unified_texts stringlengths 32 30.1k | OpenStatus_id int64 0 4 | input_ids list | token_type_ids list | attention_mask list |
|---|---|---|---|---|
Page index
===
Help me ..my page index is not working in visual studio.
my page load is as follows:
protected void Page_Load(object sender, EventArgs e)
{
if (!Page.IsPostBack)
{
CustomerView.DataSource = Customer.GetAll();
CustomerView.DataBind();
}
}
protected void CustomerView_PageIndexChanging(object sender, System.Web.UI.WebControls.GridViewPageEventArgs e)
{
int newPageNumber = e.NewPageIndex + 1;
CustomerView.DataSource = Customer.GetAll();
CustomerView.DataBind();
}
what am i doing wrong my page index in not working. | 0 | [
2,
2478,
4348,
800,
3726,
3726,
448,
55,
13,
9,
9,
915,
2478,
4348,
25,
52,
638,
19,
3458,
1120,
9,
51,
2478,
6305,
25,
28,
2415,
45,
3803,
11364,
2478,
1,
8294,
5,
23793,
2660,
106,
15,
807,
10663,
18,
13,
62,
6,
13,
1,
100... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
if you reimplemented twitter, what would you do differently?
===
I just saw the hilarious ["The Rise and Fall of Twitter"][1] which made me think:
if you reimplemented twitter, what would you do differently?
What technologies would you use? What languages?
How do you ensure that the service is scalable?
What else would you change?
[1]: http://zenhabits.tumblr.com/post/50512151/the-rise-and-fall-of-twitter | 0 | [
2,
100,
42,
302,
8983,
413,
1130,
69,
10623,
15,
98,
83,
42,
107,
12670,
60,
800,
3726,
3726,
31,
114,
441,
14,
12045,
8756,
267,
636,
7,
124,
2529,
17,
1080,
16,
10623,
7,
500,
2558,
165,
500,
56,
117,
55,
277,
45,
100,
42,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
How to deal with "java.lang.OutOfMemoryError: PermGen space" error
===
Recently I ran into this error in my web application:
java.lang.OutOfMemoryError: PermGen space
It's a typical Hibernate/JPA + IceFaces/JSF application running on Tomcat 6 and JDK 1.6.
Apparently this can occur after redeploying an application a few times. | 0 | [
2,
184,
20,
1183,
29,
13,
7,
1004,
1385,
9,
9949,
9,
1320,
1041,
790,
5171,
93,
29992,
45,
416,
79,
1863,
726,
7,
7019,
800,
3726,
3726,
1989,
31,
717,
77,
48,
7019,
19,
51,
2741,
3010,
45,
8247,
9,
9949,
9,
1320,
1041,
790,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Using the same App_Code classes across websites
===
Let's say you have a solution with two website projects, Website A and Website B. Now inside Website A's App_Code folder, there is a Class X defined in a ClassX.cs file. What do you do if Website B also needs access to ClassX.cs?
Is there any way to share this file across App_Code folders? Assume that moving the file to a common library is out of the question. | 0 | [
2,
568,
14,
205,
4865,
1,
9375,
2684,
464,
13931,
800,
3726,
3726,
408,
22,
18,
395,
42,
57,
21,
4295,
29,
81,
2271,
2314,
15,
2271,
21,
17,
2271,
334,
9,
130,
572,
2271,
21,
22,
18,
4865,
1,
9375,
19294,
15,
80,
25,
21,
718... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
How do you get selenium to recognize that a page loaded?
===
In certain unknown situations selenium does not detect that a page has loaded when using the open method. I am using the Java API. For example (This code will not produce this error. I don't know of an externally visible page that will.):
Selenium browser = new DefaultSelenium("localhost", 4444, "*firefox", "http://www.google.com");
browser.start();
browser.open("http://www.google.com/webhp?hl=en");
browser.type("q", "hello world");
When the error occurs, the call to 'open' times out, even though you can clearly see that the page has loaded successfully before the timeout occurs. Increasing the timeout does not help. The call to 'type' never occurs, no progress is made.
How do you get selenium to recognize that the page has loaded when this error occurs? | 0 | [
2,
184,
107,
42,
164,
23027,
14311,
20,
5844,
30,
21,
2478,
8572,
60,
800,
3726,
3726,
19,
1200,
2562,
8329,
23027,
14311,
630,
52,
9092,
30,
21,
2478,
63,
8572,
76,
568,
14,
368,
2109,
9,
31,
589,
568,
14,
8247,
21,
2159,
9,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
in ColdFusion 8, can you declare a function as private using cfscript?
===
Normally you create a function using cfscript like:
<cfscript>
function foo() { return "bar"; }
</cfscript>
Is there a way to declare this as a private function, available only to other methods inside the same cfc?
I know you can do it with tags:
<cffunction name="foo" access="private">
<cfreturn "bar">
</cffunction>
But I don't want to have to rewrite this large function thats already written in cfscript. | 0 | [
2,
19,
1191,
9818,
469,
15,
92,
42,
10123,
21,
1990,
28,
932,
568,
10084,
8741,
60,
800,
3726,
3726,
4147,
42,
1600,
21,
1990,
568,
10084,
8741,
101,
45,
13,
1,
8940,
8741,
1,
1990,
4310,
111,
5,
6,
13,
1,
788,
13,
7,
1850,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Time Synchronization Ubuntu Server Under Parallels
===
I've installed Ubuntu Server (8.04) into Parallels and found that the system time/clock ran fast to the extent that it would gain hours over time. | 0 | [
2,
85,
13,
16023,
1829,
287,
12968,
2473,
8128,
131,
3821,
18,
800,
3726,
3726,
31,
22,
195,
4066,
287,
12968,
2473,
8128,
13,
5,
457,
9,
3277,
6,
77,
3821,
18,
17,
216,
30,
14,
329,
85,
118,
10790,
717,
1512,
20,
14,
5674,
30... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... |
macro support in F#
===
After reading Practical Common Lisp I finally understood what the big deal about macros was, and I have been looking for a language for the .NET platform that supports this. There are a few lisp dialects for .NET but from what I have been able to gather all are either very beta or abandoned. Recently my interest has been sparked by Clojure, but it's for the java platform and while on probably could use ikvm it doesn't feel some integrated. Especially when you want to do stuff like WPF.
Recently I have been hearing whisper about F#, I tried to look at the documentation if I could find anything about macro support, but haven't found it. So does anyone know?
Thanks :) | 0 | [
2,
9069,
555,
19,
398,
5910,
800,
3726,
3726,
75,
1876,
5713,
757,
2093,
3401,
31,
722,
4014,
98,
14,
580,
1183,
88,
9069,
18,
23,
15,
17,
31,
57,
74,
699,
26,
21,
816,
26,
14,
13,
9,
2328,
2452,
30,
6747,
48,
9,
80,
50,
2... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Adding cookies to Seaside responses (without redirecting)
===
I'm making a small web application in Seaside. I have a login component, and after the user logs in I want to send along a cookie when the next component renders itself. Is there a way to get at the object handling the response so I can add something to the headers it will output?
I'm trying to avoid using WASession>>redirectWithCookies since it seems pretty kludgey to redirect only because I want to set a cookie.
Is there another way that already exist to add a cookie that will go out on the next response?
| 0 | [
2,
4721,
19396,
20,
21994,
13231,
13,
5,
14506,
302,
14706,
68,
6,
800,
3726,
3726,
31,
22,
79,
544,
21,
284,
2741,
3010,
19,
21994,
9,
31,
57,
21,
6738,
108,
5912,
15,
17,
75,
14,
4155,
18893,
19,
31,
259,
20,
2660,
303,
21,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
How best to generate a random string in Ruby
===
I'm currently using the following to generate an 8 character psydo random upper case string [A-Z]
value = ""; 8.times{value << (65 + rand(25)).chr}
but it looks junky, and since it isn't a single statement it can't be passed as an argument. To get a mixed case string [a-zA-Z] I further hack into it with
value = ""; 8.times{value << ((rand(2)==1?65:97) + rand(25)).chr}
Just looks like trash. Anyone have a better method?
| 0 | [
2,
184,
246,
20,
7920,
21,
5477,
3724,
19,
10811,
800,
3726,
3726,
31,
22,
79,
871,
568,
14,
249,
20,
7920,
40,
469,
925,
13,
16033,
537,
5477,
1508,
610,
3724,
636,
58,
8,
380,
500,
1923,
800,
13,
7,
7,
73,
469,
9,
891,
18,... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
What's the single biggest reason Senior Executives tolerate pooly-performing software development projects and teams?
===
Many senior executives of businesses that rely on custom software - within their products or underpinning their company's services - seem to tolerate or avoid tackling long standing dysfunctionalities in the projects and teams developing that software. In your personal experience, why should this be so? | 4 | [
2,
98,
22,
18,
14,
345,
3835,
1215,
1101,
17103,
14723,
3067,
93,
8,
1432,
12362,
2306,
522,
2314,
17,
952,
60,
800,
3726,
3726,
151,
1101,
17103,
16,
4426,
30,
12139,
27,
5816,
2306,
13,
8,
363,
66,
1985,
54,
131,
306,
18767,
6... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
How do I unit test an __init__() method of a python class with assertRaises()?
===
I have a class:
class MyClass:
def __init__(self, foo):
if foo != 1:
raise Error("foo is not equal to 1!")
and a unit test that is supposed to make sure the incorrect arg passed to the constructor properly raises an error:
def testInsufficientArgs(self):
foo = 0
self.assertRaises((Error), myClass = MyClass(Error, foo))
But I get...
NameError: global name 'Error' is not defined
Why? Where should I be defining this Error object? I thought it was built-in as a default exception type, no? | 0 | [
2,
184,
107,
31,
1237,
1289,
40,
13,
1,
108,
242,
1,
5,
6,
2109,
16,
21,
20059,
718,
29,
10908,
9618,
7202,
5,
6,
60,
800,
3726,
3726,
31,
57,
21,
718,
45,
718,
51,
1898,
45,
6312,
13,
1,
108,
242,
1,
5,
8411,
15,
4310,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
does elmah handle caught exceptions as well
===
I was wondering if something like elmah logged exceptions even when they do not bubble up to the application? I'd like to pop up a message when an exception occurs and still log the exception. Currently I've been putting everything in try catch blocks and spitting out messages, but this gets tedious. | 0 | [
2,
630,
931,
12016,
3053,
1383,
13392,
28,
134,
800,
3726,
3726,
31,
23,
5712,
100,
301,
101,
931,
12016,
13,
19287,
13392,
166,
76,
59,
107,
52,
10937,
71,
20,
14,
3010,
60,
31,
22,
43,
101,
20,
1675,
71,
21,
2802,
76,
40,
53... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Which is better: shipping a buggy feature or not shipping the feature at all?
===
this is a bit of a philosophical question. I am adding a small feature to my software which I assume will be used by most users but only maybe 10% of the times they use the software. In other words, the software has been fine without it for 3 months, but 4 or 5 users have asked for it, and I agree that it should be there.
The problem is that, due to limitations of the platform I'm working with (and possibly limitations of my brain), "the best I can do" still has some non-critical but noticeable bugs - let's say the feature as coded is usable but "a bit wonky" in some cases.
