Hi Eno,
In fact, I think this patch is not worth to commit as it optimizes
almost nothing, but makes code harder to read. If you think it's good,
please commit youself.
Regards,
Konstantin Triger
Atsushi Eno wrote:
Hi Kosta,
Maybe I had better take care of it by myself. I think there ar
Hi Kosta,
Maybe I had better take care of it by myself. I think there are
some cases that we can optimize. For example when tables are already
defined it does not require InferSchema and thus entire load won't
be required. However similar things also happen to XmlDataReader,
so they should be tak
Hi Eno,
I don't see it in mailing list so resending.
I made the changes we talked about, bt it turns out they probably won't
make a big diff (see attached patch)
The problem is that I have to collect almost always:
XmlReadMode.InferSchema: in order to be able to skip if "schema" is found.
Xml
Hi Eno,
I made the changes we talked about, bt it turns out they probably won't
make a big diff (see attached patch)
The problem is that I have to collect almost always:
XmlReadMode.InferSchema: in order to be able to skip if "schema" is found.
XmlReadMode.Auto: as I don't know what's gonna be
Forgot to mention. For "performance improvements", it is not
reliable without reproducible example code and compared results.
It is likely to happen that the performance improvements happen
only under GH and JProfiler (like we have seen as
http://bugzilla.ximian.com/show_bug.cgi?id=59537 ).
Atsu
Hello Konstantin,
Konstantin Triger wrote:
Hi Eno,
I committed the code before I saw your comments on Boris email. Actually
I missed it at all and Boris forwarded it to me. In any case I didn't
want to ignore it and would like, with your help, create the best
implementation possible.
The n
Hi Eno,
The main problem was that sometimes the schema was not a root element,
but its child. This is a common case when DataSet is serialized in .Net.
The existing ReadXml() failed in this case and did read the schema as a
content.
Possibly, what needs to be done is when a schema is found a
When schema inference is required, existing ReadXml() actually
did read the content into XmlDocument, inferred schema structures,
and then continued instance consumption using XmlNodeReader.
The patch just broke the performance beauty.
Anyways you committed the patch without further approval that
Hello Eno,
Some answers:
1. I need to collect the data into the XmlDocument to store the data
while infering schema. Maybe it can be done only with specific values of
XmlReadMode.
2. The WhitespaceHandling part is for optimization, as we ignore
whitespace anyway.
3. Are you sure that XML
Hi Boris,
Hello all
We're currently working on reimplementation of DataSet.ReadXml.
The need for this fist raised as a result of xml serialization bugs
in our System.WebServices testsiute, and supported by the will to
provide
more "clean" implementation.
Having clean implementation is a nic
Sorry.
The diff attached.
Boris
Atsushi Eno wrote:
Please attach the patch ;-)
Atsushi Eno
Boris Kirzner wrote:
Hello all
We're currently working on reimplementation of DataSet.ReadXml.
The need for this fist raised as a result of xml serialization bugs
in our System.WebServices testsiute
Hello all
We're currently working on reimplementation of DataSet.ReadXml.
The need for this fist raised as a result of xml serialization bugs in
our System.WebServices testsiute, and supported by the will to provide
more "clean" implementation.
Attached is a diff with current svn version.
The c
12 matches
Mail list logo
