« first day (1970 days earlier)      last day (2899 days later) » 

3:25 AM
@jcolebrand Are you talking Architecture Astronaut or hands-on DB design person?
 
 
4 hours later…
7:23 AM
@PaulWhite If that question migrated from meta gets closed, will it become a rejected migration on meta as it happens with other cross-site migrations?
 
@AndriyM Not sure :) Will keep an eye on it.
@AndriyM Actually, I headed that off at the pass. Won't happen now.
 
At first I thought it would've been better to just close it on meta, but now I'm curious what will happen when it's closed here (if it's closed, of course).
Oh, it's no longer a migrated question? How did that happen?
 
@AndriyM Yeah, maybe. Anyway, it made no sense as a meta question, and was clearly intended to be asked on main, so I migrated it.
@AndriyM That's what I did to prevent any weirdness.
I'll delete the meta original at some stage as well.
 
Nice, never heard of that. Can you do that only to questions migrated between main and meta or not only to those?
 
@AndriyM Any. Rarely needed though.
Sometimes for questions migrated to the wrong site, rejected, and then needing migration again.
 
7:57 AM
Hi there! Anyone here good with innodb recovery? I have a .frm and .ibd files for a wordpress database, and I need to dump all the data. Don't want to reconstruct the entire database/tables, just want to get as much data from it as possible. The files are from a MySQL on Windows server, and my recovery environment is linux :|
 
I have a table with 3 UNIQUE, 4 FK and 6 CHECK constraints ;)
@AwalGarg You'll have much better chances of finding help for this in either the SO MySQL chat room or in the #mysql IRC channel
 
Alright, would cross-post there. Thanks!
 
You can of course always post a question at our site. People that answer mysql questions are not regulars on this chat room
 
will do :)
 
@ypercubeᵀᴹ That's nice.
 
8:16 AM
@ypercubeᵀᴹ looks promising, thanks!
 
@AwalGarg See also this (related) answer: dba.stackexchange.com/questions/23251/…
The links at the bottom are for (his) recovery tool
Not sure how related the question is to your problem.
 
 
1 hour later…
9:39 AM
@ypercubeᵀᴹ I recovered the data successfully, thanks!
 
Great.
Now, start taking backups often ;)
 
yup. good call :)
 
9:51 AM
I don't understand your downvote. Everything I said is correct and this question is Oracle related. Noone cares about SQL-Server in this thread! — ora-600 11 mins ago
Huh.
 
let me correct that to -2
 
@ypercubeᵀᴹ He self-deleted rather than making the suggested small improvement. Goodness me.
Or perhaps he realized it was inaccurate? I guess we'll never know.
 
Yeah. If they just said, "it often uses a sort to find the duplicates in UNION", I might even upvote
 
@ypercubeᵀᴹ Same here. I would certain undownvote.
People do react strangely to a -2.
To be fair, he did edit the answer to include "In Oracle..." shortly before deleting.
 
10:11 AM
@PaulWhite Indeed. Why not wait to at least (at most?) -3.
 
@AndriyM Oh ha ha I meant the -2 rep points for a down vote.
 
I see. But still...
 
Yes, what you said is also true.
@ypercubeᵀᴹ Good edit. Surprised you didn't vote to undelete?
 
11:09 AM
Oh, thnx. I thought I did!
Done now
I was also in the middle of million things: pushing some code, talking to the boss on Skype, talking to two estate agencies ...
I'm trying to find a new home to move and the estate agencies are a pita
 
 
2 hours later…
12:44 PM
Was this any of you lot? theregister.co.uk/2016/05/27/…
 
@Phil so he is banned from Oxford Street. Can he freely use his camera in other streets?
 
1:07 PM
How can Hotels and Restaurants afford Oracle?
 
@MarkSinkinson have you seen how much hotels charge in London?
 
1:24 PM
@MaxVernon I live here, so no ;-)
 
Seen on the interwebs today ...
Tell Schroedinger I survived.
 
@ConcernedOfTunbridgeWells sweeet
 
For a kiwi, technically that should be 'Sweeet as.'
 