What to do? Is a feature that's 90% there really "better than nothing"? I know I'll get some bug reports which I won't be able to fix: what do I tell customers about those? Should I live with unanswered feature requests or unanswered bug reports? | 0 | [
2,
56,
25,
574,
45,
7405,
21,
6256,
2687,
1580,
54,
52,
7405,
14,
1580,
35,
65,
60,
800,
3726,
3726,
48,
25,
21,
1142,
16,
21,
9538,
1301,
9,
31,
589,
4721,
21,
284,
1580,
20,
51,
2306,
56,
31,
7158,
129,
44,
147,
34,
127,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
How to get Intellisense on error-marked code in Visual Studio 2005?
===
When I try to compile code on VS 2005 and it fails, the line which causes the error gets underlined blue, and mouse-hovering over it displays the error message. Fine, but you can't see object types or whatever, because Intellisense will show the error message, and not object info.
In this image, I wanted to see what type `DateTime.Subtract()` returns, but VS insists on showing the error message:
![alt text][1]
Does anyone know how to get the error message out of the way, **once you've got enough of it**?
[1]: http://img502.imageshack.us/img502/6962/vs2005errordl7.png | 0 | [
2,
184,
20,
164,
14635,
3159,
6498,
27,
7019,
8,
22048,
1797,
19,
3458,
1120,
812,
60,
800,
3726,
3726,
76,
31,
1131,
20,
26561,
1797,
27,
4611,
812,
17,
32,
13614,
15,
14,
293,
56,
4047,
14,
7019,
3049,
131,
15875,
705,
15,
17,... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
What is the best C# to VB.net converter?
===
While searching the interweb for a solution for my VB.net problems I often find helpful articles on a specific topic, but the code is C#. That is no big problem but it cost some time to convert it to VB manually.
There are some sites that offer code converters from C# to VB and vice versa, but to fix all the flaws after the code-conversion is nearly as time-consuming as doing it by myself in the first place.
Till now I am using [http://labs.developerfusion.co.uk/convert/csharp-to-vb.aspx][1]
Do you know something better?
[1]: http://labs.developerfusion.co.uk/convert/csharp-to-vb.aspx | 0 | [
2,
98,
25,
14,
246,
272,
5910,
20,
13,
20468,
9,
2328,
25893,
60,
800,
3726,
3726,
133,
5792,
14,
1480,
14113,
26,
21,
4295,
26,
51,
13,
20468,
9,
2328,
1716,
31,
478,
477,
15600,
3376,
27,
21,
1903,
8303,
15,
47,
14,
1797,
25... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Is there an elegant way to compare a checkbox and a textbox using ASP.NET validators?
===
I have an Asp.Net repeater, which contains a textbox and a checkbox. I need to add client-side validation that verifies that when the checkbox is checked, the textbox can only accept a value of zero or blank.
I would like to use one or more of Asp.Net's validator controls to accomplish this, to provide a consistent display for client side errors (server-side errors are handled by another subsystem).
The Asp:CompareValidator doesn't seem to be flexible enough to perform this kind of complex comparison, so I'm left looking at the Asp:CustomValidator.
The problem I'm running into is that there doesn't seem to be any way to pass custom information into the validation function. This is an issue because the ClientIds of the checkbox and the textbox are unknown to me at runtime (as they're part of a Repeater).
So... My options seem to be:
1. Pass the textbox and checkbox to the CustomValidator somehow (doesn't seem to be possible).
2. Find the TextBox through JavaScript based on the arguments passed in by the CustomValidator. Is this even possible, what with the ClientId being ambiguous?
3. Forget validation entirely, and emit custom JavaScript (allowing me to pass both ClientIds to a custom function).
Any ideas on what might be a better way of implementing this?
| 0 | [
2,
25,
80,
40,
11614,
161,
20,
11590,
21,
2631,
5309,
17,
21,
1854,
5309,
568,
28,
306,
9,
2328,
7394,
9922,
60,
800,
3726,
3726,
31,
57,
40,
28,
306,
9,
2328,
6830,
106,
15,
56,
1588,
21,
1854,
5309,
17,
21,
2631,
5309,
9,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Where do I start learning about GUI programming?
===
I have a pretty good knowledge of programming languages like C/C++, Java, Python. But they were all mostly learnt in a college / high school class room setting where the best user interface was a numbered menu. You know, the standard data structures, implementation of various algorithms, file handling and the like.
What I want to do now is to get into GUI programming. I am not sure if I am asking the right way, but I am looking at using the WIMP paradigm (windows icons menus pointers). I want to place buttons and forms. Event driven programming, I believe is the right word, where my application waits till the user clicks something, types something etc.
Given my background, where would be a good place to start? I am looking at the following requirements -
1> Preferably cross platform.
2> Lots of documentations, tutorials, and if possible sample code that I can learn off of
3> A good GUI builder tool, where I can drag / drop stuff the way I want them to be displayed.
Any ideas or suggestions to get me started? | 0 | [
2,
113,
107,
31,
799,
2477,
88,
9457,
3143,
60,
800,
3726,
3726,
31,
57,
21,
1772,
254,
1918,
16,
3143,
2556,
101,
272,
118,
150,
20512,
15,
8247,
15,
20059,
9,
47,
59,
46,
65,
1555,
2484,
38,
19,
21,
314,
13,
118,
183,
116,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
How do I duplicate a whole line in Emacs?
===
I saw [this same question for VIM][1] and it has been something that I myself wanted to know how to do for Emacs. In ReSharper I use CTRL-D for this action. What is the least number of commands to perform this in Emacs?
[1]: http://stackoverflow.com/questions/73319/duplicate-a-whole-line-in-vim#73357 | 0 | [
2,
184,
107,
31,
19429,
21,
979,
293,
19,
13,
62,
6893,
18,
60,
800,
3726,
3726,
31,
441,
636,
1565,
205,
1301,
26,
1790,
79,
500,
2558,
165,
500,
17,
32,
63,
74,
301,
30,
31,
992,
417,
20,
143,
184,
20,
107,
26,
13,
62,
6... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
ISS error CS0433: name collision
===
In our application we've run into an error numerous times where we get error CS0433, which complains about a name collison in two separate dlls. This is an ASP.NET app developed in C# using webforms. It always complained about A TimeLog page. Anyone have advice fpr resolving this error? | 0 | [
2,
25,
18,
7019,
272,
18,
3277,
4048,
45,
204,
11319,
800,
3726,
3726,
19,
318,
3010,
95,
22,
195,
485,
77,
40,
7019,
1548,
436,
113,
95,
164,
7019,
272,
18,
3277,
4048,
15,
56,
15310,
18,
88,
21,
204,
19830,
528,
19,
81,
1725... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
In AIML, what's the XSD-valid way to use the element <set name="it"> ?
===
In file Atomic.aiml, part of the [annotated ALICE AIML files][1], there are a lot of categories like this:
<category>
<pattern>ANSWER MY QUESTION</pattern>
<template>Please try asking <set name="it">your question</set> another way.
</template>
</category>
This code isn't valid according to the [AIML XSD][2]; the validator says that _No character data is allowed in content model_ (regarding the **your question** character data inside the set element). If I delete **your question** the error disappears, but then "it" wouldn't be defined correctly.
How do I fix the code above so it passes the validation ?
[1]: http://www.alicebot.org/aiml/aaa/Atomic.aiml
[2]: http://docs.aitools.org/aiml/schema/AIML.xsd | 0 | [
2,
19,
5226,
255,
15,
98,
22,
18,
14,
993,
18,
43,
8,
18506,
43,
161,
20,
275,
14,
4520,
13,
1,
3554,
204,
3726,
7,
242,
7,
1,
13,
60,
800,
3726,
3726,
19,
3893,
9692,
9,
3622,
8184,
15,
141,
16,
14,
636,
210,
1270,
1669,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
How can I detect if caps lock is toggled in Swing?
===
I'm trying to build a better username/password field for my workplace and would like to be able to complain when they have their caps lock on.
Is this possible? And if so I'd like to have it detected before the client types their first letter.
| 0 | [
2,
184,
92,
31,
9092,
100,
9293,
3991,
25,
20,
4572,
1294,
19,
5587,
60,
800,
3726,
3726,
31,
22,
79,
749,
20,
1895,
21,
574,
4155,
7259,
118,
6201,
9587,
575,
26,
51,
16932,
17,
83,
101,
20,
44,
777,
20,
15310,
76,
59,
57,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Project in Ruby
===
I've been coding alot of web-stuff all my life, rails lately. And i can always find a website to code, but i'm kind of bored with it. Been taking alot of courses of Java and C lately so i've become a bit interested in desktop application programming.
Problem: I can't for the life of me think of a thing to code for desktop. I just can't think of anything i can code that isn't already out there for download. So what do i do?
I need some project suggestions that i can set as a goal. | 0 | [
2,
669,
19,
10811,
800,
3726,
3726,
31,
22,
195,
74,
13,
15458,
21,
5639,
16,
2741,
8,
384,
5386,
65,
51,
201,
15,
2240,
18,
10434,
9,
17,
31,
92,
550,
477,
21,
2271,
20,
1797,
15,
47,
31,
22,
79,
825,
16,
12788,
29,
32,
9... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Flex Framework - How to tell if user is using cached framework?
===
I have a public facing application deployed with Flex. I want to switch to using the cached framework (.swz) but need to know if for my user base this is an effective solution or not (most users will only visit the site once and its just not worth it).
What I want to do is track whether or not a user has loaded the .swz/.swf file during that session - or if they are using a cached version they had previously downloaded from me or another site. If say 80% of users are downloading the framework .swz then i may as well embed the cutdown framework. But if 60% of users already have the framework I'd rather allow that cached version to be used.
The best solution I have now is to look at the web server log and count the .swz file downloads vs. the number of times my main application .swf file is loaded. This is clumsy and a pain and I havent even been able to go to the effort of doing it yet.
I cannot seem to find anything indicating what .swz or .swf files are loaded. I'd like to track against the current user session if i can determine this. | 0 | [
2,
14409,
6596,
13,
8,
184,
20,
494,
100,
4155,
25,
568,
16522,
43,
6596,
60,
800,
3726,
3726,
31,
57,
21,
317,
4325,
3010,
6698,
29,
14409,
9,
31,
259,
20,
5521,
20,
568,
14,
16522,
43,
6596,
13,
5,
9,
18,
499,
380,
6,
47,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Is There Any Advantage in Passing a UI wrapper to a view
===
Most of the MVC samples I have seen pass an instance of the view to the controller like this
public class View
{
Controller controller = new Controller(this);
}
Is there any advantage to passing a class which provides access to just the the properties and events the controller is interested in, like this:
public class UIWrapper
{
private TextBox textBox;
public TextBox TextBox
{
get {return textBox;}
}
public UIWrapper(ref TextBox textBox)
{
this.textBox = textBox;
}
public class View
{
UIWrapper wrapper = new UIWrapper(this);
Controller controller = new Controller(wrapper)
} | 0 | [
2,
25,
80,
186,
3314,
19,
2848,
21,
13,
5661,
28051,
20,
21,
1418,
800,
3726,
3726,
127,
16,
14,
307,
8990,
7855,
31,
57,
541,
1477,
40,
4851,
16,
14,
1418,
20,
14,
9919,
101,
48,
317,
718,
1418,
13,
1,
9919,
9919,
800,
78,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
KVM/QEMU network TAP problems with libvirt
===
I'm trying to use libvirt with virsh to manage my kvm/qemu vms. The problem I have is with getting it to work with public IPs. The server is running ubuntu 8.04.
libvirt keeps trying to run it as:
/usr/bin/kvm -M pc -m 256 -smp 3 -monitor pty -no-acpi -drive file=/opt/virtual-machines/calculon/root.qcow2,if=ide,boot=on -net nic,vlan=0,model=virtio -net tap,fd=10,vlan=0 -usb -vnc 127.0.0.1:0
Which boots, but does not have any network access (pings go nowhere). Running it without fd=10 makes it work right, with kvm creating the necessary TAP device for me and networking functioning inside the host. All the setup guides I've seen focus on setting up masquerading, while I just want a simple bridge and unfiltered access to the net (both the guests and host must use public IPs).
