The Whiteboard

12:14 AM

@RobertHarvey I'm not sure I'd call it a culture what's happening in the lounge... More of a...mutation....

user55340

1:17 AM

Two links: map of solar system if the moon was 1 pixel wide (scroll with arrow keys) and Nine planets without intelligent life - a philosophical road comic about two unemployed robots on an improvised interplanetary voyage of self discovery. Not sure what the better order to view them in is.

Ampt

1:27 AM

that's awesome

the 1 pixel wide moon one

user15026

1:53 AM

@MichaelT Oh,, the one pixel wide moon one was FANTASTIC but now my scrolling finger is tired.

deostroll

2:18 AM

Hi. How to "use" the library? -> jquery.vostrel.cz/reel

I mean how to generate the panaroma view? What software/device will allow me to do so...?

MetaFight

9:16 AM

@RobertHarvey programmers.stackexchange.com/questions/254659/…

Ampt

12:54 PM

@deostroll jquery.vostrel.cz/reel#how

Thomas Owens

I'm glad there's a Workplace mod that frequents this chat. It will be handy.

Ampt

mwahahahahahahhahahahahaha

also: programmers.stackexchange.com/questions/254677/…

please delete

Thank you! (I'm assuming it was you, but my thanks can be generalized to whichever glorious mod deleted that god-awful question.)

ratchet freak

@Ampt it was

< showing off my 10k rep

Ampt

oooh

I haven't had my coffee yet this morning

Ampt

1:27 PM

@gnat gah, I marked that one as asking for off site resources

candidate for the workplace?

thorsten müller

Isn't workplace more about 'social' stuff? Interaction between people things.

Ampt

that's kind of what I'm getting out of it though

they tell him its not possible

he wants to convince them to let him try

I mean I don't think anyone can provide an answer either way without knowing the system from a technical point of view

thorsten müller

Yes, the question would have some 'social' aspects too. But main point would be is it really 'impossible' or as he seems to presume only failed and where too lazy to really try. So that's not only off topic but opinion based. I told him to try chat, maybe we see him here soon :)

Akash

2:11 PM

0

Q: Design patterns: SQL

I'm aware of a few design patterns, however all of them seem to apply to object oriented programming. Is there a similar set of standardized guidelines for SQL ,data warehousing ,reporting,ETL's,etc ?

Anyone knows the answer? It was closed for being a dupe :( though the other question is an incredibly generalized version of mine

Ampt

Your question is exceptionally large

SQL is not a small field, so there are probably a thousands different ways to do things

You're going to have to narrow it down quite a bit to make it a meaningful question

TL;DR: It would take a book to answer that question.

Akash

Would be fine to get the name of a few recommended books then :)

ratchet freak

don't ask that on the site (chat is fine)

Akash

Just like for Object oriented programming, the linked WikiPedia page is a good starting point..

ohk.. I thought it was in the scope of programmers.se

Ampt

@Akash Coming right up!

ratchet freak

2:20 PM

asking for books is off topic

Ampt

@Akash Definitely not.

ratchet freak

major reason for that is because they tend to get out off date

Ampt

And it's often highly subjective. There are very few fields where there is a definitive book

and even then that book is usually just an introduction and might not cover your exact needs

Akash

@Ampt very funny ಠ_ಠ .. I wasnt looking for SQL books, but guidelines around best practices on writing that code..

Right.. ok, so its out of scope, didnt realize that.. essentially I was looking for the data management\warehousing analouge of the linked wikipedia page

Ampt

Ok. Before I send you more LMGTFY lets just step back

what are you trying to accomplish

is there a particular problem you're facing

or is this just a "I want to be the very best" endeavor

Secondly, Design Patterns are not the be-all, end-all solution you think they are. Looking to take that into non-OO situations can be even worse.

