<owen>
Shinmera: does alloy work on Mac? the glfw examples just hang, and if I switch to framebuffer it says it failed to configure Cocoa
attila_lendvai_ has quit [Ping timeout: 276 seconds]
puke has quit [Quit: WeeChat 3.8]
puke has joined #commonlisp
<owen>
maybe I can make a `define-example' macro for opengl and try that
NotThatRPG has quit [Ping timeout: 260 seconds]
NotThatRPG has joined #commonlisp
drewjose has quit [Quit: Ping timeout (120 seconds)]
drewjose has joined #commonlisp
shawnw_ has quit [Ping timeout: 255 seconds]
elderK has quit [Quit: WeeChat 4.4.2]
szkl has quit [Quit: Connection closed for inactivity]
<owen>
they work for me on linux, so ill just use that for now to try alloy
random-nick has quit [Ping timeout: 252 seconds]
paddymahoney has quit [Ping timeout: 264 seconds]
`Pixel_Outlaw has joined #commonlisp
<paulapatience>
Are *print-level* and *print-length* honored when *print-pretty* is NIL?
X-Scale has joined #commonlisp
<paulapatience>
Easy enough to check...
NotThatRPG has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
troojg has joined #commonlisp
Akbar-Birbal has joined #commonlisp
semz has quit [Quit: ZNC 1.8.2+deb2ubuntu0.1 - https://znc.in]
`Pixel_Outlaw has quit [Quit: Leaving]
semz has joined #commonlisp
Akbar-Birbal has left #commonlisp [#commonlisp]
Akbar-Birbal has joined #commonlisp
flounders has joined #commonlisp
drewjose has quit [Quit: Ping timeout (120 seconds)]
drewjose has joined #commonlisp
shawnw_ has joined #commonlisp
rtypo has quit [Ping timeout: 265 seconds]
troojg has quit [Remote host closed the connection]
decweb has quit [Ping timeout: 245 seconds]
contrapunctus has left #commonlisp [#commonlisp]
Lord_of_Life has quit [Ping timeout: 252 seconds]
shawnw_ has quit [Ping timeout: 252 seconds]
Lord_of_Life has joined #commonlisp
drewjose has quit [Quit: Ping timeout (120 seconds)]
drewjose has joined #commonlisp
contrapunctus has joined #commonlisp
chiselfuse has quit [Remote host closed the connection]
chiselfuse has joined #commonlisp
tam has joined #commonlisp
tam has quit [Quit: tam]
shawnw_ has joined #commonlisp
tam has joined #commonlisp
Pixel_Outlaw has quit [Quit: Leaving]
tam has quit [Remote host closed the connection]
meaty has joined #commonlisp
Oladon has joined #commonlisp
JuanDaugherty has joined #commonlisp
pranav has quit [Remote host closed the connection]
gorignak has quit [Quit: quit]
drewjose has quit [Quit: Ping timeout (120 seconds)]
drewjose has joined #commonlisp
gorignak has joined #commonlisp
pve has joined #commonlisp
wacki has joined #commonlisp
treflip has joined #commonlisp
shka has joined #commonlisp
szkl has joined #commonlisp
<Shinmera>
owen: how are you getting the sources for everything? If it's not via our dist then I can't help you
<Shinmera>
though I can't really help anyway given I don't have a mac
tam has joined #commonlisp
pranav has joined #commonlisp
amb007 has quit [Ping timeout: 244 seconds]
amb007 has joined #commonlisp
JuanDaugherty has quit [Quit: JuanDaugherty]
X-Scale92 has joined #commonlisp
X-Scale has quit [Ping timeout: 256 seconds]
drewjose has quit [Quit: Ping timeout (120 seconds)]
drewjose has joined #commonlisp
treflip has quit [Remote host closed the connection]
treflip has joined #commonlisp
mountainman1312 has quit [Quit: ERC 5.5.0.29.1 (IRC client for GNU Emacs 29.4)]
elderK has joined #commonlisp
<elderK>
Hey guys, recently I've been reading some articles about how to structure projects in CL. One of the articles mentioned that there was two ways that people expect their packages to be used: Some, you are expected to import what you need or use the entire package. Others, you are expected to specify the package name or use package-local nicknames.
