PHP - 2016-11-22 (page 3 of 3)

« first day (2229 days earlier) ← previous day

3:00 PM

how apache manages to untangle that mess is beyond me

can php can upload 500mb size of video ?

@jemz Slowly and carefully.

@Gordon "It doesn't" is the correct answer.

284

Q: PHP change the maximum upload file size

I have a website hosted on a PC I have no access to. I have an upload form allowing people to upload mp3 files up to 30MB big. My server side script is done in PHP Every time I try and upload a file, I receive an error claiming that the file exceeds the maximum size allowed, so I need to increas...

@Gordon but I read some comments in other thread that changing upload_max_filesize is not enought, it should use chunk ? but I have no idea how

31

Q: Upload 1GB files using chunking in PHP

I have a web application that accepts file uploads of up to 4 MB. The server side script is PHP and web server is NGINX. Many users have requested to increase this limit drastically to allow upload of video etc. However there seems to be no easy solution for this problem with PHP. First, on the ...

php upload large-files

3:04 PM

@Gordon plupload ?

@jemz gesundheit

@Gordon I can't get you

@jemz blobs can be pretty big, but thats sql. Make a table with a longblob, you get max 4G file size, must be the old 32 bit limit.

aright y'all this has been educational, have a good day/night/whatever-time-of-day-by-you

I am heading home now, too. cya

3:06 PM

later @BIU

github-status minor: We're investigating an increase in 404 responses when serving GitHub Pages as of 2016-11-22T15:07:05Z

I really like that thing @JayIsTooCommon

do you have some reference using chunk upload of video ?

Did @Jay make that? Awesome :D

!!version

v1.3.0-469-gf04ee5d

3:09 PM

yeah the kid has its moments

What keeps it checking for status changes?

Nvm, found it

Triggers a job on enabling for a room.

thoughts? github.com/amphp/aerys/issues/109

Hi guyz can I ask please how does the php upload the video that this will send chunk by chunk to server ?

@Patrick github.com/amphp/aerys/blob/master/lib/…

Also tight coupling

3:17 PM

@bwoebi that's what I mean, side effect

@Patrick I mean the exception

@bwoebi yeah but that's just a band-aid on the problem. It should not even be possible with the change that I suggested

31

Q: Upload 1GB files using chunking in PHP

I have a web application that accepts file uploads of up to 4 MB. The server side script is PHP and web server is NGINX. Many users have requested to increase this limit drastically to allow upload of video etc. However there seems to be no easy solution for this problem with PHP. First, on the ...

php upload large-files

@PeeHaa thanks ...

np

3:21 PM

docs.google.com/spreadsheets/d/…

can you see it ?

Not without login

@PeeHaa <3

@JoeWatkins no with login

@JoeWatkins need to request access

> You need permission

Want in? Ask for access, or switch to an account with permission. Learn more

You are signed in as [email protected].

@JoeWatkins

docs.google.com/spreadsheets/d/…

that's the same link, refresh ?

3:23 PM

aye see it

$resu = mysql_query($query, $conn);

$fila = mysql_fetch_row($resu);

@Patrick replied in the issue.

yes @JoeWatkins

mysql_fetch_row(): supplied argument is not a valid MySQL result resource in // line 2

Any ideas ?

That's odd. I would expect a Fatal error: Uncaught Error: Call to undefined function mysql_fetch_row() in ...

3:25 PM

mysql_connect has been called previously

Yeah was just trolling (ish)

Gfu.

!!urban Gfu

[ gfu ] 1.Go fuck yourself 2.Good for you

lol. Not sure what I should pick :P

3:26 PM

@Patrick this issue is that, while each individual request+response pair is stateless to the outside, the HTTP response itself is actually stateful. It has state: expecting headers, state: expecting body data, with order mattering.

my current thinking is that we need a combination of a higher (than 2/3) bar, and a rethink about who can vote ... but when you look at this, it looks like some pretty nice stuff wouldn't have got into php7, aswell as some stuff that was really required to fix endless sec issues with unserialize

if anything is the kind of change that requires a majority, "64 bit platform improvements for string length and integer in zval" that should, but would also fail with a higher bar ...