Akash

2:26 PM

hmm.. essentially, I know how to write SQL, I know about the common pitfalls,etc that you can fall into while writing SQL, but at the end of the day I have < 2 years experience writing it.. there are folks with way more experience and just like Design patterns evolved for OO languages, I assumed that something similar would exist for SQL, essentially a nearly exhaustive list of common scenarios and best practices to go about solving them instead of rolling my own

(going back to your prev link, wouldnt "software architecture and design" cover my question as posted originally?)

Ampt

@Akash the on-topic vs off-topic link?

Akash

yes

Ampt

and it is not about [...] where to find a software library, tool or other resource, what book you should read next

small caveat on that one :)

There was a day, long ago, where just putting "As a programmer..." in the title would make your question on topic for this site. Fortunately that's no longer the case.

ratchet freak

sounds like you should read up on database design

how to structure tables and such

Thomas Owens

2:42 PM

Can someone explain why people are voting this question for closure? It's not off-topic - requirements elicitation is explicitly on-topic.

Is there something I'm not seeing?

ratchet freak

first glance looks like a people/social question

Ampt

It looks like the problem is more about dealing with a scatter-brain co-worker than actually getting the Reqs, but if you say it's on topic...

ratchet freak

I agree with ampt, sound like it would be good for workplace but they might have a dupe there already...

Thomas Owens

I don't see the difference between a coworker with requirements and a customer with requirements.

Ampt

I don't see the difference between this and getting a manager to give you clear tasks. It's more about the communication between him and his manager than requirements

he could be trying to ask his boss where he wants to eat for lunch

ratchet freak

2:46 PM

or designing a bridge

Ampt

cough @enderland cough

ratchet freak

the comment thread seems to support the workplace issue over the req gathering issue

Thomas Owens

The question I see is this: "How do you wring clear, correct requirements from someone who understands them but does not have the inclination to take the time/effort to explain what they actually are?"

I guess I just don't see how that's not a Programmers question.

Ampt

Ah, ok. I see "How can I communicate with my manager, and is it my fault that the communication has broken down this far?"

ratchet freak

The question I see is "How can I get my coworker to communicate his requirements with me clearly instead of going of tangents?"

Ampt

2:51 PM

@DocBrown and Adel: My apologies - this is more a people question than a programming question. I have no problem with the question being migrated. RE: communication: I have not spoken with the co-worker about this concern yet. This is probably the way to go. The person will not know it's seen as a problem if it's not brought to their attention, and expecting them to change behavior all of a sudden is not likely to happen. Plus, they might have suggestions for me. I will consider talking to the co-worker about the concern if I can manage to talk to them one-on-one, privately. — Daniel Neel 8 hours ago

user41796

@ThomasOwens It is on-topic, but I think it's a dupe of the one I suggested

user41796

Can I get some votes to delete an answer please? It's a poll disguised as an answer.

user41796

-2

A: Why doesn't libxml2 support XPath 2.0?

Sorry to be late arriving with this post (the question asked Jul 21 '13). I am looking forward to use Saxon/C 1.0 and XSLT2 with PHP... But, while we wait, let's understand better the "forces" that perhaps move Libxml2. I will check here by your votes (downvotes also, but please comment it) if...

Ampt

he made it community too so he didnt take the rep hit

technically we could just edit it...

user41796

I think "coward" is the appropriate word.

ratchet freak

3:00 PM

there is not much more value that we can edit it too

Ampt

@ratchetfreak I was thinking of editing a half assed answer and removing the rest

user41796

@Ampt Better just to get it deleted off.

Ampt

This isn't an answer. Please remove this. — Ampt 20 secs ago

lets see if being nice works

user15026

@Ampt I dunno, being chill wasn't as effective as I wanted... ;)

Ampt

@AshleyNunn You clearly weren't yelling it loud enough :P

user15026

3:04 PM

I've learned my lesson for next time :)

ratchet freak

I'm surprised it wasn't removed soon after it was posted

user41796

@ratchetfreak We're a little light on our 20k+ community members at the moment, OTW we could have taken it out already.

user41796

And our mods are getting soft since the community is so vigilant. :-D (Gonna get kicked for that one I bet.... :-) )

Ampt

It takes 20k to delete an answer? Man that's ridiculous

ratchet freak

I still got 8k to go for that benchmark

user41796

3:06 PM

@Ampt I dropped a delete vote on it, so it should make it's way to the low quality review queue.

ratchet freak

its on the queue programmers.stackexchange.com/review/low-quality-posts/75431

Ampt

19

Q: Ethics of assisting other programmers, where do we draw the line?

