Comments by PeejayAdams (Top 62 by date)

What kind of time variations are you getting? Is there any kind of pattern relating to which users are running the procedure?

Horses and water.

Try looking at the SelectedValue property rather than SelectedItem. && is definitely the right operator.

Would 52 be valid or would it need to be entered as 052? If both are valid, then you're simply looking at any number under 1000. If it has to be 052, then really you're looking for a string rather than an int, a regular expression ([0-9][0-9][0-9]) would be the way to go.

Thanks, Richard. I certainly hadn't thought about STUFF. In this case, I still need to substring the description field down to the actual code but that certainly takes some of the complexity out of the equation. In the end, I used a CTE and applied a subsequent transfer on the final SELECT:

CASE
WHEN PATINDEX ('%J[A-Z][0-9]0%', [Phase Code]) > 0
THEN
REPLACE ([Phase Code], SUBSTRING ([Phase Code], PATINDEX('%[1-9]%', [Phase Code]), 1), '0')
ELSE
[Phase Code]
END
AS [Phase Code]

(creating a function wasn't an available option as it isn't our database) which certainly does the job but I can't help finding it a little untidy. Untidiness, I guess, is sometimes something we just have to live with in SQL - I'd sure hate to see an UNstructured Query Language!

Hint: You've accidentally written most of the code required to return false in your "What I haven't been arsed to try" section - take three of the words, drop one of them, add a semi-colon and think about your life choices.

It's very hard to say without seeing your definition for IDCursor and your declaration of @ID.

Their documentation looks pretty good to me. They also have a forum which includes an FAQ section.

Can you use something like: where !x.Contains("SearchPattern")?

If you agree to paint my house, I'll agree to do your homework :) Seriously, you're not going to learn anything by copying my code or anyone else's (and it would pretty darned obvious to your tutor that you haven't written it yourself, anyway). Think about what you've been asked to do and have a go at doing it - that's what developers do for a living. If you get stuck on something specific, there are plenty of us here who are more than happy to help but at least TRY writing some code to get to that position.

I don't think that your tutor wants you to write a program in multiple styles, he wants you to write some code that employs all of the basic OOP concepts.

Yes, that looks fine to me, too. Make sure your <connectionstrings> segment is within the <configuration> tab and as F-ES suggests make sure you're dealing with the correct app.config file.

I have both languages on my CV and I get about a 6:1 ratio of C#:VB.NET recruitment emails - obviously there's a regional element to that but it does seem broadly indicative. Whilst I take those "must-have IT skills" lists with a pinch of salt, it's notable that C# appears prominently on all of them whereas VB.NET rarely makes an appearance these days.

The question asked which of the two languages is better to learn not where their commonality lies.

That's an 8 year instant advantage for C#! Most languages hang around for a long, long time (I know a couple of people who still work with COBOL which is all the proof we could ever need) but I would imagine that there are far more green-field developments been undertaken in C# than there are in VB.

WPF is awesome but it's quite a steep learning curve. Don't consider using WPF without the MVVM pattern, I've seen people try and it always ends in tears. If you're moving from VB to C# and trying to learn XAML, data bindings and MVVM at the same time, that year will fly by but you may not have too much to show at the end of it. If the app requirement doesn't dictate that you use WPF, I'd advise you to go with WinForms until you become more au fait with C#.

That really is ridiculously slow, such things should be measured in milliseconds rather than minutes. Tempting as it might be to paper over the cracks, it really is unreasonable to expect the user to wait for that long - someone really needs to address the performance issue or you won't have 30,000 accounts for very long.

I'd start by addressing the issue of why it's taking so long to perform the calculation.

There are plenty of people here who are more than happy to help people out if they're stuck with something but not when they can't be bothered to even have a go themselves. Just getting someone to write your assignments for you isn't going to teach you a damned thing and it's unfair on your class mates who are doing it themselves rather than drafting in professional help. If you posted some code and said "this is what I've done but I'm struggling with aspect X", you'd get plenty of help on this site but as has been pointed out already, this is no different from me asking for someone to come and paint my house with no offer of payment (if anyone would like to do that, please give me a shout but I'm not holding my breath).

Could we qualify that and maybe say "Yes, anyone could providing that they've done their homework?"

