C# VSTO: Highlighting E-mail Body Text - ClearHitHighlight() not clearing? - c#

Using VSTO and C#, I'm trying to get Outlook to highlight specific words in an e-mail body. So far I've been able to get this done using the code below:
Outlook.MailItem mailItem = this.inspector.CurrentItem as Outlook.MailItem;
if (inspector.IsWordMail())
{
var outlookWordDocument = inspector.WordEditor as Word.Document;
if (outlookWordDocument == null || outlookWordDocument.Application.Selection == null)
{ return; }
var wordRange = outlookWordDocument.Application.Selection.Range;
Word.Find find_highlight = wordRange.Find;
find_highlight.HitHighlight("apples", Word.WdColor.wdColorDarkRed);
find_highlight.ClearHitHighlight(); // trying to clear for testing purposes, but does nothing
}
My problem is that the ClearHitHighlight() function isn't clearing anything. The only way I can clear is if I perform another search right after. See comments below:
find_highlight.HitHighlight("apples"); //highlights "apples"
find_highlight.HitHighlight("oranges"); //highlights "oranges" too
find_highlight.ClearHitHighlight(); //does nothing
find_highlight.HitHighlight("pears"); //clears previous highlights, adds pears
As an alternative I could highlight the text by formatting the actual body of the e-mail, but this HitHighlight function seems to be more appropriate - if only I could figure out how to clear the markings when finished!
Any help would be appreciated.

Instead of calling find on the Word Range, call find on the document's content variable. Not sure exactly why, but this change causes the correct behaviour.
Change this:
var wordRange = outlookWordDocument.Application.Selection.Range;
Word.Find find_highlight = wordRange.Find;
To this:
Word.Find find_highlight = outlookWordDocument.Content.Find;

Related

How to point to the correct Store in Outlook automation by C#?

I have a lot of VBA automation that interlinks an Outlook and Word solution; it is fine, but time is inexorable... so, I'm start to decorating and extending that old solution, wraping it with C#/VS2017.
Through a conventional Winform I can choose my patients, and from this action I do a lot of actions, including open the correct Outlook contact; that's the problem, because I can't get the correct Store; the patients.pst, depending on the machine, may be the 1st, 2nd, 3rd...
In VBA I do this:
WhichStoreNameToPointAt="patients"
Set myNamespace = myolApp.GetNamespace("MAPI")
For i = 1 To myNamespace.Stores.Count Step 1
If myNamespace.Stores.item(i).DisplayName = WhichStoreNameToPointAt Then
intOutlookItemStore = i
End if
End If
Set myFolderPatients = myNamespace.Stores.item(intOutlookItemStore).GetDefaultFolder(olFolderContacts)
And it always functions like a charm.
In C# I tried a lot of variations, and could not point to the correct store:
public void OpenPatientContact(string patientName)
{
Outlook.Store whichStore = null;
Outlook.NameSpace nameSpace = OlkApp.Session;
int i = 1;
foreach (Outlook.Folder folder in nameSpace.Folders)
{
bool p = false;
if (whichStoreNameToPointAt == folder.Name)
{
p = true;
whichStore = folder.Store;
//Correct Store selected; I can tell because of this watch:
//whichStore.displayname == whichStoreNameToPointAt
}
i++;
if (p)
break;
}
var contactItemsOlk = whichStore.Session.GetDefaultFolder
(Outlook.OlDefaultFolders.olFolderContacts).Items;
// The problem is below; always the first Store
Outlook.ContactItem contact = (Outlook.ContactItem)contactItemsOlk
.Find(string.Format("[FullName]='{0}'", patientName)); //[1];
if (contact != null)
{
contact.Display(true);
}
else
{
MessageBox.Show("The contact information was not found.");
}
}
Unfortunately, it keeps pointing ever to the same first Store, the one that has no patients...
If I change the Store order I can get past this and test other stuff, but of course it is not the right way.
Any other heads/eyes to see the light?
TIA
While seated writing the question, looking at a yellow rubber duck - and a lot of other stuff that belongs to my 1 yo daughter ;), I realized that whichStore.Session.GetDefaultFolder is a little strange in this context. I only changed this
var contactItemsOlk = whichStore.Session.GetDefaultFolder
(Outlook.OlDefaultFolders.olFolderContacts).Items;
To that:
var contactItemsOlk = whichStore.GetDefaultFolder
(Outlook.OlDefaultFolders.olFolderContacts).Items;
Voilá! Magic happens with C# too!
Session returns the default NameSpace object for the current session.
PS: yellow rubber duck; guys of The Pragmatic Programmer really knows some secrets and tricks ;)
Thanks Thomas and Hunt!