@ConcernedOfTunbridgeWells Ironically, he didn't :)
 
2:10 PM
@MaxVernon Pretty big ask, Max.
 
@PaulWhite aaaarrrrgggghhhhh!!!
 
@MaxVernon Also, that E & LU answer is from almost six years ago.
 
@PaulWhite and your point is? ASK IS NOT A NOUN! What's wrong with saying "that's a pretty big request" ?
@PaulWhite that's photoshopped, isn't it, lol :-)
 
@MaxVernon Well I'm saying is it is pretty common use these days. Especially in my country, can't say about yours.
 
I know I won't be getting any big promotions if I keep using "request" instead of "ask". People will think I'm some kind of throw-back to the dark ages.
 
2:20 PM
And the financial sense of bid/ask has been around forever.
 
@PaulWhite oh man, everyone uses it here. It drives me insane.
@PaulWhite true enough.
 
Well then. :)
 
> ‘Sometimes one is forced to consider the possibility that affairs are being conducted in a manner which, all things being considered and making all possible allowances is, not to put too fine a point on it, perhaps not entirely straightforward.’
 
Sounds like a quote from Sir Humphrey Appleby.
 
@PaulWhite lol, you're just way too smart
 
2:22 PM
 
although, that itshambles post is from 2012, so it's quite likely very out of date.
 
"submit a fish passage plan" sounds quite nasty. I'm pretty sure I've had one or two of those.
 
Still reading it. Is quite long.
Oh I saw you commented on a Karen Lopez blog. Brave!
Hm. "ask" fails the "In general, don’t use a short word where a long one (ideally, four or five long ones) will do." test.
 
@PaulWhite hmmmm refresh me?
 
2:31 PM
@PaulWhite I just happened to see a re-run of Yes, Minster the other day and it's still a classic.
It was one of the ones with the Israeli ambassador that Jim Hackere went to LSE with.
 
@MaxVernon Sorry I confused you with Mike Walsh for some reason.
 
@PaulWhite ahhh. no worries mate!
 
@ConcernedOfTunbridgeWells It came to the St James theatre last year. Was excellent.
 
@MaxVernon That's 'Sweet as, Mate.'
 
@ConcernedOfTunbridgeWells lol
 
2:33 PM
@ConcernedOfTunbridgeWells Foreigners eh.
 
@PaulWhite Anybody would think they didn't know the difference between us and Australians!
 
@ConcernedOfTunbridgeWells what you mean you're not Australians? ;-)
lol
 
le sigh
 
that's like people calling Canadians Americans.
Or Scottish English.
it might be a fitting chapter in "How to win friends, and influence enemies"
 
Except if you get bashed, we won't apologise immediately after.
 
2:35 PM
lol
 
2:53 PM
This is going to sound weird but what would you guys consider "a line of code" for the purpose of counting lines of code?
for example how many lines is this ...
var newInvoices = await api.PostAsJsonAsync("CLeX/SIHead/ByRefs()?$expand=SupplierCompanyAux", refs)
    .ContinueWith(t => t.Result.Content.ReadAsAsync<ODataCollection<SIHead>>())
    .Unwrap()
    .ContinueWith(t => t.Result.Value);
is that 1, 4 ... some other number?
 
@Darth_Wardy Does Visual Studio not have line numbering?
:)
 
it does ... just trying to build some guidance notes for later improving our coding standards
are you thinking "go with the VS definition of a line"
 
what language is this? C# ?
 
the bit between two Environment.Newline chars
@ypercubeᵀᴹ yeh it is
 
I guess the answer is: the way everyone else on your team counts lines of code. If you are setting that standard, ask them, then come to an agreement. Using the VS definition seems easy, yes.
 
2:56 PM
@PaulWhite yeh im just trying to pre-empt the pedantic ones
 
@Darth_Wardy I would say that is 1 line, but really you should go with whatever the common IDE is in your team
 
@Darth_Wardy That is doomed to failure. No one can anticipate a true pedant.
4
 
@MarkSinkinson most the methods I write these days contain about 2 to 4 statements like that
so technically I write on average (depending on how you look at it) 3 line methods or like 12 line methods
 
Perhaps the number of lines of code in that example is the Schrödinger number [1:4].
 