In general, not just in relation to stackoverflow, sometimes I'm asked a programming question for which the answer will probably be used for good purposes, but there is a chance that it could not be. Two recent examples which reminded me of some real world questions I've been asked are: Send e...

ethics

that could almost be a meta question

user41796

@Ampt I put one of the 2 VTCs on that after reading over the answer

Ampt

yeah.. it's old, has very few views and too many answers

user41796

First off, Ethics are localized. Second, it was the broad boogeyman of "what's ethical" without any constraints. Ergo, primarily opinion based.

Ampt

3:14 PM

Yeah Jeff Atwood, get your crap answers out of here!

(pleasedon'tbanme)

user41796

@Ampt I'm not certain how active he is anymore

user41796

I wonder if he has a measure of inventor's syndrome with the overall site

ratchet freak

let's look it up: stackexchange.com/users/1/jeff-atwood?tab=activity

user41796

Inventor's syndrome == The inventor has a particular vision for their creation. When the users innovate and take the invention in a different direction, the creator sometimes gets upset about that and can't handle what their creation has morphed into.

Ampt

Have you seen the kind of crap that goes on around here? I'd be pissed!

Jimmy Hoffa

3:19 PM

@Ampt I'm a programmer not an artist, I don't draw lines

user41796

@Ampt Keep in mind that he fought having an MSE for a long, long time.

Ampt

@JimmyHoffa But can you draw red lines with this green pen?

user41796

And it was only the persistent amount of meta questions popping on SO that convinced him to allow it.

Ampt

Again, I ask, Have you seen the kind of crap that goes on around here?

hahaha

user41796

@JimmyHoffa You don't draw lines? You just use vectors instead?

ratchet freak

3:21 PM

@GlenH7 lines are glorified triangles anyway

user41796

@ratchetfreak only to the enlightened.

user41796

18 hours ago, by Jimmy Hoffa

@GlenH7 what's the sig of the Func<> you passed to .Select() ? What's the sig of MyMapperFunc() ? (note they're the same -> .Select(rec.MyMapperFunc) should work)

user41796

Apparently not. "rec" isn't in context at that point in time.

user41796

> The name 'rec' does not exist in the current context

user41796

Here's the actual line: .Select(rec.GetOneTrueClass());

Ampt

3:27 PM

6

A: Would you hire programmers without a 4-year college degree?

Why wouldn't you hire someone without a college degree or long term experience who just seems to be the guy for the job? Another candidate with similar skills plus a degree or experience. This primarily. A requirement to not only code, but also articulate your process in terms a degree or...

holy grave digging

Jimmy Hoffa

@GlenH7 oh right you're in an ienumerable, you had it right

user41796

@JimmyHoffa I had to run it both ways to be certain. So it's not like I knew ahead of time which was the right way.

user41796

@Ampt arguably it adds context. However, crud like that should just be left alone.

ratchet freak

needs a lot of delete votes to get purged though

user41796

What I would love to see is a C# explained from a C point of view.

user41796

3:34 PM

I understand pointers, function pointers, and other voodoo within C really well.

user41796

And when I map C# constructs onto that point of view, it helps me understand what the compiler is actually doing. So I have a more intuitive sense of what the code needs to look like.

ratchet freak

magic VM interprets some bytecode with GC and OO stuff

user41796

@ratchetfreak See, I want to pull back the curtain. I'm no longer content with ignoring the man standing behind the curtain.