King_julian has joined #commonlisp
<elderK>
I was wondering if it's just like, a subjective personal thing or if there is a consensus on which approach is favored.
<aeth>
it really depends... the problem with USEing more than CL is that it's really easy to get name clashes.
<jackdaniel>
using USE or importing symbols makes it also quite hard to restructure code
<aeth>
but that means that you can safely USE one thing other than CL, and more if you take care to avoid name clashes (i.e. your own packages)
<aeth>
Importing symbols is the third way, as jackdaniel mentions. It's kind of OK. Verbose. The problem is, implementations don't warn you for unused imports afaik. This means you're going to get out of sync with what you need when you restructure code. And not notice.
<jackdaniel>
say you move a symbol FOOBAR from X to Y; consumers of X won't really know that FOOBAR is missing until they hit a runtime issue that something is undefined
<jackdaniel>
while if you had used X:FOOBAR, then you'd hit the error during compilation
<jackdaniel>
from the implementation perspective, I think that projects should be structured as follows: API packages are USEd by the implementation package; and whole implementation takes place in a single namespace
X-Scale92 has quit [Quit: Client closed]
<jackdaniel>
because if you have too many packages internally, then moving symbols between them is a major pita, while if whole implementation resides in a FOO-INTERNALS, then a) you avoid name clashes (otherwise it won't work), b) it is not visible to consumers of api packages which may be fine-grained
<jackdaniel>
and a third side benefit, people who like to shoot themselves in the foot, can USE your api packages in any combination
<jackdaniel>
for people *
<elderK>
Thanks guys: Maybe my work so far has been using too many packages.
<jackdaniel>
perhaps I should write a blog post with clearer argumentation
<elderK>
jackdaniel: I'd find that very interesting.
<elderK>
One thing that I found nice about the "use via packages" was that you don't have to worry about prefixes.
<elderK>
Like, say you are writing a bunch of queues or something.
<elderK>
When I was doing the Advent of Code, for instance, I had a circular queue. A lot of my functions had circular-queue- as a prefix, but then shortened to cqueue-
<elderK>
But I dislike abbreviating stuff so...
<elderK>
But if I used the package approach, I could've just said cq:function-name
<jackdaniel>
I think that in a single project there shouldn't be two distinct objects with the same name but in different packages; although I'm aware that there are other opinions on the matter
<jackdaniel>
that said my opinion is implicit from the advice to have a single implementation package
<jackdaniel>
two distinct objects in the same namespace [*]
mgl has joined #commonlisp
<jackdaniel>
and there's somewhat vaguely formed thought, that packages should separate codebase horizontally, not vertically
<jackdaniel>
i.e from higher level to lower level, not from one variant to another
X-Scale has joined #commonlisp
simendsjo has joined #commonlisp
<aeth>
I think the only time you should have the same name in a different package in the same project is if you are doing a DSL or something similarly fancy.
<aeth>
e.g. +
<elderK>
Thanks guys, that's good advice.
<elderK>
So, don't use packages just for convenience.
<elderK>
So, if you have a package that has say, lists or queues or whatever, it's best to just put them all in the same package, iono, "queues" or "adts" or whatever.
<elderK>
This is a contrived example but still. :)
varjag has joined #commonlisp
<elderK>
The articles I read were like, it's great to be able to reuse the same names! You can have as many unrelated PUSH functions as you want!