Running ifconfig on the host gives this, the bridge is manually setup in my /etc/network/interfaces file. :
br0 Link encap:Ethernet HWaddr 00:1e:c9:3c:59:b8
inet addr:12.34.56.78 Bcast:12.34.56.79 Mask:255.255.255.240
inet6 addr: fe80::21e:c9ff:fe3c:59b8/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:3359 errors:0 dropped:0 overruns:0 frame:0
TX packets:3025 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:180646 (176.4 KB) TX bytes:230908 (225.4 KB)
eth0 Link encap:Ethernet HWaddr 00:1e:c9:3c:59:b8
inet6 addr: fe80::21e:c9ff:fe3c:59b8/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:6088386 errors:0 dropped:0 overruns:0 frame:0
TX packets:3058 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:680236624 (648.7 MB) TX bytes:261696 (255.5 KB)
Interrupt:33
Any help would be greatly appreciated. | 0 | [
2,
13571,
79,
118,
1251,
62,
3677,
982,
5526,
1716,
29,
13,
8326,
8145,
38,
800,
3726,
3726,
31,
22,
79,
749,
20,
275,
13,
8326,
8145,
38,
29,
8701,
1635,
20,
4705,
51,
13571,
79,
118,
1251,
62,
3677,
13,
20147,
18,
9,
14,
144... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
How to do a "where in values" in LINQ-to-Entities
===
Does anybody know how to apply a "where in values" type condition using LINQ-to-Entities? I've tried the following but it doesn't work:
var values = new[] { "String1", "String2" }; // some string values
var foo = model.entitySet.Where(e => values.Contains(e.Name));
I believe this works in LINQ-to-SQL though? Any thoughts? | 0 | [
2,
184,
20,
107,
21,
13,
7,
2798,
19,
4070,
7,
19,
6294,
1251,
8,
262,
8,
2291,
3808,
800,
3726,
3726,
630,
11181,
143,
184,
20,
5645,
21,
13,
7,
2798,
19,
4070,
7,
1001,
2874,
568,
6294,
1251,
8,
262,
8,
2291,
3808,
60,
31,... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Trying to load files from github through a firewall is impossibly slow. Any suggestions for workarounds?
===
I'm a little hesitant to post this, as I'm not completely sure what I'm doing. Any help would be wonderful.
I'm on a computer with a firewall/filter on it. I can download files without any difficulty. When I try to clone files from Github, though, the computer just hangs. Nothing happens. It creates a git file in the folder, but the key files don't get loaded in. For context, I'm working on a Rails app, trying to load in Restful Authentication.
Have any of you dealt with this? Any suggestions for getting the clone to work? Disabling the firewall might be an option, but if I can do something without going through that process, I'd appreciate it. | 0 | [
2,
749,
20,
6305,
6488,
37,
13,
10404,
20926,
120,
21,
535,
6051,
25,
28921,
2276,
9,
186,
18389,
26,
170,
10037,
18,
60,
800,
3726,
3726,
31,
22,
79,
21,
265,
22821,
20,
678,
48,
15,
28,
31,
22,
79,
52,
1524,
562,
98,
31,
2... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Can Adobe AIR display SVG?
===
I see that Adobe AIR uses WebKit as its render and I see that WebKit (at least the most current build) has some SVG support. Does this mean (and has anyone specifically tried) that an Adobe AIR application could render SVG on an HTML page? | 0 | [
2,
92,
20299,
282,
3042,
13,
18,
22955,
60,
800,
3726,
3726,
31,
196,
30,
20299,
282,
2027,
2741,
13703,
28,
82,
16535,
17,
31,
196,
30,
2741,
13703,
13,
5,
721,
639,
14,
127,
866,
1895,
6,
63,
109,
13,
18,
22955,
555,
9,
630,... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Tell if a Javascript function is defined - is this a good way?
===
This is a follow up question to [this](http://stackoverflow.com/questions/85815/how-to-tell-if-a-javascript-function-is-defined) question.
I like (and understand) the solution there. However, in the code I am working in, another way to solve the same problem is used:
function exist(sFN) {
if(self[sFN]) return true;
return false;
}
It seems to work fine, although I don't understand how. Does it work? How? What are minuses of this approach? Should I switch to solution from the other question? | 0 | [
2,
494,
100,
21,
8247,
8741,
1990,
25,
2811,
13,
8,
25,
48,
21,
254,
161,
60,
800,
3726,
3726,
48,
25,
21,
1740,
71,
1301,
20,
636,
1565,
500,
5,
21127,
6903,
25325,
2549,
9990,
9,
960,
118,
24652,
18,
118,
3058,
3980,
10551,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
getting a DrawingContext for a wpf WriteableBitmap
===
Is there a way to get a `DrawingContext` (or something similar) for a `WriteableBitmap`? I.e. something to allow you to call simple `DrawLine`/`DrawRectangle`/etc kinds of methods, rather than manipulate the raw pixels directly. | 0 | [
2,
1017,
21,
3533,
1126,
11969,
26,
21,
619,
7721,
2757,
579,
3326,
15022,
800,
3726,
3726,
25,
80,
21,
161,
20,
164,
21,
13,
1,
3030,
3546,
1126,
11969,
1,
13,
5,
248,
301,
835,
6,
26,
21,
13,
1,
23716,
579,
3326,
15022,
1,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
When is a browser considered "Dead"?
===
Since it's been a while since my last web job, I decided to do this latest site up from scratch. Now I have to decide *again* what to support in terms of browsers. Certainly I have a list of what I'd like to support, but the decisions that went into that list seem to be a little arbitrary to me. Where can I go to get a reliable picture of browser usage and what seems to be a good point at which to cut off an old version of a browser from support?
Keep in mind that I'm not looking for a list of current browsers to support, I'm looking for logical ways to make that list. | 0 | [
2,
76,
25,
21,
16495,
724,
13,
7,
14260,
7,
60,
800,
3726,
3726,
179,
32,
22,
18,
74,
21,
133,
179,
51,
236,
2741,
1205,
15,
31,
868,
20,
107,
48,
5736,
689,
71,
37,
12395,
9,
130,
31,
57,
20,
4073,
1637,
18762,
2483,
98,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
throwing exceptions in asp.net c#
===
IS there a difference between just saying throw; and throw ex; assuming ex is the exception you're catching? | 0 | [
2,
6033,
13392,
19,
28,
306,
9,
2328,
272,
5910,
800,
3726,
3726,
25,
80,
21,
2841,
128,
114,
1148,
3814,
73,
17,
3814,
1396,
73,
11704,
1396,
25,
14,
5391,
42,
22,
99,
9436,
60,
3,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... |
How to replace $*=1 with an alternative now $* is no longer supported
===
I'm a complete perl novice, am running a perl script using perl 5.10 and getting this warning:
$* is no longer supported at migrate.pl line 380.
Can anyone describe what $* did and what the recommended replacement of it is now?
Alternatively if you could point me to documentation that describes this that would be great.
The script I'm running is to migrate a source code database from vss to svn and can be found here:
[http://www.x2systems.com/files/migrate.pl.txt][1]
The two snippets of code that use it are:
$* = 1;
$/ = ':';
$cmd = $SSCMD . " Dir -I- \"$proj\"";
$_ = `$cmd`;
# what this next expression does is to merge wrapped lines like:
# $/DeviceAuthority/src/com/eclyptic/networkdevicedomain/deviceinterrogator/excep
# tion:
# into:
# $/DeviceAuthority/src/com/eclyptic/networkdevicedomain/deviceinterrogator/exception:
s/\n((\w*\-*\.*\w*\/*)+\:)/$1/g;
$* = 0;
and then some ways later on:
$cmd = $SSCMD . " get -GTM -W -I-Y -GL\"$localdir\" -V$version \"$file\" 2>&1";
$out = `$cmd`;
# get rid of stupid VSS warning messages
$* = 1;
$out =~ s/\n?Project.*rebuilt\.//g;
$out =~ s/\n?File.*rebuilt\.//g;
$out =~ s/\n.*was moved out of this project.*rebuilt\.//g;
$out =~ s/\nContinue anyway.*Y//g;
$* = 0;
many thanks,
- Rory
(sorry, I previously posted this to perl.perl5.porters but think that was probably the wrong group)
[1]: http://www.x2systems.com/files/migrate.pl.txt | 0 | [
2,
184,
20,
3934,
5579,
2483,
3726,
165,
29,
40,
2676,
130,
5579,
2483,
25,
90,
1045,
1827,
800,
3726,
3726,
31,
22,
79,
21,
1279,
416,
255,
21999,
15,
589,
946,
21,
416,
255,
3884,
568,
416,
255,
331,
9,
1036,
17,
1017,
48,
3... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Best way to convert a decimal value to a currency string for display in HTML.
===
I wanting to show prices for my products in my online store.
I'm currently doing:
<span class="ourprice">
<%=GetPrice().ToString("C")%>
</span>
Where GetPrice() returns a decimal. This is rendering fine in most browsers, e.g. £12.00 (The server is in the UK), but some browsers (Mozilla) show this as $12.00.
Is the below an improvement, or is there a better way?
<span class="ourprice">
<%=GetPrice().ToString("C").Replace("£","£")%>
</span>
| 0 | [
2,
246,
161,
20,
8406,
21,
26380,
1923,
20,
21,
10507,
3724,
26,
3042,
19,
13,
15895,
9,
800,
3726,
3726,
31,
4638,
20,
298,
7334,
26,
51,
1985,
19,
51,
2087,
1718,
9,
31,
22,
79,
871,
845,
45,
13,
1,
18,
3206,
718,
3726,
7,... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Business Objects, Validation And Exceptions
===
I’ve been reading a few questions and answers regarding exceptions and their use. Seems to be a strong opinion that exceptions should be raised only for exception, unhandled cases. So that lead me to wondering how validation works with business objects.
Lets say I have a business object with getters/setters for the properties on the object. Let’s say I need to validate that the value is between 10 and 20. This is a business rule so it belongs in my business object. So that seems to imply to me that the validation code goes in my setter. Now I have my UI databound to the properties of the data object. The user enters 5, so the rule needs to fail and the user is not allowed to move out of the textbox. . The UI is databound to the property so the setter is going to be called, rule checked and failed. If I raised an exception from my business object to say the rule failed, the UI would pick that up. But that seems to go against the preferred usage for exceptions. Given that it’s a setter, you aren’t really going to have a ‘result’ for the setter. If I set another flag on the object then that would imply the UI has to check that flag after each UI interaction.
So how should the validation work?
| 0 | [
2,
508,
3916,
15,
27999,
17,
13392,
800,
3726,
3726,
31,
1,
195,
74,
1876,
21,
310,
2346,
17,
6709,
3467,
13392,
17,
66,
275,
9,
2206,
20,
44,
21,
966,
4052,
30,
13392,
378,
44,
1127,
104,
26,
5391,
15,
367,
3203,
1294,
1871,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
In Python, how do you take tokenized input such as with the C++?
