Re: Osproc input stream hang
A hang in OS i/o can relate to buffering. You'd need to post a working (hanging) example.
Re: What prevents you from using Nim as your main programming language?
So, I already attempted to make a reply to tell you guys why I'm not using Nim as my programming language, even though, I really would like too. I'm unable to describe the problem I have with Nim, but I don't have to, because somebody else already did and I'm so glad that I'm not alone with this thought and mindset. I had planed writing a massive orchestration software and I'd love to write it in Nim. But as a maintainer of such a large-scale codebase I imagine it to be a nightmore doing code reviews on GitHub, not knowing what is coming from where. You could argue with "Use qualified imports blah blah blah" but who really does that. And it doesn't cooperate well with the general Nim coder with this "freedom mindset". In my point of view Nim is great for prototyping, CLI tools, "scripting" and maybe even minor-ish web applications. But you just can't do large-scale stuff with it and I'm not gonna believe anything else if not proven otherwise. (And please don't come up with Status' Ethereum Client "Nimbus". Have a look first.) It's actually pretty sad. I'd really like to use Nim for something large-scale but Nim, in its current state, feels more like a scripting language with a "AoT facade".
Re: Using async Nim procedures in Python
This is a cool idea and doing this might actually not be too difficult. The main pieces to get this working are likely: * Calling Nim's async event loop from Python's periodically * Interoperability between Nim's futures and Python's (not strictly necessary, but would allow awaiting Nim's async procs in Python which would be pretty nice) I know that it is possible to ask epoll to let you know when another epoll FD is ready to be polled. Python should allow you to add arbitrary FDs to its event loop in order to receive readable/writeable events on it, you can then grab Nim's asyncdispatcher's FD and ask Python to tell when it should be polled. Then simply poll it. I take this approach in my HTTP server: [https://github.com/dom96/httpbeast/blob/master/src/httpbeast.nim#L168-L171](https://github.com/dom96/httpbeast/blob/master/src/httpbeast.nim#L168-L171). Happy to help further if you're interested in pursuing this.
Re: question on autolayout in wNim
Your rule 2 and rule 4 have contradictions. Rule 2: |-50-[text]| Rule 4: |[text]|
Using async Nim procedures in Python
Hello, I have been looking on Nim for couple of months now, and I have to say that I'm very intrigued. I'm currently working for a company that is mainly using python. Our core application is almost entirely written in it (some small parts are written in cython). It is quite large (close to 100k lines of python). Lately we have been having some difficulties with performance (we usually resort to horizontal scaling), but also with mypy (static typing). We are using asyncio heavily, basically our entire application is asynchronus. We are utilizing static type annotations (mypy) extensively so our code sometimes comes close to Nim syntax (if you ignore usage of class hierarchies). There are some parts that could be rewritten easily, but I'm afraid that we would inevitably run into limitations of missing support for async. I'm aware of [nimpy](https://github.com/yglukhov/nimpy) and I was wondering if there is any way to write asynchronous Nim code that would be await-able in python. I'm guessing that general solution that would automagically allow usage of Nim async procedures to be run in python event loop would be to much to ask. But I was thinking that there might be something that can be implemented for specific use-cases. Do you have any idea what it would entail? I was searching online and I think I found something analogues for rust lang, but I cannot find it now. The similarity to python was the thing that brought me to Nim in the first place. So I would really like some more effort put towards interoperability with python. I think that many company's that are using python for something other than Django applications, would appreciated it. I think that there are a lot of potential Nim customers.
Re: Help to create a template/macro that reads small chunks of data from a stream.
Thanks @Hlaaftana but, unfortunately it doesn't work! /home/hdias/dev/tmp/test.nim(32, 25) template/generic instantiation of `async` from here /home/hdias/dev/test.nim(43, 10) template/generic instantiation of `forStream` from here /home/hdias/dev/tmp/test.nim(26, 24) template/generic instantiation of `await` from here /usr/local/programs/x86_64/nim-1.0.6/lib/pure/asyncmacro.nim(448, 10) Error: Await only available within .async Run I don't know if 'async' and the 'await' works inside templates!
