Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
  • Dec 03 2017 05:53
    @PeterWAWood banned @matrixbot
  • Sep 28 2016 12:19
    @PeterWAWood banned @TimeSeriesLord
  • Aug 13 2016 03:23
    @PeterWAWood banned @Vexercizer
Vladimir Vasilyev
@9214
:wave: have fun.
lepinekong
@lepinekong_twitter
@9214 then all red functions were malformed if that was the case. Except it wasn't malformed because if you are now obliged to remove docstring for return it means you cannot document it anymore. That's not a good evolution.
@ne1uno see above: it works but you cannot document anymore.
Vladimir Vasilyev
@9214
@lepinekong_twitter instead of wasting your breath on me, go on and write a clear, succinct and technical proposal, which will address this issue. Oh, wait, I already did that! See, not that hard at all.
lepinekong
@lepinekong_twitter
@ne1uno I'm learning red at the moment by developping the stuff alone because I know how hard it is to version things correctly so I don't want anybody to messup with my code at the moment :)
nedzadarek
@nedzadarek
@lepinekong_twitter but you know... you are using alpha version ;)
lepinekong
@lepinekong_twitter
@nedzadarek it seems it's not even alpha now I know :)
@9214 it is you who started the flame with your habit of being insulting.
Vladimir Vasilyev
@9214
"A beam in one's own eye", sweetie.
I'm not the one who parasites on our community responsiveness, giving in return nothing but false accusations mixed with pure FUD and nonsensical, bold claims.
lepinekong
@lepinekong_twitter
@9214 continue like that ;)
nedzadarek
@nedzadarek
Is there a way safely load some code into the Red? Or do I have to check it?
I'm trying to serialize some data to the file and load it later.
Vladimir Vasilyev
@9214
@nedzadarek load/trap, perhaps?
load does not evaluate anything, if that's what you mean.
ne1uno
@ne1uno
NP hard? run it in a sandbox. check for reads writes net access etc etc. create a checksum so you know if it was modified
ne1uno
@ne1uno
lepinekong_twitter I just saw a note that return in function header is a hint for the compiler, maybe doc string error is a regression you found?
Interpreter actually completely ignores return spec.
nedzadarek
@nedzadarek

@9214 so I load it o: load "...". Can a hacker "malform" it so I do something dangerous by doing operation on o (for example o/...)?

I guess I shouldn't store function's body or any strings that I want to do but am I safe without doing this?

Vladimir Vasilyev
@9214
@nedzadarek you're safe as long as you don't do whatever you just loaded.
nedzadarek
@nedzadarek
@9214 I see, thank you.
Vladimir Vasilyev
@9214
@nedzadarek always double-check the input that came from the untrusted source, even when you use only a part of it, like select o 42 or something like that. load just gives you a block of Red values, nothing more. Validity of these values is another matter.
lepinekong
@lepinekong_twitter
@ne1uno I'm not sure to understand the note you refer to and I'm not knowledgeable enough in red to qualify as regression, I just know what behavior I would need :)
nedzadarek
@nedzadarek
@9214 yes, it can be anything, e.g. string instead of integer, negative integer ( < 0) instead of positive integer (> 0). But that kind of things just give me incorrect return value or wrong argument type error, so it's good enough.
nedzadarek
@nedzadarek
@GiuseppeChillemi as @9214 said: view [a: area button "copy" [a2/text: copy/part at a/text a/selected/1 a/selected/2 ] a2: area]
  1. Select text in the first area. 2. Click button.
Vladimir Vasilyev
@9214

@nedzadarek I'd rather go with:

view [
    in:  area 
    out: area 
    return below

    button "copy!" [
        if size: attempt [1 + in/selected/y - in/selected/x][
            out/text: take/part at trim/all/lines copy in/text in/selected size
        ]
    ]
]

... which gives me an obscure error whenever I try to set out/text. :confused:

And handling newlines is a tricky business, my version is not quite sane in this regard.
Vladimir Vasilyev
@9214
Ah, probably overriding in is the culprit.
ne1uno
@ne1uno
rich-text but it's nowhere near ready, I think toomasv work on that scared everyone is still in shock
also copy/paste is text only
nedzadarek
@nedzadarek

Ah, probably overriding in is the culprit.

@9214 you are right

nedzadarek
@nedzadarek
The function correctly keep it local: f: function [] [view [in: area]] f ? in.
Gregg Irwin
@greggirwin

As most of you know, @9214 is part of our extended Team Red group. I've chatted with him about some of his recent comments here, and apologize. We expect everyone to remain civil, though things will sometimes get heated. We understand that.

@lepinekong_twitter, while not defending @9214's behavior, I understand what sets him off.

You ask a lot of questions. All well and good. But then you don't seem to listen to advice given, beyond using working code or fixes others provide. You do things in quite a different way, also fine, but it makes it harder to help you. That means more effort on our part. Most frustrating is that you seem to blame Red first, and cast aspersions, which makes us (myself included) defensive. I admit that your idea of Red being stable, or subject to semantic versioning in its current state, made me wonder where you got those ideas. It's unfair to us, to claim an expectation for something we never committed to. Do you agree? Red is Alpha, and things will break. Sometimes big things. Expect that for a while.

