Code Question: 04/17/11

Sunday, April 17, 2011

How to specify notificationSender? (Cocoa Touch)

All of the examples I can find of setting up a notification look like this:

[nc addObserver:self selector:@selector(keyboardWillShow:) name: UIKeyboardWillShowNotification object:nil];

where object is always being set to nil. object is the notificationSender, which is defined thusly in the documentation:

"The object whose notifications the observer wants to receive; that is, only notifications sent by this sender are delivered to the observer. When nil, the notification center doesn’t use a notification’s sender to decide whether to deliver it to the observer."

I want to receive the keyboard notification only for one particular text view, so I want to tell addObserver that, but I don't know what to give it for the object. I tried the outlet variable that's bound to the text view I'm interested in, but that just resulted in my getting no notifications at all.

Any suggestions?

From stackoverflow

In this case you can not get the notification only for one particular text view, as the docu for UIKeyboardWillShowNotification says "The notification object is nil.".

You have to check in your keyboardWillShow impl if your particular text view isFirstResponder.
The UIKeyboardWillShowNotification is coming from your window instance, and for all intents and purposes is a "system" notification. The keyboard is either showing or not showing, it's not really something that is tied to a specific control.

If you want to do something when a user enters a specific text field, you should probably control that in the text field's delegate instead.
I had two competing needs - I had to use the keyboard notification because I needed to get the keyboard height, which appears to be only available that way, but I also needed to know which text view I was in, which meant also using the textViewDidBeginEditing delegate. After much messing around I finally resorted to getting the keyboard height in the notification method and storing it in an instance variable, which was then available to use in the delegate method (I need to scroll the view up so the bottom text view is not mostly hidden under they keyboard when they start typing). A bit inelegant, but it works.

Thanks for the pointers!

Capture who clicked a link from Website A to Website B?

Assume I have a website called Website A and it has some links on it to other websites. One of these links could be a link to Website B. Is there a way, using some existing solution, open source, or something you have to pay for where you can track who, what, where, etc clicked the link to go from Website A to Website B?

Website B could be Stackoverflow and my website is www.websiteA.com, so on my website, I have a link to www.stackoverflow.com. If a user clicks on www.stackoverflow.com from my page, I want to capture information, such as name, phone, etc (not sure if I can do this unless the user is actually logged into my website). In this case, this is a public page which anyone from anywhere can access. If I can't do it from my website, is there a way I can tell stackoverflow that this particular person clicked on my link to get to your website and I need their information.

From stackoverflow

Try using HTTP_REFERER

Chad Birch : Would only work if he also controls Website B, which he didn't specify.

Xaisoft : I have no control over Website B, for example, imagine Website B is stackoverflow and I have a link to stackoverflow. If someone clicked a link on my site to get to stackoverflow, could I get their information.

Gero : I see. So you can consider using the methods explained in other answers =)
You could always check the referer, but this can be spoofed.
```
Request.UrlReferrer.ToString
```
Other than that, you could append an additional query parameter on the links from Website A so you know where they came from:
```
<a href="www.websiteb.com?from=websitea">
```
The method most sites use for this is to have a script on your site handle the redirect. So instead of the link being directly to:

http://websiteB.com

the link goes to something like:

http://websiteA.com/link/websiteB.com

The "link" script logs the request and then forwards them to the address (this should be done instantly, don't give them a "forwarding you to Website B in 3 seconds!" page).

Note that as a visitor to your site, I much prefer a method where the link's destination is apparent when I hover over the link, as opposed to the method that some sites use where all you see is an id. What I mean is, instead of the link being something like:

http://websiteA.com/link/websiteB.com

they instead have a link like:

http://websiteA.com/link/483

Showing the destination can be achieved in a few ways. One way is something like I described above, where the destination is part of the query string. Alternatively, you can use some Javascript for this. The status bar's value can be changed with Javascript by setting the window.status variable, but many browsers prevent this by default (Firefox, for example).

Google has found a sneaky way around this problem by actually using Javascript to change the link instead of the status bar. When the page is loaded, the links in the results list go to their actual destinations, but every link has an onmousedown javascript event that changes the link when you click on it. You can see this in action by right-clicking on a link. After you've done that, hovering over the link will now show the true destination (Google's click-tracking script) in the status bar instead of the result's real URL.

Of course the information displayed with any method could be a misrepresentation of where I'm actually going to be sent, but I still prefer it over seeing a meaningless id number.

Andrew Hare : +1 For a very nice explanation.

Xaisoft : I am not to up-to-date on javascript, do you have an example of how this is done.

Chad Birch : I will edit some information into my answer about it.

Frank Crook : That's incredibly neat and something I've never noticed about the links Google provided. I had to play with the links by clicking down and not releasing before I believed you.

Xaisoft : Is the Click-Tracking script open source to view?
If you control website B, then you just use the http_referer portion in logging. Otherwise, you will want to have a redirect page on site A, that records the url, then redirects the browser through to the other site/page.
```
http://siteA/redirect?url=siteB/page
```
the redirect page in site a redirects to..
```
"http://" + request['url']
```

MySQL performance optimization: order by datetime field

I have a table with roughly 100.000 blog postings, linked to a table with 50 feeds via an 1:n relationship. When I query both tables with a select statement, ordered by a datetime field of the postings table, MySQL always uses filesort, resulting in very slow query times (>1 second). Here's the schema of the postings table (simplified):

+---------------------+--------------+------+-----+---------+----------------+
| Field               | Type         | Null | Key | Default | Extra          |
+---------------------+--------------+------+-----+---------+----------------+
| id                  | int(11)      | NO   | PRI | NULL    | auto_increment |
| feed_id             | int(11)      | NO   | MUL | NULL    |                |
| crawl_date          | datetime     | NO   |     | NULL    |                |
| is_active           | tinyint(1)   | NO   | MUL | 0       |                |
| link                | varchar(255) | NO   | MUL | NULL    |                |
| author              | varchar(255) | NO   |     | NULL    |                |
| title               | varchar(255) | NO   |     | NULL    |                |
| excerpt             | text         | NO   |     | NULL    |                |
| long_excerpt        | text         | NO   |     | NULL    |                |
| user_offtopic_count | int(11)      | NO   | MUL | 0       |                |
+---------------------+--------------+------+-----+---------+----------------+