Re: question on autolayout in wNim
Most probably a typo error in your layout proc()
proc layout() =
panel.autolayout """
H:|{groupButton:[btnClick(btnClose)]-[btnClose]}|
H:|{groupText:[-50-[text]]}|
V:|[label]-[groupText]-[groupButton]|
H:|[label, groupText]|
Run
This template works, at least no error after compilation. By the way not sure
it is the layout you want.
Is this a bug? result variable vs last statement
Hello everyone!
I'm not sure if this is a bug, so I'm writing this post to avoid creating an
unnecessary github issue.
In the following proc the result variable is being assigned to, but then there
is a last statement that would act as the return value. This causes an error as
expected:
proc foo(name: string): string =
result = "Hello " & name
"this is an error"
echo foo("test")
Run
Output:
> Error: expression '"this is an error"' is of type 'string' and has to be
> discarded
But when you change the assignment to .add, the last statement gets returned
instead of causing an error:
proc foo(name: string): string =
result.add("Hello " & name)
"this should be an error"
echo foo("test")
Run
Output:
> this should be an error
$ nim --version
Nim Compiler Version 1.0.6 [MacOSX: amd64]
Compiled at 2020-01-29
Copyright (c) 2006-2019 by Andreas Rumpf
active boot switches: -d:release -d:useLinenoise
Run
Re: Generate XML file with xmltree [nesting format issue]
I have attempted to fix this issue in this PR: [https://github.com/nim-lang/Nim/pull/13482/files](https://github.com/nim-lang/Nim/pull/13482/files) XML/xmltree experts: Please review that PR and help me make it better!
ANY TWEAKERS IN THE HOUSE, GOT THAT FIRE CARTEL METH ON DECK DM ME
Are you just new into the dark web and wondering where and how to start going about buying drug stashes, then here’s the solution for you. Get your VPN connected and contact me for more details. Telegram: Drugporn01 WhatsApp: +1 (408) 412-0793 Wickr: Drugporn01 Email: drugp...@protonmail.com I noticed many people don’t know how to use the Dark Web so I took it upon me and asides being a vendor in the various markets, (you can check me out in the following ) @Cryptonia -drugporn01 @Apollon \- drugporn01 @Grey Market drugporn01- (new) @Avior \- drugporn01 (new) @Empire\- drugporn01 I decided to go to forums and other onion sites which are easily accessible to put up an ad to make buying on the dark web easy . Not everyone can use the markets and all the processes so here’s the real deal for you guys. Are you looking to buy-Good quality crystal methamphetamine,| ---|--- -Sealed pints of Codeine cough syrup| -Pills, pops, tablets (ask for those available)| -Documentation services| -Getting a job done, then you’re at the right spot.| CONTACT ME THROUGH Telegram: Drugporn01 WhatsApp: +1 (408) 412-0793 Wickr: Drugporn01 Email: drugp...@protonmail.com
ROXICODONE/OXYCODONE- 30MG TOP GRADE 500 PILLS AVAILABLE. Telegram: Drugporn01
> If you lost your regular drug connect, he got killed, locked or jailed, went > missing or travelled and youre tired of waiting because you need your shit, > then i believe you need an honest, loyal, fast and trustworthy person to keep > supplying them to you no matter your location. Dont get scammed by these > scumbags online, Real should recognize real. So here's the deal If you need some pretty bloody good cold ice packs for your esky (crystal meth not leaving out anything white), then get in touch, ONLY MEANS OF CONTACT Telegram: Drugporn01 WhatsApp: +1 (408) 412-0793 Wickr: Drugporn01 Email: lemybea...@protonmail.com Will sometime need a bit of notice depending on request. Karma is coming to all of us one day. You get what you give. Straight up no bs, smooth and effective from both me and you. The following can change, but is an example for the dusty and icey. 1/4 oz =$250 So don't come telling me ma price is too good to be true and vice versa cause i won't take that. This is no organization but a personal and recognized business- cartel on his own. The amount depends on how it rates and how much demand there is for a specific batch. E.g -good batch= Inflated rates on| ---|--- -average batch= discounted rates.