We Keep Coding

Sort List by date values

I have the following list -
List<string> finalMessageContent
where
finalMessageContent[0] = "<div class="mHr" id="mFID">
<div id="postedDate">11/12/2015 11:12:16</div>
</div>" // etc etc
I am trying to sort the list by a particular value located in the entires - postedDate tag.
Firstly I have create an new object and then serialized it to make the html elements able to be parsed -
string[][] newfinalMessageContent = finalMessageContent.Select(x => new string[] { x }).ToArray();
string json = JsonConvert.SerializeObject(newfinalMessageContent);
JArray markerData = JArray.Parse(json);
And then used Linq to try and sort using OrderByDescending -
var items = markerData.OrderByDescending(x => x["postedDate"].ToString()).ToList();
However this is failing when trying to parse the entry with -
Accessed JArray values with invalid key value: "postedDate". Array position index expected.
Perhaps linq is not the way to go here however it seemed like the most optimised, where am I going wrong?

First, i would not use string methods, regex or a JSON-parser to parse HTML. I would use HtmlAgilityPack. Then you could provide such a method:
private static DateTime? ExtractPostedDate(string inputHtml, string controlID = "postedDate")
{
var doc = new HtmlAgilityPack.HtmlDocument();
doc.LoadHtml(inputHtml);
HtmlNode div = doc.GetElementbyId(controlID);
DateTime? result = null;
DateTime value;
if (div != null && DateTime.TryParse(div.InnerText.Trim(), DateTimeFormatInfo.InvariantInfo, DateTimeStyles.None, out value))
result = value;
return result;
}
and following LINQ query:
finalMessageContent = finalMessageContent
.Select(s => new { String = s, Date = ExtractPostedDate(s) })
.Where(x => x.Date.HasValue)
.OrderByDescending(x => x.Date.Value)
.Select(x => x.String)
.ToList();

Don't know if I get your question right.
But did you know that you can parse HTML with XPath?
foreach (var row in doc.DocumentNode.SelectNodes("//div[#id="postedDate"]"))
{
Console.WriteLine(row.InnerText);
}
this is just an example from the top of my head you might have to double-check the XPath query depending on your document. You can also consider converting it to array or parsing the date and do other transformations with it.
Like I said this is just from the top of my head. Or if the html is not so compley consider to extract the dates with an RegEx but this would be a topic for another question.
HTH

Json Serializer serializes JSON typed strings. Example here to json
To parse HTML I suggest using HtmlAgility https://htmlagilitypack.codeplex.com/
Like this:
HtmlAgilityPack.HtmlDocument htmlparsed = new HtmlAgilityPack.HtmlDocument();
htmlParsed.LoadHtml(finalMessageContent[0]);
List<HtmlNode> OrderedDivs = htmlParsed.DocumentNode.Descendants("div").
Where(a => a.Attributes.Any(af => af.Value == "postedDate")).
OrderByDescending(d => DateTime.Parse(d.InnerText)); //unsafe parsing

Using Linq to XML, how can I select the elements that have exactly x number of parent elements?

I am trying to use Linq to XML in my c# program to extract element names from an auto-generated XSD file. This means I cannot do queries based on element names/ contents, and instead need to find elements based on their position in the xsd.
I'm trying to do this with a function that returns an element's 'depth'. Somehow, if I use it in my query, I get not only the elements I need but also all underlying elements/attributes. I am guessing I need an "exclude" function in some way, but I don't know how to do this. This is the code I have at the moment:
static public void XsdReader()
{
var xsd = XDocument.Load(#"c:\XsdToLoad.xml");
var elementsAtDepthEight = from e in xsd.Descendants()
where GetElementDepth(e) == 8
select e;
foreach (var p in elementsAtDepthEight)
{
Console.WriteLine(p.ToString());
}
}
static int GetElementDepth(XElement element)
{
int result = 1;
//always return 1 as the root node has depth of 1.
//in addition, return the maximum depth returned by this method called on all the children.
if (element.Parent != null)
{
result += GetElementDepth(element.Parent);
}
return result;
}

