performance problem with SelectSingleNod e

[John R.]

I have an application written in C# and i am using MS XML
DOM!

I have a document with the following structure (only the
<DicEntry> - Elements are important):

<NewDataSet xmlns:xsi="http ://www.w3.org/2001/XMLSchema-
instance">

[Christoph Schittko [MVP]]

Where are you using XmlDataDocument ? It is not obvious from your code.
Anyway, XmlDataDocument is not nearly as performance optimized as an
XPathDocument for example. Have you tried reading the Xml from the DataSet
into an XPath doc or a regular XmlDocument to do the merge?

--
HTH
Christoph Schittko [MVP]
Software Architect, .NET Mentor

"John R." <[e-mail]> wrote in message
news:10******** *************** *****@phx.gbl.. .

I have an application written in C# and i am using MS XML
DOM!

I have a document with the following structure (only the
<DicEntry> - Elements are important):

<NewDataSet xmlns:xsi="http ://www.w3.org/2001/XMLSchema-
instance">
.
<DicEntry TextKey="Test1" KeyDependency=" True">
<default>Test 1</default>
<de>Button Germanmanuell</de>
<it />
</DicEntry>
<DicEntry TextKey="Dialog 1" KeyDependency=" True">
<default>Dialog 1</default>
<de />
<it />
</DicEntry>
.
</NewDataSet>

I now want to merge such a document with a document of
type XmlDataDocument with the same structure in memory!

There is the main code:

XmlElement root = this.Dictionary Document.Docume ntElement;
//bei Formbezogene "TextKey" = "default"

ls = @"/*/DicEntry[@TextKey=" +
TTUtility.GetXP athConcatParamL ist(this.TextKe y) + "]
[@KeyDependency= " + TTUtility.GetXP athConcatParamL ist
(this.CurrentMe rgeNode.Attribu tes.GetNamedIte m
("KeyDependency ").Value.ToStri ng()) + "]/*[name()=" +
TTUtility.GetXP athConcatParamL ist(mNode.Name) + "]";

lNode = root.SelectSing leNode(ls);

The method GetXPathConcatP aramList builds a parameterlist
for the Xpath method concat.

/// <summary>
/// this method builds a parameterlist for the Xpath
method concat,
/// so that the method SelectSingleNod e can also work
(search) on
/// elements, attributes with double quotation marks
in their names.
/// </summary>
/// <param name="aValue">a string</param>
/// <returns>conc at condition for SelectSingleNod e
method</returns>
public static string GetXPathConcatP aramList(string
aValue)
{
char[] lSplit = new char[1] {'"'};
string[] lSplitValues={" "};
int lSplitValuesLen ;
string lReturn="";
string lValue="";
string lHelp="";

//replace &quot; in aValue with "
lValue = aValue.Replace( "&quot;", "\"");

//split with "
lSplitValues = lValue.Split(lS plit);
lSplitValuesLen = lSplitValues.Le ngth;
//if there is no double quotation mark then return
//and there is no build for concat
if(lSplitValues Len == 1)
{
lReturn = "\"" + aValue + "\"";
}
else //build parameterlist for method concat
{
for (int i = 0; i <= lSplitValuesLen - 1; i++)
{
if(i==lSplitVal uesLen - 1)
lHelp = lHelp + "\"" + lSplitValues[i] + "\"";
else
lHelp = lHelp + "\"" + lSplitValues[i] + "\""
+ "," + "'\"'" + ",";
}
lReturn = "concat(" + lHelp + ")";
}
return lReturn;
}
Performance results:

Merging 10 <DicEntry> - Elements into 100 in
memory => ca. 1 sec.
Merging 10 <DicEntry> - Elements into 500 in
memory => ca. 6 sec.
Merging 10 <DicEntry> - Elements into 1000 in
memory => ca. 12 sec.
Merging 100 <DicEntry> - Elements into 100 in
memory => ca. 21 sec.
Merging 100 <DicEntry> - Elements into 1000 in
memory => ca. 190 sec.
Merging 1000 <DicEntry> - Elements into 1000 in
memory => ca. over 3600 sec.

The method "SelectSingleNo de" takes time of 96 % of the
applications running time!!!!!!!!!!! !!!!!!!!!!

Please can you help on this problem?

thanks,
john

[John R.]

I have XML data in a XmlDataDocument and i want to merge
them into another existing XmlDataDocument .
I have to insert and update data in the existing
XmlDataDocument .

In my code this.Dictionary Document is of type
XmlDataDocument (the existing one).