I've asked @9214 not to respond to you anymore, not even to be helpful. I will also ask other team members to make sure their time is distributed equitably. We do want to help and support you but, for all the help you receive, I don't often see you helping others in return. Until we can put RED tokens to use, our informal ledger has to be balanced based on human judgement.

And, by using FOSS software, be aware that if you say "it should...", we may say "You're right! Make that happen." :^)

I do wish you luck and Happy Reducing.

Petr Krenzelok
@pekr

Thanks Gregg. In fact I wanted to ask yesterday, if Vladimir is the same 9214 user we used to know, as some of the reactions to questions (no matter how technically funded), were quite harsh and personal. In fact, I do remember many Rebol gurus from the past (who are not with us for quite some time), we went thru so many concept discussions, but it never slipped to almost personal level.

I agree, that precise technical knowledge and understanding of all concepts is a big advantage, but I do remember, with Rebol, even gurues argued over the concepts, R2 being a closed sources. Yet it did not prohibit ppl from writing a useful tools. Simply Rebol allowed even not so much skilled ppl to write a useful app. Yes, I know, it might turn into a disaster later, but ....

Simply put, we should be friendlier to both newcomers / oldtimers, who seek for a help. If answers are dismissive and discouraging, ppl will refrain from asking in the end, to just not feel dumb. As the saying goes - there is no dumb question. Of course if someone is trolling the language or would like to turn it into another language xy he/she already knows, that's a bit different story, but I think we have handled even such cases in the past in a civilised way ...

Thanks again ....

Gregg Irwin
@greggirwin
Thanks for weighing in @pekr. I agree.
lepinekong
@lepinekong_twitter
@greggirwin I'm not blaming Red first, it's rather @9214 always is on the defensive of taking it as a blame. As I said to @ne1uno I perfectly know how difficult industrialising software is a difficult process, so when I report something that is not working anymore instead of jumping on the roof and taking it as an attack you should rather take it as a contribution because right testing the wrong thing is easy but testing the right thing is not. I have more a BDD than TDD approach which is said it is TDD done right: focusing on user's need (the user of the product vs the designer of the product).
lepinekong
@lepinekong_twitter
@ne1uno to answer your remark about why I don't seem to have TDD tests, first I don't like TDD in its strict sense (see my remark about BDD vs TDD) and I prefer Deming PDSA cycles (Plan Do Study Act which is in essence the scientific method of aristotle circle :) ) but I keep it for myself only not in the public repo for learning purpose as I record every little stupid microsteps I made as you can see in the screenshot below:
https://i.imgur.com/De9ccrx.png
GiuseppeChillemi
@GiuseppeChillemi

@nedzadarek

@GiuseppeChillemi as @9214 said: view [a: area button "copy" [a2/text: copy/part at a/text a/selected/1 a/selected/2 ] a2: area]
Better than COPY, I need to CUT and INSERT.
Also, is there a way to do this without resetting the current text position ?

lepinekong
@lepinekong_twitter
In VSCode Powershell keeps a session available so that one can reuse its variable, why can't red seem not being able to do the same ?
https://i.imgur.com/vrXTGTk.png
Boleslav Březovský
@rebolek
@lepinekong_twitter you mean that you start Red, type a: 10, close Red, start it again and a will be there?
Linker Storm
@linker_storm_twitter
Hi. I can't compile "test-wpf.red" in Red sources red\red\bridges\dotnet\.
The error message is :
-=== Red Compiler 0.6.3 ===-

Compiling D:\Temp\red\red\bridges\dotnet\test-wpf.red ...
Compiling libRedRT...
...compilation time : 1621 ms

Compiling to native code...
...compilation time : 47266 ms
...linking time     : 554 ms
...output file size : 933888 bytes
...output file      : D:\Temp\red\red\bridges\dotnet\libRedRT.dll

...compilation time : 195 ms

Target: MSDOS

Compiling to native code...
*** Compilation Error: undefined symbol: red/unicode/to-utf16-len
*** in file: %/D/Temp/red/red/bridges/dotnet/test-wpf.red
*** in function: exec/clr-search-class
*** at line: 28
*** near: [unicode/to-utf16-len as red-string! name :len false
    ret: 0
]
Vladimir Vasilyev
@9214
@linker_storm_twitter .NET bridge wasn't maintained in a long while, and I doubt that it's fully functional at this point. @koba-yu and @iceflow19 have .NET related projects that they might share, if you are interested.
Still, try to compile with -r flag instead of -c, and with -t Windows.
@GiuseppeChillemi for cutting and inserting, you can use take instead of copy, that will modify original text in-place.
Not sure what you mean by "resetting current position" though.
Linker Storm
@linker_storm_twitter
-t Windows forces me to add Needs: View in the header of main source because Windows target is a GUI app.
-r switch works like a charm ! I didn't know it, what is it standing for ?
Vladimir Vasilyev
@9214
@linker_storm_twitter use -h flag and find out yourself :wink:
Linker Storm
@linker_storm_twitter
I see : release flag...
Vladimir Vasilyev
@9214
@linker_storm_twitter yes, it will bake a stand-alone executable for you, instead of linking to a dynamic library with Red runtime.