egg is now known as egg|zzz|egg
<RockyTV> meh my raspi2 just stopped working
aradapilot has joined #spacedock
<RockyTV> ;tell Thomas why not use BSON for files, so we decrease the data size?
<RockyTV> !tell Thomas why not use BSON for files, so we decrease the data size?
<Qboid> RockyTV: I'll redirect this as soon as they are around.
<RockyTV> or just some compressed jon?
<VITAS> poor jon
<VITAS> RockyTV, webservers use zip to compress http traffic
<RockyTV> we can't zip the json unfortunately
<VITAS> (i think its bidirectional compression)
<VITAS> no its transparent
<VITAS> look up apache compression or mod zip
<VITAS> all you can do to minimize traffic is a) to minimize connection count and b) minimize traffic
<VITAS> pretty much standart on all installs (i do)
<VITAS> but it seems that its only for server-> client
<VITAS> so you might have a valid point if its client -> server
<VITAS> but always think about slow devices like phones and old/embeded computers
politas has quit [Ping timeout: 201 seconds]
politas has joined #spacedock
aradapilot has quit [Ping timeout: 201 seconds]
aradapilot has joined #spacedock
politas has quit [Ping timeout: 200 seconds]
politas has joined #spacedock
<Thomas> !tell RockyTV Sure, we could use BSON or message pack for the requests
<Qboid> Thomas: I'll redirect this as soon as they are around.
<Qboid> Thomas: RockyTV left a message for you in #spacedock [05.08.2017 06:38:48]: "why not use BSON for files, so we decrease the data size?"
<VITAS> !tell VITAS Hello sexy!
<Qboid> VITAS: I'll redirect this as soon as they are around.
<Qboid> VITAS: VITAS left a message for you in #spacedock [05.08.2017 11:16:25]: "Hello sexy!"
<VITAS> :D
<Thomas> hi
<VITAS> !tell Thomas Hi
<Qboid> VITAS: I'll redirect this as soon as they are around.
<Thomas> ¯\_(ツ)_/¯
<VITAS> I wonder if there is a ASON
<VITAS> anyways what data is so large it needs compression?
<VITAS> images and mdos are allready compressed
<Thomas> Well, my concern is that we end up making lots of smaller requests
<Thomas> which would result in lots of data being transmitted
<VITAS> if i wouldnt fear websockets to be blocked in places....
<VITAS> but you could make some sort of broker that at least aggreagates all req at the init load of a page/view and sends them in bluk to the backend and distributs the response
<Thomas> This seems promising http://api.jquery.com/jQuery.when/
<VITAS> the global question is always: how to keep the user entertained
<VITAS> in this case entertained while youre laoding new stuff
<Thomas> minigames on the loading screen? :P
<VITAS> or content that is important /at the top
<VITAS> in other words: its no problem laoding stuff later that a user only gets to at an later stage
<VITAS> e.g. troughs crolling etc.
<SpaceDock> [OpenDock] StollD pushed 1 new commit to master: https://git.io/v7wdm
<SpaceDock> OpenDock/master 9917edb Dorian Stoll: Preparing the request Methods for parallel requests.
Lartza has quit [*.net *.split]
Snoozee has quit [*.net *.split]
NBones has quit [*.net *.split]
DuoDex has quit [*.net *.split]
DuoDex has joined #spacedock
NBones has joined #spacedock
Lartza has joined #spacedock
Snoozee has joined #spacedock
Snoozee is now known as Majiir
<SpaceDock> [OpenDock] StollD pushed 1 new commit to master: https://git.io/v7wNt
<SpaceDock> OpenDock/master 80eced7 Dorian Stoll: Parallel requests
Azander has quit [Quit: I'm gone!]
dpbk has joined #spacedock
dpbk has left #spacedock [#spacedock]
* Thomas feels like he is rewriting the whole ajax part of jquery because nothing does what he wants :P
<VITAS> normal
<VITAS> the cycle of software dev
<VITAS> you have a plan, you implement it, you learn while duing it, you find the flaws in your plan, you have a new plan
<Thomas> Why does it have to fail completely when the status code says "error" :/
<Thomas> That breaks the parallel execution
<VITAS> because it isnt a warning
<Thomas> Yes, but when someone wants to read the message sent?
<VITAS> huh?
<VITAS> then save it before sending?