@Darth_Wardy From my point of view, Line 2 in your statement doesn't do anything. It is nothing without Line 1, so should not be considered a separate line
 
2:59 PM
Or count the number of characters and divide by the Standard Statement Length :)
Ok I'm not being helpful. The site seems quiet, so I'm going to grab some sleep.
 
@MarkSinkinson yeh i kinda look at that as a "multiline definition of a task" that interpreter / compilers sees as a single statement
i guess I was wondering if its best to consider a line as a line or a statement as a line
 
@PaulWhite this is similar to the definition of surreal numbers (and games which is a generalized type of numbers)
@Darth_Wardy not exactly relevant but Python has a style guide which is almost a whole book: PEP8
 
@Darth_Wardy why are you counting lines of code? Surely not as any kind of reliable metric for performance?
neither performance of the code nor performance of the developer
 
@MaxVernon right now its just about code analysis, i'm trying to determine our throughput of actual work vs boilerplate / whitespace type work
 
@Darth_Wardy so performance of the developer then. is more lines a good thing or a bad thing?
 
3:07 PM
@PaulWhite If you can find a book about Surreals, take a look. I think you'll enjoy the elegant definitions
 
@MaxVernon yeh i'm probably going down the wrong road here
 
reason I ask is you could say the line count is 4 or 1 depending on if you want more or less
 
I was thinking of taking a look and seeing if Icould set a standard for "max number of lines in a method" or something
 
@Darth_Wardy ok.
 
it's just about figuring out guidance notes for junior devs
 
3:08 PM
two words. code review.
 
personally i'm overkill on SOLID so if it remotely breaks any of those principles i'm like "do it again"
@MaxVernon yeh this is code review stuff
just trying to figure out some happy mediums for the team
 
I'm a really big opponent of setting any kind of hard and fast rules about development.
Development sometimes needs to be meticulous, and sometimes needs to be fast. How do you rectify that?
I'd be much more interested in understanding if a piece of code successfully meets the criteria for that piece of code
which admittedly is much harder than simply creating a "standard"
 
yeh that's why i'm thinking of having a standard which is something like "aim for a method that fits on your screen that should be roughly 30 lines (according to VS)"
based on our standard screen sizes and resolutions
its more a should than a must
and the code reviewer would make the call as to weather or not the standard was broken
many things of course would come in to play
based on that, the same above would be treated as 4 lines
 
@Darth_Wardy see, if you prefer readability of a single line over readability of a single function, that could go either way. If I was told all my code had to fit the height of code window in Visual Studio, I'd make all my multi-line code into single lines. The unfortunate side-effect there would be reduced readability of that code down the road.
 
@MaxVernon yeh but there would be another standard to prevent that
fluent api's already have a standard to account for that (namely each fluent call should begin on a new line)
also we have line length limits
in theory if you have to either write a 100 line method or 30 really long lines I would questio what it was you were trying to do
in other words ... should that code be broken down a little better?
that's what code reviews are for ... to raise that question
 
3:17 PM
@Darth_Wardy so that drives indirection then?
 
i'm not out to have this used as an iron stick to beat people with
 
@Darth_Wardy I've found split bamboo works much better for that :-)
 
it's more about "if you write code this way it will be easier to maintain"
just trying to be flexible i guess and allow for different situations
 
@Darth_Wardy anything that makes code easier to maintain has to be going in the right direction, I suppose.
 
I had a junior tell me the other day that "SRP was not worth apply in code because it took longer to create new classes in new files than simply throwing a bunch of methods in the main program class"
this is coming from the guy that wrote a whole business process in a single method
I'm looking to avoid that
I basically threw SOLID at him and said ... how do we unit test this?
and he shrugged at me and said ... you don't
 
3:19 PM
morning heapers
 
@Darth_Wardy no doubt. SRP is a good thing. Having said that it can, like anything, be taken too far.
mornin' oh blue mighty one!
 