The simplest way to measure "depth" is to count ancestors:
var elementsAtDepth8 = xsd.Descendants()
.Where(x => x.Ancestors().Count() == 8);
No need to write your own recursive code at all.
Another slightly odd alternative would be to write:
var elementsAtDepth = xsd.Elements()
.Elements()
.Elements()
.Elements()
.Elements()
.Elements()
.Elements()
.Elements();
You could obviously write a pair of recursive methods to do this:
public static IEnumerable<XElement> ElementsAtDepth(this XNode node, int depth)
{
return node.Elements().ElementsAtDepth(depth - 1);
}
public static IEnumerable<XElement> ElementsAtDepth(
this IEnumerable<XElement> elements, int depth)
{
// TODO: Validate that depth >= 0
return depth == 0 ? elements : elements.Elements().ElementsAtDepth(depth - 1);
}

Linq-to-xml to get child nodes

I am having trouble identifying how to use linq-to-xml to extract total price and individual prices from the xml below (e.g I want to get the fare price and also sum of all prices). Any help would be much appreciated especially with using the method syntax of linq-to-xml
I use the following code to get the data loaded into an xDocument and work with the xmlResponse object to parse the response.
var xmlResponse = from element in xdoc.Descendants()
select element;
and get data like
xmlResponse.SingleOrDefault(x => x.Name.LocalName == "Registration")
Below is a subset of thwe xml response :-
<StateList>
<State>
<SourceJobID>J999999999999</SourceJobID>
<TargetJobState>Complete</TargetJobState>
<TargetJobID>11111111</TargetJobID>
<TargetSystem>TESTSYSTEM</TargetSystem>
<VehicleDetails>
<Registration>TESTREGISRATION</Registration>
<Plate>11111111111</Plate>
<CO2Rating>160</CO2Rating>
<Badge>1111111</Badge>
<Description>TEST DESCRIPTION</Description>
</VehicleDetails>
<CompleteDetails>
<CompletedOn>2015-09-15T13:39:11+01:00</CompletedOn>
<JobDistance>0</JobDistance>
<WaitingTime />
<CO2Usage>0</CO2Usage>
<ChargeList>
<Charge>
<Name>Airport Pickup</Name>
<Currency>GBP</Currency>
<Price>0.00</Price>
</Charge>
<Charge>
<Name>Fare</Name>
<Currency>GBP</Currency>
<Price>0.00</Price>
</Charge>
<Charge>
<Name>Extra Stops</Name>
<Currency>GBP</Currency>
<Price>0.00</Price>
</Charge>
</ChargeList>
</CompleteDetails>
</State>

Assuming you only have a single state like in your example, you could do something like the following:
decimal fare = decimal.Parse(xml.Descendants("Charge").Single(x => x.Element("Name").Value == "Fare").Element("Price").Value);
decimal total = xml.Descendants("Charge").Sum(x => decimal.Parse(x.Element("Price").Value));
Although if you have a series of elements in your list you will have to modify that.
EDIT: If, as you say in the comments, you would like to sum only certain charges:
// Valid names of charges to sum.
string[] names = { "Airport Pickup", "Fare" };
// Iterate over every state.
foreach (var state in xml.Descendants("State"))
{
// Get all charge elements in the current state whose names are contained in 'names' - then convert their 'Price' element to decimal and sum them.
decimal stateTotal = state.Descendants("Charge").Where(x => names.Contains(x.Element("Name").Value)).Sum(x => decimal.Parse(x.Element("Price").Value));
}

if(doc.Descendants("Charge").Any())
{
var FarePrice = doc.Descendants("Charge")
.Where(x => x.Descendants("Name").First().Value.Equals("Fare")).First().Element("Price").Value;
var Sum = doc.Descendants("Charge")
.Select(x => Convert.ToDouble(x.Descendants("Price").First().Value))
.Sum();
Console.WriteLine("Fare price:{0} Sum:{1}",FarePrice,Sum);
}
It returns 35 as sum for 10 and 25 inputs.
Fiddle here : https://dotnetfiddle.net/cuHXBn