===
In C++, I can have take input like this:
cin >> a >> b >> c;
And a can be int, b can be float, and c can be whatever... How do I do the same in python?
input() and raw_input(), the way I'm using them, don't seem to be giving me the desired results. | 0 | [
2,
19,
20059,
15,
184,
107,
42,
247,
20,
2853,
1333,
6367,
145,
28,
29,
14,
272,
20512,
60,
800,
3726,
3726,
19,
272,
20512,
15,
31,
92,
57,
247,
6367,
101,
48,
45,
13,
4484,
13,
1,
21,
13,
1,
334,
13,
1,
272,
73,
17,
21,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
How do I create a node from a cron job in drupal?
===
In a custom module for drupal 4.7 I hacked together a node object and passed it to node_save($node) to create nodes. This hack appears to no longer work in drupal 6. While I'm sure this hack could be fixed I'm curious if there is a standard solution to create nodes without a form. In this case the data is pulled in from a custom feed on another website. | 0 | [
2,
184,
107,
31,
1600,
21,
15421,
37,
21,
13,
19587,
1205,
19,
15708,
6720,
60,
800,
3726,
3726,
19,
21,
5816,
12613,
26,
15708,
6720,
268,
9,
465,
31,
11835,
69,
429,
21,
15421,
3095,
17,
1100,
32,
20,
15421,
1,
19863,
5,
4403,... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Can regex capture and substitution be used with an Apache DirectoryMatch directive?
===
Does anyone know if it's possible to use regex capture within Apache's DirectoryMatch directive? I'd like to do something like the following:
<DirectoryMatch ^/home/www/(.*)>
AuthType Basic
AuthName $1
AuthUserFile /etc/apache2/svn.passwd
Require group $1 admin
</DirectoryMatch>
but so far I've had no success. Thanks for any help! | 0 | [
2,
92,
7953,
1706,
3683,
17,
19910,
44,
147,
29,
40,
17140,
16755,
12280,
15626,
60,
800,
3726,
3726,
630,
1276,
143,
100,
32,
22,
18,
938,
20,
275,
7953,
1706,
3683,
363,
17140,
22,
18,
16755,
12280,
15626,
60,
31,
22,
43,
101,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Any problems running SharpDevelop 3.0 and Visual Studio 2008 side by side?
===
I have been asked to lend a hand on a hobby project that a couple friends are working on, they are using SharpDevelop 3.0 (Beta 2 I think, but it might be Beta 1) is there any hassle for me to install and use this IDE given that I have Visual Studio 2008 installed? | 0 | [
2,
186,
1716,
946,
2641,
26051,
203,
9,
387,
17,
3458,
1120,
570,
270,
34,
270,
60,
800,
3726,
3726,
31,
57,
74,
411,
20,
13866,
21,
224,
27,
21,
18229,
669,
30,
21,
1335,
954,
50,
638,
27,
15,
59,
50,
568,
2641,
26051,
203,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Should I use an exception specifier in C++?
===
In C++, you can specify that a function may or may not throw an exception by using an exception specifier. For example:
void foo() throw(); // guaranteed not to throw an exception
void bar() throw(int); // may throw an exception of type int
void baz() throw(...); // may throw an exception of some unspecified type
I'm doubtful about actually using them because of the following:
1. The compiler doesn't really enforce exception specifiers in any rigorous way, so the benefits are not great. Ideally, you would like to get a compile error.
2. If a function violates an exception specifier, I think the standard behaviour is to terminate the program.
3. In VS.Net, it treats throw(X) as throw(...), so adherence to the standard is not strong.
Do you think exception specifiers should be used? Please answer with "yes" or "no" and provide some reasons to justify your answer. | 0 | [
2,
378,
31,
275,
40,
5391,
12737,
16292,
19,
272,
20512,
60,
800,
3726,
3726,
19,
272,
20512,
15,
42,
92,
19077,
30,
21,
1990,
123,
54,
123,
52,
3814,
40,
5391,
34,
568,
40,
5391,
12737,
16292,
9,
26,
823,
45,
11364,
4310,
111,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Structure of a PDF file?
===
For a small project I have to parse pdf files and take a specific part of them (a simple chain of characters). I'd like to use python to do this and I've found several libraries that are capable of doing what I want in some ways.
But now after a few researches, I'm wondering what is the real structure of a pdf file, does anyone know if there is a spec or some explanations anywhere online? I've found a link on adobe but it seems that it's a dead link :( | 0 | [
2,
1411,
16,
21,
13,
11124,
3893,
60,
800,
3726,
3726,
26,
21,
284,
669,
31,
57,
20,
2017,
870,
13,
11124,
6488,
17,
247,
21,
1903,
141,
16,
105,
13,
5,
58,
1935,
2858,
16,
1766,
6,
9,
31,
22,
43,
101,
20,
275,
20059,
20,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
What algorithm can you use to find duplicate phrases in a string?
===
Given an arbitrary string, what is an efficient method of finding duplicate phrases? We can say that phrases must be longer than a certain length to be included.
Ideally, you would end up with the number of occurrences for each phrase. | 0 | [
2,
98,
9083,
92,
42,
275,
20,
477,
19429,
18492,
19,
21,
3724,
60,
800,
3726,
3726,
504,
40,
17237,
3724,
15,
98,
25,
40,
8243,
2109,
16,
3007,
19429,
18492,
60,
95,
92,
395,
30,
18492,
491,
44,
1045,
119,
21,
1200,
1476,
20,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
How do you include a JavaScript file from within a SharePoint WebPart?
===
We have a medium sized .js file that we include in our web framework that I am porting over to SharePoint. However, I'm not sure how to go about this or what the best practice is. This is for a framework solution that will be used by other client projects, so it's best for it to be self contained and deploy-able, rather than requiring manually deploying files to the webserver.
My current thinking to put the JavaScript into an embedded resource and then use the script manager to write out the file. Does this seem reasonable? Or does anyone have any other recommendations? | 0 | [
2,
184,
107,
42,
468,
21,
8247,
8741,
3893,
37,
363,
21,
1891,
3132,
2741,
3091,
60,
800,
3726,
3726,
95,
57,
21,
4152,
13,
6560,
13,
9,
728,
18,
3893,
30,
95,
468,
19,
318,
2741,
6596,
30,
31,
589,
1295,
68,
84,
20,
1891,
3... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
What do you think about the loss of MIT's 6.001 (SICP) course?
===
MIT abandoned its legendary [6.001 (Structure and Interpretation of Computer Programs) course][1] and replaced it with 6.00, 6.01 and 6.02 in the new curriculum. They are AFAIK about Python and robots. What is your opinion about the loss of SICP and Scheme in computer science education? Is it a necessary step in the right direction or a bad mistake?
[1]: http://ocw.mit.edu/OcwWeb/Electrical-Engineering-and-Computer-Science/6-001Spring-2005/CourseHome/ | 4 | [
2,
98,
107,
42,
277,
88,
14,
1526,
16,
6414,
22,
18,
400,
9,
18951,
13,
5,
7820,
306,
6,
674,
60,
800,
3726,
3726,
6414,
3092,
82,
9082,
636,
379,
9,
18951,
13,
5,
13971,
17,
6973,
16,
1428,
1726,
6,
674,
500,
2558,
165,
500... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Best platform for learning embedded programming?
===
I'm looking to learn about embedded programming (in C mainly, but I hope to brush up on my ASM as well) and I was wondering what the best platform would be. I have some experience in using Atmel AVR's and programming them with the stk500 and found that to be relatively easy. I especially like AVR Studio and the debugger that lets you view that state of registers.
However, If I was to take the time to learn, I would rather learn about something that is prevalent in industry. I am thinking ARM, that is unless someone has a better suggestion.
I would also be looking for some reference material, I have found the books section on the ARM website and if one is a technically better book than another I would appreciate a heads up.
The last thing I would be looking for is a prototyping/programming board like the STK500 that has some buttons and so forth.
Thanks =]
| 4 | [
2,
246,
2452,
26,
2477,
12138,
3143,
60,
800,
3726,
3726,
31,
22,
79,
699,
20,
2484,
88,
12138,
3143,
13,
5,
108,
272,
2011,
15,
47,
31,
1376,
20,
7588,
71,
27,
51,
28,
79,
28,
134,
6,
17,
31,
23,
5712,
98,
14,
246,
2452,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
How do you do performance testing in Ruby webapps?
===
I've been looking at ways people test their apps in order decide where to do caching or apply some extra engineering effort, and so far httperf and a simple sesslog have been quite helpful.
What tools and tricks did you apply on your projects? | 0 | [
2,
184,
107,
42,
107,
956,
4431,
19,
10811,
2741,
7753,
18,
60,
800,
3726,
3726,
31,
22,
195,
74,
699,
35,
2847,
148,
1289,
66,
4865,
18,
19,
389,
4073,
113,
20,
107,
1658,
7192,
54,
5645,
109,
2230,
1552,
2150,
15,
17,
86,
46... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Best way of getting notifications in SQL Server Reporting Services
===
Is it possible to get notifications using [SQL Server Reporting Services][1]? Say for example I have a report that I want by mail if has for example suddenly shows more than 10 rows or if a specific value drop below 100 000. Do I need to tie Notification Services into it and how do I do that?
Please provide as much technical details as possible as I've never used [Notification Services][2] before.
Someone also told me that Notifications Services is replaced by new functionality in Reporting Services in Sql Server 2008 - is this the case?
[1]: http://www.microsoft.com/sql/technologies/reporting/default.mspx
[2]: http://www.microsoft.com/sql/technologies/notification/default.mspx | 0 | [
2,
246,
161,
16,
1017,
52,
4634,
18,
19,
4444,
255,
8128,
6670,
687,
800,
3726,
3726,
25,
32,
938,
20,
164,
52,
4634,
18,
568,
636,
18,
22402,
8128,
6670,
687,
500,
2558,
165,
500,
60,
395,
26,
823,
31,
57,
21,
1330,
30,
31,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
What's the best way to serialize a HashTable for SOAP/XML?
===
What's the best way to serialize a HashTable (or a data best navigated through a string indexer) with SOAP/XML?
Let's say I have an Foo that has an property Bar[] Bars. A Bar object has a key and a value. By default, this serializes to the following XML:
`<Foo>
<Bars>
<Bar key="key0" value="value0"/>
...
</Bars>
</Foo>`
For JSON, this serializes to
`{"Foo":["Bars":[{"Key":"key0","Value":"key1} ... ]}]}`
What I'd really like to have is this serialize to better reflect the underlying relationship. E.g.,
`<Foo>
<Bars>
<Key0 value="value0"/>
<Key1 value="value1"/>
...
</Bars>
</Foo>`
I realize there are some challenges with serializing to SOAP in this way, but what's the best approach to providing a schema that better reflects this?
I've tried creating a BarsCollection object and defining custom serialization on that, but it doesn't seem to actually invoke the serialization on that object. E.g.,
` void ISerializable.GetObjectData(SerializationInfo info, StreamingContext context)
{
foreach (Bar bar in Bars)
info.AddValue(bar.Key. bar);
}
`
Any suggestions? What's the best practice here?
| 0 | [
2,
98,
22,
18,
14,
246,
161,
20,
5956,
2952,
21,
19170,
5924,
26,
6447,
118,
396,
8184,
60,
800,
3726,
3726,
98,
22,
18,
14,
246,
161,
20,
5956,
2952,
21,
19170,
5924,
13,
5,
248,
21,
1054,
246,
20782,
43,
120,
21,
3724,
4348,... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
what's the best app to draw wireframes on the mac? (And why)
===
need something less print-oriented than omnigraffle. think pixels!
also, need good building blocks (aka stencils). | 0 | [
2,
98,
22,
18,
14,
246,
4865,
20,
2003,
6511,
8361,
18,
27,
14,
1572,
60,
13,
5,
290,
483,
6,
800,
3726,
3726,
376,
301,
787,
4793,
8,
6800,
119,
14677,
12461,
10812,
9,
277,
18146,
18,
187,
67,
15,
376,
254,
353,
5198,
13,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... |
How Does Relational Theory Apply in Ways I can Care About while Learning it?