C# Get Values from two different websites

I am using HTMLElementCollection, HtmlElement to iterate through a website and using Get/Set attributes of a website HTML and returning it to a ListView. Is it possible to get values from website a and website b to return it to the ListView?
HtmlElementCollection oCol1 = oDoc.Body.GetElementsByTagName("input");
foreach (HtmlElement oElement in oCol1)
{
if (oElement.GetAttribute("id").ToString() == "search")
{
oElement.SetAttribute("value", m_sPartNbr);
}
if (oElement.GetAttribute("id").ToString() == "submit")
{
oElement.InvokeMember("click");
}
}
HtmlElementCollection oCol1 = oDoc.Body.GetElementsByTagName("tr");
foreach (HtmlElement oElement1 in oCol1)
{
if (oElement1.GetAttribute("data-mpn").ToString() == m_sPartNbr.ToUpper())
{
HtmlElementCollection oCol2 = oElement1.GetElementsByTagName("td");
foreach (HtmlElement oElement2 in oCol2)
{
if (oElement2 != null)
{
if (oElement2.InnerText != null)
{
if (oElement2.InnerText.StartsWith("$"))
{
string sPrice = oElement2.InnerText.Replace("$", "").Trim();
double dblPrice = double.Parse(sPrice);
if (dblPrice > 0)
m_dblPrices.Add(dblPrice);
}
}
}
}
}
}
As one of the comments mentioned the better approach would be to use HttpWebRequest to send a get request to www.bestbuy.com or whatever site. What it returns is the full HTML code (what you see) which you can then parse through. This kind of approach keeps you from seinding too many requests and getting blacklisted. If you need to click a button or type in a text field its best to mimic human input to avoid being blacklisted also. I would suggest injecting a simple javascript into the page header or body and execute it from the app to send a 'onClick' event from the button (which would then reply with a new page to parse or display) or to modify the text property of something.
this example is in c++/cx but it originally came from a c# example. the script sets the username and password text fields then clicks the login button:
String^ script = "document.GetElementById('username-text').value='myUserName';document.getElementById('password-txt').value='myPassword';document.getElementById('btn-go').click();";
auto args = ref new Platform::Collections::Vector<Platform::String^>();
args->Append(script);
create_task(wv->InvokeScriptAsync("eval", args)).then([this](Platform::String^ response){
//LOGIN COMPLETE
});
//notes: wv = webview
EDIT:
as pointed out the absolute best approach would be to get/request an api. I was surprised to see that site mason pointed out for bestbuy developers. Personally I have only tried to work with auto part stores who either laugh while saying I can't afford it or have no idea what I'm asking for and hang up (when calling corporate).
EDIT 2: in my code the site used was autozone. I had to use chrome developer tools (f12) to get the names of the username, password, and button name. From the developer tools you can also watch what is sent from your computer to the site/server. This allows you to recreate everything and mimic javascript input and actions using post/get with HttpWebRequest.