Ampt

Well if you take C and go up a half step, you get c#

user41796

(subtle Wizard of Oz reference there)

Ampt

3:36 PM

Another half step gets you to D

ratchet freak

@Ampt going up a semitone

Ampt

Semitone

A semitone, also called a half step or a half tone, is the smallest musical interval commonly used in Western tonal music, and it is considered the most dissonant when sounded harmonically. It is defined as the interval between two adjacent notes in a 12-tone scale (e.g. from C to C♯). This implies that its size is exactly or approximately equal to 100 cents, a twelfth of an octave. In a 12-note approximately equally divided scale, any interval can be defined in terms of an appropriate number of semitones (e.g. a whole tone or major second is 2 semitones wide, a major third 4 semitones, and a perfect...

note the first sentence

ratchet freak

I know

Ampt

Ok?

@GlenH7 Wait a minute. Thats OP with that answer

user41796

@Ampt Doesn't really matter. Answers aren't meant for polls. Questions aren't meant for polls either.

Ampt

3:41 PM

I just didn't realize that

user41796

I hadn't either. The opening comment implied it wasn't the same person.

Jimmy Hoffa

@GlenH7 want to pull back the curtain? Go learn to use WinDbg and start trawling .NET memory dumps - that helped me learn a lot about the GC and how the objects and other things are laid out in memory and partitioned etc

user41796

I can well imagine

Ampt

@JimmyHoffa that actually sounds fun

Jimmy Hoffa

not to mention being a very useful tool for a .NET developer to dig into the dirtiest of bugs hiding away

Ampt

3:45 PM

have any memory dumps lying around?

Jimmy Hoffa

@Ampt It really is pretty neat. You can take a memory dump anytime, go to your task manager in windows, right click your .NET app under processes and click "Create memory dump" or something like that on the menu

user41796

But I just wrote a 17 line function (with whitespace) the reflects the properties of my source and target classes; traverses the source properties; finds the target property; and assigns the source value to the target.

Jimmy Hoffa

the tooling to dig into the memory dumps is all free too

user41796

I think my head is going to explode at how powerful and easy that was.

Ampt

@GlenH7 in C#?

or C?

user41796

3:47 PM

C#

user41796

impossible to do in C

Jimmy Hoffa

@Ampt Have you used reflection in C much?

ratchet freak

@GlenH7 it can be surprising how elegantly short some code can be

@JimmyHoffa reflection in C?

Jimmy Hoffa

@GlenH7 did you account for parameterized types? Reflection passes often require separate logic for generic types - they don't tend to behave the same in the scope of a reflected instance

@ratchetfreak my point exactly

Ampt

@JimmyHoffa I've done some fairly... dangerous things in C. Mostly just hard coding memory and overwriting it on the fly. This method just changed kthx.

nothing in production mind you. Just for fun

Jimmy Hoffa

3:48 PM

@Ampt it was a joke, you can't do reflection in C - do you know what reflection really is?

Ampt

@JimmyHoffa I thought it was just reading in the raw file of the code to do modifications on it. Sort of a hack-ish meta programming

user41796

scrubbed version:
private static OneTrueClass SourceToOneTrueClassTransform(Source source)
{
	OneTrueClass otc = new OneTrueClass();

	var sourceProps = source.GetType().GetProperties();
	var otcProps = otc.GetType().GetProperties();

	foreach(var sourceProp in sourceProps)
	{
		var otcProp = (from prop in otcProps
						where prop.Name.Equals(sourceProp.Name)
						select prop).First();

		otcProp.SetValue(otc, sourceProp.GetValue(source));
	}

	return otc;
}

whatsisname

WinDbg kind of sucks at debugging .net though

Jimmy Hoffa

@Ampt yes sorta except not the source code - it looks at the running assembly because it's done runtime when the only "source" avaiulable is the assembly itself. This works in languages which implement descriptive information of everything in the assembly in the assembly usually by using IL