I don't know how to interpret this data

@JoeWatkins What would you change about voting eligibility?

I don't know yet, I don't know what is effecting outcomes

@JoeWatkins tHAT WAS EXACTLY WHAT i WAS AFRAID OF

Sorry caps ^

I don't necessarily think the votes are being influenced by people who should not be voting, but what we should do is make clear these rules so that when a contentious RFC comes along, we don't spend forever arguing about who can vote on it ...

3:30 PM

@JoeWatkins Why did I vote against that? o_O

so did I

Ah yeah

Because of things like php_int_t, php_size_t

which were ultimately changed post-RFC

It was going to happen one day... I have a question, and when I find what I want, it turns out I wrote the answer on SO

iirc it was accompanied by a bunch of stupid changes that were since reverted

@Leigh lol

3:32 PM

@bwoebi and Z_INT_P or something like that ... if I remember ...

@JoeWatkins yep … I think if that RFC failed, it probably would have been attempted again afterwards, with the changes reverted and passed then

that's true, it needed more work

how are we going to rescue stuff like scalars though ?

Common sense still is prevailing when we merge things.

@JoeWatkins hmm?

would fail with a higher bar

@JoeWatkins Yeah, don't try introducing a higher bar for language changes

3:34 PM

but nothing to do with defects in the patch, it was just a contentious idea ...

also, can't believe we don't have a builtin for longest common prefix

@bwoebi chat crashed.. answered in the issue

@bwoebi but I think it will do some good

a lot of good, actually ...

right now there is a tendency to just announce, and keep pushing forward with your idea, in the hope that you can convince enough people to vote in the affirmative ...

that's wrong, when you experience resistance, of any kind, you should act, not argue the case, if there are changes to be made to reduce resistance, you should do it rather than spend time on internals arguing your case ...

@Patrick Can you please elaborate why exactly this is an issue for you to use Aerys at all? … Anyway, you ultimately could write your own response object with a method being passed the Aerys\Response object which you call where you'd return instead. I.e. instead of return $myResponse; you'd write $myResponse->send($AerysResponse);

@JoeWatkins There are some instances where there's nothing to act on. If I do a, the one half complains; if I do b, the other half complains

usually it is possible to come up with C ...

even if it isn't, that's exactly the sort of thing we shouldn't let in, isn't it ?

3:41 PM

@JoeWatkins Example, short closures. By-ref or by-val? What's option C here?

@JoeWatkins That scalar types vote is an excellent example that we should rather rethink who can vote.

@bwoebi Because the API doesn't make sense to me. request in - response out is what I want. If I have to worry about side effects in my request handlers (which is not some one off code, it's code that will be touched often), that's a deal breaker for me

@bwoebi yeah sometimes there isn't one ...

@Patrick In 99% of cases, you'll only ever call $response->end() as the very last statement of your request handler

What does exactly mean Withdraw as RFC status?

the author gave up

3:43 PM

@Patrick Ultimately, when you return a Response object, you trigger the implicit side effect of replying to the client with the returned response.

I was looking at nested classes @JoeWatkins RFC this was yours. Do you remember why you gave up?

@bwoebi it's difficult to know how to define who can vote ... without researching people that I don't think should be able to vote ...

In that case you may even want to argue that explicitly triggering the side effect is more cleanly @Patrick

I think it might be great feature

@JoeWatkins Instead research who you think should be able to vote

3:45 PM

@JoeWatkins At some point it was discussed to be looking at karma. I.e. exclude docs people iirc

I disagree. If I return some other stuff might happen too. Caching for example. I don't care what happens inside my handler, that responsibility is moved outside.
How would you implement caching with end()?

but that's extremely unfair, and probably harmful, doc maintainers can be the most knowledgeable people ...

@Patrick have a middleware read the data to be streamed.

@JoeWatkins What about active doc maintainers?

Evenings

@JoeWatkins Yes not saying it's a good idea. Just venting I heard something like that a while back

3:47 PM

@brzuchal yes, because there was no consensus about whether we should have module, package, namespace level, or nested classes ...

