p-lang thread: (now (have you (problems two)))

The Something Awful Forums > Discussion > Serious Hardware/Software Crap > YOSPOS > p-lang thread: (now (have you (problems two)))

«‹›1784 »

Shaggar: Apr 26, 2006

web services is unrelated to web "development"

# ? Sep 23, 2013 19:49

Adbot: ADBOT LOVES YOU

# ? Jun 11, 2024 02:50

Notorious b.s.d.: Jan 25, 2003; by Reene

if everything in your application and everything your application communicates with are all under your control and written for a single loving platform

yeah, go buck wild. write your own serialiser that formats all data as ascii dicks

i don't care, any god drat thing will work. it's the degenerate case

# ? Sep 23, 2013 19:50

Notorious b.s.d.: Jan 25, 2003; by Reene

xml solves a lot of hard problems, and if you don't understand why xml is good, it means you haven't figured out the problems yet

# ? Sep 23, 2013 19:50

Posting Principle: Dec 10, 2011; by Ralp

OBAMA BIN LinkedIn posted:

web services

Java code:

String accept = request.getHeader("accept");
if (accept != null && accept.contains("xml")) {
	throw new HTTPException(HttpServletResponse.SC_NOT_ACCEPTABLE);
}

# ? Sep 23, 2013 19:51

X-BUM-RAIDER-X: May 7, 2008

trigger warning for people who do not know what a web service is

# ? Sep 23, 2013 19:52

double riveting: Jul 5, 2013; look at them go

sorry i'm not keeping track, but didn't w3c or whoever decide that xhtml was a dumb idea?

# ? Sep 23, 2013 20:00

Zlodo: Nov 25, 2006

Notorious b.s.d. posted:

xml solves a lot of hard problems, and if you don't understand why xml is good, it means you haven't figured out the problems yet

i still dont see how it is such help with interoperability when you are allowed to just shove data haphazardly in it or could perfectly get away with writing an adhoc parser that could break if you do as much as inserting a tag somewhere

# ? Sep 23, 2013 20:01

prefect: Sep 11, 2001; No one, Woodhouse.
No one.; Dead Man’s Band

double riveting posted:

sorry i'm not keeping track, but didn't w3c or whoever decide that xhtml was a dumb idea?

it said that tags had to be lower-case instead of upper-case, which was all it took to get me to endorse it :thumbsup:

# ? Sep 23, 2013 20:02

Shaggar: Apr 26, 2006

Zlodo posted:

i still dont see how it is such help with interoperability when you are allowed to just shove data haphazardly in it or could perfectly get away with writing an adhoc parser that could break if you do as much as inserting a tag somewhere

you cant get away with writing an adhoc parser and the entire point of xml is it prevents you from shoving in data haphazardly.

# ? Sep 23, 2013 20:03

Zlodo: Nov 25, 2006

Shaggar posted:

you cant get away with writing an adhoc parser and the entire point of xml is it prevents you from shoving in data haphazardly.

no, it doesnt, because you can shove data in an xml file without defining a schema which is what most people do in my experience

and i dont see what guarantees you have that people who consume your data will actually use your schema and not just write some adhoc parser that doesnt care about anything in your format that is not yet used (and which will break their parser when it is some time down the line)

# ? Sep 23, 2013 20:21

Shaggar: Apr 26, 2006

i don't care about how they consume my well formed xml and i wouldn't accept non-well formed xml. the point is they can do it the right way and doing it the right way is easy if you aren't using lovely tools (and also easier than doing it any other way). with every other serialization format there is no right way

# ? Sep 23, 2013 20:23

X-BUM-RAIDER-X: May 7, 2008

Zlodo perhaps you should leave the forums for a while and come back once you've caught up on what a "standard" is

X-BUM-RAIDER-X fucked around with this message at 20:32 on Sep 23, 2013

# ? Sep 23, 2013 20:29

Janitor Prime: Jan 22, 2004; PC LOAD LETTER

What da fuck does that mean; Fun Shoe

Soricidus posted:

a format that's deliberately designed not to scale? sounds good to me

How so?

Also I don't see what problems xml is solving that protobuf hasn't also solved but better.

# ? Sep 23, 2013 20:51

Zlodo: Nov 25, 2006

OBAMA BIN LinkedIn posted:

Zlodo perhaps you should leave the forums for a while and come back once you've caught up on what a "standard" is

yeah, it is a standard. yeah, you can avoid pitfalls when using it if people respect a standardized way to use it. yes, you can build a standard on top of it.
it has nothing to do with whether it is poo poo or not. perhaps you could say its serviceable poo poo? there's lot of serviceable poo poo such as p-langs, but they're poo poo nonetheless

# ? Sep 23, 2013 21:11

X-BUM-RAIDER-X: May 7, 2008

Zlodo posted:

yeah, it is a standard. yeah, you can avoid pitfalls when using it if people respect a standardized way to use it. yes, you can build a standard on top of it.
it has nothing to do with whether it is poo poo or not. perhaps you could say its serviceable poo poo? there's lot of serviceable poo poo such as p-langs, but they're poo poo nonetheless

the point of a standard is that you can verify that the stuff your library is writing conforms to what another thing expects which also uses the same standard, while completely ignoring non standard poo poo that peasants fire to your service because it's bad and poo poo. XML and SOAP have these standards. Json has nothing of the sort.

