freenode/#lisp - IRC Chatlog
Search
5:07:45
fiddlerwoaroof
But, to do that right, I'd need an implementation that made the code in the image the source of truth
5:08:09
fiddlerwoaroof
e.g. it'd be more like a smalltalk environment than like modern CL environments
5:10:36
fiddlerwoaroof
One advantage of this would be that, it wouldn't matter _where_ I define a function: if I type it in the repl, or enter it through a buffer, the code will end up in the same place and I could extract it into a buffer using a hypothetical query language for source code
5:11:47
PuercoPop
But with current xref's capabilities you could create a buffer that is the 'view' current loaded methods associated with a generic functions today
5:12:59
fiddlerwoaroof
Yeah, the problem with that is that if I edit the from that buffer, I'd have to manage patching those functions back into their source files
5:13:14
PuercoPop
Except the prospect of writing complex UI controls in elisp is not an enticing one
5:14:42
fiddlerwoaroof
I think so, although from what I gather, combining parts of several files into one file doesn't work very reliably in emacs.
5:15:42
fiddlerwoaroof
Also, the other thing is, if the source of truth for the code is whatever is stored in the image, it might be less painful to maintain customizations of other people's code
5:17:25
PuercoPop
yeah, an indirect-buffers can only show content from one buffer. But you a buffer to select the method and a buffer to display its contents. Similar to how the Smalltalk Browser shows protocols and methods. Although that is mostly the same to the current xref capabilities in sl{y,ime}
5:19:29
fiddlerwoaroof
I'm actually a bit curious if it's possible to tell something like sqlite to embed a database at the end of a file
5:20:58
oni-on-ion
i was looking at org-mode attachments for a similar purpose; to compound a given project source tree into a single org file
9:45:18
francogrex
hi is there a link that shows how to cross compile build sbcl for windows on a linux (from linux to make win64 executable sbcl)? I have problems building directly on windows (with mingw)
9:52:36
Shinmera
Download an SBCL from sourceforge, then use MSYS 2's MinGW console. Works just fine
10:02:40
jackdaniel
and all I had to do was to type "cltl2 ansi common lisp differences" in the search engine! :-)
10:05:14
jackdaniel
careful reader would notice "ANSI deleted the functions variable-information, function-information, declaration-information, augment-environment, parse-macro, and enclose as well as the macro define-declaration. This
10:05:47
jackdaniel
even more careful reader would find in the first paragraph: "In fact, this FAQ has many known but minor omissions that are too awkward to add with pencil to CLTL2."
10:59:46
no-defun-allowed
tbh environment information would have been really nice to see in the ansi spec
11:00:36
jackdaniel
sbcl and ccl have it, ecl will have it next release (I have written regression tests already)
11:01:06
jackdaniel
(in a form suggested in CLtL2) - I believe there is even a portability library for that
11:02:52
beach
If I may guess, the reason it didn't make it is that it is insufficient, distinctly non-CLOS-y and can not be extended.
11:03:47
jackdaniel
otoh if it were standarized it would make code walkers easier to implement (and provide better compiler macros)
11:04:55
beach
I would much rather see a compatibility protocol based on generic functions. TRIVIAL-ENVIRONMENTS or something.
11:07:28
jackdaniel
no-defun-allowed: either way, here it goes: https://github.com/alex-gutev/cl-environments/tree/master
11:08:27
jackdaniel
I've found some problems with sbcl which I've used as a reference for testing CLtL2 api (I don't quite remember what they were, but SBCL doesn't conform in some edge cases)
11:21:08
scymtym
i wonder why https://common-lisp.net/downloads mentions allegro but not lispworks. that would have been the obvious choice given that the two most popular open source implementations are listed
11:28:58
scymtym
i mean, i'm not saying every implementation should be listed, but those two are pretty obvious omissions
12:23:13
hjudt
does anyone know this -> if i fork a project that is under apache 2 license, do i really need to add a notification to every modified file stating that it has been changed? There is actually no copyright information included in (most of) the files. Or is meta-information from vcs enough?
12:39:49
ogamita
hjudt: 4. b) is clear: "You must cause any modified files to carry prominent notices stating that You changed the files;".
12:40:20
ogamita
hjudt: I think that "prominent notices" can be implemented in a separate file, but must not be hidden in a database such as git logs.
12:40:46
ogamita
hjudt: Of course, you can generate this PROMINENT-NOTICE.txt file from the git logs; write a script!
12:44:43
hjudt
ogamita: thanks. if i add a new section in the readme stating that the project has been forked from another project and provide the reference and download location of that will it be enough?
12:46:34
ogamita
It depends on each license. You need to read them to know what the requirements are, when you distribute modified forks. https://www.apache.org/licenses/LICENSE-2.0
12:47:51
ogamita
Giving credit to the original author is not a requirement of the Apache v2. (but 4.d is an approximation).
12:48:44
ogamita
Giving the reference and download location is not a requirement either of Apache v2.
13:58:57
jcowan
The point of marking files as modified by you is reputational: the original author does not want to be blamed for bugs you introduced into them!
14:04:52
drdo
I don't even want to try to understand how this works in a project with several tens or hundreds of contributors
17:03:41
stylewarning
(Echoing in here from #sbcl): We are running into a problem where we compile and load an entire system (a vanilla ASD), and we get some performance baseline of the application, then we do (load (compile-file X)), performance improves by about 20%, then (load (compile-file Y)), performance regresses back to normal. Y depends on X, and X precedes Y in the system definition. I looked for compiler policy leakage and that
17:06:05
stylewarning
(It might also help to know neither X nor Y were modified between compilations/loading, cache has been cleared, and X just contains structures+functions, Y just contains variables and functions)
17:06:47
phoe
beach: also, I think that Shinmera has done work that is superior to mine - he has done some work on something he calls a wire protocol that is usable for serializing data over the network into S-expressions.
17:07:10
phoe
His implementation is much better than mine and I would recommend to base any Eclector-related code on his work.