And here's the feed table:

+-------------+--------------+------+-----+---------+----------------+
| Field       | Type         | Null | Key | Default | Extra          |
+-------------+--------------+------+-----+---------+----------------+
| id          | int(11)      | NO   | PRI | NULL    | auto_increment |
| type        | int(11)      | NO   | MUL | 0       |                |
| title       | varchar(255) | NO   |     | NULL    |                |
| website     | varchar(255) | NO   |     | NULL    |                |
| url         | varchar(255) | NO   |     | NULL    |                |
+-------------+--------------+------+-----+---------+----------------+

And here's the query that takes >1 second to execute. Please note that the post_date field has an index, but MySQL isn't using it to sort the postings table:

SELECT 
    `postings`.`id`, 
    UNIX_TIMESTAMP(postings.post_date) as post_date, 
    `postings`.`link`, 
    `postings`.`title`, 
    `postings`.`author`, 
    `postings`.`excerpt`, 
    `postings`.`long_excerpt`, 
    `feeds`.`title` AS feed_title, 
    `feeds`.`website` AS feed_website
FROM 
    (`postings`)
JOIN 
    `feeds` 
ON 
    `feeds`.`id` = `postings`.`feed_id`
WHERE 
    `feeds`.`type` = 1 AND 
    `postings`.`user_offtopic_count` < 10 AND 
    `postings`.`is_active` = 1
ORDER BY 
    `postings`.`post_date` desc
LIMIT 
    15

The result of the explain extended command on this query shows that MySQL is using filesort:

+----+-------------+----------+--------+---------------------------------------+-----------+---------+--------------------------+-------+-----------------------------+
| id | select_type | table    | type   | possible_keys                         | key       | key_len | ref                      | rows  | Extra                       |
+----+-------------+----------+--------+---------------------------------------+-----------+---------+--------------------------+-------+-----------------------------+
|  1 | SIMPLE      | postings | ref    | feed_id,is_active,user_offtopic_count | is_active | 1       | const                    | 30996 | Using where; Using filesort |
|  1 | SIMPLE      | feeds    | eq_ref | PRIMARY,type                          | PRIMARY   | 4       | feedian.postings.feed_id |     1 | Using where                 |
+----+-------------+----------+--------+---------------------------------------+-----------+---------+--------------------------+-------+-----------------------------+

When I remove the order by part, MySQL stops using filesort. Please let me know if you have any ideas on how to optimize this query to get MySQL to sort and select the data by using indexes. I have already tried a few things such as creating a combined index on all where/order by fields, as suggested by a few blog postings, but this didn't work either.

From stackoverflow

MySQL has two filesort algorithms: an older filesort that sorts records on disk, and a new version that works in memory.

If it cannot use an index on the first table in the join to sort the query, it will have to do a filesort. If resultset before sorting converted to fixed-width format is greater than the sort buffer OR if it contains any text fields, it will have to use the slower on-disk filesort algorithm (the second condition is satisfied since your query has a text field).

MySQL is choosing to use the is_active column, ostensibly because it thinks that column is most selective in eliminating rows before it continues with the other joins and where conditions. The first thing I would suggest would be to try creating composite indexes with post_date, feed_id, and the columns in the where condition, e.g. (is_active, user_offtopic_count, post_date, feed_id).

Dennis G. : Thanks for the explanation!

Create a composite index either on postings (is_active, post_date) (in that order).

It will be used both for filtering on is_active and ordering by post_date.

MySQL should show REF access method over this index in EXPLAIN EXTENDED.

Note that you have a RANGE filtering condition over user_offtopic_count, that's why you cannot use an index over this field both in filtering and in sorting by other field.

Depending on how selective is your user_offtopic_count (i. e. how many rows satisfy user_offtopic_count < 10), it may be more useful to create an index on user_offtopic_count and let the post_dates be sorted.

To do this, create a composite index on postings (is_active, user_offtopic_count) and make sure the RANGE access method over this index is used.

Which index will be faster depends on your data distribuion. Create both indexes, FORCE them and see which is faster:

CREATE INDEX ix_active_offtopic ON postings (is_active, user_offtopic_count);
CREATE INDEX ix_active_date ON postings (is_active, post_date);

SELECT 
    `postings`.`id`, 
    UNIX_TIMESTAMP(postings.post_date) as post_date, 
    `postings`.`link`, 
    `postings`.`title`, 
    `postings`.`author`, 
    `postings`.`excerpt`, 
    `postings`.`long_excerpt`, 
    `feeds`.`title` AS feed_title, 
    `feeds`.`website` AS feed_website
FROM 
    `postings` FORCE INDEX (ix_active_offtopic)
JOIN 
    `feeds` 
ON 
    `feeds`.`id` = `postings`.`feed_id`
WHERE 
    `feeds`.`type` = 1 AND 
    `postings`.`user_offtopic_count` < 10 AND 
    `postings`.`is_active` = 1
ORDER BY 
    `postings`.`post_date` desc
LIMIT 
    15

/* This should show RANGE access with few rows and keep the FILESORT */

SELECT 
    `postings`.`id`, 
    UNIX_TIMESTAMP(postings.post_date) as post_date, 
    `postings`.`link`, 
    `postings`.`title`, 
    `postings`.`author`, 
    `postings`.`excerpt`, 
    `postings`.`long_excerpt`, 
    `feeds`.`title` AS feed_title, 
    `feeds`.`website` AS feed_website
FROM 
    `postings` FORCE INDEX (ix_active_date)
JOIN 
    `feeds` 
ON 
    `feeds`.`id` = `postings`.`feed_id`
WHERE 
    `feeds`.`type` = 1 AND 
    `postings`.`user_offtopic_count` < 10 AND 
    `postings`.`is_active` = 1
ORDER BY 
    `postings`.`post_date` desc
LIMIT 
    15

/* This should show REF access with lots of rows and no FILESORT */

Dennis G. : That did the trick for me, thanks a lot! I had to use force index to get the best index used. We are now using multiple combined indexes for the different queries.

Also, it's important to remember that MySQL won't use an index if the column you're ordering by has a function applied to it.

You should also try aliasing postings.post_date as something else. This will tell MySQL to order by the unaltered column, and you'll still select the unix timestamp.

Is there a way to add an onclick event to an ASP.NET Label server control?