Ampt

which is essentially what I had done. I edited the execution code during runtime, inserting new code like goto

then popping back in when I was done

whatsisname

3:50 PM

because you often get a dead end at the .net runtime and windbg can't really see into that

Jimmy Hoffa

@whatsisname it totally does, but for those hard to snag bugs, it really is one of the best ways to get an authoritative source of truth

whatsisname

but for native stuff though

Ampt

You can do that in C... not seeing what the problem is?

whatsisname

it's very useful

user41796

@JimmyHoffa probably not, I don't understand the term

Jimmy Hoffa

3:51 PM

@Ampt in C you can read your assembly file and identify the types and functions in it - aside from what the FFI exposes?

whatsisname

you could have reflection in C

Jimmy Hoffa

@whatsisname wouldn't that just be a disassembler?

whatsisname

not necessarily

Ampt

@JimmyHoffa Well, not if you don't know explicitly where the code exists in memory, but provided you know where you want to insert stuff (I.e. have the dissassembly code)

Jimmy Hoffa

@whatsisname you mean if the compiler put manifest information in the assembly to make reflection available?

whatsisname

3:51 PM

yeah

Jimmy Hoffa

@Ampt not in memory - on disk

whatsisname

noone does that to my knowledge

but you could if you wanted to

ratchet freak

@whatsisname self modifying code?

whatsisname

if you were making your own compiler/interpreter

Ampt

@JimmyHoffa No disk on a microcontroller. EEPROM->Memory, then modify your code in memory

Jimmy Hoffa

3:53 PM

@Ampt yeah the key is having the disassembly code - languages like .NET and Java building to IL effectively have that in their assemblies - which is how they support reflection, languages that compile to native as a rule usually don't package that disassembly information into their binaries

Ampt

so long as you're not dealing with a Harvard Architecture, your program resides in the same memory as your heap.

@JimmyHoffa yeah but a lot of the time you can either A) specify where in memory a function is located or B) inference based on known areas

Jimmy Hoffa

@Ampt have you ever used reflection in .NET to load an assembly from disk that wasn't in memory before?

Ampt

@JimmyHoffa No?

Jimmy Hoffa

in .NET and Java as well likely, you can reflect assemblies arbitrarily that aren't a part of your in memory program

@Ampt go use ildasm and open a few .NET assemblies and read their manifests, neat to see if you never have before

whatsisname

.NET gives spiffy tools for doing stuff like that

I don't remember what they are, but there are classes that you can use to inspect a bunch of assemblies in a directory

and it will look for classes that match an interface by name and type

and report those back for you, and it's provided by MS to make spiffy plugins without having to roll your own loader

Jimmy Hoffa

3:58 PM

@whatsisname yep, they come in handy once in a while when you know about this - you can use ildasm to figure out if a .NET assembly was built to run 64-bit only, 32-bit only, or Any CPU which is sometimes necessary to check

@whatsisname are you talking about the Managed Extensibility Framework (MEF) - the .NET plugin libraries?

whatsisname

Maybe, I don't remember what it was called

Jimmy Hoffa

.NET's reflection lets you do all that you describe, but MEF is the plugin library that abstracts over reflection to give something nice and easy to use for plugin systems in .NET. It's a nice clean lib and works well, I rather like it

whatsisname

just that it was spiffy and didn't depend on the plugin assemblies using the interface defined in your assembly, just had to match by name and type

Jimmy Hoffa

@whatsisname yep

sounds like MEF

whatsisname

brb

user41796

4:08 PM

@JimmyHoffa - what did you mean when you said "Reflection passes often require separate logic for generic types - they don't tend to behave the same in the scope of a reflected instance"

user41796

Can you give me an example of a generic type so I can figure out if I have to worry about that?

Jimmy Hoffa

@GlenH7 I don't remember that precise details, let me go dig into something real quick...

user41796

ok, thanks. That's leading me to think I won't have to deal with that.