Pulling Data from Word Form

Using C#, I need to pull data from a word document. I have NetOffice for word installed in the project. The data is in two parts.
First, I need to pull data from the document settings.
Second, I need to pull the content of controls in the document. The content of the fields includes checkboxes, a date, and a few paragraphs. The input method is via controls, so there must be some way to interact with the controls via the api, but I don't know how to do that.
right now, I've got the following code to pull the flat text from the document:
private static string wordDocument2String(string file)
{
NetOffice.WordApi.Application wordApplication = new NetOffice.WordApi.Application();
NetOffice.WordApi.Document newDocument = wordApplication.Documents.Open(file);
string txt = newDocument.Content.Text;
wordApplication.Quit();
wordApplication.Dispose();
return txt;
}
So the question is: how do I pull the data from the controls from the document, and how do I pull the document settings (such as the title, author, etc. as seen from word), using either NetOffice, or some other package?
I did not bother to implement NetOffice, but the commands should mostly be the same (except probably for implementation and disposal methods).
Microsoft.Office.Interop.Word.Application word = new Microsoft.Office.Interop.Word.Application();
string file = "C:\\Hello World.docx";
Microsoft.Office.Interop.Word.Document doc = word.Documents.Open(file);
// look for a specific type of Field (there are about 200 to choose from).
foreach (Field f in doc.Fields)
{
if (f.Type == WdFieldType.wdFieldDate)
{
//do something
}
}
// example of the myriad properties that could be associated with "document settings"
WdProtectionType protType = doc.ProtectionType;
if (protType.Equals(WdProtectionType.wdAllowOnlyComments))
{
//do something else
}
The MSDN reference on Word Interop is where you will find information on just about anything you need access to in a Word document.
UPDATE:
After reading your comment, here are a few document settings you can access:
string author = doc.BuiltInDocumentProperties("Author").Value;
string name = doc.Name; // this gives you the file name.
// not clear what you mean by "title"
As far as trying to understand what text you are getting from a "legacy control", I need more information as to exactly what kind of control you are extracting from. Try getting a name of the control/textbox/form/etc from within the document itself and then look up that property on the Google.
As a stab in the dark, here is an (incomplete) example of getting text from textboxes in the document:
List<string> textBoxText = new List<string>();
foreach (Microsoft.Office.Interop.Word.Shape s in doc.Shapes)
{
textBoxText.Add(s.TextFrame.TextRange.Text); //this could result in an error if there are shapes that don't contain text.
}
Another possibility is Content Controls, of which there are several types. They are often used to gather user input.
Here is some code to catch a rich text Content Control:
List<string> contentControlText = new List<string>();
foreach(ContentControl CC in doc.ContentControls)
{
if (CC.Type == WdContentControlType.wdContentControlRichText)
{
contentControlText.Add(CC.Range.Text);
}
}