I wanted to do something like this:

<asp:Label ID="lblMyLabel" onclick="lblMyLabel_Click" runat="server">My Label</asp:Label>

I know that in Javascript I can do:

<span onclick="foo();">My Label</span>

So I'm wondering why I can't do that with a Label object.

From stackoverflow

You can use Attributes to add onclick client side callback.

I didn't know you can do this on span tags, but if it works you can add 'onclick' by lblMyLabel.Attributes.Add("onclick", "foo();");

But foo(); would need to be a client side javascript function.

System.Web.UI.WebControls.Label does NOT have OnClick server event. You could look into using AJAX if you want server callback with example above.

You could also use LinkButton like other say. You can make it not look like a link by using CSS, if it is just that underline you are concerned about.

ASPX:
```
<asp:LinkButton ID="LinkButton1" runat="server" 
    CssClass="imjusttext" OnClick="LinkButton1_Click">
LinkButton
</asp:LinkButton>
```
CSS:
```
a.imjusttext{ color: #000000; text-decoration: none; }
a.imjusttext:hover { text-decoration: none; }
```

You can do it in code in the page_load eg:

void Page_Load(object sender, EventArgs e) 
{
      lblMyLabel.Attributes.Add("onclick",
           "javascript:alert('ALERT ALERT!!!')");
}

I think you can, but it's a client-side onclick handler, not server side. It will complain about the attribute not being supported (or some such) but I think it renders correctly. If you want to to a server-side handler, I think you'll need to do a LinkButton.
As far as I know that's impossible. Label control emits <span> element which is “unclickable” on the server side. You would need to replace your Label control with a LinkButton.
If you want an onclick event, why don't you use a linkbutton, which has the onclientclick event:
```
<asp:linkbutton id="lblMyLink" onclientclick="lblMyLink_Click" runat="server">My Label</asp:linkbutton>
```
You can use CSS to make the link look like whatever you want

fuentesjr : I didn't want it to look like a link and I didn't want to hack a link to look like plain text using CSS, but I guess that's what I have to do. :( Thanks though.
Your question doesn't specify if you mean to raise the click event on the server (VB or c#) or the client (javascript.) If you're looking for a server-side event you should use a link button with css that makes the link appear as a label. You can then use the link button's server-side click event. If you're looking for a client-side click event - just type it into your server control markup (<asp:label id="MyLabel" runat="server" onclick="javascript:alert('hello');" Text="Click Me" />) ASP.NET will emit additional attributes in the html markup that generates.

Another hack would be to use a hidden link or button with style="display:none;" and trigger the click on the server control from a javascript function in the span.

Something like this:

<asp:linkbutton id="lblMyLink" onClick="lblMyLink_Click" runat="server" style="display:none;">My Link</asp:linkbutton>
<span onclick="document.getElementById('lblMyLink').click();">My Label</span>

you could always roll out your own control which produces a span, with the .net's standard postback javascript, but as stated earlier using a linklabel with a CSS class would be easier
Just to chime in on this issue, I used a label in my .aspx page that was only to be visible in my DataList Item template if there were child records in the dataset.

I added a onclick function to the label:

moreOptionsLabel.Attributes.Add("onclick", string.Format("toggle_visibility('{0}')", div.ClientID));

in my .cs file. It will now control a div tag in the .aspx page to show or hide the records - because the onclick points to a client side javascript function. Notice the div.ClientID, that makes this workable in a datalist.

As noted above - the span tag does indeed become functional with "onclick". And since the label control is rendered as a span after the page request, using the Addtribute.Add("onclick".... does work.

The result is show/hide functionality of data with out doing a postback. If you use the LinkButton, or simlar controls - the postback and page reload is unavoidable - unless you want to get into some Ajax stuff.

NOTE: the span tag won't look clickable unless you style it with an underline and hand cursor.

Credit to this idea comes from Will Asrari over at http://www.willasrari.com/blog/display-nested-repeaters-and-gridviews-asynchronously/000292.aspx
You need to create a class object that inherits from the label and onclick event handler which will be a method by yourslef and use your object as a custom label.
Thanks KiwiBastard, you helped me.

Find the last value in a "rolled-over" sequence with a stored procedure?

Suppose I had a set of alpha-character identifiers of a set length, e.g. always five letters, and they are assigned in such a way that they are always incremented sequentially (GGGGZ --> GGGHA, etc.). Now, if I get to ZZZZZ, since the length is fixed, I must "roll over" to AAAAA. I might have a contiguous block from ZZZAA through AAAAM. I want to write a sproc that will give me the "next" identifier, in this case AAAAN.

If I didn't have this "rolling over" issue, of course, I'd just ORDER BY DESC and grab the top result. But I'm at a bit of a loss now -- and it doesn't help at all that SQL is not my strongest language.

If I have to I can move this to my C# calling code, but a sproc would be a better fit.

ETA: I would like to avoid changing the schema (new column or new table); I'd rather just be able to "figure it out". I might even prefer to do it brute force (e.g. start at the lowest value and increment until I find a "hole"), even though that could get expensive. If you have an answer that does not modify the schema, it'd be a better solution for my needs.

From stackoverflow

You'd have to store the last allocated identifier in the sequence.

For example, store it in another table that has one column & one row.
```
CREATE TABLE CurrentMaxId (
    Id CHAR(6) NOT NULL
);

INSERT INTO CurrentMaxId (Id) VALUES ('AAAAAA');
```
Each time you allocate a new identifier, you'd fetch the value in that tiny table, increment it, and store that value in your main table as well as updating the value in CurrentMaxId.

The usual caveats apply with respect to concurrency, table-locking, etc.
I think I'd have tried to store the sequence as an integer, then translate it to string. Or else store a parallel integer column that is incremented at the same time as the alpha value. Either way, you could sort on the integer column.
A problem here is that you can't really tell from the data where the "last" entry is unless there is more detail as to how the old entries are deleted.

If I understand correctly, you are wrapping around at the end of the sequence, which means you must be deleting some of your old data to make space. However if the data isn't deleted in a perfectly uniform manner, you'll end up with fragments, like below:
```
ABCD   HIJKL NOPQRS   WXYZ
```
You'll notice that there is no obvious next value...D could be the last value created, but it might also be L or S.