@JoeWatkins We have a lot of people which are never showing up, just sometimes quietly voting.

It's hard to come up with some other line to cut off people I think

Would this be the best or applicable place to inquiry about pdo?

yes

evenings

3:49 PM

evening westminster

@JoeWatkins Also, we should separate technical and not purely technical votes … I.e. give core committers some more possibility to veto a RFC based on implementation. And hold a general vote on the proposed semantics.

@JoeWatkins so it is possible to pick it up again and discuss find some consensus and improve RFC

evening peeholee

@brzuchal sure, run with it ... I don't think there's a working patch any more ...

boy. that's escalated quickly

3:50 PM

lol

sorry, couldn't think of something softer

:P

PeeHole, that's gonna stick :D

@bwoebi inb4 stas vetos everything

Oh god wes. What have you done

@PeeHaa based on the implementation

I.e. justified implementational veto

@PeeHaa he always has something to say that is worth listening too, even if it's always negative ... you cannot objectively call him wrong ...

3:52 PM

@bwoebi I could live with that if it's indeed used like that

given a definition of active, it seems superficially reasonable to say that only active contributors should be allowed to vote ...

@JoeWatkins I wouldn not say "always" though

Hello, Im a noob thats highly frustrated with tutorials showing plain sql which is nothing less than useless unless your only on phpmyadmin. Does anyone know a good pdo tutorial for a noob. Im in a position where the material "your common sense" provided is not quite clicking in my head, but all general php and sql functions i feel i have a sturdy grasp on

@Patrick You'd just have another handler before your actual handler. If that handler does nothing, the actual handler will be called.

we can't just pick on doc maintainers ... can we ?

and there's not enough core maintainers for (imaginary) quorum ...

3:55 PM

@JoeWatkins With equal weight or dependent weight?

but wait a minute, there's a big hole in that idea ... what happens when bob has to leave us for a couple of years for real life stuff, or niki, it's going to happen ... are we then going to say they don't have the right to vote on a thing they helped create, and probably still know more about than most of the people voting at the time ?

@Andymechanical Have you tried phpdelusions.net/pdo?

@JoeWatkins Bob shouldn't be allowed to vote anyway

3

or maybe we can reason thus: the reason a doc maintainer is allowed to vote at all is because they have special knowledge of the current state of php, precisely because they documented it ... when you become inactive you loose that special knowledge ?

@PeeHaa hehe

The results in the next follow up question: what is active?

3:59 PM

we don't have that definition yet

Also what about community "leaders" or whatever monkeys we let vote?

That is a thing right?

@PeeHaa seldaek probably for example.

@JoeWatkins your knowledge is not up to date though

@kelunik Yeah wasn't sure who was the people tbh

@JoeWatkins It is pretty unlikely that you have no time at all, not even for a small bugfix or such

4:03 PM

@MadaraUchiha Yes i have been there i cant quite comprehend it however. its slightly over my head.

@bwoebi :o

@JoeWatkins I mean if you were two years away

oh right I see, yeah

@Andymechanical PDO is a way for PHP to abstract database access on a low level

@bwoebi you haven't had children yet ... that could happen ...

4:04 PM

Unlike mysqli which is an abstraction only over MySQL, PDO is an abstraction over a large array of databases.

@JoeWatkins Well, in that case you probably won't have time to vote much either.

that's also true

When you're back and actively contributing again, I see no reason why you shouldn't be allowed to vote either.

I don't want to make rules that any of us are going to curse in 5 years time as short sighted or silly

also, I'd be quite annoyed if I was stopped from voting just because I was dealing with real life, even if I had no time to vote, I should have the option ...

@JoeWatkins That's true … I'm just giving ideas, needs dueful debate too.

4:06 PM

@MadaraUchiha can you define abstraction for this application please? I took from PDO that it out quotes input into unsyntaxed strings and sends 2 types of data the actual function then the values?

@JoeWatkins … you are dealing with real life for a year, okay … but for multiple consecutive years? muh.

@Andymechanical An abstraction is a way to describe something in a programming language, while ignoring the small implementation details.

you could be sick, or incarcerated ...

