChar is typedef of "XMLCh".
For some reason xerces 3.1.1 generates XMLCh as "wchar_t", but xerces 3.2.2 - as "char16_t". And xalan has some code that is oriented on "wchar_t"...
It seems they all run on Python 2.x.
(Actually I want a more widely-used engine that has Python 3.x support.)
Just released Jinja 2.3 has experimental support for Python 3. http://lucumr.pocoo.org/2010/2/10/jinja-2-3-released
If you don't mind using beta software, the newly-released moody-templates was written for Python 3 from the ground up.
https://github.com/etianen/moody-templates
Maybe very outdated, but Mako supports python 3.
The first search result is the templating wiki page; which includes at least one templating engine which says it works on Python 2.4, 2.5, 2.6 and 3.0.