At best you could look for the first or last missing element (use a stored procedure to perform a x+1 check just like you would to find a missing element in an integer sequence), but it's not going to provide any special result for rolled-over lists.

Coderer : The sequence doesn't always start at the beginning. Sometimes, it starts at BBBB, while sometimes it starts at YYYY, so there's already "room" at the beginning of the alphabet
Since I don't feel like writing code to increment letters, I'd create a table of all valid IDs (AAAAAA through ZZZZZZ) with an integer from 1 to X for those IDs. Then you can use the following:
```
SELECT @max_id = MAX(id) FROM Possible_Silly_IDs

SELECT
    COALESCE(MAX(PSI2.silly_id), 'AAAAAA')
FROM
    My_Table T1
INNER JOIN Possible_Silly_IDs PSI1 ON
    PSI1.silly_id = T1.silly_id
INNER JOIN Possible_Silly_IDs PSI2 ON
    PSI2.id = CASE WHEN PSI1.id = @max_id THEN 1 ELSE PSI1.id + 1 END
LEFT OUTER JOIN My_Table T2 ON
    T2.silly_id = PSI2.silly_id
WHERE
    T2.silly_id IS NULL
```
The COALESCE is there in case the table is empty. To be truly robust you should calculate the 'AAAAAA' (SELECT @min_silly_id = silly_id WHERE id = 1) in case your "numbering" algorithm changes.

If you really wanted to do things right, you'd redo the database design as has been suggested.

Here's code that I think will give you your Next value. I created 3 functions. The table is just my simulation of the table.column with your alpha ids (I used MyTable.AlphaID). I assume that it's as you implied and there is one contiguous block of five-character uppercase alphabetic strings (AlphaID):

IF OBJECT_ID('dbo.MyTable','U') IS NOT NULL
    DROP TABLE dbo.MyTable
GO
CREATE TABLE dbo.MyTable (AlphaID char(5) PRIMARY KEY)
GO
-- Play with different population scenarios for testing
INSERT dbo.MyTable VALUES ('ZZZZY')
INSERT dbo.MyTable VALUES ('ZZZZZ')
INSERT dbo.MyTable VALUES ('AAAAA')
INSERT dbo.MyTable VALUES ('AAAAB')
GO
IF OBJECT_ID('dbo.ConvertAlphaIDToInt','FN') IS NOT NULL
    DROP FUNCTION dbo.ConvertAlphaIDToInt
GO
CREATE FUNCTION dbo.ConvertAlphaIDToInt (@AlphaID char(5))
RETURNS int
AS
BEGIN
RETURN 1+ ASCII(SUBSTRING(@AlphaID,5,1))-65
              + ((ASCII(SUBSTRING(@AlphaID,4,1))-65) * 26)
              + ((ASCII(SUBSTRING(@AlphaID,3,1))-65) * POWER(26,2))
              + ((ASCII(SUBSTRING(@AlphaID,2,1))-65) * POWER(26,3))
              + ((ASCII(SUBSTRING(@AlphaID,1,1))-65) * POWER(26,4))
END
GO 

IF OBJECT_ID('dbo.ConvertIntToAlphaID','FN') IS NOT NULL
    DROP FUNCTION dbo.ConvertIntToAlphaID
GO
CREATE FUNCTION dbo.ConvertIntToAlphaID (@ID int)
RETURNS char(5)
AS
BEGIN
RETURN CHAR((@ID-1) / POWER(26,4) + 65)
      + CHAR ((@ID-1) % POWER(26,4) / POWER(26,3) + 65)
      + CHAR ((@ID-1) % POWER(26,3) / POWER(26,2) + 65)
      + CHAR ((@ID-1) % POWER(26,2) / 26 + 65)
      + CHAR ((@ID-1) % 26 + 65)

END
GO 
IF OBJECT_ID('dbo.GetNextAlphaID','FN') IS NOT NULL
    DROP FUNCTION dbo.GetNextAlphaID
GO
CREATE FUNCTION dbo.GetNextAlphaID ()
RETURNS char(5)
AS
BEGIN
    DECLARE @MaxID char(5), @ReturnVal char(5)
    SELECT @MaxID = MAX(AlphaID) FROM dbo.MyTable
    IF @MaxID < 'ZZZZZ'
        RETURN dbo.ConvertIntToAlphaID(dbo.ConvertAlphaIDToInt(@MaxID)+1)
    IF @MaxID IS NULL
        RETURN 'AAAAA'
    SELECT @MaxID = MAX(AlphaID) 
    FROM dbo.MyTable 
    WHERE AlphaID < dbo.ConvertIntToAlphaID((SELECT COUNT(*) FROM dbo.MyTable))
    IF @MaxID IS NULL
        RETURN 'AAAAA'
    RETURN dbo.ConvertIntToAlphaID(dbo.ConvertAlphaIDToInt(@MaxID)+1)
END
GO

SELECT * FROM dbo.MyTable ORDER BY dbo.ConvertAlphaIDToInt(AlphaID)
GO
SELECT  dbo.GetNextAlphaID () AS 'NextAlphaID'

By the way, if you don't want to assume contiguity, you can do as you suggested and (if there's a 'ZZZZZ' row) use the first gap in the sequence. Replace the last function with this:

IF OBJECT_ID('dbo.GetNextAlphaID_2','FN') IS NOT NULL
    DROP FUNCTION dbo.GetNextAlphaID_2
GO
CREATE FUNCTION dbo.GetNextAlphaID_2 ()
RETURNS char(5)
AS
BEGIN
    DECLARE @MaxID char(5), @ReturnVal char(5)
    SELECT @MaxID = MAX(AlphaID) FROM dbo.MyTable
    IF @MaxID < 'ZZZZZ'
        RETURN dbo.ConvertIntToAlphaID(dbo.ConvertAlphaIDToInt(@MaxID)+1)
    IF @MaxID IS NULL
        RETURN 'AAAAA'
    SELECT TOP 1 @MaxID=M1.AlphaID
    FROM dbo.Mytable M1
    WHERE NOT EXISTS (SELECT 1 FROM dbo.MyTable M2 
                      WHERE AlphaID = dbo.ConvertIntToAlphaID(dbo.ConvertAlphaIDToInt(M1.AlphaID) + 1 )
                     )
    ORDER BY M1.AlphaID
    IF @MaxID IS NULL
        RETURN 'AAAAA'
    RETURN dbo.ConvertIntToAlphaID(dbo.ConvertAlphaIDToInt(@MaxID)+1)
END
GO

To return the next ID for a given ID (with rollover), use:

SELECT  COALESCE
        (
        (
        SELECT  TOP 1 id
        FROM    mytable
        WHERE   id > @id
        ORDER BY
                id
        ),
        (
        SELECT  TOP 1 id
        FROM    mytable
        ORDER BY
                id
        )
        ) AS nextid

This query searches for the ID next to the given. If there is no such ID, it returns the first ID.