It's an auto-generated partial class for your XAML and you can't stop it being generated. If you're having issues with one, do a clean and rebuild and it will get rebuilt. Why do you want to stop it being generated? If the problems appears to be coming from there it's most likely an issue in your code with missing/incorrect namespace references rather than with anything in the actual .g.cs file but you'd need to provide a few details as to what your actual issue is before anyone can help you with that.

The camel/banana problem traditionally involves 3000 bananas - I'm almost certain that there's no solution if there are only 2000.

Yes. Assignment goes to the left, so x = z would give x the value of z whereas z = x would give the value of x to z.

If we were assigning a value from a value rather than a variable we'd say, for example, "x = 2;" which effectively reads as "Let x have the value of 2." Similarly, when we say "x = z;" we are saying "Let x have the value currently assigned to z."

You need to loop through PostedFiles (plural) rather than looking at PostedFile (singular).

Try this http://stackoverflow.com/questions/4019831/how-do-you-center-your-main-window-in-wpf

It simply can't be done in a UDF. You could try using a CLR function.

http://stackoverflow.com/questions/13948897/not-able-to-open-vb-project-getting-error-c-windows-system32-mscomctl-ocx-co

Don't take my word for it because I don't know C++ but I'd try "if (f.compare("add") == 0)" in place of "if(f == "add")"

Many thanks for that, PureNsanity. I've been somewhat sidetracked from the issue in question so haven't had a chance to readdress but I very much suspect that you've put me on the right track - will come back and accept as answer when I get the chance to go back and look at it but in the meantime, thanks again and +5!

As pointed out in solution 2, you're replacing rather than concatenating. Your actual command text is "System.Data.SqlClient.SqlCommandwhere date_TripDate ='2016-08-24' and trip.int_
TripStatusID not in (0)" which is clearly not what you're after. If you really must do it this way, it should be cmd.CommandText += "something"; rather than cmd.CommandText += cmd + "something"; - cmd is a SqlCommand not the command text (also if doing a heap of string manipulation, do it in a StringBuilder would be a a better approach). As Solution 1 suggests - this really should be a stored procedure call, anyway.

I would ask yourself the following questions:

1) Is Excel really the tool for the job? If yes, then you'll need to look into SSIS as a way to export from Excel to SQLServer. It won't be dynamic but you can automate the export of saved spreadsheets. It's not entirely clear what your requirement is but if you're trying to allow someone to maintain data, it's far, far better to do that via a form based app where you have control over validation etc.

2) Do you really want users to decide what fields get exported? (Hint: the answer you're looking for is a very firm "no!")

3) Do you need to update all 160 columns in a single procedure? I very much hope that your database doesn't have a 160+ column table. If it does, rewrite it/get it rewritten before you go any further.

I can see plenty of problems there. Firstly, don't validate a number directly from a string - you check for "0" but "0.00" ".0" and many other things equate to zero. Your test for "0" is actually negated anyway as you effectively say "if (empty) ... if ([something we already know is true] or [it doesn't matter what goes here because we already have a true]) Your try ... catch block will only catch exceptions and you could be generating a fair few of those - how would you know where the exception came from? What about negative values?

In all honesty, I'm a bit confused by the whole thing and your question doesn't really give much clue as to what you're trying to achieve but if all you want to validate is that the product of the three is positive, you only need to test the product not look for individual zeroes.

IIS User is a guest account so probably won't have sufficient permissions, try setting the application pool identity to a built-in account - LocalSystem should work.

What account are you using on the application pool?

You need to look at PIVOT (https://technet.microsoft.com/en-us/library/ms177410(v=sql.105).aspx)

Thanks, I did look at that route - and may have to go back there in some shape or form - it's not ideal because I simply can't predict all the places where the user might want to insert something. It's also problematic as it makes any global formatting changes somewhat difficult - this is a customer-facing document so needs to be WYSIWYG.

The document in question is a summary of an estimate - it's potentially a very large document containing a combination of header details, text blocks, itemised activities with hours and prices and various comment fields. At present the summary goes out as it is generated from the data entered elsewhere in the application but we'd like people to be add to the generated parts of the document but not touch anything that relates to the stored data - e.g. we don't want people inadvertently adding a digit to a price.