@bluefeet morning
 
@MaxVernon yeh i was literally just saying stuff like "why is your parsing code making business decisions" and "why is validation writing xml files"
and when you get to line 2000 in a single mthod and you are 5 loops deep I felt i was completely lost as to what private variables I had and what state those variables were in
 
> The blue voice then says her famous line "Blue is beautiful, blue is best. I'm blue, I'm beautiful, I'm best!".
 
so he broke it all down in to a ton of static methods in the program class ech about 30 to 70 lines long and was like "there ya go SRP" ... i was like ... eh???
 
3:22 PM
@MaxVernon <3
 
@Darth_Wardy hmmmm fun.
 
yeh so i want to define some guidelines to prevent this
 
@bluefeet you're welcome!
in case anyone's wondering what that's all about, see this
fond memories from my childhood
@Darth_Wardy that is frightening when you see it. It's really frightening when you have to go back and fix legacy code written by someone 5 years ago that looks like that. Good luck is all I can say.
 
@MaxVernon well that's why i basically sat there and told him that under no condition unless he could prove it was both 1 working and 2 testable was it ever going anywhere near production
and the boss already gave him a deadline to get it production ready
 
@Darth_Wardy some lessons are hard :-)
 
3:28 PM
well considering our entire web stacks core framework is about 2000 lines of code I was like ... WTF are you doing here?
he was like "well there's all these odd conditions and this and that i have to check" I was like uh ok ... show me ...
to which we found that half his code belongs in 1 of our web API's and the other half in another with about 10 lines to call them both in what he has
 
@Darth_Wardy maybe he's a follower of the Flying Spaghetti Monster
 
so i'm iteratively teaching him what I can already see
 
@Darth_Wardy sounds like he's going to get a good lesson or two!
 
@MaxVernon he's not happy because the boss keeps changing the requirements on him either which doesn't help but I told him if he followed SOLID more rigidly then each refactor might only be in a couple of methods instead each time he has to dig through about 2000 lines of mess
he so far is in agreement but says stuff like "well that's where we want to be but aren't there yet"
and i'm like ... well we will be when you do as your told instead of complaining
 
man, having the requirements change mid stream is a bit of a bitch, but yes, modular code is much easier to change to fit those fluid requirements.
 
3:32 PM
then he starts moaning about time ... and i'm like fine ... new backlog item ... this is your only task to do until done ... get on with it
@MaxVernon the thing is the requirements do shift but he's working directly off the requirements doc instead of the backlog
he's his own worst enemy in that right
refuses to work from TFS backlog items and insists he needs all this constantly shifting requirements docs
im like ... just build a version then do a release, then build another version (in theory having only to build bits for v2 as v1 has it mostly right)
 
3:49 PM
little manageable chunks are definitely the way to go. You don't try to eat a steak in one bite, and if you do, you'll most likely die from choking.
 
yeh thats why i aim for SOLID based code models
 
 
1 hour later…
5:14 PM
just call me king of the obvious™ dba.stackexchange.com/questions/139758/…
 
 
2 hours later…
7:15 PM
Don't forget to tell PASS how many PCs your organisation uses so that you're eligible to vote
Actually n/m. Not sure that that is one of the required questions
 
7:43 PM
Nice piece of hardware: 22TB RAM : serverwatch.com/server-news/…?
 
@ypercubeᵀᴹ and, if you have to ask "how much", you can't afford it!
So, while you may think I'm not being constructive, I'm actually trying to ensure you get the best answer possible. Presumably you are aware of the XY problem and are simply adding the 3rd column to slow things down? If you're interested, look at this question, which provides more details. — Max Vernon 30 mins ago
 
@MaxVernon the reason (or one of them) is probably an ORM. See the first version of the question.
The edit made the q less clear. "implementing business logic"? Hah? How is this related to surrogate IDs?
(TLDR: I agree with your comments)
 
@ypercubeᵀᴹ ahh that old tired bit, "I'm too lazy to actually code the database layer myself". Next questions will be "why is this so slow".
@ypercubeᵀᴹ no doubt!
 
 
1 hour later…
8:53 PM
@Erik Hi, Erik, how are you? I haven't seen you in The Heap lately. Your votes (in case you haven't voted already, of course) and/or opinions regarding this meta question would be much appreciated. Cheers.
 

« first day (1970 days earlier)      last day (2899 days later) »