Here are the results:

WITH mytable AS
        (
        SELECT  'AAA' AS id
        UNION ALL
        SELECT  'BBB' AS id
        UNION ALL
        SELECT  'CCC' AS id
        UNION ALL
        SELECT  'DDD' AS id
        UNION ALL
        SELECT  'EEE' AS id
        )
SELECT  mo.id,
        COALESCE
        (
        (
        SELECT  TOP 1 id
        FROM    mytable mi
        WHERE   mi.id > mo.id
        ORDER BY
                id
        ),
        (
        SELECT  TOP 1 id
        FROM    mytable mi
        ORDER BY
                id
        )
        ) AS nextid
FROM    mytable mo

id      nextid
-----   ------
AAA     BBB
BBB     CCC
CCC     DDD
DDD     EEE
EEE     AAA

, i. e. it returns BBB for AAA, CCC for BBB, etc., and, finally, AAA for EEE which is last in the table.

Coderer : Could you edit to explain what this actually does?

I think the lowest-impact solution for my needs is to add an identity column. The one thing I can guarantee is that the ordering will be such that entries that should "come first" will be added first -- I'll never add one with identifier BBBB, then go back and add BBBA later. If I didn't have that constraint, obviously it wouldn't work, but as it stands, I can just order by the identity column and get the sort I want.

I'll keep thinking about the other suggestions -- maybe if they "click" in my head, they'll look like a better option.

Looking for pointers on MIDI driver development

I am looking for resources, documentation and general advices on writing a virtual MIDI device (see my previous question for reasons) and basics of generating MIDI events from a VST plugin.

The target platform is Windows, but Mac compatibility would be a plus.

From stackoverflow

I hope you don't mind me editing the subject of your question -- but I think that you shouldn't be so worried about the VST part of your software here, and the nature of your question is more about driver development than VST development.

That said, you are essentially trying to write a normal MIDI driver, but you don't actually need to connect to any hardware. So writing a virtual driver is really not as hard as you anticipate... you just need to find the driver API's (see the edit to my answer on your previous question).

One more thing I should add here, is that the best way to accomplish what you are trying to use a separate application with Rewire. So that is to say, you write an application which initializes a MIDI device when started, and connects to the sequencer via Rewire. Your users would start up the sequencer first, then your application, then send the audio on a bus track to your app, and configure their sequencer to receive MIDI from the device driver which you would write. The app, conversely, streams audio from rewire, does some type of FFT to get the pitches (or whatever you want to do to the audio stream), and then pushes those events out to the host through the MIDI driver API. IMO, that's probably the best way to solve this problem here, if I understand your project correctly.

Luper Rouch : Yes, you understood what I want to do perfectly, but I'm looking for documentation material now.
Couldn't you just use existing virtual Midi devices such as Midi Yoke?

Emile Vrijdags : Does midi yoke work on 64bit windows?

SQL Server 2005 - Trigger Loop?

I am using triggers for the first time.

If I update a field in a table by an update trigger on the same table, with this spark a loop? Does sql server guard against this recursive behavior?

Thanks

From stackoverflow

This page (search for RECURSIVE_TRIGGERS) describes some of the database settings you can use to modify this behavior. Also, one way to safeguard your procedures is to use either the UPDATE() function or the COLUMNS_UPDATED() function.

If, for example, you have a table with columns A, B, and C, and you want the value of C to change automagically when the value in column B is updated, you can protect the call in the trigger:
```
CREATE TRIGGER Whatever ON TableName AFTER UPDATE
AS
BEGIN
    IF UPDATE(B)
    BEGIN
        /* Update column C here */
    END
END
```
This way you avoid calling the trigger recursively when column C is updated by your trigger. COLUMNS_UPDATED() is also useful, but I find it to be fragile (relies on position of column instead of column name).

Sung Meister : +1 great link and an example in the answer.
You can control recursion of triggers at the DB level via the RECURSION_TRIGGER option; it's turned off by default. Even if this option is turned on, there is a limit of 32 nested levels of triggers; all changes will be rolled back if your exit condition didn't stop the recursion before reaching the limit of 32 levels.

NSIndexPath - how to?

Hi, I want to implement index feature in UITableView as is in the standart contacts application.

Is there an elaborate tutorial how to achieve this goal? What is the right way to do so?

Data in my table is sorted alphabetically and when a user presses on a certain letter in the index I want to scroll to the same letter in the table.

How do I return number of rows in each section? Currently I have a function that calculates number of items in every section. Is there a way to calculate it automatically? When I get to a method cellForRowAtIndexPath I need to know an actual number of an item in the table, not the one in a row. How do I get it? i.e. I have 50 items in my table and 26 sections. When it comes to an item 30, I want to get 30, not "section:3 row:2". Is there a way to do so?

Thank you in advance.

From stackoverflow

As for the list on the side, use the UITableViewDataSource method sectionIndexTitlesForTableView:

Keep in mind it only works if you're using UITableViewStylePlain

Handling missing image files when displaying XAML

I'm using XamlReader.Load() to load an external XAML file which may contain some images and other media.

I'm using ParserContext to set the baseUri, so if the paths inside the XAML file are correct, the media is loaded successfully.

// works
var pc = new ParserContext();
pc.BaseUri = new Uri(baseUri, UriKind.RelativeOrAbsolute);

using (FileStream stream = new FileStream(filename, FileMode.Open, FileAccess.Read))
    return XamlReader.Load(stream, pc);

Is there a way for me to figure out which media files will fail to load? That is, if the XAML file contains an Image tag with the source set to a path that doesn't exist, can I tell at runtime that an image is not being displayed correctly? (Images will be the most common, but other controls like MediaElement might load external files too.)