===
So I'm taking the Discrete Math course from MIT's OpenCourseWare and I'm wondering... I see the connection between relations and graphs but not enough to "own" it. I've implemented a simple state machine in SQL as well so I grok graphs pretty well, just not the more rigorous study of how relations and sets compeltely apply. Should I just be following the Yegge train of thought where I just glance over the stuff that I'm not grokking readily and come back when I've learned more? I'd like to be able to better analyze the graph structures I create on a day to day basis (sounds fun) and I want to make sure I'm not passing up valuable information right now.
Any good resources where I could learn more about this? I'm using the 5th edition of Discrete Mathematics and Its Applications by Rosen in case it matters.
Thanks! | 0 | [
2,
184,
630,
5827,
192,
1639,
5645,
19,
2847,
31,
92,
781,
88,
133,
2477,
32,
60,
800,
3726,
3726,
86,
31,
22,
79,
741,
14,
18563,
5057,
674,
37,
6414,
22,
18,
368,
17556,
5011,
17,
31,
22,
79,
5712,
9,
9,
9,
31,
196,
14,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Reporting Services Deployment
===
I need to create a repeatable process for deploying SQL Server Reporting Services reports. I am not in favor of using Visual Studio and or Business Development Studio to do this. The rs.exe method of scripting deployments also seems rather clunky. Does anyone have a very elegant way that they have been able to deploy reports. | 0 | [
2,
6670,
687,
10475,
800,
3726,
3726,
31,
376,
20,
1600,
21,
6830,
579,
953,
26,
17617,
68,
4444,
255,
8128,
6670,
687,
2813,
9,
31,
589,
52,
19,
3654,
16,
568,
3458,
1120,
17,
54,
508,
522,
1120,
20,
107,
48,
9,
14,
13,
1224,... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
How to concatenate icons into a single image with ImageMagick?
===
I want to use [CSS sprites][1] on a web site instead of separate image files, for a large collection of small icons that are all the same size. How can I concatenate (tile) them into one big image using [ImageMagick][2]?
[1]: http://alistapart.com/articles/sprites
[2]: http://www.imagemagick.org/ | 0 | [
2,
184,
20,
1065,
9530,
8820,
9801,
18,
77,
21,
345,
1961,
29,
1961,
21200,
197,
60,
800,
3726,
3726,
31,
259,
20,
275,
636,
6824,
18,
27902,
18,
500,
2558,
165,
500,
27,
21,
2741,
689,
700,
16,
1725,
1961,
6488,
15,
26,
21,
3... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Loading DLLs into a separate AppDomain.
===
I want to load one or more DLLs dynamically so that they run with a different security or basepath than my main application. How do I load these DLLs into a separate AppDomain and instantiate objects from them? | 0 | [
2,
12797,
13,
43,
211,
18,
77,
21,
1725,
4865,
537,
6232,
9,
800,
3726,
3726,
31,
259,
20,
6305,
53,
54,
91,
13,
43,
211,
18,
7782,
1326,
86,
30,
59,
485,
29,
21,
421,
1221,
54,
1000,
8353,
119,
51,
407,
3010,
9,
184,
107,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
CPAN/gem-like repository for Objective-C and Cocoa?
===
Is there any centralized repository of useful Objective-C / Cocoa libraries as there is for Perl, Ruby, Python, etc.?
In building my first iPhone app, I'm finding myself implementing some very basic functions that would be just a quick "gem install" away in Ruby.
| 0 | [
2,
272,
3206,
118,
20231,
8,
1403,
24869,
26,
7038,
8,
150,
17,
24507,
60,
800,
3726,
3726,
25,
80,
186,
25176,
24869,
16,
4811,
7038,
8,
150,
13,
118,
24507,
8649,
28,
80,
25,
26,
416,
255,
15,
10811,
15,
20059,
15,
2722,
9,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Best way to get into an open source project
===
I really want to up my programming experience and I've heard that one of the best ways to do this is to get into an open source project and try and contribute. But what is the best way of doing that? How do you pick a project? Are there certain unspoken rules or obligations one should know before joining a project? What is the best way to contribute?
I'm just looking for what one should know before joining an open source project. | 0 | [
2,
246,
161,
20,
164,
77,
40,
368,
1267,
669,
800,
3726,
3726,
31,
510,
259,
20,
71,
51,
3143,
1496,
17,
31,
22,
195,
752,
30,
53,
16,
14,
246,
2847,
20,
107,
48,
25,
20,
164,
77,
40,
368,
1267,
669,
17,
1131,
17,
8606,
9,... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Using jmockit expectations with matchers and primitive types
===
I'm using jmockit for unit testing (with TestNG), and I'm having trouble using the Expectations class to mock out a method that takes a primitive type (boolean) as a parameter, using a matcher. Here's some sample code that illustrates the problem.
/******************************************************/
import static org.hamcrest.Matchers.is;
import mockit.Expectations;
import org.testng.annotations.Test;
public class PrimitiveMatcherTest {
private MyClass obj;
@Test
public void testPrimitiveMatcher() {
new Expectations(true) {
MyClass c;
{
obj = c;
invokeReturning(c.getFoo(with(is(false))), "bas");
}
};
assert "bas".equals(obj.getFoo(false));
Expectations.assertSatisfied();
}
public static class MyClass {
public String getFoo(boolean arg) {
if (arg) {
return "foo";
} else {
return "bar";
}
}
}
}
/******************************************************/
The line containing the call to invokeReturning(...) throws a NullPointerException.
If I change this call to not use a matcher, as in:
invokeReturning(c.getFoo(false), "bas");
it works just fine. This is no good for me, because in my real code I'm actually mocking a multi-parameter method and I need to use a matcher on another argument. In this case, the Expectations class requires that **all** arguments use a matcher.
I'm pretty sure this is a bug, or perhaps it's not possible to use Matchers with primitive types (that would make me sad). Has anyone encountered this issue, and know how to get around it? | 0 | [
2,
568,
487,
79,
5668,
242,
11870,
29,
730,
445,
17,
11473,
2551,
800,
3726,
3726,
31,
22,
79,
568,
487,
79,
5668,
242,
26,
1237,
4431,
13,
5,
1410,
1289,
2723,
6,
15,
17,
31,
22,
79,
452,
2572,
568,
14,
11870,
718,
20,
10506,... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
What's a good FREE c++ API for reading/writing to FTP sites
===
What's a good FREE c++ API for writing a Windows XP application that can read/write to FTP sites? Free is key!
Thanks in advance! | 0 | [
2,
98,
22,
18,
21,
254,
551,
272,
20512,
21,
2159,
26,
1876,
118,
12646,
20,
1187,
306,
3259,
800,
3726,
3726,
98,
22,
18,
21,
254,
551,
272,
20512,
21,
2159,
26,
1174,
21,
1936,
23045,
3010,
30,
92,
1302,
118,
23716,
20,
1187,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
0,
0,
0,
0,
0,
0,
0... |
What Is ASP.Net MVC?
===
When I first heard about StackOverflow, and heard that it was being built in ASP.Net MVC, I was a little confused. I thought ASP.Net was always an example of an MVC architecture. You have the .aspx page that provides the view, the .aspx.vb page that provides the controller, and you can create another class to be the model. The process for using MVC in ASP.Net is described in this [Microsoft article][1].
So my question is. What Does ASP.Net MVC provide that you wouldn't be able to do with regular ASP.Net (even as far back as ASP.Net 1.1)? It is just fancy URLs? Is it just for bragging rights for MS to be able to compare themselves with new technologies like Ruby On Rails, and say, "We can do that too"? Is there something more that ASP.Net MVC actually provides, rather than a couple extra templates in the File->New menu?
I'm probably sounding really skeptical and negative right now, so I'll just stop. But I really want to know what ASP.Net MVC actually provides. Also, if anybody can tell me why it's Model-View-Controller and not in order of the layers of View-Controller-Model or Model-Control-View depending on whether you are going top to bottom, or vice versa, I'd really appreciate that too.
[1]: http://msdn.microsoft.com/en-us/library/ms998540.aspx | 0 | [
2,
98,
25,
28,
306,
9,
2328,
307,
8990,
60,
800,
3726,
3726,
76,
31,
64,
752,
88,
7566,
2549,
9990,
15,
17,
752,
30,
32,
23,
142,
392,
19,
28,
306,
9,
2328,
307,
8990,
15,
31,
23,
21,
265,
4230,
9,
31,
289,
28,
306,
9,
2... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Disable (Politely) a website when the sql server is offline
===
I work at a college and have been developing an ASP.NET site with many, many reports about students, attendance stats... The basis for the data is a ms sql server db which is the back end to our student management system. This has a regular maintenance period on thursday mornings for an unknow length of time (dependent on what has to be done).
Most of the staff are aware of this but the less regular users seem to be forever ringing me up. What is the easiest way to disable the site during maintenance obviously I can just try a db query to test if it is up but am unsure of the best way to for instance redirect all users to a "The website is down for maintenance" message, bearing in mind they could have started a session prior to the website going down.
Hopefully something can be implemented globally rather than per page. | 0 | [
2,
1460,
579,
13,
5,
8147,
591,
102,
6,
21,
2271,
76,
14,
4444,
255,
8128,
25,
168,
1143,
800,
3726,
3726,
31,
170,
35,
21,
314,
17,
57,
74,
3561,
40,
28,
306,
9,
2328,
689,
29,
151,
15,
151,
2813,
88,
573,
15,
3648,
12819,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Do you think ASP.NET MVC will compete with ASP.NET Webforms?
===
Do you think ASP.NET MVC will ever have a significant share of the Microsoft web development market? Or will it be more like 10-15% of the market? | 0 | [
2,
107,
42,
277,
28,
306,
9,
2328,
307,
8990,
129,
3975,
29,
28,
306,
9,
2328,
2741,
4190,
18,
60,
800,
3726,
3726,
107,
42,
277,
28,
306,
9,
2328,
307,
8990,
129,
462,
57,
21,
1505,
1891,
16,
14,
7099,
2741,
522,
1135,
60,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
0,
0,
0,
0... |
What do you think of using properties as object initializers in C#;
===
I was wondering what people thought of using properties as object initializers in C#. For some reason it seems to break the fundamentals of what constructors are used for.
An example...
public class Person
{
string firstName;
string lastName;
public string FirstName
{
get { return firstName; }
set { firstName = value; }
}
public string LastName
{
get { return lastName; }
set { lastName= value; }
}
}
Then doing object intialization with.....
Person p = new Person( FirstName = "Joe", LastName = "Smith" );
Person p = new Person( FirstName = "Joe" );
| 0 | [
2,
98,
107,
42,
277,
16,
568,
3704,
28,
3095,
2104,
11907,
18,
19,
272,
5910,
73,
800,
3726,
3726,
31,
23,
5712,
98,
148,
289,
16,
568,
3704,
28,
3095,
2104,
11907,
18,
19,
272,
5910,
9,
26,
109,
1215,
32,
2206,
20,
1442,
14,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
VS2008 Crashing on Project Load
===
I am unable to load any existing projects after starting VS2008. When I try to open an existing project VS2008 will crash. It looks like it is crashing when trying to load a floating window in VS but I cant tell which one.