| ONLY Crypto payments accepted (bitcoin is an example) Those in LA can link up for the goodies. Happy to negotiate, PICK UP and DELIVERY. Happy to travel within season. (Meth head, others call it tweaker) Get in touch if you wanna get your toke on! TAKE NOTE PLEASE A lot have been hitting me up to buy stuff. If you're hitting me up, be serious, don't waste my time. Like think to yourself for a brief moment that YOU'RE HITTING ME UP, not the other way around. How do you honestly think it will work? I'll give you my location and we can meet. UhNO. If you want something shipped, OBVIOUSLY payment is going to be needed first. I swear there are some mfs that can't use their heads. If you are going to a random stranger and asking for product, then you should have thought of everything including risk. I'm not trying to fuck anyone over. As ALL you can see I DON'T NEED TO based on my weekly pictures. I'M GOOD. If you think I got all these Pills, powder, and much more by fucking people over, then you trippin' Don't hit me up to tell me your previous stories with the play boys playing with your dollars or telling me how you been scammed before, like "motherfucker you were just so fucking dumb not to recognize real from a fake" If you serious let's do business, if not then keep moving along bro. F T P Stay wise Stay low Hit me up NB: Once you receive product, If not then keep moving
Re: Osproc input stream hang
Since the process streams has not implemented peak functionallity, it seems the only way to do this is with threads.
Re: question on autolayout in wNim
I suppose you should ask on wNim's github issue tracker.
Re: Capture Problem
I realised you don't actually need toSeq here import random, sugar, math proc foo[T, U](a: openArray[T], cdf: openArray[U]): auto = let (a, cdf) = (@a, @cdf) (() => random.sample(a, cdf.cumsummed)) echo foo([1,2], [1,2])() Run
Re: Generate XML file with xmltree [nesting format issue]
It's probably because you are using `newXmlTree` for both **channel** and **rss** tags. May be you should use `newElement` for **channel** too? _PS: I am just guessing; I haven 't used this library._
Re: How to use integer generics in types?
There's no easy way to provide default values for such integer generic parameters, is there? E.g. (this doesn't work): type MatchPool[N: static[int] = 3] = object Run
Re: Capture Problem
Thanks a lot!!!
Re: Capture Problem
import random, sugar, math from sequtils import toSeq proc test1[I, T, U](a: array[I, T], cdf: array[I, U]): auto = (() => random.sample(a, cdf.cumsummed)) proc test2[T, U](a: openArray[T], cdf: openArray[U]): auto = (() => random.sample(a.toSeq, cdf.toSeq.cumsummed)) assert compiles test1([1,2], [1,2]) assert compiles test2([1,2], [1,2]) Run .
Capture Problem
The following code import random, sugar from math import cumsummed type Generator[T] = () -> T TrafficLight = enum red, yellow, green proc genCDF[T, U](a: openArray[T], cdf: openArray[U]): Generator[T] = assert(a.len == cdf.len) result = proc(): T = random.sample(a, cdf.cumsummed) proc sample[T](gen: Generator[T], len: Natural = 10): seq[T] = result = newSeq[T](len) for i in 0.. which cannot be captured as it would violate memory safety Run Is there any chance to solve this error?
Re: Modify old code using new/finalizer to compile with gc:arc
`defined(gcDestructors)` is fine, but undocumented.
Re: How to use integer generics in types?
To straight up give an example it's like this: type MatchPool[N: static[int]] = object vals: array[N, int] var mp = MatchPool[3](vals: [2, 4, 1]) Run
Re: Generate XML file with xmltree [nesting format issue]
Might be a minor bug, but does it matter? XML is not whitespace sensitive.
Re: How to use integer generics in types?
You use static[int], as answered in my thread. If N is not a compiletime constant, you need to use seq or unchecked_array.
