I have a three List in c# ,the variable names are l_lstData1, l_lstData2, l_lstData3
.
File structure is
<FileDetails>
<Date FileModified="29/04/2010 12:34:02" />
<Data Name="Data_1" DataList="India" Level="2" />
<Data Name="Data_2" DataList="chennai" Level="2" />
<Data Name="Data_3" DataList="hyderabad" Level="2" />
<Data Name="Data_4" DataList="calcutta" Level="2" />
<Data Name="开发者_开发技巧Data_5" DataList="vijayawada" Level="1" />
<Data Name="Data_6" DataList="cochin" Level="1" />
<Data Name="Data_7" DataList="madurai" Level="0" />
<Data Name="Data_8" DataList="trichy" Level="0" />
</FileDetails>
The Values of 3 Lists are as follows :
l_lstData1[0] = "India";
l_lstData1[1] = "chennai";
l_lstData1[2] = "hyderabad";
l_lstData1[3] = "calcutta";
so the level attribute of the above XML(element : Data) has value="2".
l_lstData2[0] = "vijayawada";
l_lstData2[1] = "cochin";
so the level attribute of the above XML(element : Data) has value="1".
l_lstData3[0] = "madurai";
l_lstData3[1] = "trichy";
so the level attribute of the above XML (element: Data) has value="0".
It's not clear exactly why the "Level" attributes are as specified, but this would create the relevant XML for you:
// Used for side-effects in the XElement constructor. This is a bit icky. It's
// not clear what the "Name" part is really for...
int count = 1;
var doc = new XDocument(
new XElement("FileDetails",
new XElement("Date", new XAttribute("FileModified", DateTime.UtcNow)),
l_lstData1.Select(x => new XElement("Data",
new XAttribute("Name", "Data_" + count++),
new XAttribute("DataList", x),
new XAttribute("Level", 2))),
l_lstData2.Select(x => new XElement("Data",
new XAttribute("Name", "Data_" + count++),
new XAttribute("DataList", x),
new XAttribute("Level", 1))),
l_lstData3.Select(x => new XElement("Data",
new XAttribute("Name", "Data_" + count++),
new XAttribute("DataList", x),
new XAttribute("Level", 0)))));
It would probably be neater if you could extract the projections from a list item to its element, but the "Data_" + count
bit makes that tricky. It's not clear why you need such a thing to be honest... if you could get away without that, the code could be cleaner.
I suppose one alternative would be to create the document without the Name
attributes, and then populate them afterwards. For example:
private static IEnumerable<XElement> ProjectList(IEnumerable<string> list,
int level)
{
return list.Select(x => new XElement("Data",
new XAttribute("DataList", x),
new XAttribute("Level", level)));
}
then:
var doc = new XDocument(
new XElement("FileDetails",
new XElement("Date", new XAttribute("FileModified", DateTime.UtcNow)),
ProjectList(l_lstData1, 2),
ProjectList(l_lstData2, 1),
ProjectList(l_lstData3, 0)));
int count = 1;
foreach (var element in doc.Descendants("Data"))
{
element.SetAttributeValue("Name", "Data_" + count++);
}
What about:
XDocument doc = new XDocument();
var total = (from a in list1 select new { Name = a, Level = 2 }).Concat(
from b in list2 select new { Name = b, Level = 1 }).Concat(
from c in list3 select new { Name = c, Level = 0 });
XElement root = new XElement("FileDetails", from i in Enumerable.Range(0, total.Count())
let element = total.ElementAt(i)
let name = new XAttribute("Name", String.Format("Data_{0}", i + 1))
let level = new XAttribute("Level", element.Level)
let datalist = new XAttribute("DataList", element.Name)
select new XElement("Data", name, datalist, level),
new XElement("Date", new XAttribute("FileModified", DateTime.Now)));
精彩评论