<elderK>
That was not very compelling, though :)
tam has quit [Ping timeout: 252 seconds]
pranav has quit [Remote host closed the connection]
tam has joined #commonlisp
mgl has quit [Ping timeout: 244 seconds]
tam has quit [Ping timeout: 252 seconds]
elderK has quit [Quit: WeeChat 4.4.2]
X-Scale has quit [Ping timeout: 256 seconds]
X-Scale has joined #commonlisp
pranav has joined #commonlisp
tam has joined #commonlisp
X-Scale has quit [Ping timeout: 256 seconds]
tam has quit [Ping timeout: 246 seconds]
treflip has quit [Ping timeout: 276 seconds]
tam has joined #commonlisp
tam has quit [Ping timeout: 252 seconds]
X-Scale has joined #commonlisp
anticomputer_ has quit [Remote host closed the connection]
anticomputer has joined #commonlisp
mountainman1312 has joined #commonlisp
mountainman1312 has quit [Client Quit]
mountainman1312 has joined #commonlisp
tam has joined #commonlisp
augfab has joined #commonlisp
tam has quit [Ping timeout: 276 seconds]
X-Scale has quit [Ping timeout: 256 seconds]
tam has joined #commonlisp
wacki has quit [Ping timeout: 265 seconds]
wacki has joined #commonlisp
tam has quit [Ping timeout: 260 seconds]
attila_lendvai_ has joined #commonlisp
wacki has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
tam has joined #commonlisp
<Josh_2>
I have been playing with Allegro
<Josh_2>
I don't like the IDE very much. Needs proper emacs keybindings
tam has quit [Ping timeout: 252 seconds]
triffid has quit [Remote host closed the connection]
triffid has joined #commonlisp
King_julian has quit [Ping timeout: 244 seconds]
tam has joined #commonlisp
<edgar-rft>
Josh_2: who won, you or allegro? :-)
tam has quit [Ping timeout: 246 seconds]
X-Scale has joined #commonlisp
X-Scale has quit [Write error: Broken pipe]
tam has joined #commonlisp
mountainman1312 has quit [Remote host closed the connection]
brokkoli_origin has quit [Remote host closed the connection]
brokkoli_origin has joined #commonlisp
tam has quit [Ping timeout: 252 seconds]
mwnaylor has quit [Ping timeout: 248 seconds]
mistivia has quit [Ping timeout: 252 seconds]
tam has joined #commonlisp
mistivia has joined #commonlisp
tam has quit [Ping timeout: 265 seconds]
wacki has joined #commonlisp
donleo has joined #commonlisp
tam has joined #commonlisp
simendsjo has quit [Remote host closed the connection]
tam has quit [Ping timeout: 248 seconds]
simendsjo has joined #commonlisp
luis has quit [Ping timeout: 245 seconds]
tam has joined #commonlisp
<Josh_2>
Me
tam has quit [Ping timeout: 252 seconds]
mgl has joined #commonlisp
<Josh_2>
The IDE supports a stripped down Emacs keyset for the editor windows, but you have to manually set the keybinds in the config file for everything else :skull: Fortunately its pretty easy to setup SLIME + Allegro. Would be nice if Eli (Allegro's version of Slime) used Slimes keybindings :joy:
<Josh_2>
I have used some of the built in debug tools and they are really nice, you can inspect objects that are traced through the call stack which is really nice
<Josh_2>
Saves adding conditional breaks in your code
<Josh_2>
Best part of Allegro is the docs are very similar to the hyperspec, no graphics required, just text and some hyperlinks :sunglasses:
edgar-rft_ has joined #commonlisp
edgar-rft has quit [Ping timeout: 246 seconds]
tam has joined #commonlisp
triffid has quit [Remote host closed the connection]
triffid has joined #commonlisp
tam has quit [Ping timeout: 252 seconds]
gorignak has quit [Quit: quit]
gorignak has joined #commonlisp
random-nick has joined #commonlisp
mgl has quit [Ping timeout: 255 seconds]
puke has quit [Quit: WeeChat 3.8]
tam has joined #commonlisp
heisig has joined #commonlisp
rtypo has joined #commonlisp
tam has quit [Ping timeout: 252 seconds]
tam has joined #commonlisp
<paulapatience>
I like to partition an API into several packages when it makes sense, e.g., there is a principal API that users are expected to use, and some secondary APIs that might be extensions to the primary one.
rainthree has joined #commonlisp
<paulapatience>
Another example is for unrelated implementations of a common protocol. The protocol might be the main API, then users can load different implementations in a fine grained manner.
<paulapatience>
For instance, a protocol for dictionary data structures might be the dictionary package, and then dictionary-hash-table could be the implementation based on hash-tables.
<Josh_2>
Allego has these 'fwrappers' which are sort of like :around methods for functions, pretty cool
<Josh_2>
I am not trying to shill Allegro btw just my experience coming from using SBCL exclusively
decweb has joined #commonlisp
tam has quit [Ping timeout: 255 seconds]
<beach>
Josh_2: Do they work on the function object or on the function binding?