using C# and UI Automation to grab contents of unknown control-type

In the image below there is an area, which has an unknown (custom) class. That's not a Grid or a Table.
I need to be able:
to select Rows in this area
to grab a Value from each cell
The problem is since that's not a common type element - I have no idea how to google this problem or solve it myself. So far the code is following:
Process[] proc = Process.GetProcessesByName("programname");
AutomationElement window = AutomationElement.FromHandle(proc [0].MainWindowHandle);
PropertyCondition xEllist2 = new PropertyCondition(AutomationElement.ClassNameProperty, "CustomListClass", PropertyConditionFlags.IgnoreCase);
AutomationElement targetElement = window.FindFirst(TreeScope.Children, xEllist2);
I've already tried to threat this Area as a textbox, as a grid, as a combobox, but nothing solved my problem so far. Does anybody have any advice how to grab data from this area and iterate through rows?
EDIT: sorry I've made a wrong assumption. Actually, the header(column 1, column 2, column 3) and the "lower half" of this area are different control-types!!
Thanks to Wininspector I was able to dig more information regarding these control types:
The header has following properties: HeaderControl 0x056407DC (90441692) Atom: #43288 0xFFFFFFFF (-1)
and the lower half has these: ListControl 0x056408A4 (90441892) Atom: #43288 0x02A6FDA0 (44498336)
The code that I've showed earlier - retrieved the "List" element only, so here is the update:
Process[] proc = Process.GetProcessesByName("programname");
AutomationElement window = AutomationElement.FromHandle(proc [0].MainWindowHandle);
//getting the header
PropertyCondition xEllist3 = new PropertyCondition(AutomationElement.ClassNameProperty, "CustomHeaderClass", PropertyConditionFlags.IgnoreCase);
AutomationElement headerEl = XElAE.FindFirst(TreeScope.Children, xEllist3);
//getting the list
PropertyCondition xEllist2 = new PropertyCondition(AutomationElement.ClassNameProperty, "CustomListClass", PropertyConditionFlags.IgnoreCase);
AutomationElement targetElement = window.FindFirst(TreeScope.Children, xEllist2);
After giving it a further thought I've tried to get all column names:
AutomationElementCollection headerLines = headerEl.FindAll(TreeScope.Children, new PropertyCondition(AutomationElement.ControlTypeProperty, ControlType.HeaderItem));
string headertest = headerLines[0].GetCurrentPropertyValue(AutomationElement.NameProperty) as string;
textBox2.AppendText("Header 1: " + headertest + Environment.NewLine);
Unfortunately in debug mode element count in "headerLines" is 0 so the program throws an error.
Edit 2: Thanks to the answer below - I've installed Unmanaged UI Automation, which holds better possibilities than the default UIA. http://uiacomwrapper.codeplex.com/
How do you use the legacy pattern to grab data from unknown control-type?
if((bool)datagrid.GetCurrentPropertyValue(AutomationElementIdentifiers.IsLegacyIAccessiblePatternAvailableProperty))
{
var pattern = ((LegacyIAccessiblePattern)datagrid.GetCurrentPattern(LegacyIAccessiblePattern.Pattern));
var state = pattern.Current.State;
}
Edit 3. IUIAutoamtion approach (non-working as of now)
_automation = new CUIAutomation();
cacheRequest = _automation.CreateCacheRequest();
cacheRequest.AddPattern(UiaConstants.UIA_LegacyIAccessiblePatternId);
cacheRequest.AddProperty(UiaConstants.UIA_LegacyIAccessibleNamePropertyId);
cacheRequest.TreeFilter = _automation.ContentViewCondition;
trueCondition = _automation.CreateTrueCondition();
Process[] ps = Process.GetProcessesByName("program");
IntPtr hwnd = ps[0].MainWindowHandle;
IUIAutomationElement elementMailAppWindow = _automation.ElementFromHandle(hwnd);
List<IntPtr> ls = new List<IntPtr>();
ls = GetChildWindows(hwnd);
foreach (var child in ls)
{
IUIAutomationElement iuiae = _automation.ElementFromHandle(child);
if (iuiae.CurrentClassName == "CustomListClass")
{
var outerArayOfStuff = iuiae.FindAllBuildCache(interop.UIAutomationCore.TreeScope.TreeScope_Children, trueCondition, cacheRequest.Clone());
var outerArayOfStuff2 = iuiae.FindAll(interop.UIAutomationCore.TreeScope.TreeScope_Children, trueCondition);
var countOuter = outerArayOfStuff.Length;
var countOuter2 = outerArayOfStuff2.Length;
var uiAutomationElement = outerArayOfStuff.GetElement(0); // error
var uiAutomationElement2 = outerArayOfStuff2.GetElement(0); // error
//...
//I've erased what's followed next because the code isn't working even now..
}
}
The code was implemented thanks to this issue:
Read cell Items from data grid in SysListView32 of another application using C#
As the result:
countOuter and countOuter2 lengths = 0
impossible to select elements (rows from list)
impossible to get ANY value
nothing is working
You might want to try using the core UI automation classes. It requires that you import the dll to use it in C#. Add this to your pre-build event (or do it just once, etc):
"%PROGRAMFILES%\Microsoft SDKs\Windows\v7.0A\bin\tlbimp.exe" %windir%\system32\UIAutomationCore.dll /out:..\interop.UIAutomationCore.dll"
You can then use the IUIAutomationLegacyIAccessiblePattern.
Get the constants that you need for the calls from:
C:\Program Files\Microsoft SDKs\Windows\v7.1\Include\UIAutomationClient.h
I am able to read Infragistics Ultragrids this way.
If that is too painful, try using MSAA. I used this project as a starting point with MSAA before converting to all UIA Core: MSSA Sample Code
----- Edited on 6/25/12 ------
I would definitely say that finding the proper 'identifiers' is the most painful part of using the MS UIAutomation stuff. What has helped me very much is to create a simple form application that I can use as 'location recorder'. Essentially, all you need are two things:
a way to hold focus even when you are off of your form's window Holding focus
a call to ElementFromPoint() using the x,y coordinates of where the mouse is. There is an implementation of this in the CUIAutomation class.
I use the CTRL button to tell my app to grab the mouse coordinates (System.Windows.Forms.Cursor.Position). I then get the element from the point and recursively get the element's parent until I reach the the desktop.
var desktop = auto.GetRootElement();
var walker = GetRawTreeWalker();
while (true)
{
element = walker.GetParentElement(element);
if (auto.CompareElements(desktop, element) == 1){ break;}
}
----- edit on 6/26/12 -----
Once you can recursively find automation identifiers and/or names, you can rather easily modify the code here: http://blog.functionalfun.net/2009/06/introduction-to-ui-automation-with.html to be used with the Core UI Automation classes. This will allow you to build up a string as you recurse which can be used to identify a control nested in an application with an XPath style syntax.