You need to do it anywhere after the for loop. Check that word is non-zero so that you don't get a result of 1 where you should be getting zero so if (word > 0) word += 1; would do the job.

Hi shukla, by counting the spaces your always going to miss the last word so you'd need to increment your count by one for the final result.

You're always checking the whole string to be a space (if (str = " ") rather than the specific character (something like if (Char.IsWhiteSpace(str[i]) ) - that's why your current code doesn't work. Split() would also do the job as Mehdi suggests but it wouldn't be suitable for large bodies of text as you'd be creating a huge number of objects.

Have you tried ms-appx:///... (3 slashes rather than two)?

There is something very wrong with your database design if you have a hundred columns in a table. If you don't address that first, anything you try to do in code is going to be a nightmare. Look at the basics of RDBMS and normalization before you go any further.

Your code's kind of unreadable - proper casing and indentation would make a huge difference as would meaningful table aliases. Without any context or even an error message, it's somewhat hard to say what's causing the error.

Thanks Nathan. It's certainly worth investigating, though I suspect it's going to turn out to be a very small needle in a very large haystack ...

The VDI is running .NET 4.6 whereas my local is on 4.0 but that's never given rise to any issues in the past (both are running Windows 7 Pro). I'd expect .NET incompatibilities to give flat out errors rather than performance issues - though I could well be wrong on that.

Thanks, CHill - sorry for the delayed reply - I'm suspecting that I'd used the non-SysWOW64 version of regsvr32 the first time round.

Can you not just put it the Page_Load event handler?

Does your FTP server support Unicode?

That was the inspiration I needed, thank you!

I've changed my code to attach an XmlReader to the XmlWriter's FileStream and it does just what I want:

private static string GetRow<t>(object _data)
{
StringBuilder xml = new StringBuilder();
MemoryStream stream = new MemoryStream();

T data = (T)_data;

XmlSerializer serialiser = new XmlSerializer(typeof(T));

XmlWriter writer = XmlWriter.Create(stream);

serialiser.Serialize(writer, data, null);
writer.Close();

stream.Seek(0, SeekOrigin.Begin);

XmlReader xmlReader = XmlReader.Create(stream);

xmlReader.MoveToContent();

return xmlReader.ReadInnerXml();
}

9839([0-9])\1{3}(?!\1)

would stop the next character being a repeat of the previous group by using a negative backreference but you'd need a separate check for the string being all numeric.

I suspect that someone out there will know how to combine both checks into a single regex.

I don't know VB too well but shouldn't you be using a Byte array rather than a Byte for ProductPhoto?

Thanks, Tadit. Your solution is certainly tidier than mine but the end result is exactly the same. I still get the updated panel appearing in the wrong place.

Hi Tadit,

Yes it's happening in the AddParameterClick

protected void AddParameterClick(object sender, EventArgs e)
{
if (_parameters == null)
_parameters = new List<ProcedureParameter>();

foreach (RepeaterItem item in rptParameters.Items)
{
TextBox name = (TextBox)item.FindControl("tbParameterName");
TextBox value = (TextBox)item.FindControl("tbParameterValue");

_parameters.Add(new ProcedureParameter(name.Text, value.Text));
}

_parameters.Add(new ProcedureParameter());

rptParameters.DataSource = _parameters;
rptParameters.DataBind();
pnlSproc.Update();
}

Reason for my vote of 2 \n Hi Mafaz, I certainly don't want to be too disparaging about anyone's first article but the points raised about terminology should be noted - the title is very misleading. As a historical aside, the ?: operator predates OO, it was part of the C language

Thanks, Piebald but that runs into the same issue - I also get the same "invalid object" error if I run the commands via SMS. The problem really stems from going via a linked server. For my current purposes, I can work around it by making a direct connection to the server/database that contains the sproc in question and creating the temporary table there but that's a slightly untidy solution as I'll have to maintain multiple connection strings etc.

Thanks tgrt - sorry for the delay in replying, I've been on holiday.

I've tried creating a stand-alone test form and everything works as expected as it does for you. So there's something in my app that's manipulating the IsDropDown property indirectly (maybe via the IsOpen property on the internal Popup) but I'm just not seeing it. I'll let you know when I find something.

Thanks again for having a look - it's much appreciated.