# ? Sep 23, 2013 21:35

MononcQc: May 29, 2007

It's just a very much relaxed standard (E: do you count RFCs as standards? I guess so?)

# ? Sep 23, 2013 21:53

Brain Candy: May 18, 2006

Notorious b.s.d. posted:

xml solves a lot of hard problems, and if you don't understand why xml is good, it means you haven't figured out the problems yet

# ? Sep 23, 2013 22:05

Brain Candy: May 18, 2006

i looked at the 8 page RFC for JSON. i looked at the errata too

http://www.ietf.org/rfc/rfc4627.txt?number=4627 posted:

A JSON text can be safely passed into JavaScript's eval() function
(which compiles and executes a string) if all the characters not
enclosed in strings are in the set of characters that form JSON
tokens. This can be quickly determined in JavaScript with two
regular expressions and calls to the test and replace methods.

var my_JSON_object = !(/[^,:{}\[\]0-9.\-+Eaeflnr-u \n\r\t]/.test(
text.replace(/"(\\.|[^"\\])*"/g, ''))) &&
eval('(' + text + ')');

ahahahahhahahah

# ? Sep 23, 2013 22:09

X-BUM-RAIDER-X: May 7, 2008

Brain Candy posted:

i looked at the 8 page RFC for JSON. i looked at the errata too

ahahahahhahahah

# ? Sep 23, 2013 22:16

Notorious b.s.d.: Jan 25, 2003; by Reene

MononcQc posted:

It's just a very much relaxed standard (E: do you count RFCs as standards? I guess so?)

RFCs are the perfect example of why real standards issued by committees are so great

# ? Sep 23, 2013 22:22

Suspicious Dish: Sep 24, 2011; 2020 is the year of linux on the desktop, bro; Fun Shoe

Brain Candy posted:

i looked at the 8 page RFC for JSON. i looked at the errata too

ahahahahhahahah

the comedy stylings of douglas crockford. he legitimately believes that's a safe way of evaluating json

# ? Sep 23, 2013 22:30

Soricidus: Oct 21, 2010; freedom-hating statist shill

Hard NOP Life posted:

How so?

Also I don't see what problems xml is solving that protobuf hasn't also solved but better.

"Protocol Buffers are not designed to handle large messages. As a general rule of thumb, if you are dealing in messages larger than a megabyte each, it may be time to consider an alternate strategy."

"Protocol Buffers do not include any built-in support for large data sets."

"If you want to write multiple messages to a single file or stream, it is up to you to keep track of where one message ends and the next begins. The Protocol Buffer wire format is not self-delimiting, so protocol buffer parsers cannot determine where a message ends on their own."

in other words, protobufs are intended for sending short messages over the internet. they do that pretty well, but that's all they're good for. if you need to handle non-trivial quantities of data, you'd have to invent your own proprietary file format that maybe uses protobufs internally to store chunks of data (and then you still have to worry about whether those chunks might hit the size limit).

alternatively you could just get over your irrational fear of angle brackets and use xml.

# ? Sep 23, 2013 23:30

tef: May 30, 2004; -> some l-system crap ->

Brain Candy posted:

i looked at the 8 page RFC for JSON. i looked at the errata too

ahahahahhahahah

read the bit about unicode

quote:

To escape an extended character that is not in the Basic Multilingual
Plane, the character is represented as a twelve-character sequence,
encoding the UTF-16 surrogate pair. So, for example, a string
containing only the G clef character (U+1D11E) may be represented as
"\uD834\uDD1E".

and of course we all know that json isn't a subset of javascript, right?

# ? Sep 23, 2013 23:33

Nomnom Cookie: Aug 30, 2009

So is array() a JSON list or a JSON object?

# ? Sep 24, 2013 00:26

double sulk: Jul 2, 2010

xml is fine and better than a lot of programming related things out there

# ? Sep 24, 2013 00:49

Shaggar: Apr 26, 2006

Nomnom Cookie posted:

So is array() a JSON list or a JSON object?

yes

# ? Sep 24, 2013 00:54

Nomnom Cookie: Aug 30, 2009

Shaggar posted:

yes

i need a response from tiny bug tef

# ? Sep 24, 2013 00:55

double sulk: Jul 2, 2010

json is ok when you are only using it for document store, for example when you are utilizing the sheer power of mongodb in order to reach web scale.