2 mins ago, by bwoebi

@JoeWatkins Well, in that case you probably won't have time to vote much either.

In this case, it means that you have the same, similar syntax and calls, while not caring about how queries get executed or even which database is being connected to.

4:08 PM

@MadaraUchiha ok that puts some immediate understanding on that part forthwith thank you for that.

no one put it like that before

@Andymechanical Here's a simple example for abstraction using object oriented concepts:

yeah it's true that you wouldn't have time, but should that be enough to say that if you made time, broke out of jail, or heaved yourself out of bed to vote, that we should actually say you cannot vote ...

function whatever(Logger $logger) {
  $logger->log("Hello!");
}

whatever doesn't care how logging is done. It just asks the Logger to log a message. whatever can ignore the implementation details completely.

Logger will deal with them, internally.

@JoeWatkins In that case you'll also be easily able to resume working on PHP and rightfully have the privilege to vote.

it's silly to talk about people in jail, I know that, I'm just trying to think of what might happen, and how we might get called stupid 5 years hence for not seeing these things ...

let's have some suggestions for what we think should qualify someone as active, then we can check who meets those criteria today ?

I have no idea how to define that ...

without pulling numbers out of the air anyway ...

4:12 PM

@JoeWatkins that's the hard problem… I'd say at least a significant contribution or multiple smaller ones in the last two years … but "multiple" and "significant" are wishywashy words in that context :-/

BTW semi related:

Apr 1 '15 at 14:02, by rdlowrey

I wouldn't mind voting for a three-person BDFL triumvirate for each year. Or maybe two years.

if we leave core voters to the side, and just concentrate on what makes a doc maintainer active, is that easier for anyone ?

@PeeHaa when the leadership is bound to change every X years, I think you create instability ...

it's not easier for me .. I totally agree with the sentiment, significant or amounting to significant, but I've no clue what that means in terms of loc or commits, or if that's even a reasonable metric ...

@JoeWatkins I would not call myself an active docs maintainer. Can you see somehow how many times I've done something this year?

oh wait. svn probably

we're putting ourselves in an odd position, because we should prefer less commits

@JoeWatkins a) needs to show up sometimes on-list and b) make significant contributions?

4:17 PM

sometimes and significant

@JoeWatkins commits is a bad quantification … measure contribution.

@bwoebi yeah I was just realising that ...

@bwoebi That's hard I think

I think I'll rewrite my Go app in PHP.

@MadaraUchiha Im going to digest what you've provided carefully for now so your efforts are not wasted, thank you so much. One last thing should i have a better understanding of object oriented php code before working with pdo "opinion of you"?

4:18 PM

@PeeHaa things which actually add value and do not just format and typo and create stubs etc.

Yes again. That is REALLY hard to measure

@Andymechanical Yes

At least somewhat.

And you would need to maintain the list manually

maybe we're looking for a way to define activity, rather than contribution ... we should not exclude those people who spend long time on bugs resolving doc bugs but not actually needing to make changes, or making few changes

I know it helped me figure it out.

4:19 PM

@bwoebi Even fixing a lot of typos helps.

@MadaraUchiha i wish i could upvote you for your support, i know its frustrating dealing with people that know nothing.

but then it still suffers from the same thing, someone might be very active because they wrote bad documentation and are required to keep changing it ...

@kelunik A lot of typos, yes.

is this an already solved problem for any other large project ?

@JoeWatkins good question

But many languages are not that openly self-governed like PHP…

Some languages have benevolent dictators, others are funded by a company influencing it heavily, others are designed by a committee…

4:24 PM

yeah it's difficult to find something with a comparable model ... but maybe it wouldn't have to be programming related, it could be anything, we are just looking for a way to determine the activeness of a member of a group ... maybe from a computer game or something in that realm ...

« first day (2229 days earlier) ← previous day

Transcript for

Nov22

PHP

Support group for those afflicted with PHP. Don't ask to ask, ...

join this room
about this room

00:00

06:00

12:00

18:00

all times are UTC

site design / logo © 2016 Stack Exchange Inc; legal

mobile