C# Reading certain element / attribute from xml file

Im trying to read a certain attributes from following xml file (as console program)
http://api.openweathermap.org/data/2.5/forecast?q=lahti,fin&mode=xml
As you see inside 'forecast' element there are multiple 'time' elements. What I want is to pick certain 'time' element and then pick given thing inside of it (lets say 'symbol') and print all/any attributes it has.
I want to be able to control which 'time' element I pick and which attributes I want to print.
This far all I have managed to do is to print every 'time' element and their attributes and also I managed to print every attribute inside of given 'time' element. But I just can't figure how to control it.
With the following code, I can print everything inside the first 'time' element. Item(0) is the index of element and the for loop makes sure that I don't get empty lines. As you can see from xml file, some 'time' elements has different amount of attributes inside of them so I guess I need to call them By name insted of index.
static void xmlReader()
{
int i;
XmlDocument xmlDoc = new XmlDocument();
xmlDoc.Load(parseLink());
foreach (XmlNode xmlNode in xmlDoc.DocumentElement.GetElementsByTagName("time").Item(0))
for (i = 0; i < xmlNode.Attributes.Count; i++)
{
Console.WriteLine(xmlNode.Attributes[i].Value);
}
Console.ReadKey();
}

Use Linq2Xml, it's much easier and convenient.
public static void Main()
{
var forecast = XDocument.Load(#"http://api.openweathermap.org/data/2.5/forecast?q=lahti,fin&mode=xml")
.Root
.Element("forecast");
foreach (var time in forecast.Elements("time")
.Where(e => e.Element("clouds")
.Attribute("value")
.Value == "overcast clouds"))
{
Console.WriteLine(time.Element("symbol").Attribute("name").Value);
}
}

Using XDocument is a bit easier here...
private static void XmlOutput()
{
var filePathAndName = #"http://api.openweathermap.org/data/2.5/forecast?q=lahti,fin&mode=xml";
var xmlDoc = XDocument.Load(filePathAndName);
// Get list of Nodes matching the "time" name or any other filters you wish.
var nodes = xmlDoc.Descendants().Where(nd => nd.Name.LocalName == "time");
// Filter the node list to only those where the date is as specified (and any other filters you wish).
// This could be done in the initial linq query - just making it clearer here.
nodes = nodes.Where(nd => nd.Attributes().Any(cnd => cnd.Name.LocalName == "from" && cnd.Value.Contains("2015-03-07")));
foreach (XElement element in nodes)
{
// For each descendant node where named "symbol"...
foreach(var node in element.Descendants().Where(nd => nd.Name.LocalName == "symbol"))
{
// Write out these particular attributes ("number" and "name")
string output = "";
output += node.Attributes().FirstOrDefault(nd => nd.Name.LocalName == "number").Value;
output += ", " + node.Attributes().FirstOrDefault(nd => nd.Name.LocalName == "name").Value;
Console.WriteLine(output);
}
}
Console.ReadKey();
}

Similar #aush's response, but with some formatting
var doc = XDocument.Load(#"http://api.openweathermap.org/data/2.5/forecast?q=lahti,fin&mode=xml");
var forecastEl = doc.Root.Element(XName.Get("forecast"));
var timeNodes = from e in forecastEl.Elements("time")
where e.Element("symbol")
.Attribute(XName.Get("name"))
.Value == "light snow"
select e;
var colFormat = "{0,-20} {1,-20} {2,-30}";
Console.WriteLine(colFormat, "TimeFrom", "TimeTo", "SymbolName");
foreach(var time in timeNodes)
Console.WriteLine(colFormat
, time.Attribute("from").Value
, time.Attribute("to").Value
, time.Element("symbol").Attribute("name").Value);
Results:
TimeFrom TimeTo SymbolName
2015-03-07T12:00:00 2015-03-07T15:00:00 light snow
2015-03-07T15:00:00 2015-03-07T18:00:00 light snow