I can think of a few possible approaches:

Parse the XAML file to find any filenames, and check each of them directly.
When displaying the resulting visual tree, look for some sort of RoutedEvent that would indicate a problem loading external files

Any tips?

Edit: I'm thinking the XamlReader stuff may be a distraction; I'm getting a valid visual tree, but it's the controls themselves that may have a problem when trying to load files that don't exist. Those are the errors I want to catch.

From stackoverflow

If you are linking your image via binding them to an background object, you will find errors with binding in WPF do not raise exceptions. Have a look at this blog article which may help get around this limitation.

The Media Element however, is simpler as it has an event (MediaFailed) which you can catch in order to determine if it has failed.
```
<MediaElement Source="filePath.wmv" MediaFailed="EventHandler"/>
```

What are some of your oldest programming books that you still use?

Most books on a programmer's bookshelf are very new. What are some books that have stood the test of time? What are some of the oldest programming books you still refer to?

Please list one book per answer so they can be voted on individually.

From stackoverflow

The Mythical Man-Month, Fred Brooks, 1975.

Kb : Still this is relevant +1

Novelocrat : Note that there is a newer edition published in 1995, with some new essays, some retrospective, and better/cleaner typesetting.
"The C Programming Language" (K&R2) by Brian W. Kernighan and Dennis M. Ritchie.

1988 (not 1978! I wrote K&R2 for a reason!)

It is indispensable.

Anthony Cuozzo : I'm sorry if the post seems confusing, but someone edited it with the year of the first edition (instead of the year of the second edition).

Simucal : @Anthony Cuozzo, I didn't think it was confusing really.. I just wanted to add a date to keep in line with the rest of the posts, and then I gave it +1!

Cervo : I have the second edition as well, but a large part of even the first edition is still relevant.

Anthony Cuozzo : @Simucal: I understand. No offense meant or taken.

Anthony Cuozzo : @Cervo: I definitely agree.
Not that old, but I still like thumbing through The Mythical Man-Month

Though my copy is from 1995 not 1975 :'(

Greg Hewgill : I have the 1995 edition, but only because I loaned my 1975 edition to somebody along the way and never saw it again.

MaxVT : Oh, the 20th Anniversary Edition! Have one.

Jonathan Leffler : There is a higher-voted MMM.
Algorithms in C++, Robert Sedgewick, 1992.

RBerteig : +1, but I have the earlier Algorithms in C version here. Nice clear exposition, with well thought-out illustrations.
Programming Perl (O'Reilly Camel book), 1991. Always good for a refresher in perl.

Cervo : I have never read the entire book, but this book is great, not only is it a pretty comprehensive reference on the Perl language, but it is a pretty interesting read instead of being dry like most references....
"Dragon Book" by Alfred V. Aho, Ravi Sethi, and Jeffrey D. Ullman
The Art of Computer Programming, Donald Knuth, 1973 (and earlier).

Gabe Moothart : I know that this is a classic, but (at the risk of being a heretic), what do you use it for? I found it mostly full of a lot of low-level algorithms that just aren't necessary in a modern programming environment. _Introduction to Algorithms_, by Cormen et al, seems more relevant
Domain-Driven Design: Tackling Complexity in the Heart of Software (2003) by Eric Evans
Programming Pearls, by Jon Bentley. First edition was published 1985.

hughdbrown : Awesome book...
I keep my copy of the LaTeX user's guide (1986) on my desk at home.

timday : Yup, that's one book I had to repurchase because I got rid of my original copy in the 90s after using FrameMaker/SGML a lot and thinking LaTeX was obsolete. Big mistake!
Design Patterns: Elements of Reusable Object-Oriented Software (1995) by GoF
"The Theory of Parsing, Translation, and Compiling" (volumes 1 and 2) by Alfred V. Aho and Jeffrey D. Ullman. Published in 1972!
"Advanced Programming in the Unix Environment (APUE) by W. Richard Stevens
The C++ Programming Language - 1986 (although I also have the third edition from 1997).

Dmitri Nesteruk : Same here, but I use 3rd edition.

Rob K : I have the second and the special edition.
Numerical recipes in C, 2nd ed, by Press/Teukolsky/Vetterling/Flannery. 1992.

MarkJ : Yep - although I use the Fortran version. (Gasp!)

Ken : That book has brought me nothing but pain -- see http://www.uwyo.edu/buerkle/misc/wnotnr.html
The Elements of Programming Style by Kernighan & Plauger (2nd Edn, 1978) - reread periodically to remind me what I should be doing.
Introduction to Algorithms (1990) from MIT Press

Cervo : Not the easiest read, but it is a pretty comprehensive algorithm study. I have the second edition but not that many changes over the first.

Gabe Moothart : +1 Best algorithms resource ever

elhoim : Third edition is out BTW: http://www.amazon.com/Introduction-Algorithms-Third-Thomas-Cormen/dp/0262033844
Structure and Interpretation of Computer Programs (The Wizard Book) by Abelson and Sussman. (1985, MIT Press)

Cervo : I'm still going through this book, a completely amazing read!!!
Data Structures by Sahni
# 1996. Smalltalk Best Practice Patterns. Prentice Hall.

Even though it's been over 10 years since I programmed smalltalk, I occasionally read it to be inspired by beautiful code and Kent Beck's amazing thoughts on what constitutes good code. Really a subtle book with a huge story to tell.

The book has been more or less re-released as a java version, which I also have, But if I want something beautiful it's always the original (much nice can be said about java, but it's not a beautiful language like smalltalk)
Code Complete Steve McConnell, 1993. It's not that old, but it is definitely a classic.
Jerry Weinberg's "The Psychology of Computer Programming" I'm now on to my fifth copy, the new updated 25th anniversary edition (sanitised Amazon link), as people keep borrowing them and then I never see it again! (-:

I first read about it in Ed Yourdon's book "The Decline and Fall of the American Programmer" (sanitised Amazon link). This book is rather dated now with it's emphasis on CASE tools, but the appendix about the programmer's bookshelf is definitely worth the purchase price which is currently $0.01 for a used copy! There are quite a few books listed here that you wouldn't normally see on a programmer's bookshelf. Best thing is the accompanying text where Ed talks about why he's included the books.