Jimmy Hoffa

@GlenH7 plausibly not. You'll find out if you do as soon as you pass a generic type to the method

user41796

everything in my source class is a primitive or nullable primitive

user41796

4:12 PM

Throwing words out, I'd say everything I'm going to pass in is going to be concrete. ie. I'll have a specific class for it with specific properties.

user41796

The sources are the classes that EF built around my tables.

Jimmy Hoffa

@GlenH7 perhaps so long as you have the type it's fine, it's just when you're instancing a type from nothing but a string like "Some.NameSpace.MyType" you can instance easily from reflection, but constructing a type object and interacting with an instance of Some.NameSpace.MyType<SomeOther.NameSpace.OtherType> can be tricky because the type object essentially is describing two types and the behaviours get a little different

user41796

ah, ok. I'm not ready for that level of voodoo. :-)

Jimmy Hoffa

@GlenH7 yeah is no matter

user41796

I'm trying to decide if I want .First() or .FirstOrDefault()

Jimmy Hoffa

4:13 PM

probably won't effect you.. I think it happens when you're doing reflection construction

whatsisname

back

Jimmy Hoffa

@GlenH7 .First() -> Doesn't match predicate (or there's no items in list) throws exception, .FirstOrDefault() gives you a null you can either coalesce if you have a default value or bubble the null up the stack if it's acceptable to you and your consumers

whatsisname

@GlenH7: reflection does not recognize generics

a List<int> and a List<string> are two completely different 'normal' types

user41796

Ok, I want .FirstOrDefault() as I'd rather check for null than catch an exception and continue processing

whatsisname

and it's compiler voodoo that makes them work

there is nothing special about generics once you are at the IL stage

user41796

4:15 PM

@whatsisname ok, that makes sense

user41796

Since I only have a class that I'm passing in, not a generic container of classes, I ought to be okay

whatsisname

somewhat similar to C++ templates in that it results in generating the code twice, sorta

Jimmy Hoffa

@whatsisname yes and no... as a type, the parameterization of the type is something you can inspect through reflection and identify, but yes I know what you meant in that it's all turned into a non-parameterized types in the IL and heap

whatsisname

right, hence the "sorta"

user41796

4:37 PM

@JimmyHoffa - Another odd question for you

user41796

Why is it that I get an exception when I put a .ToList<int>() at the end of a .Select(blah), but I can put the same .ToList<int>() on the IQueryable that I get back from the .Select and it works just fine?

user41796

Maybe I should ask that on SO and get some repz

whatsisname

imaginary internet points ftw

psr

@whatsisname Reflection does recognize generics. There are ways to get both the unbound generic type and the generic type with the type parameters bound to specific types (though with some weirdness, especially that a generic method reflecting on an unbound type with a type parameter used in the generic method itself can reference an unbound type at compile time and a bound type at run time).

Jimmy Hoffa

@GlenH7 When you parameterize your .ToList call, it tries to do a cast on it. Doing it on an IEnumerable will execute the IEnumerable's implementation which does it in the .NET process and probably barfs on the cast. Doing it on an IQueryable will use the implementation on that IQueryable which is to generate an expression tree, translate that to T-SQL and the result is your cast happens in the SQL server which has more forgiving cast behaviour

user41796

4:55 PM

That makes sense

user41796

I ran some more tests

user41796

.ToList works okay with a non-null int field

user41796

although it throws an exception when I do the same thing against my OTC results that I got off of transforming a select

user41796

This time it threw: Message "Unable to create a constant value of type 'System.Object'. Only primitive types or enumeration types are supported in this context." string

user41796

And I'm pretty sure I'm going to get the same exception when I use a ToList against a null field in the table.