# ? Sep 24, 2013 00:57

Brain Candy: May 18, 2006

tef posted:

read the bit about unicode

and of course we all know that json isn't a subset of javascript, right?

i went in expecting unicode literals, was not disappointed

# ? Sep 24, 2013 01:05

Vanadium: Jan 8, 2005

I just don't get how the rust people manage to get stuff done when the language can't even have multiple non-const pointers to the same object without requiring the GC (or refcounting since they don't have a GC yet, afaik), and the type checker has the weirdest notion of pointer lifetime (like here https://github.com/mozilla/rust/issues/6613) so basically anything gets you "hold on there's too many pointers to this thing already" errors.

Yet they have a somewhat working browser engine

they must be wizards

# ? Sep 24, 2013 21:35

Bloody: Mar 3, 2013

Soricidus posted:

"Protocol Buffers are not designed to handle large messages. As a general rule of thumb, if you are dealing in messages larger than a megabyte each, it may be time to consider an alternate strategy."

"Protocol Buffers do not include any built-in support for large data sets."

"If you want to write multiple messages to a single file or stream, it is up to you to keep track of where one message ends and the next begins. The Protocol Buffer wire format is not self-delimiting, so protocol buffer parsers cannot determine where a message ends on their own."

in other words, protobufs are intended for sending short messages over the internet. they do that pretty well, but that's all they're good for. if you need to handle non-trivial quantities of data, you'd have to invent your own proprietary file format that maybe uses protobufs internally to store chunks of data (and then you still have to worry about whether those chunks might hit the size limit).

alternatively you could just get over your irrational fear of angle brackets and use xml.

this is funny because im using gigabyte+ protobufs in a project

# ? Sep 24, 2013 22:06

double sulk: Jul 2, 2010

thanks tim for installing a default postgres or some poo poo making it a pita to work with until you add localhost and install postgres.app

# ? Sep 24, 2013 22:15

Malcolm XML: Aug 8, 2009; I always knew it would end like ｔｈｉｓ．

gucci void main posted:

thanks tim for installing a default postgres or some poo poo making it a pita to work with until you add localhost and install postgres.app

brew postgres works4me

# ? Sep 24, 2013 22:19

double sulk: Jul 2, 2010

Malcolm XML posted:

brew postgres works4me

dunno, something got hosed and it wasn't tracking the psql until i installed postgres.app. apparently it's not an uncommon issue? idk

# ? Sep 24, 2013 22:21

Malcolm XML: Aug 8, 2009; I always knew it would end like ｔｈｉｓ．

gucci void main posted:

dunno, something got hosed and it wasn't tracking the psql until i installed postgres.app. apparently it's not an uncommon issue? idk

did it move to brazil without telling you in advance

# ? Sep 24, 2013 22:22

Brain Candy: May 18, 2006

Bloody posted:

this is funny because im using gigabyte+ protobufs in a project

clearly it's impossible to use protobuffs by themselves. i'm guessing you encode each message into the filesystem like /foo/v4/bar/fartz/seg192131 that way you have all the coherent data under /foo/v4/bar.

# ? Sep 24, 2013 22:23

Soricidus: Oct 21, 2010; freedom-hating statist shill

Bloody posted:

this is funny

holy poo poo

# ? Sep 24, 2013 22:34

Janitor Prime: Jan 22, 2004; PC LOAD LETTER

What da fuck does that mean; Fun Shoe

Bloody posted:

this is funny because im using gigabyte+ protobufs in a project

God drat, I use megabyte+ regularly but Jesus Christ a gigabyte is just :eyepop:

Also I don't know what the big deal is about not being able to send multiple protos back to back, you can just nest them inside of a repeated field of another message and problem solved.

Soricidus posted:

alternatively you could just get over your irrational fear of angle brackets and use xml.

It's not a fear, the performance and overhead are too much for our apps. Yes we do send a lot of small < 50k messages over the internet and it fit our needs way better than the stupid hand rolled binary format they were doing before.

# ? Sep 24, 2013 23:10

MrMoo: Sep 14, 2000

Soricidus posted:

"If you want to write multiple messages to a single file or stream, it is up to you to keep track of where one message ends and the next begins. The Protocol Buffer wire format is not self-delimiting, so protocol buffer parsers cannot determine where a message ends on their own."

The C# API does though and you have to write your own code to match it in other languages.

Managed a 6GB compressed protobuf today :woop:

# ? Sep 25, 2013 00:59

Adbot: ADBOT LOVES YOU

# ? Jun 11, 2024 02:50

Nomnom Cookie: Aug 30, 2009

Non-standard extensions, what a great idea.

# ? Sep 25, 2013 01:08

The Something Awful Forums > Discussion > Serious Hardware/Software Crap > YOSPOS > p-lang thread: (now (have you (problems two)))

«‹›1784 »