HTH

cheers,

Rob
Rapid Development - Steve McConnell- Amazon Link Published 1996. Provides a lot of good advice about managing the process of software development.
Peopleware by Tom de Marco and Timothy Lister. First published 1987. Insights into the people aspects of development
Computer Graphics: Principles and Practice in C (2nd Edition) (1995) by James D. Foley, Andries van Dam, Steven K. Feiner and John F. Hughes.
"Algorithms + Data Structures = Programs", Niklaus Wirth, 1978
Besides a lot of the books other people have mentioned: The Black Art of 3D Game Programming, Lamothe, 1995. I learned the basic math of 3D graphics from this book, and still use it as a reference all the time.

Jasper Bekkers : I immediately bought a copy after reading your post. This book seems very informative and useful. Thanks!
Hart, John. "Computer Approximations", 1968 (reprinted 1978). Invaluable for calculation of nontrivial functions like sqrt/cos/sin/log/erf etc., though it's out of print. (see Jack Ganssle's website for some discussion.)
Expert C Programming - Deep C Secrets - 1993

This is one of my favorites that I like to re-read every now and then.
Martin Fowler's "UML Distilled" is the still the best UML book out there. It's a rarity in today's publishing world: thin, but packed with good information.

I see it's in its third edition now, which is testament to its staying power, but I bought the first edition back in 1997.
P. J. Brown, writing interactive compilers and interpreters, early 1970s. (Mine is in a box because my office shelves are not installed yet; thus the inexact date.)
Fundamentals of Operating Systems, 5th edition - A.M Lister and R.D. Eager (1993)

Don't actually program for OS's but I enjoyed reading it the first few times, and I like to refresh on some topics now and then.
Design Patterns Elements of Reusable Object-Oriented Software - GOF (1995) Introduction to Algorithms, 2nd edition (2001) - MIT Press

Grouping these tow as they have already been mentioned, but I thought those are really the books I refer to often.
Here's another classic: Numerical Methods that Work. First published in 1970, reprinted in 1990. Still one of my favorite books on numerical analysis.
Interesting question. Made me realize that (1) I have no "new" programming books and (2) I don't use many "old" ones anymore. Almost exclusively I use the Internet as my reference material for programming and system administration.
Fred Brooks' The Mythical Man-Month, 2nd ed. published 1995. I have read it twice and still plan to read it again.

Jonathan Leffler : There are two higher-voted MMM entries.
The UNIX Programming Environment by Kernghan & Pike, from 1984. I'm surprised no-one else mentioned it before.

Darius Bacon : I considered mentioning it, but I've only "used" it recently in the sense of using what I learned long ago. So I didn't post any book at all -- it's pretty fuzzy.

Darius Bacon : (There are plenty of 70's books like that, though I'm having trouble thinking of any from the 60s that haven't been updated.)

Bratch : Mostly I work on .Net stuff, but when I need to work in UNIX I refer to this book, or the only other UNIX book I own, UNIX In a Nutshell, with the referee on it. My copy of this from the early 90's still has green P-150 primer paint smeared on the back of it from my Navy days...
Deitel and deitel c how to program
The UNIX C Shell Field Guide -- Great read!

1986
‘Faster than thought’, ed. B.V. Bowden, London 1953. Published by Pitman Publishing.

The chapter ‘Digital computers applied to games’ is by Alan Turing.

I got this at a garage sale in 1980 for 25 cents. Now thanks to the Internet everybody can read it!

http://www.turingarchive.org/browse.php/B/7
Paradigms of Artificial Intelligence Programming: Case Studies in Common Lisp. Peter Norvig/1992
Practical C Programming. It was my very first computer book.
The Visual Display of Quantitative Information, by Edward Tufte. Actually, all the rest of his books are good too, but the first is the classic. At first glance this isn't a programming book. But it really is about the display of information, and good GUI design is really about the guiding the user's eye to the important information and controls.

It doesn't hurt that the production value on this book is impressive, either. It can be left out as a decent coffee table art book when not needed as a reference. ;-)
The Design of Everyday Things (Donald A Norman), first published in 1988.

It's not specifically a programming book, but obviously similar rules apply to anything with which a user has to interact, be it a GUI or a teapot.
Professional Java Security Wrox Press; 1st edition (May 2001)

I was reading it this morning.....Some things are dated but still a pretty good book on Java Cryptography APIs...
(I have no idea why anyone would vote this up, but here goes...)

Sitting here on my bookshelf is a copy of A Book on C, (first edition) by Al Kelley and Ira Pohl. Copyright 1984, bought 1985. The back cover has fallen off, as has part of the spine. I don't actually use C, but enough of the basics of C++ are the same that this book is useful if I want to refresh my memory of things like how to declare a ragged array, what "%" code to use for a double with no less than 5 characters, which term gets returned by the comma operator, etc.

Also, I know right where to turn in it to get at the ASCII table. Its way quicker than Google. :-)
Refactoring: Improving the Design of Existing Code. Martin Fowler, 1999

I've got plenty of older books, but honestly I don't refer back to them much at all.
Kernighan and Plauger's Software Tools.
Forget the books, read blogs dude!
Principles of Interactive Computer Graphics by William M. Newman and Robert F. Sproull, 1978, second edition, 1979.

While much of what is covered is probably already implemented in third party libraries these days, there is a lot of good information in there for those who prefer to (or must, for whatever reason) roll their own graphics routines.

It also gives you pretty good insight on how computer graphics works.
FORTH, a text and reference. Mahlon G. Kelly and Nicholas Spies. ISBN 0-13-326349-5 / ISBN 0-13-326331-2. 1986 Prentice-Hall.

Leo Brodie's books are good but this book is even better. For instance it covers defining words and the interpreter in depth.
"C++: The Complete Reference" by Herbert Schildt (third edition) has been on my shelf for about 9 years, and I still pull it down and look something up about once a week.
The Pragmatic Programmer, Andrew Hunt & Dave Thomas, 1999
Sams "Teach Yourself C in 21 * 150 Days"

More useful than the title suggests ;-)

Indications this year suggest I might have graduated...

ASP.NET MVC using UpdateModel to post child records

Continuing from this question, I've got a form with all the Vehicles listed for a Person, the vehicle fields are editable, and they are now successfully posting back to my Save action.