Jimmy Hoffa

4:59 PM

@GlenH7 do a .ToList() on your IQueryable immediately unless you need to do a complex query for performance reasons or otherwise then doing all that extra composition to alter the SQL to be executed is a necessity... but it sounds like you're running into a delayed-execution issue with the placement of your first IQueryable iteration vs. the scope of your connection

(perhaps)

user41796

@JimmyHoffa Yeah, I think that's what's happening

Jimmy Hoffa

just because I've never heard of an IQueryable returning an IEnumerable of elements where any of the elements are null...

the concept seems very odd

Do your .ToList() early to force iteration

user41796

I'm not getting null elements, but I'll get a runtime exception

user41796

I start with a promise (select expression from EF) and make another promise with that result. Chaining a .ToList<> off the second promise is where I'm seeing the exception. And it seems like when I make the second promise off of the first, I don't have anything to force the evaluation of the first promise.

Jimmy Hoffa

forcing iteration means "this computation of a query - I need the results now - so actually expand it and execute it" -> then make sure all references to that IQueryable disappear lest a later point it get's iterated with a .Where() or other clause that will cause another query execution incidentally

user55340

5:04 PM

Gah... was up late last night... poked on P.SE in the early hours and... gah... it was filled with crap questions.

user55340

Another example separated by 10 minutes: 04:27, 04:38 — MichaelT 12 hours ago

user55340

Those are 10k links now too.

Jimmy Hoffa

@GlenH7 that .ToList will force it, but again because you're hanging it off another promise (assuming that other promise is doing something that can't be translated to T-SQL like a C# select execution) you're .ToList<> cast is happening in .NET (is that what you wanted or do you want to cast in T-SQL?)

user41796

@JimmyHoffa Yes, that's what I need to have happen.

user41796

This all goes back to that table pivoting crap I've been dealing with.

user41796

5:07 PM

Fortunately, I don't have to pivot anymore. But I have to break out the data points from the humongous tables.

user41796

In this last particular case, I need to extract out distinct years from my smaller result set that I'm operating upon.

user41796

So I fetch target; convert target to OTC (props to you!); and then get distinct years off of OTC results.

Jimmy Hoffa

@GlenH7 OH - distinct could be your problem. Distinct implementations in LINQ are a touch counter-intuitive sometimes

user41796

Looking at the variables as I'm stepping through, the distinct years request is still a promise because I haven't done anything with it.

user55340

btw, could we get two more reviews: programmers.stackexchange.com/review/close/75342

Jimmy Hoffa

5:10 PM

@GlenH7 you mean when it blows up that distinct hasn't happened yet?

2

A: LINQ join with distinct resultset

Linq distinct is annoying, you have to implement the IEqualityComparer interface on the class you want to Distinct(), then the way you get your distinct selection you're referring to would be: IEnumerable<Person> people = GetPeople(); people.SelectMany((person) => person.LastName).Distinct(); ...

user55340

@Snowman btw, thank you for regularly showing up in the close review queue. It helps a lot.

user41796

@JimmyHoffa I'm not quite sure of the trigger yet.

user41796

As it just blew up when I did a ToList from the Iqueryable of the IQueryable.

Jimmy Hoffa

@GlenH7 you know how to use your exceptions filter in the debugger to break on handled exceptions?

user41796

@JimmyHoffa Oh Fark.

Jimmy Hoffa

5:12 PM

@GlenH7 that answer is SHIT

I just looked at it... gah wth?

user41796

Tell me it's not true?

Jimmy Hoffa

@GlenH7 iduno. The bits about distinct are right, but I have no idea why I mentioned SelectMany and used it wrong to begin with

O_o

feckin SO archiving my idiocy to remind me about it years later...

user41796

Rather than worry about IEqualityCompare, I'm going to push the distinct operation over to T-SQL. :-)

Jimmy Hoffa

@GlenH7 wise choice

will be more efficient anyway

user41796

I can be lazy too.

user41796

5:16 PM

:-)

Ampt

Alright, I'm back. I know I've been gone a while but I'll have you all know I was busy.

Jimmy Hoffa

@GlenH7 the fact that you know how to do those shows you're getting your head around that oh so nebulous "LINQ" thing people are always vaguely waving their hands about...

