I was checking out some random tutorials, and found this little nugget. I think someone is a tad bit confused about how things work in C/C++.

 #include <shellapi.h></shellapi.h>

 I know nothing whatsoever about programming, and even my eyes are watering in sympathy here. I mean, who's the guy around here who says 'Nuke it from orbit!" about pretty much anything that sucks, and can I borrow his catchphrase?

Wow.

Just... Wow.

Please tell me someone accidentially put that file through some sort of XML cleanup program... 

And this, gentlemen, is the true meaning of WTF.

What's next, if (i &gt; 0) ?

Does that actually compile? 

 warning: extra tokens at end of #include directive

 

Otherwise, it compiles and runs in g++. 

movzx:

 #include <shellapi.h></shellapi.h>

The very best wtf's truly don't really require a long explanation.

 O...M...G...

 

Google led me to Tomy Chen's blog, which unfortunately confirms this wasn't a one-time mistake

#include "stdafx.h"
#include <shellapi.h></shellapi.h>
#include <stdio.h></stdio.h>

believe it or not, I think that somebody has a screwed up stylesheet generating the webpage and a validator added the ending tags.

billhead:

Google led me to Tomy Chen's blog, which unfortunately confirms this wasn't a one-time mistake

#include "stdafx.h"
#include <shellapi.h></shellapi.h>
#include <stdio.h></stdio.h>

So that's how Chinese C/C++ is different than the English one. (Just kidding)

: IF` 0=GOTO` ORIG ;
: THEN` HERE SWAP ! ;
: ELSE` GOTO` ORIG SWAP THEN` ;

 mmm... so I was reading through that paula thread, and the last page or so of comments appears to have been linkbotted.

Is there a way to easily report that abuse?

 -MBirchmeier

MBirchmeier:

Is there a way to easily report that abuse?

Check the link at the bottom right corner of these posts .. the that says "report abuse" Feel free to use it :D

OzPeter:

MBirchmeier:

Is there a way to easily report that abuse?

Check the link at the bottom right corner of these posts .. the that says "report abuse" Feel free to use it :D

 

The one that doesn't exist in front page comments you mean?

This HAS to be a misguided auto-formatter. The rest of the code there is too sane to have been written by someone who would write that atrocity.

Simply clbuttic!!

 How the fuck do you screw up <stdio.h> and such O_O;

 

Even copy and paste code should have that right.

 

I'm not sure what the bigger WTF is; that horrible horrible include exists, or it compiles ....

movzx:

 #include <shellapi.h></shellapi.h>

Well, still much better than the tutorials that just seem to have "#include", until you happen to look at the page source code.

Take a look here: http://www.shield.on.ca/Blog/?p=105 There you'll find the same thing, but it also has does strange things to the ampersand. However, I don't understand why, since the tags are clearly not part of the HTML/whatever code.

Ah, I get the joke. It's supposed to be this, right?

#include <shellapi.h />

 They are both using some html code colorizer that is obviously overzealously closing tags.

rbowes:

Ah, I get the joke. It's supposed to be this, right?

#include <shellapi.h />

 

The most correct application would be:

<includes>
    <include name="shellapi.h" />
</includes>

;)

Soviut:

The most correct application would be:

<includes>
    <include name="shellapi.h" />
</includes>

I disagree. The usage of attributes for content is becoming deprecated. Attributes should be descriptive of the content within the tag.

More Correct Example:

<?xml version="1.0" encoding="UTF-8"?>
<includes xmlns="http://forums.thedailywtf.com/xml/includes">
<include type="header">shellapi.h</include>
</includes>

joe.edwards:

More Correct Example:

<?xml version="1.0" encoding="UTF-8"?>

<includes xmlns="http://forums.thedailywtf.com/xml/includes">
<include type="header">shellapi.h</include>
</includes>

 

Ah, ah, but this doesn't even has a schema... how are you supposed to validate it?

Also, in case you wanted to extend the "type" attribute, how would you avoid name collisions?

Obviously, the correct example is:

<includes
     xmlns = "http://forums.thedailywtf.com/xml/includes"
     xmlns:cinc = "http://forums.thedailywtf.com/xml/C-include-types"
     xmlns:xsi = "http://www.w3.org/2001/XMLSchema-instance"
     xsi:schemaLocation = "http://forums.thedailywtf.com/xml/includes http://forums.thedailywtf.com/xml/includes.xsd"
>
  <include type="cinc:header">shellapi.h</include>
</includes>

PSWorx:

joe.edwards:

More Correct Example:

<?xml version="1.0" encoding="UTF-8"?>

<includes xmlns="http://forums.thedailywtf.com/xml/includes">
<include type="header">shellapi.h</include>
</includes>

 

Ah, ah, but this doesn't even has a schema... how are you supposed to validate it?

Also, in case you wanted to extend the "type" attribute, how would you avoid name collisions?

Obviously, the correct example is:

<includes
     xmlns = "http://forums.thedailywtf.com/xml/includes"
     xmlns:cinc = "http://forums.thedailywtf.com/xml/C-include-types"
     xmlns:xsi = "http://www.w3.org/2001/XMLSchema-instance"
     xsi:schemaLocation = "http://forums.thedailywtf.com/xml/includes http://forums.thedailywtf.com/xml/includes.xsd"
>
  <include type="cinc:header">shellapi.h</include>
</includes>

O...M...G...

I think you just invented ... ACAX.

Nice, all it needs now is a header saying whether it's UTF-8, UTF-32 or some other encoding.

edit: missed it from a previous post XD