Now I'd like to use UpdateModel to save the data, but I'm not sure how to construct it. Here's my Save action right now:

<ActionName("Edit"), AcceptVerbs(HttpVerbs.Post)> _
Function Save(ByVal form As Person, ByVal vehicles() As Vehicle) As ActionResult
    Dim original = From v In MyDataContext.Vehicles Where v.person_id = Person.person_id
    For Each item In original
        For i = 0 To vehicles.Count - 1
            If vehicles(i).vehicle_id = item.vehicle_id Then
                UpdateModel(item, New String() {"license_nbr", "color"})
                Exit For
            End If
        Next
    Next
    MyDataContext.SubmitChanges()
    Return RedirectToAction("Index", "Home")
End Function

When I run this, it doesn't save anything, and UpdateModel doesn't throw any errors. I'm assuming I have to give it a little more direction to get the magic to work, because UpdateModel doesn't know which item in the vehicles array to use for each update.

Do I need to specify a ValueProviderResult as the third parameter to UpdateModel? If so, how do I create one out of vehicles(i)? Am I completely off base in how I have this set up?

From stackoverflow

Why use UpdateModel -- which just updates the properties from form fields -- when you already have the form fields processed into model data? Can't you just assign the values from vehicle to item directly?
```
For Each item In original
    For i = 0 To vehicles.Count - 1
        If vehicles(i).vehicle_id = item.vehicle_id Then 
           item.license_nbr = vehicles(i).license_nbr
           item.color = vehicles(i).color
           Exit For
        End If
    Next
Next
```
gfrizzle : I could definitely do that, but can't you make that argument against UpdateModel across the board? Not arguing for or against, just learning.

tvanfosson : Normally I would use UpdateModel when not using model binding, i.e., there are no parameters or only an id parameter to the method and I get the values via UpdateModel from the ValueProvider directly. In the case where you already have a bound model I don't think it makes sense to use UpdateModel.

How virtual events work in C# ?

Below is the program I used for the test. It prints (as expected):

Raise A
Event from A
Raise B
Event from B

Now, if we change first two lines of the Main to be:

        A a = new B();
        B b = new B();

the Program will print:

Raise A
Raise B
Event from B

which is also expected, as overriding event hides the private backing field in the base class and therefore events fired by the base class are not visible to clients of the derived class.

Now I am changing the same lines to:

 B b = new B();
 A a = b;

and the program starts printing:

Raise A
Raise B
Event from A
Event from B

What's going on?

class A
{
    public virtual event EventHandler VirtualEvent;
    public void RaiseA()
    {
        Console.WriteLine("Raise A");
        if (VirtualEvent != null)
        {
            VirtualEvent(this, EventArgs.Empty);
        }
    }
}
class B : A
{
    public override event EventHandler VirtualEvent;
    public void RaiseB()
    {
        Console.WriteLine("Raise B");             
        if (VirtualEvent != null)
        {
            VirtualEvent(this, EventArgs.Empty);
        }
    }
}

class Program
{
    static void Main(string[] args)
    {
        A a = new A();
        B b = new B();

        a.VirtualEvent += (s, e) => Console.WriteLine("Event from A");
        b.VirtualEvent += (s, e) => Console.WriteLine("Event from B");

        a.RaiseA();
        b.RaiseB();
    }
}

From stackoverflow

You hooked up two event handlers to the same event. Since A and B are pointing to the same object, when you call b.RaiseB() both event handlers get fired. So first you're calling RaiseA which is a basecalss method. That prints Raise A. It then doesn't actually fire off the event because it's null. Then, you're raising B but TWO handlers are hooked up to it, therefore it first prints Raise B, and when the event fires, both handlers get called.
We have a single instance (of B) which has the following fields:
- A.VirtualEvent: null
- B.VirtualEvent: Two event handlers
The call to a.RaiseA() just prints "Raise A" - but nothing more, because the private field in A is null.

The call to b.RaiseB() prints the remaining three lines, because the event has been subscribed to twice (once to print "Event from A" and once to print "Event from B").

Does that help?

EDIT: To make it clearer - think of the virtual event as a pair of virtual methods. It's very much like this:
```
public class A
{
    private EventHandler handlerA;

    public virtual void AddEventHandler(EventHandler handler)
    {
        handlerA += handler;
    }

    public virtual void RemoveEventHandler(EventHandler handler)
    {
        handlerA -= handler;
    }

    // RaiseA stuff
}

public class B : A
{
    private EventHandler handlerB;

    public override void AddEventHandler(EventHandler handler)
    {
        handlerB += handler;
    }

    public override void RemoveEventHandler(EventHandler handler)
    {
        handlerB -= handler;
    }

    // RaiseB stuff
}
```
Now is it clearer? It's not quite like that because as far as I'm aware you can't override just "part" of an event (i.e. one of the methods) but it gives the right general impression.

Prankster : It definitely helps, except it is unclear how a.VirtualEvent += (s, e) => Console.WriteLine("Event from A"); was even able to see B.VirtualEvent to subscribe to it?

Michael Meadows : +1, bang! that was the sound of my mind snapping back. I stretched my brain to its limit staring at OP's code trying to wrap my brain around it. Even if I saw the bug, I wouldn't have been able to word it so lucidly.

Michael Meadows : B.VirtualEvent hides A.VirtualEvent, because events are not polymorphic.

Prankster : @Michael Meadows: That's not what http://msdn.microsoft.com/en-us/library/ms173152(VS.80).aspx says about polymorphism.

Michael Meadows : @prankster ok, now my brain just exploded. I never knew you could qualify events with the virtual keyword. I apologize for the misinformation.

Jon Skeet : @prankster: a.VirtualEvent(...) was able to "see" B.VirtualEvent *precisely* because it was virtual. It was overriding A.VirtualEvent.

Firoso : as "JS" converts to C# in realtime.
Try making your RaiseA function protected + virtual.

A rule of thumb: If derived class overrides event accessors, it must also override the function that invokes the event.

Prankster : "If derived class overrides event accessors, it must also override the function that invokes the event.". Can you point me to the section of language specification that states this, please?

Koistya Navin : Take a look at this page: http://msdn.microsoft.com/en-us/library/8627sbea(VS.80).aspx

Koistya Navin : There is a statement "Wrap the event in a protected virtual method to enable derived classes to raise the event."

Koistya Navin : Also check this blog post: http://blogs.msdn.com/samng/archive/2007/11/26/virtual-events-in-c.aspx

Sunday, April 17, 2011

Blog Archive