<Thomas> No, when the server sends a message with status code 403, ajax fails. But I want the message, and I dont care about the error code
<VITAS> in that case log it on the server?
<VITAS> or do you want to use it in your code? then implement your own error handling
<Thomas> "then implement your own error handling" --> /me feels like he is rewriting the whole ajax part of jquery because nothing does what he wants :P
<Thomas> normal
<Thomas> Wut
<Thomas> I didnt want to copy the normal :P
<VITAS> :)
<Thomas> Anyway: I had my own handling, but now it isnt used anymore because of $.when
<VITAS> what i usualy do is use the ajax error handeling for hard errors and my own more fine for state handeling and feedback
<VITAS> you usualy want some descriptive message to go with your errors so what i would do is wrap your messages in another layer of json
<VITAS> something like {status:3,errorcode:8932,msg:"bla",data:{your data}}
<Thomas> sdb already has those messages. My only problem is that ajax doesn't ignore the error codes.
<Thomas> I think I am simply going to remove all the 4xx codes from sdb
<VITAS> i think theres something like onerror functions
<VITAS> yes that would make the rrors "softer"
<Thomas> Yes, but the problem with the on error functions is: They dont return data. The problem is, that with the parallel execution of the request functions, every function is expected to return something that gets passed to a single big callback
<VITAS> ok you could use a more lowlevel jquery ajax function
<VITAS> or do as you said: remove http errors and rely on your own
<VITAS> there still can be 404 and alike if its realy the problem
<VITAS> (aka 4040 if the route doesnt exist)
<VITAS> -0
<Thomas> Hmm, I could just copy the source of the ajax function and replace the check :D
<VITAS> you can write new jquery functions using existing lower level jquery funcrtions
<VITAS> so no need to copy them
<VITAS> you can simply extend or overload them
<VITAS> (at least i remember doing stuff liek that at some point)
<Thomas> The problem is: how do I overload a local variable inside of a 200 lines function :D
<VITAS> noo
<VITAS> you can do $.fn or something similar
<VITAS> it enables you to do $.thomasGet or even $.get
<VITAS> using the normal get foo but with your own twist
<VITAS> and the same for other functions
<VITAS> look up custom jquery functions
<VITAS> or somethign like that
<Thomas> I know that I can define own jquery functions
<Thomas> My problem was that I would need to rewrite the ajax function completely
<Thomas> But I think I found a solution
<VITAS> ok...
<VITAS> i still think you dont entirly
<Thomas> Hah! It works \o/
<SpaceDock> [SpaceDock-Backend] StollD pushed 1 new commit to master: https://git.io/v7rT4
<SpaceDock> SpaceDock-Backend/master 63d71a6 Dorian Stoll: Send the users confirmation in the admin data
<Qboid> Starting build #76 for job SpaceDock-Backend (previous build: SUCCESS)
<SpaceDock> [OpenDock] StollD pushed 1 new commit to master: https://git.io/v7rTB
<SpaceDock> OpenDock/master 1a774db Dorian Stoll: More fake jQuery
<Qboid> Project SpaceDock-Backend build #76: SUCCESS in 2 min 13 sec: https://jenkins.tmsp.io/job/SpaceDock-Backend/76/
<Qboid> dorian: Send the users confirmation in the admin data
egg|zzz|egg is now known as egg|nomz|egg
<RockyTV> o/
<Qboid> RockyTV: Thomas left a message for you in #spacedock [05.08.2017 11:10:26]: "Sure, we could use BSON or message pack for the requests"
<Thomas> \o
egg|nomz|egg is now known as egg
<Thomas> Note to myself: split sdb.js into multiple files
<Thomas> This is getting ridiculous xD
<VITAS> !tell Thomas split sdb.js into multiple files
<Qboid> VITAS: I'll redirect this as soon as they are around.
<SpaceDock> [OpenDock] StollD pushed 1 new commit to master: https://git.io/v7rc4
<SpaceDock> OpenDock/master 75bcd13 Dorian Stoll: Functionality for user profile view
SilverFoxy has joined #spacedock
SilverFox has quit [Ping timeout: 383 seconds]
SilverFoxy is now known as SilverFox
SilverFox has quit [Ping timeout: 204 seconds]
politas has quit [Ping timeout: 204 seconds]
politas has joined #spacedock
SilverFox has joined #spacedock