<beach>
Like if you do (defparameter *f* (fdefinition 'foo)) and then wrap foo, will (funcall *f* ...) be wrapped as well?
<Josh_2>
I believe it looks up the symbol-function. But I will double check
<beach>
Well, that would be unfortunate for (SETF FOO) functions. :)
dnhester26 has joined #commonlisp
<Josh_2>
okay. Docs say 'the act of encapsulating a function produces a nwe, non-eq function object' meaning that if #'foo is encapsulated but previously captured and stored, the funcall function does not execute the encapsulated functionality
<beach>
I see. So it is a pretty trivial thing to implement if you would want to do that.
<Josh_2>
I believe the primary purpose of their fwrap functionality is for tracing
<jackdaniel>
(setf foo) always annoyed me; either allow arbitrary pairs as function names or give it a break and disallow (setf foo) names (I personally prefer the latter)
Lord_of_Life has joined #commonlisp
<jackdaniel>
like this exceptions sticks like a sore thumb, and that shows when you work on a compiler
<Josh_2>
You define the wrapper independently and then you choose when you wish to invoke the wrapper using a function. Pretty cool
<beach>
Josh_2: If I understand you right, this is a trivial thing that you can implement in around 5 lines of code.
<jackdaniel>
Josh_2: aren't fwrappers basically advices as seen in ccl and cmucl? (and lw I think)
<Josh_2>
Yes its their replacement for advices
<Josh_2>
doc is at franz.com/support/documentation/fwrappers-and-advice.html
tam has joined #commonlisp
<jackdaniel>
I think that unless the advice (or wrap) is stored directly as part of the function (you could that i.e with standard funcallable objects), it is unworthy clutter
<jackdaniel>
* the method combination is real, the advice to use it is a joke
tam has quit [Ping timeout: 265 seconds]
cage has joined #commonlisp
cage has quit [Excess Flood]
cage has joined #commonlisp
bjorkintosh has joined #commonlisp
bjorkintosh has quit [Changing host]
bjorkintosh has joined #commonlisp
<Josh_2>
I wonder if the allegro implementation of date-time also uses the Naggum paper
<dnhester26>
beach: you mentioned magit. they made a CL version of it for lem called legit which may be useful to second climacs. Though I think that's not necessary at all for a code editor, the most important thing I think is just better CL interactivity, I don't mind switching tabs to use git...
<Josh_2>
Magit makes git very easy though
<dnhester26>
yeah, but you can just switch tabs and use it in emacs, that's not the point of a code editor
<beach>
dnhester26: I see, thanks. But what I would like is a CLIM-based application that integrates with the other planned components of an IDI.
<beach>
IDE
tam has joined #commonlisp
<dnhester26>
I would imagine that if it's straightforward to contribute and the features are really cool that people will be happy to jump in and contribute like many started in lem
<rainthree>
I've been busy figuring out why I can't use multiple ssl certificates on hunchentoot (the so called "virtual hosts" functionality so that it uses different certificates for different domain names) and in a book I came across this explanation, that SNI (Server Name Identification) must be implemented: https://pastebin.com/2pwY3neM I wanted to avoid having to install nginx or apache on top of hunchentoot... But I don't
<rainthree>
have the time and experience/I am not in the mood, to implement SNI either. I installed nginx and to make it work is complex as I expected. My idea was to host a bunch of different sites for different domains, on the same lisp image. Succeeded, up to the ssl part at which I am stuck now.