I am using XmlDataDocument because i have a GUI where i
present the resulting data and a grid where i can bind the
dataset property of the XmlDataDocument !

with the code fragment:
"XmlElement root =
this.Dictionary Document.Docume ntElement;
ls = @"/*/DicEntry[@TextKey=" +
TTUtility.GetXP athConcatParamL ist(this.TextKe y) + "]
[@KeyDependency= " + TTUtility.GetXP athConcatParamL ist
(this.CurrentMe rgeNode.Attribu tes.GetNamedIte m
("KeyDependency ").Value.ToStri ng()) + "]/*[name()=" +
TTUtility.GetXP athConcatParamL ist(mNode.Name) + "]";
lNode = root.SelectSing leNode(ls);"

....i get the node in the existing XmlDataDocument that i
want to update (before i testet if there is one!)

thanks,
John

-----Original Message-----
Where are you using XmlDataDocument ? It is not obvious from your code.Anyway, XmlDataDocument is not nearly as performance optimized as anXPathDocumen t for example. Have you tried reading the Xml from the DataSetinto an XPath doc or a regular XmlDocument to do the merge?
--
HTH
Christoph Schittko [MVP]
Software Architect, .NET Mentor

"John R." <[e-mail]> wrote in message
news:10******* *************** ******@phx.gbl. ..

I have an application written in C# and i am using MS XML DOM!

I have a document with the following structure (only the
<DicEntry> - Elements are important):

<NewDataSet xmlns:xsi="http ://www.w3.org/2001/XMLSchema-
instance">
.
<DicEntry TextKey="Test1" KeyDependency=" True">
<default>Test 1</default>
<de>Button Germanmanuell</de>
<it />
</DicEntry>
<DicEntry TextKey="Dialog 1" KeyDependency=" True">
<default>Dialog 1</default>
<de />
<it />
</DicEntry>
.
</NewDataSet>

I now want to merge such a document with a document of
type XmlDataDocument with the same structure in memory!

There is the main code:

XmlElement root = this.Dictionary Document.Docume ntElement; //bei Formbezogene "TextKey" = "default"

ls = @"/*/DicEntry[@TextKey=" +
TTUtility.GetXP athConcatParamL ist(this.TextKe y) + "]
[@KeyDependency= " + TTUtility.GetXP athConcatParamL ist
(this.CurrentMe rgeNode.Attribu tes.GetNamedIte m
("KeyDependency ").Value.ToStri ng()) + "]/*[name()=" +
TTUtility.GetXP athConcatParamL ist(mNode.Name) + "]";

lNode = root.SelectSing leNode(ls);

The method GetXPathConcatP aramList builds a parameterlist for the Xpath method concat.

/// <summary>
/// this method builds a parameterlist for the Xpath
method concat,
/// so that the method SelectSingleNod e can also work (search) on
/// elements, attributes with double quotation marks
in their names.
/// </summary>
/// <param name="aValue">a string</param>
/// <returns>conc at condition for SelectSingleNod e
method</returns>
public static string GetXPathConcatP aramList(string
aValue)
{
char[] lSplit = new char[1] {'"'};
string[] lSplitValues={" "};
int lSplitValuesLen ;
string lReturn="";
string lValue="";
string lHelp="";

//replace " in aValue with "
lValue = aValue.Replace( """, "\"");

//split with "
lSplitValues = lValue.Split(lS plit);
lSplitValuesLen = lSplitValues.Le ngth;
//if there is no double quotation mark then return
//and there is no build for concat
if(lSplitValues Len == 1)
{
lReturn = "\"" + aValue + "\"";
}
else //build parameterlist for method concat
{
for (int i = 0; i <= lSplitValuesLen - 1; i++)
{
if(i==lSplitVal uesLen - 1)
lHelp = lHelp + "\"" + lSplitValues[i] + "\""; else
lHelp = lHelp + "\"" + lSplitValues[i] + "\"" + "," + "'\"'" + ",";
}
lReturn = "concat(" + lHelp + ")";
}
return lReturn;
}
Performance results:

Merging 10 <DicEntry> - Elements into 100 in memory => ca. 1 sec.
Merging 10 <DicEntry> - Elements into 500 in memory => ca. 6 sec.
Merging 10 <DicEntry> - Elements into 1000 in
memory => ca. 12 sec.
Merging 100 <DicEntry> - Elements into 100 in memory => ca. 21 sec.
Merging 100 <DicEntry> - Elements into 1000 in
memory => ca. 190 sec.
Merging 1000 <DicEntry> - Elements into 1000 in
memory => ca. over 3600 sec.

The method "SelectSingleNo de" takes time of 96 % of the
applications running time!!!!!!!!!!! !!!!!!!!!!

Please can you help on this problem?

thanks,
john

.

[Oleg Tkachenko]

John R. wrote:

I have XML data in a XmlDataDocument and i want to merge
them into another existing XmlDataDocument .
I have to insert and update data in the existing
XmlDataDocument .

Obviously you can get better XPath expression evaluation performance using
XPathDocument instead, which is specially optimized for that. But you cannot
update XPathDocument, because it's read-only by definition.
Also you can try to simplify your XPath expression.
--
Oleg Tkachenko
http://www.tkachenko.com/blog
Multiconn Technologies, Israel