When I launch the debugger on the crashed instance I get the following message which is not very useful.
Unhandled exception at 0x00740078 in devenv.exe: 0xC0000005: Access violation writing location 0x7e429ed9.
I have previously had SP1 installed but have now removed this. I also have used Resharper 4.0 but have uninstalled this as well and am still getting the problem.
Does anyone have tips on how to solve VS2008 crashing problems? I really dont want to have to do a reinstall of the product.
As a work around I have found that if I create a new class library project it will fail because of a write to protected memory error. If i try and create another new class library project it will work and then I can load an existing project. | 0 | [
2,
4611,
2753,
14604,
27,
669,
6305,
800,
3726,
3726,
31,
589,
2343,
20,
6305,
186,
3149,
2314,
75,
1422,
4611,
2753,
9,
76,
31,
1131,
20,
368,
40,
3149,
669,
4611,
2753,
129,
4597,
9,
32,
1879,
101,
32,
25,
14604,
76,
749,
20,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
How do I restore from a drop database command using a mysql binary log?
===
How can I restore a mysql database that was dropped using a "drop database" command? I have access to binary logs which should make this type of rollback possible. | 0 | [
2,
184,
107,
31,
8454,
37,
21,
2804,
6018,
1202,
568,
21,
51,
18,
22402,
14171,
6738,
60,
800,
3726,
3726,
184,
92,
31,
8454,
21,
51,
18,
22402,
6018,
30,
23,
1539,
568,
21,
13,
7,
12361,
6018,
7,
1202,
60,
31,
57,
1381,
20,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
0,
0,
0,
0,
0... |
What are the advantages and disadvantages of the Session Façade Core J2EE Pattern?
===
What are the advantages and disadvantages of the Session Façade Core J2EE Pattern?
What are the assumptions behind it?
Are these assumptions valid in a particular environment? | 0 | [
2,
98,
50,
14,
14683,
17,
13387,
18,
16,
14,
3723,
7224,
2884,
487,
135,
2851,
3732,
60,
800,
3726,
3726,
98,
50,
14,
14683,
17,
13387,
18,
16,
14,
3723,
7224,
2884,
487,
135,
2851,
3732,
60,
98,
50,
14,
20751,
439,
32,
60,
50... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
0,
0,
0,
0,
0,
0,
0,
0,
0... |
How to convert Strings to and from UTF8 byte arrays in Java
===
In Java, I have a String and I want to encode it as a byte array (in UTF8, or some other encoding). Alternately, I have a byte array (in some known encoding) and I want to convert it into a Java String. How do I do these conversions? | 0 | [
2,
184,
20,
8406,
7887,
20,
17,
37,
287,
11720,
457,
34,
591,
7718,
18,
19,
8247,
800,
3726,
3726,
19,
8247,
15,
31,
57,
21,
3724,
17,
31,
259,
20,
20523,
32,
28,
21,
34,
591,
7718,
13,
5,
108,
287,
11720,
457,
15,
54,
109,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Has anyone got a polynomial time algorithm for find a Hamiltonian walk in a graph
===
My algorithm is N factorial and is really slow. | 4 | [
2,
63,
1276,
330,
21,
15039,
85,
9083,
26,
477,
21,
3715,
806,
1466,
19,
21,
7210,
800,
3726,
3726,
51,
9083,
25,
13,
103,
3932,
2815,
17,
25,
510,
2276,
9,
3,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... |
Does the Java VM move objects in memory, and if so - how?
===
Does the Java virtual machine ever move objects in memory, and if so, how does it handle updating references to the moved object?
I ask because I'm exploring an idea of storing objects in a distributed fashion (ie. across multiple servers), but I need the ability to move objects between servers for efficiency reasons. Objects need to be able to contain pointers to each-other, even to objects on remote servers. I'm trying to think of the best way to update references to moved objects.
My two ideas so far are:
1. Maintain a reference indirection somewhere that doesn't move for the lifetime of the object, which we update if the object moves. But - how are these indirections managed?
2. Keep a list of reverse-references with each object, so we know what has to be updated if the object is moved. Of course, this creates a performance overhead.
I'd be interested in feedback on these approaches, and any suggestions for alternative approaches. | 0 | [
2,
630,
14,
8247,
13,
20147,
780,
3916,
19,
1912,
15,
17,
100,
86,
13,
8,
184,
60,
800,
3726,
3726,
630,
14,
8247,
6599,
1940,
462,
780,
3916,
19,
1912,
15,
17,
100,
86,
15,
184,
630,
32,
3053,
71,
43,
1880,
7231,
20,
14,
38... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Maven Jdepend report contains no data
===
I'm running the jdepend maven plugin on my project and whether I run "mvn site:site" or "mvn jdepend:generate" the report that gets generated says "There are no package used." There are no errors in the maven output. Other plugins (cobertura, findbugs, etc.) run fine. My pom is configured like this:
<reporting>
<plugins>
<plugin>
<groupId>org.codehaus.mojo</groupId>
<artifactId>jdepend-maven-plugin</artifactId>
</plugin>
Any ideas? | 0 | [
2,
1216,
3124,
487,
19038,
1330,
1588,
90,
1054,
800,
3726,
3726,
31,
22,
79,
946,
14,
487,
19038,
1216,
3124,
10922,
108,
27,
51,
669,
17,
1472,
31,
485,
13,
7,
79,
16578,
689,
45,
9097,
7,
54,
13,
7,
79,
16578,
487,
19038,
4... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
YUI Autocomplete renders under other page elements in IE7
===
I'm working now on a page that has a column of boxes styled with sexy shadows and corners and whatnot using the example [here][1]. I have to admit, I don't fully understand how that CSS works, but it *looks* great.
Inside the topmost box is a text-type input used for searching. That search box is wired up to a [YUI autocomplete][2] widget.
Everything works fine in FF3 on Mac, FF2 on Windows, Safari on Mac. In IE7 on WinXP, the autocomplete suggestions render underneath the round-cornered boxes, making all but the first one unreadable (although you can still see enough peeking out between boxes that I'm comfortable IE7 really is getting more than one suggestion).
Where could I start looking to correct the problem?
[1]: http://www.schillmania.com/content/projects/even-more-rounded-corners/
[2]: http://developer.yahoo.com/yui/autocomplete/ | 0 | [
2,
2600,
49,
3108,
15990,
16535,
18,
131,
89,
2478,
2065,
19,
13,
660,
465,
800,
3726,
3726,
31,
22,
79,
638,
130,
27,
21,
2478,
30,
63,
21,
4698,
16,
8120,
1034,
43,
29,
7819,
5327,
17,
8894,
17,
98,
1270,
568,
14,
823,
636,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Recommended Java wizard library?
===
I'm looking for a wizard library/framework for Java, but it seems all the ones I can find are abandoned and/or lacking in enough documentation to use quickly. Here's the ones I've found, and all appear to be no longer maintained.
- [Java Wizard Framework](http://sourceforge.net/projects/jwf/)
- [SwingLabs Wizard](https://wizard.dev.java.net/)
- [Swing Wizard Framework](https://wizard-framework.dev.java.net/)
I tried using SwingLab's Wizard, but I ran into an issue that was not explained in the documentation and I received no response from the maintainer. I'm hoping someone else knows of a good replacement before I need to re-invent the wheel.
| 4 | [
2,
5773,
8247,
10276,
1248,
60,
800,
3726,
3726,
31,
22,
79,
699,
26,
21,
10276,
1248,
118,
8361,
3783,
26,
8247,
15,
47,
32,
2206,
65,
14,
53,
18,
31,
92,
477,
50,
3092,
17,
118,
248,
11690,
19,
511,
13945,
20,
275,
976,
9,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
"Error Creating Window Handle"
===
We're working on a very large .NET WinForms composite application - not CAB, but a similar home grown framework. We're running in a Citrix and RDP environment running on Windows Server 2003.
We're starting to run into random and difficult to reproduct "Error creating window handle" error that seems to be an old fashion handle leak in our application. We're making heavy use of 3rd Party controls (Janus GridEX, Infralution VirtualTree, and .NET Magic docking) and we do a lot of dynamic loading and rendering of content based on metadata in our database.
There's a lot of info on Google about this error, but not a lot of solid guidance about how to avoid issues in this area.
Does the stackoverflow community have any good guidance for me for building handle-friendly winforms apps?
Thanks in advance
-kellyb | 0 | [
2,
13,
7,
29992,
2936,
1463,
3053,
7,
800,
3726,
3726,
95,
22,
99,
638,
27,
21,
253,
370,
13,
9,
2328,
628,
4190,
18,
12639,
3010,
13,
8,
52,
4846,
15,
47,
21,
835,
213,
3651,
6596,
9,
95,
22,
99,
946,
19,
21,
5486,
17224,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Zend Framework - ErrorHandler does not seem to be working as expected.
===
This is my first experience using the Zend Framework. I am attempting to follow the [Quick Start][1] tutorial. Everything was working as expected until I reached the section on the [Error Controller and View][2]. When I navigate to a page that does not exist, instead of receiving the error page I get the Fatal Error screen dump (in all it's glory):
Fatal error: Uncaught exception 'Zend_Controller_Dispatcher_Exception' with message 'Invalid controller specified (error)' in /home/.fantasia/bcnewman/foo.com/library/Zend/Controller/Dispatcher/Standard.php:249 Stack trace: #0 /home/.fantasia/bcnewman/foo.com/library/Zend/Controller/Front.php(946): Zend_Controller_Dispatcher_Standard->dispatch(Object(Zend_Controller_Request_Http), Object(Zend_Controller_Response_Http)) #1 /home/.fantasia/bcnewman/foo.com/public/index.php(42): Zend_Controller_Front->dispatch() #2 {main} thrown in /home/.fantasia/bcnewman/foo.com/library/Zend/Controller/Dispatcher/Standard.php on line 249
I do not believe this is caused by a syntax error on my part (a copied and pasted the example file's content from the tutorial) and I believe I have the application directory structure correct:
./application
./application/controllers
./application/controllers/IndexController.php
./application/controllers/ErrorHandler.php
./application/views
./application/views/scripts
./application/views/scripts/index
./application/views/scripts/index/index.phtml
./application/views/scripts/error
./application/views/scripts/error/error.phtml
./application/bootstrap.php
./public
./public/index.php
And finally, the IndexController and index.phtml view does work. Many thanks in advance. ~b
[1]: http://framework.zend.com/docs/quickstart/introduction
[2]: http://framework.zend.com/docs/quickstart/create-an-error-controller-and-view
| 0 | [
2,
10526,
43,
6596,
13,
8,
7019,
3203,
1252,
630,
52,
2260,
20,
44,
638,
28,
1727,
9,
800,
3726,
3726,
48,
25,
51,
64,
1496,
568,
14,
10526,
43,
6596,
9,
31,
589,
6314,
20,
1740,
14,
636,
25865,
799,
500,
2558,
165,
500,
29724... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
What is a heuristic fencepost?