@Ampt nobody wants to hear about your morning constitutional

Ampt

@JimmyHoffa ... that was only like 20% of it.

user15026

@Ampt You were gone for long enough for me to go for a walk and eat crepes.

user41796

@JimmyHoffa Yeah, I'm still pretty dang proud of that 17 line function I wrote to feed source into target.

Ampt

5:18 PM

@AshleyNunn That constitues ~ 1.28 days of work for me.

user55340

@Ampt btw, make sure to use mathjax in your comments to piss people off...

user55340

@dmckee you mean you don't like pimped comments? — MichaelT 12 hours ago

user55340

Chase those links for more mathjax fun.

Ampt

Those are glorious!

We need mathjax on programmers

user15026

Okay, that is awesome.

Miklas

5:19 PM

Requesting for a tag to be merged or created as a synonym can be posted on meta?

Ampt

Yes.

user55340

And chat? 💩 isn't enough?

Ampt

tag-requests Disclaimer: May or may not be a real tag.

user41796

@Ampt but only for comments

user41796

@Miklas yes

Jimmy Hoffa

5:20 PM

@AshleyNunn eat creeps? I thought canadians were supposed to be polite... shouldn't you just say "Excuse me" and hand them your wallet next time?

Ampt

@GlenH7 Definitely not Ever for titles.

user55340

user15026

@JimmyHoffa Next time. This time I was too hungry :P

Ampt

No wonder they get so many stupid questions over there. you can't search for anything

user55340

@Miklas On smaller sites (read: anything other than SO), there are rarely enough people with the ability to vote on tag synonyms... meta is the best place for that.

user55340

5:23 PM

(You need 4 people with +5 or more score for answers for a tag... the less used tag rarely have the 4 users with +5 answer score)

psr

@JimmyHoffa Eat crepes is a French Canadian time unit, equal to a third of a Poutine.

user55340

@psr I thought they were metric?

user55340

btw, I am a proponent of power of 2 time... 64 seconds/min, 64 min/hour, 32 hour/day... just need to redefine a few things... but picture how easy that would make it for programmers.

user15026

@MichaelT Would make for an interesting use case. I wonder how it would affect a day/night cycle.

user55340

@AshleyNunn One day night cycle is still 32 hours.

user15026

5:27 PM

@MichaelT Well, yes, but I more mean with daylight and such

psr

@MichaelT I meant a hundred centicrepes.

user55340

A Deepnes in the Sky had base 10 seconds as its unit...

user55340

Incidentally, its epoch time was unix epoch time.

user55340

> Take the Traders' method of timekeeping. The frame corrections were incredibly complex - and down at the very bottom of it was a little program that ran a counter. Second by second, the Qeng Ho counted from the instant that a human had first set foot on Old Earth's moon. But if you looked at it still more closely ... the starting instant was actually about fifteen million seconds later, the 0-second of one of Humankind's first computer operating systems.

user114359

@MichaelT you're welcome. I try to help.

user55340

5:36 PM

I recall an old version of bind had a TTL configuration parameter that was based on the power of 2. You could set the cache timeout to N where the value was 2^N. Thus, if you wanted about a minute for the TTL, you set it to 6. (2^6 = 64 seconds)... a quarter hour or so was 10 (2^10 = 1024 seconds vs 900 seconds). An hour was 2^12 seconds or there abouts... A day was 16, a week was 20... Two weeks was 21...

user55340

This let you use the configuration for a DNS cache TTLs with reasonable values in a single byte for time.

user41796

→ 1 message moved to Trash

user55340

@GlenH7 btw, lots more engineering fun in that photo sensor question...

user55340

19

A: What limits the size of digital imaging sensors?

You can make some very large CCDs. An older press release talks of a CCD that was made for the US Naval Observatory that is 4" x 4" and 10,560 pixels x 10,560 pixels. Thats 111 megapixels on one sensor. Thats kind of not small. (From above press release) The first restriction that the sens...

user55340

Someone tried calling me on the "but there are 3 gigapixel telescopes..." and I pointed out how they were designed...

Transcript for