<ixelp>
I've been busy figuring out why I can't use multiple ssl certificates on hunchen - Pastebin.com
tam has joined #commonlisp
King_julian has joined #commonlisp
johnjaye has joined #commonlisp
tam has quit [Ping timeout: 248 seconds]
triffid has quit [Ping timeout: 260 seconds]
tam has joined #commonlisp
gorignak has quit [Quit: quit]
triffid has joined #commonlisp
gorignak has joined #commonlisp
rainthree2 has joined #commonlisp
tam has quit [Ping timeout: 252 seconds]
rainthree has quit [Quit: Leaving]
tam has joined #commonlisp
mgl has joined #commonlisp
yitzi has joined #commonlisp
<Shinmera>
rainthree2: it's best to put something more robust in front of HT anyway
<Shinmera>
let nginx handle ssl and just have it proxy hunchentoot
brokkoli_origin has quit [Ping timeout: 252 seconds]
rainthree2 has quit [Ping timeout: 246 seconds]
tam has left #commonlisp [ERC 5.5.0.29.1 (IRC client for GNU Emacs 29.4)]
treflip has joined #commonlisp
skeemer has joined #commonlisp
donleo has quit [Quit: Leaving]
mjoerg has joined #commonlisp
prokhor has quit [Ping timeout: 264 seconds]
donleo has joined #commonlisp
mjoerg has quit [Client Quit]
<yottabyte>
is that called a REVERSE PROXY?
Lord_of_Life has quit [Excess Flood]
<dlowe>
yes
<dlowe>
almost all the "web servers" in any language should really be considered "backend webapp servers"
<yottabyte>
wow, nice
<yottabyte>
is nginx hard to configure?
<dlowe>
Not really. Just search for a config file and modify it
<yottabyte>
so if I'm running nginx, my lisp server, and then a node server that serves my frontend, do I need a lot of ram/cpu?
<dlowe>
I would run both the lisp and node server behind nginx
<dlowe>
depends on what you consider a lot
<yottabyte>
is like 2gb of ram enough?
<dlowe>
nginx won't add much ram or cpu though
puke has joined #commonlisp
Lord_of_Life has joined #commonlisp
<dlowe>
well, you have a completely undescribed lisp and node server so I really can't answer that
<yottabyte>
hahahaha true
<yottabyte>
it's a simple crud app. login, click around, make some small requests. not many users. nothing computationally expensive
Th30n has quit [Quit: going off]
<Shinmera>
it'll be fine.
<owen>
Shinmera I am using your ql dist, cloned the main alloy repo though for convenience
<Shinmera>
ok.
<Shinmera>
then yeah, no clue
<Shinmera>
possible some main thread nonsense, that's always a thing
<owen>
I saw that issue and tried running right in terminal, had a different bug, I can message you that later but gotta work
treflip has quit [Ping timeout: 265 seconds]
<owen>
I like the style of alloy and want to use it though, clean API
<dnhester26>
dlowe: Shinmera: if running a single site why use nginex? with HT it's possible to do SSL, I made a compression cache to do gzip and set proper headers, is there anything else besides that?
<Josh_2>
yottabyte: I have many webapps running behind nginx, you wont notice nginx ram usage because a single lisp image uses so much :joy:
treflip has joined #commonlisp
<owen>
idk about nginx but i always use caddy so i can route subdomains and never think about certs
<Shinmera>
dnhester26: nginx is an extremely battle-tested server against all sorts of weirdness. hunchentoot is not.
<Shinmera>
especially when it comes to ssl
<Shinmera>
you can also easily make nginx cache stuff for you to get faster serving for static files
<dnhester26>
Shinmera: thanks, so basically it's just because of weird issues that may show up, but in essence besides the caching there's nothing major
<dnhester26>
I am also deploying a server, was trying to stay all lisp, but probably should use nginex or caddy. any recommendations? just go with nginex because it's been more battle tested?
King_julian has quit [Ping timeout: 244 seconds]
<dnhester26>
owen: how's caddy going? why caddy instead of nginex?
<Shinmera>
I don't know caddy, but I've used nginx for years without any issues, so
<Josh_2>
There is also Traefik
<owen>
just because I learned it first really
<Shinmera>
nginx also has plugins for blocking malicious actors and other crap if you care about that
<owen>
not sure if nginx has this but I like to have a declarative Caddyfile to route all the subdomains, and it has fileservers and all that
<owen>
and automagically does SSL with no extra work
<Shinmera>
sure does have that
NotThatRPG has joined #commonlisp
<dnhester26>
ok, ok, so I'll use nginx, just follow the instructions here: https://nginx.org/en/docs/install.html Shinmera any settings you recommend for automating letsencrypt and other things?