===
And why does gdb seem to "hit" it? | 0 | [
2,
98,
25,
21,
24,
911,
3771,
6545,
6962,
60,
800,
3726,
3726,
17,
483,
630,
489,
9007,
2260,
20,
13,
7,
10242,
7,
32,
60,
3,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... |
lining up function parameter lists with vim
===
When defining or calling functions with enough arguments to span multiple lines, I want vim to line them up. For example,
` def myfunction(arg1, arg2, arg, ...`
`_______________argsN-1, argN)`
I'm using undrescores because I don't know how to make it work with markdown either, but the idea is for argsN-1 to have its 'a' lined up with args1.
Does anyone have a way to have this happen automatically in vim? I've seen the align plugin for lining equal signs (in assignment statements) and such, but I'm not sure if it can be made to solve this problem? | 0 | [
2,
13,
8930,
71,
1990,
18906,
7227,
29,
1790,
79,
800,
3726,
3726,
76,
14684,
54,
2555,
3719,
29,
511,
10553,
20,
6289,
1886,
1560,
15,
31,
259,
1790,
79,
20,
293,
105,
71,
9,
26,
823,
15,
13,
1,
6312,
51,
22359,
5,
10663,
165... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Why should Python PEP-8 specify a maximum line length of 79 characters?
===
Why in this millenium should Python [PEP-8][1] specify a maximum line length of 79 characters?
Pretty much every code editor under the sun can handle longer lines. What to do with wrapping should be the choice of the content consumer, not the responsibility of the content creator.
Are there any (legitimately) good reasons for adhering to 79 characters in this age?
[1]: http://www.python.org/dev/peps/pep-0008/ | 4 | [
2,
483,
378,
20059,
11841,
8,
457,
19077,
21,
2979,
293,
1476,
16,
9933,
1766,
60,
800,
3726,
3726,
483,
19,
48,
2238,
219,
2187,
378,
20059,
636,
1664,
306,
8,
457,
500,
2558,
165,
500,
19077,
21,
2979,
293,
1476,
16,
9933,
1766,... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Is there any reason to not ship the pdb's with your application?
===
Since you can use reflector to reverse-engineer a .Net app, is there any reason to NOT ship the pdb files with the app? If you do ship them with it, then your stack trace will include the line number with the problem, which is useful if it crashes.
Please only enter 1 reason per comment for voting. | 0 | [
2,
25,
80,
186,
1215,
20,
52,
995,
14,
351,
9007,
22,
18,
29,
154,
3010,
60,
800,
3726,
3726,
179,
42,
92,
275,
6907,
248,
20,
7006,
8,
19153,
21,
13,
9,
2328,
4865,
15,
25,
80,
186,
1215,
20,
52,
995,
14,
351,
9007,
6488,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
What Does {0} Mean in C?
===
When {0} is used to initialize an object in C/C++, what does it mean? I can't find any references to {0} anywhere, and because of the curly braces Google searches are not helpful.
Example code:
SHELLEXECUTEINFO sexi = {0};
sexi.cbSize = sizeof(SHELLEXECUTEINFO);
sexi.hwnd = NULL;
sexi.fMask = SEE_MASK_NOCLOSEPROCESS;
sexi.lpFile = lpFile.c_str();
sexi.lpParameters = args;
sexi.nShow = nShow;
if(ShellExecuteEx(&sexi))
{
DWORD wait = WaitForSingleObject(sexi.hProcess, INFINITE);
if(wait == WAIT_OBJECT_0)
GetExitCodeProcess(sexi.hProcess, &returnCode);
}
Without it, the above code will crash on runtime....
Thanks. | 0 | [
2,
98,
630,
13,
1,
387,
1,
884,
19,
272,
60,
800,
3726,
3726,
76,
13,
1,
387,
1,
25,
147,
20,
2104,
2952,
40,
3095,
19,
272,
118,
150,
20512,
15,
98,
630,
32,
884,
60,
31,
92,
22,
38,
477,
186,
7231,
20,
13,
1,
387,
1,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Preventing bad data input
===
Is it good practice to delegate data validation entirely to the database engine constraints?
Validating data from the application doesn't prevent invalid insertion from another software (possibly written in another language by another team). Using database constraints you reduce the points where you need to worry about invalid input data.
If you put validation both in database and application, maintenance becomes boring, because you have to update code for who knows how many applications, increasing the probability of human errors.
I just don't see this being done very much, looking at code from free software projects. | 0 | [
2,
11520,
896,
1054,
6367,
800,
3726,
3726,
25,
32,
254,
1345,
20,
11300,
1054,
27999,
2894,
20,
14,
6018,
1406,
16747,
60,
7394,
1880,
1054,
37,
14,
3010,
1437,
22,
38,
2501,
16671,
24245,
37,
226,
2306,
13,
5,
21628,
642,
19,
22... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Asynchronous file IO in .Net
===
I'm building a toy database in C# to learn more about compiler, optimizer, and indexing technology.
I want to maintain maximum parallelism between (at least read) requests for bringing pages into the buffer pool, but I am confused about how best to accomplish this in .NET.
Here are some options and the problems I've come across with each:
1. Use `System.IO.FileStream` and the `BeginRead` method
But, the position in the file isn't an argument to `BeginRead`, it is a property of the `FileStream` (set via the `Seek` method), so I can only issue one request at a time and have to lock the stream for the duration. (Or do I? The documentation is unclear on what would happen if I held the lock only between the `Seek` and `BeginRead` calls but released it before calling `EndRead`. Does anyone know?) I know how to do this, I'm just not sure it is the best way.
2. There seems to be another way, centered around the `System.Threading.Overlapped` structure and P\Invoke to the `ReadFileEx` function in kernel32.dll.
Unfortunately, there is a dearth of samples, especially in managed languages. This route (if it can be made to work at all) apparently also involves the `ThreadPool.BindHandle` method and the IO completion threads in the thread pool. I get the impression that this is the sanctioned way of dealing with this scenario under windows, but I don't understand it and I can't find an entry point to the documentation that is helpful to the uninitiated.
3. Something else? | 0 | [
2,
21,
16023,
1291,
3893,
13,
1963,
19,
13,
9,
2328,
800,
3726,
3726,
31,
22,
79,
353,
21,
20,
93,
6018,
19,
272,
5910,
20,
2484,
91,
88,
21486,
15,
22864,
139,
15,
17,
4348,
68,
1099,
9,
31,
259,
20,
4027,
2979,
3821,
756,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Produce conditional compile time error in Java
===
I do not mean the compile errors because I made a syntax mistake or whatever. In C++ we can create compile time errors based on conditions as in the following example:
template<int> struct CompileTimeError;
template<> struct CompileTimeError<true> {};
#define STATIC_CHECK(expr, msg) { CompileTimeError<((expr) != 0)> ERROR_##msg; (void)ERROR_##msg; }
int main(int argc, char* argv[])
{
STATIC_CHECK(false, Compile_Time_Failure);
return 0;
}
In VS 2005 this will output:
------ Build started: Project: Test, Configuration: Debug Win32 ------
Compiling...
Test.cpp
f:\temp\test\test\test.cpp(17) : error C2079: 'ERROR_Compile_Time_Failure' uses undefined struct 'CompileTimeError<__formal>'
with
[
__formal=0
]
Build log was saved at "file://f:\temp\Test\Test\Debug\BuildLog.htm"
Test - 1 error(s), 0 warning(s)
========== Build: 0 succeeded, 1 failed, 0 up-to-date, 0 skipped ==========
Is there any way to achieve this in Java?
| 0 | [
2,
2213,
21206,
26561,
85,
7019,
19,
8247,
800,
3726,
3726,
31,
107,
52,
884,
14,
26561,
11908,
185,
31,
117,
21,
22649,
5643,
54,
2099,
9,
19,
272,
20512,
95,
92,
1600,
26561,
85,
11908,
432,
27,
2039,
28,
19,
14,
249,
823,
45,... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Copyright and Fair Use in Distributable Software
===
At what length of text and/or length of audio snippet does a piece of commercially distributable software pass the threshold of fair use and violate the included work's copyright? Does attribution absolve the developer from infringement? An example would be a quote from a novel used on a start-up screen. | 0 | [
2,
8398,
17,
1768,
275,
19,
1460,
13202,
5924,
2306,
800,
3726,
3726,
35,
98,
1476,
16,
1854,
17,
118,
248,
1476,
16,
4023,
13,
29061,
630,
21,
1855,
16,
11845,
1460,
13202,
5924,
2306,
1477,
14,
11361,
16,
1768,
275,
17,
21957,
1... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Place To get EULA and Other Legalese For Software?
===
I was curious if anyone out there has experience getting the necessary legal documents (user agreements, privacy policies, disclaimers, etc.) for a small software business. For example if you just want to have a software 'company' that sells a few piece of software that you have written, are there cheap solutions for something small scale like that? | 0 | [
2,
209,
20,
164,
2898,
531,
17,
89,
1517,
4635,
26,
2306,
60,
800,
3726,
3726,
31,
23,
7686,
100,
1276,
70,
80,
63,
1496,
1017,
14,
2378,
1517,
4374,
13,
5,
16704,
12267,
15,
9226,
4845,
15,
1460,
22661,
445,
15,
2722,
9,
6,
2... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Queuing actions (not effects) to execute after an amount of time.
===
What I'd like to know is if there is a nice way to queue jQuery functions to execute after a set amount of time. Perhaps an example of what I'd envisage it would look like would illustrate:
$('#alert')
.show()
.wait(5000) // <-- this bit
.hide()
;
I know that it's possible by using timeouts, but it seems like that's a messy way to do it, especially compared to the above example (if it were real).
So, is something like this already built-in to jQuery, and if not, what is the best way to emulate it? | 0 | [
2,
9386,
291,
68,
3078,
13,
5,
1270,
2292,
6,
20,
15644,
75,
40,
2006,
16,
85,
9,
800,
3726,
3726,
98,
31,
22,
43,
101,
20,
143,
25,
100,
80,
25,
21,
2210,
161,
20,
22521,
487,
8190,
93,
3719,
20,
15644,
75,
21,
309,
2006,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
How do you read c declarations?
===
I have heard of some methods, but none of them have stuck. Personally I try to avoid complex types in C and try to break them into component typedef.
I'm now faced with maintaining some legacy code from a so called 'three star programmer', and I'm having a hard time reading some of the ***code[][].
How do you read complex C declarations? | 0 | [
2,
184,
107,
42,
1302,
272,
7098,
18,
60,
800,
3726,
3726,
31,
57,
752,
16,
109,
3195,
15,
47,
2369,
16,
105,
57,
4549,
9,
7200,
31,
1131,
20,
2658,
1502,
2551,
19,
272,
17,
1131,
20,
1442,
105,
77,
5912,
1001,
13862,
9,
31,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Does "scrum" work in off shoring?
===
I do a lot of work as an architect with offshore teams and some of these teams use Scrum development practices in their purest form.
I'm finding problems when developers are spread so far apart with pure scrum and wondered if anyone has solutions. 80% development is remote but some team members are US side.
The biggest issue is a lack of documentation - I'm no big believer in lots of docs but I like use cases or something that resembles a spec even if small. The remote developers just don't get a clear enough understanding of what is required even with all the meetings. New developers to the projects don't stand any chance of being productive.
Personally I'm pushing some teams back to just agile. Thanks! | 2 | [
2,
630,
13,
7,
18,
25241,
7,
170,
19,
168,
5799,
2090,
60,
800,
3726,
3726,
31,
107,
21,
865,
16,
170,
28,
40,
3338,
29,
13609,
952,
17,
109,
16,
158,
952,
275,
26562,
522,
5242,
19,
66,
4267,
384,
505,
9,
31,
22,
79,
3007,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
What happens to your time slice if you get pre-empted in vxWorks?