You can use XmlReader (tutorial) as flowing,
Here I get from attribute from time element and name attribute sybol element.. And they are for same Time element. Also added extracting example for value of cloud
using (XmlReader reader = XmlReader.Create(#"http://api.openweathermap.org/data/2.5/forecast?q=lahti,fin&mode=xml"))
{
// Walk through all Elements
while (reader.Read())
{
// If we meet time element ; go inside and walk
if (reader.Name == "time")
{
Console.WriteLine("A new TIME ");
// Extract from attribute
String from = reader["from"];
if (from != null)
{
Console.WriteLine("\t From : " + from);
}
// Now walk through all elements inside same Time element
// Here I use do-while ; what we check is End element of time : </time> .. when we walk till we meet </time> we are inside children of same Time
// That mean we start from <time> and walk till we meet </time>
do
{
reader.Read();
if (reader.Name == "symbol")
{
// You can use this approach for any Attribute in symbol Element
String name = reader["name"];
if (name != null)
{
Console.WriteLine("\t Symbol name :" + name);
}
}
if (reader.Name == "clouds")
{
String clouds = reader["value"];
if (clouds != null)
{
Console.WriteLine("\t\t Clouds value : " + clouds);
}
}
} while (reader.NodeType != XmlNodeType.EndElement && reader.Name != "time");
}
}
}
Simply try this on your Console program..

Using my xml library here, you can search for an actual DateTime value like this:
XElement root = XElement.Load(#"http://api.openweathermap.org/data/2.5/forecast?q=lahti,fin&mode=xml");
var search = DateTime.Parse("2015-03-07T21:00:00");
XElement time = root.XPathElement("//time[#from >= {0} and #to > {1}]", search, search);
var value = time.ToString();

How to write XPath expression to select node name from its value

I'm trying to write an XPath expression to select the name of a node from its value in "qualities" and then select in "qualityNames" the value inside node whose name has previously captured.
E.g. In "qualities" - got value "4", take name "rarity3" then in "qualityNames" I got node named "rarity3" and take value "amazingrarity"
<result>
<status>1</status>
<qualities>
<Normal>0</Normal>
<rarity1>1</rarity1>
<rarity2>2</rarity2>
<vintage>3</vintage>
<rarity3>4</rarity3>
<rarity4>5</rarity4>
</qualities>
<qualityNames>
<Normal>Normal</Normal>
<rarity1>Genuine</rarity1>
<rarity2>rarity2</rarity2>
<vintage>Vintage</vintage>
<rarity3>amazingrarity</rarity3>
<rarity4>Unusual</rarity4>
</qualityNames>
</result>
I'm doing this in C# (It's a MVC App) and I'd prefer to use XPath because I'm indexing the XML and I haven't found a fastest way to query in-memory technique (this XML file has ~3MB and I'm using IndexingXPathNavigator).

Use the local-name() and text() functions + predicates. For value "4" it will be
//qualityNames/*[local-name()=local-name(//qualities/*[text() = '4'])]
Tested with http://www.xpathtester.com

Sounds like you want to create a dictionary of key/value pairs (assuming the node names are only needed to find matches and aren't important to your code).
If so, you can use the following:
var doc = XElement.Parse(#"<result>
<status>1</status>
<qualities>
<Normal>0</Normal>
<rarity1>1</rarity1>
<rarity2>2</rarity2>
<vintage>3</vintage>
<rarity3>4</rarity3>
<rarity4>5</rarity4>
</qualities>
<qualityNames>
<Normal>Normal</Normal>
<rarity1>Genuine</rarity1>
<rarity2>rarity2</rarity2>
<vintage>Vintage</vintage>
<rarity3>amazingrarity</rarity3>
<rarity4>Unusual</rarity4>
</qualityNames>
</result>");
var query = from quality in doc.XPathSelectElements("qualities/*")
join qualityName in doc.XPathSelectElements("qualityNames/*")
on quality.Name equals qualityName.Name
select new { Key = quality.Value, Value = qualityName.Value };
var qualities = query.ToDictionary(a => a.Key, a => a.Value);
var quality3 = qualities["3"];
// quality3 == "Vintage"
var quality4 = qualities["4"];
// quality4 == "amazingrarity"
EDIT: example of how to cache this dictionary
// add reference to System.Web dll
public Dictionary<string, string> GetQualities()
{
// assuming this code is in a controller
var qualities = this.HttpContext.Cache["qualities"] as Dictionary<string, string>;
if (qualities == null)
{
// LoadQualitiesFromXml() is the code above
qualities = LoadQualitiesFromXml();
this.HttpContext.Cache["qualities"] = qualities;
}
return qualities;
}

I think this is what you asked
var rarity3ValueInQualities = xml.SelectSingleNode("/result/qualities/rarity3").InnerText;
var rarity3ValueInqualityNames = xml.SelectSingleNode("/result/qualityNames/rarity3").InnerText;