Suppressing Outlook Field Printing

We have written an add-on for Outlook that files emails into our CRM system. Int he process of this, it saves the Outlook Message ID as a UserField on the Message itself.
eg.
currentUserProperty = Constants.APPLICATION_NAME + "EntryID";
mailItem.UserProperties.Add(currentUserProperty,
Microsoft.Office.Interop.Outlook.OlUserPropertyType.olText,
Missing.Value,
Missing.Value).Value = entryId;
Unfortunately, this is a HUUUGGEE number, much like:
"00000000D502D779150E2F4580B1AADDF04ECDA6070097EF5A1237597748A4B4F9BFF540020800000006E9E4000068BB5B6DFC36924FAEC709A17D056583000002DE0E350000"
The problem is that when the user prints the message off, Outlook insists on including this field (beneath the From/To) and because it has no spaces, cannot wrap the ID and compresses the A4 page until it can fit on horizontally. This produces teeny-tiny email printouts.
Is there any way I can correct this? I had thought of overwriting the field OriginalEntryID (which is the one causing the problem) with one delimited by spaces but I get an exception from the COM layer. My next stop is to try and suppress the output of this and other user-defined fields on Outlook stationary.
Does anyone know how this can be achieved?
You must use .NET Reflection to fix this (as recommended by Microsoft Support). Hopefully this will be fixed in future versions of the VSTO SDK.
Suppress Outlook User Field Printing
static void SuppressUserPropertyPrinting(Outlook.MailItem message)
{
try
{ // Late Binding in .NET: https://support.microsoft.com/en-us/kb/302902
Type userPropertyType;
long dispidMember = 107;
long ulPropPrintable = 0x4; // removes PDO_PRINT_SAVEAS
string dispMemberName = String.Format("[DispID={0}]", dispidMember);
object[] dispParams;
if (message.UserProperties.Count == 0) return; // no props found (exit)
// marks all user properties as suppressed
foreach (Outlook.UserProperty userProperty in message.UserProperties.Cast<Outlook.UserProperty>())
{
if (userProperty == null) continue; // no prop found (go to next)
userPropertyType = userProperty.GetType(); // user property type
// Call IDispatch::Invoke to get the current flags
object flags = userPropertyType.InvokeMember(dispMemberName, BindingFlags.GetProperty, null, userProperty, null);
long lFlags = long.Parse(flags.ToString()); // default is 45 - PDO_IS_CUSTOM|PDO_PRINT_SAVEAS|PDO_PRINT_SAVEAS_DEF (ref: http://msdn.microsoft.com/en-us/library/ee415114.aspx)
// Remove the hidden property Printable flag
lFlags &= ~ulPropPrintable; // change to 41 - // PDO_IS_CUSTOM|PDO_PRINT_SAVEAS_DEF (ref: http://msdn.microsoft.com/en-us/library/ee415114.aspx)
// Place the new flags property into an argument array
dispParams = new object[] { lFlags };
// Call IDispatch::Invoke to set the current flags
userPropertyType.InvokeMember(dispMemberName, BindingFlags.SetProperty, null, userProperty, dispParams);
}
}
catch { } // safely ignore if property suppression doesn't work
}
I think this might be able to help me:
http://www.add-in-express.com/forum/read.php?FID=5&TID=5071#postform

Categories