===
If you have round robin enabled in Vxworks and your task gets preempted by a higher priority task, what happens to the remaining time slice? | 0 | [
2,
98,
5531,
20,
154,
85,
12982,
100,
42,
164,
782,
8,
14228,
69,
19,
566,
396,
7684,
60,
800,
3726,
3726,
100,
42,
57,
560,
4804,
9338,
19,
566,
396,
7684,
17,
154,
3005,
3049,
782,
14228,
69,
34,
21,
1184,
9857,
3005,
15,
98... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... |
Computer Hardware website
===
I'm looking for a website that can help me find the best new hardware options to build a new computer. I'm looking for reviews and features. | 2 | [
2,
1428,
7610,
2271,
800,
3726,
3726,
31,
22,
79,
699,
26,
21,
2271,
30,
92,
448,
55,
477,
14,
246,
78,
7610,
6368,
20,
1895,
21,
78,
1428,
9,
31,
22,
79,
699,
26,
2783,
17,
967,
9,
3,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... |
Microsoft Sharepoint
===
I hear the name Microsoft Sharepoint being thrown around a lot yet I am still not sure what its for and why I hear a lot of big companies use it. In short, the questions are, what is sharepoint and why is it popular? | 4 | [
2,
7099,
1891,
3132,
800,
3726,
3726,
31,
990,
14,
204,
7099,
1891,
3132,
142,
6027,
140,
21,
865,
768,
31,
589,
174,
52,
562,
98,
82,
26,
17,
483,
31,
990,
21,
865,
16,
580,
1532,
275,
32,
9,
19,
502,
15,
14,
2346,
50,
15,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
0,
0,
0,
0,
0,
0... |
Diagnosing Deadlocks in Win32 Program
===
What are the steps and techniques to debug an apparent hang due to a deadlock in a Win32 production process. I heard that WinDbg can be used for this purpose but could you please provide clear hints on how this can be accomplished? | 0 | [
2,
926,
21887,
18,
68,
828,
3966,
18,
19,
628,
3125,
625,
800,
3726,
3726,
98,
50,
14,
2382,
17,
4212,
20,
121,
16254,
40,
5982,
4546,
397,
20,
21,
828,
3966,
19,
21,
628,
3125,
637,
953,
9,
31,
752,
30,
1511,
19924,
92,
44,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Apache gives me 403 Access Forbidden when DocumentRoot points to two different drives
===
I am getting an 403 access forbidden when attempting to open a page under a vhost where the document root is sitting on a different drive than where apache is sitting. I installed using the apachefriends release. This is my httpd-vhosts.conf file:
<pre><code>
NameVirtualHost 127.0.0.1
<VirtualHost 127.0.0.1>
ServerName foo.localhost
DocumentRoot "C:/xampp/htdocs/foo/public"
</VirtualHost>
<VirtualHost 127.0.0.1>
ServerName bar.localhost
DocumentRoot "F:/bar/public"
</VirtualHost>
</pre></code>
When opening bar.localhost in my browser, Apache is giving me 403 Access Forbidden. I tried setting lots of different access rights, even full rights to everyone, but nothing I tried helped. | 0 | [
2,
17140,
2352,
55,
13,
25821,
1381,
11369,
76,
4492,
14032,
819,
20,
81,
421,
10396,
800,
3726,
3726,
31,
589,
1017,
40,
13,
25821,
1381,
11369,
76,
6314,
20,
368,
21,
2478,
131,
21,
566,
11694,
113,
14,
4492,
5900,
25,
1805,
27,... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Pursue carrer as "Software Engineer" or a "Lang/Framework-X Developer"?
===
I am currently a Software Engineering undergrad and recently begun my senior year. I see on the horizon a choice that must be made: choosing a career path.
For the past several months, I have been of the mindset that I should:
- Find a language and maybe a web app framework that I like
- Learn it
- Build some web apps to learn it better
- Try to learn as much as possible about the Lang-X/Framework
- Start job searches on Monster for "Lang-X/Framework Developer" and applying.
So for several months I have been trying out and playing with several different language / framework combinations to get a sense of them and decide which one I like best.
Recently though, I have begun to think that maybe I'm going about this all wrong. When I do a search for "Lang-X Developer", regardless of the language/framework I generally get between 150-300 hits.
But if I do a search for "Software Engineer," I get 2-3 thousand hits. I understand, I'm getting the hits for "Software Engineer" for all lang/framework combinations and much more, but that's the root of my concern.
My questions:
1. Am I severely limiting my career choices by focusing on solely web development? If so, is that bad?
2. Is focusing on one language/framework a good idea, in your opinion? Or should I spread my time/energy around and always be trying new languages/frameworks?
3. When hiring junior developers that are fresh-out-college CS / SE majors, what are you looking for?
If you can, please elaborate. Feel free to answer any one or all of them, as you see fit. Thanks
| 2 | [
2,
5760,
7123,
106,
28,
13,
7,
12980,
5011,
2335,
7,
54,
21,
13,
7,
9949,
118,
8361,
3783,
8,
396,
10058,
7,
60,
800,
3726,
3726,
31,
589,
871,
21,
2306,
1552,
131,
7691,
17,
1989,
4349,
51,
1101,
159,
9,
31,
196,
27,
14,
75... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Benefits of using short-circuit evaluation
===
boolean a = false, b = true;
if ( a && b ) { ... };
In most languages, `b` will not get evaluated because `a` is false so `a && b` cannot be true. My question is, wouldn't short circuiting be slower in terms of architecture? In a pipeline, do you just stall while waiting to get the result of a to determine if b should be evaluated or not? Would it be better to do nested ifs instead? Does that even help?
Also, does anyone know what short-circuit evaluation is typically called? This question arose after I found out that my programming friend had never heard of short-circuit evaluation and stated that it is not common, nor found in many languages, and is inefficient in pipeline. I am not certain about the last one, so asking you folks!
Thanks | 0 | [
2,
5800,
16,
568,
502,
8,
14022,
11193,
8670,
800,
3726,
3726,
9827,
413,
210,
21,
800,
4997,
15,
334,
800,
1151,
73,
100,
13,
5,
21,
279,
1569,
334,
13,
6,
13,
1,
13,
9,
9,
9,
13,
1,
73,
19,
127,
2556,
15,
13,
1,
220,
1... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Autocomplete Textbox on Gridview editing
===
How do you implement autocomplete on ASP.Net Gridview? Can anyone point me where to go to achieve this? I'm willing to use non-.Net ajax controls if that what it takes. | 0 | [
2,
3108,
15990,
1854,
5309,
27,
7354,
4725,
9510,
800,
3726,
3726,
184,
107,
42,
8713,
3108,
15990,
27,
28,
306,
9,
2328,
7354,
4725,
60,
92,
1276,
454,
55,
113,
20,
162,
20,
4689,
48,
60,
31,
22,
79,
4452,
20,
275,
538,
8,
9,... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
0,
0,
0,
0,
0,
0,
0,
0... |
Are there any good and/or famous examples of AGPL project?
===
Or even more specific - profitable AGPL projects
I'm considering a startup and it is interesting how can it benefit from being AGPL. Any real world example could bring some light. | 0 | [
2,
50,
80,
186,
254,
17,
118,
248,
1561,
3770,
16,
11342,
5727,
669,
60,
800,
3726,
3726,
54,
166,
91,
1903,
13,
8,
17506,
11342,
5727,
2314,
31,
22,
79,
5154,
21,
20205,
17,
32,
25,
4883,
184,
92,
32,
4234,
37,
142,
11342,
57... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
0,
0,
0,
0,
0,
0,
0... |
How to conduct a sucessful code review?
===
So I work in a shop that doesn't conduct code reviews. What are some of the best practices out there that *actually* work? | 0 | [
2,
184,
20,
3989,
21,
21792,
160,
18,
1566,
1797,
1487,
60,
800,
3726,
3726,
86,
31,
170,
19,
21,
2546,
30,
1437,
22,
38,
3989,
1797,
2783,
9,
98,
50,
109,
16,
14,
246,
5242,
70,
80,
30,
1637,
15661,
2483,
170,
60,
3,
0,
0,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... |
In Python, what is the fastest algorithm for removing duplicates from a list so that all elements are unique *while preserving order*?
===
For example:
>>> x = [1, 1, 2, 'a', 'a', 3]
>>> unique(x)
[1, 2, 'a', 3]
**Assume list elements are hashable.**
| 0 | [
2,
19,
20059,
15,
98,
25,
14,
7518,
9083,
26,
9096,
19429,
18,
37,
21,
968,
86,
30,
65,
2065,
50,
2619,
1637,
10144,
17936,
389,
2483,
60,
800,
3726,
3726,
26,
823,
45,
13,
1,
993,
800,
636,
165,
15,
137,
15,
172,
15,
13,
22... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
CVS Checkout to a directory
===
How do i check out a specific directory from CVS and omit the tree leading up to that directory?
EX.
Id like to checkout to this directory
C:/WebHost/MyWebApp/www
My CVS Project directory structure is
MyWebApp/Trunk/www
How do i omit the Trunk and MyWebApp directories?
| 0 | [
2,
13,
12732,
18,
2631,
1320,
20,
21,
16755,
800,
3726,
3726,
184,
107,
31,
2631,
70,
21,
1903,
16755,
37,
13,
12732,
18,
17,
13,
8560,
38,
14,
1541,
1005,
71,
20,
30,
16755,
60,
1396,
9,
4924,
101,
20,
2631,
1320,
20,
48,
167... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Get Flex app's position on a web page?
===
Is it possible to get the x,y coordinates of a Flex app within an HTML page? I know you can use ExternalInterface.ObjecID to get the "id attribute of the object tag in Internet Explorer, or the name attribute of the embed tag in Netscape" but I can't seem to get past that step. It seems like it should be possible to get a handle on that embed object. Any suggestions?
Thanks. | 0 | [
2,
164,
14409,
4865,
22,
18,
649,
27,
21,
2741,
2478,
60,
800,
3726,
3726,
25,
32,
938,
20,
164,
14,
993,
15,
93,
13714,
16,
21,
14409,
4865,
363,
40,
13,
15895,
2478,
60,
31,
143,
42,
92,
275,
4886,
6280,
6413,
9,
4995,
1969,... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Does LINQ-to-SQL Support Composable Queries?
===
Speaking as a non-C# savvy programmer, I'm curious as to the evaluation semantics of LINQ queries like the following:
var people = from p in Person
where p.age < 18
select p
var otherPeople = from p in people
where p.firstName equals "Daniel"
select p
Assuming that `Person` is an ADO entity which defines the `age` and `firstName` fields, what would this do from a database standpoint? Specifically, would the `people` query be run to produce an in-memory structure, which would then be queried by the `otherPeople` query? Or would the construction of `otherPeople` merely pull the data regarding the query from `people` and then produce a new database-peered query? So, if I iterated over both of these queries, how many SQL statements would be executed? | 0 | [
2,
630,
6294,
1251,
8,
262,
8,
18,
22402,
555,
6479,
759,
579,
9386,
2829,
60,
800,
3726,
3726,
3324,
28,
21,
538,
8,
150,
5910,
13220,
5903,
17968,
15,
31,
22,
79,
7686,
28,
20,
14,
8670,
17220,
18,
16,
6294,
1251,
9386,
2829,
... | [
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0,
0... | [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.