<ixelp>
Installing nginx
<Shinmera>
iunno I have my own stuff for that
<Josh_2>
I believe that letsencrypt will handle grabbing new certs for you automatically
<Shinmera>
What I believe is that at this point this is rather off-topic.
treflip` has joined #commonlisp
<dnhester26>
ok, sorry
treflip has quit [Ping timeout: 244 seconds]
mgl has quit [Ping timeout: 255 seconds]
mgl has joined #commonlisp
varjag has quit [Quit: ERC 5.5.0.29.1 (IRC client for GNU Emacs 29.3)]
jonatack has quit [Quit: WeeChat 4.4.2]
<dlowe>
dnhester26: there are a *lot* of malicious agents out there that will abuse any http endpoints they find
pve has quit [Ping timeout: 272 seconds]
<dnhester26>
so that makes it better to use nginx so that they don't abuse the lisp server? I would be surprised if they can figure out any vulnerabilities, how easy is it to spot them? I thought usually they just reuse known vulnerabilites that people haven't updated their systems for, hence I thought using nginx is more of a risk, because it's so popular that there will be a lot of interest whereas a lisp server is just not worth their trouble
<dnhester26>
I thought using any lisp system would be a benefit in terms of obscurity
<dnhester26>
perhaps it's easy to find new vulnerabilities, is there a way to test HT and woo for security vulnerabilities in a bruteforce approach?
<yacin>
security by obscurity is generally frowned upon
<dnhester26>
really? Meaning I protected web web routes with a lot of validation of parameters, recaptcha, ssl auth, etc for any request for which it's reasonable, but outside of that the vulnerabilities would be Hunchentoot related. My app is a small one man project, and it doesn't seem to be cost effective for anyone to invest time in figuring out vulnerabilities for that since the profit is basically nothing, which is not the case for nginx f
<dnhester26>
or which I can guarantee that every single state actor has been making scripts targeting any possible vulnerability they can find, and even regular crackers probably have those since it's so prevalent
NotThatRPG_ has joined #commonlisp
NotThatRPG_ has quit [Client Quit]
NotThatRPG has quit [Ping timeout: 252 seconds]
NotThatRPG_ has joined #commonlisp
pranav has quit [Remote host closed the connection]
<yacin>
you're doing the standard things, which is the right thing to do
<yacin>
i'm just saying the security community typically doesn't consider obscurity to give true hardness gains, unlike the protection you're already implementing
<yacin>
and there's nothing preventing a general vulnerability (e.g., slowloris ddos attacks) from working on hunchentoot, and we're in the _worse_ situation because no one knows that it's vulnerable since it's less frequently used
<yacin>
but yeah, prob off-topic. my bad here
brokkoli_origin has joined #commonlisp
<kagevf>
There's a #lispweb channel fwiw
heisig has quit [Quit: Leaving]
dnhester26 has quit [Remote host closed the connection]
dnhester26 has joined #commonlisp
dnhester26 has quit [Read error: Connection reset by peer]
dnhester26 has joined #commonlisp
dnhester26 has quit [Remote host closed the connection]
dnhester26 has joined #commonlisp
dnhester26 has quit [Read error: Connection reset by peer]
dnhester26 has joined #commonlisp
dnhester26 has quit [Read error: Connection reset by peer]
dnhester26 has joined #commonlisp
<yottabyte>
Josh_2: haha, good to know
dnhester_ has joined #commonlisp
dnhester_ has quit [Read error: Connection reset by peer]
dnhester_ has joined #commonlisp
dnhester_ has quit [Read error: Connection reset by peer]
dnhester_ has joined #commonlisp
dnhester_ has quit [Remote host closed the connection]
dnhester_ has joined #commonlisp
dnhester_ has quit [Read error: Connection reset by peer]
dnhester_ has joined #commonlisp
dnhester_ has quit [Remote host closed the connection]
dnhester_ has joined #commonlisp
dnhester_ has quit [Remote host closed the connection]
dnhester_ has joined #commonlisp
dnhester_ has quit [Remote host closed the connection]
dnhester_ has joined #commonlisp
dnhester_ has quit [Remote host closed the connection]
dnhester_ has joined #commonlisp
dnhester_ has quit [Remote host closed the connection]
dnhester26 has quit [Ping timeout: 255 seconds]
NotThatRPG_ has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]