Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Repo info
Activity
  • 10:35
    bitbegin assigned #4004
  • 10:35
    bitbegin unassigned #4004
  • 10:28
    bitbegin commented #3952
  • 09:53
    dockimbel labeled #3999
  • 09:53
    dockimbel assigned #3999
  • 09:52
    dockimbel labeled #3952
  • 09:52
    dockimbel closed #3952
  • 09:52
    dockimbel commented #3952
  • 09:52
    dockimbel commented #3952
  • 08:45
    9214 commented on 84f55a2
  • 08:29
    meijeru commented on 84f55a2
  • 07:24
    bitbegin assigned #4004
  • 06:48
    bitbegin commented #3077
  • 06:18
    bitbegin opened #4004
  • 06:08
    bitbegin commented #3837
  • 05:49
    qtxie review_requested #3996
  • 05:48
    qtxie commented #3996
  • 05:48
    qtxie commented #3996
  • 05:45
    qtxie commented #3837
  • 05:41
    qtxie commented #3837
Vladimir Vasilyev
@9214
@Oldes so far it happened only with Google.
Oldes Huhuman
@Oldes
It can happen everywhere. It is just that Google is used quite a lot. What I was looking last time, the error is in some stupid inlined javascript.
Vladimir Vasilyev
@9214

It can happen everywhere.

In such case you should try to reproduce it with any other web resource or file.

Oldes Huhuman
@Oldes
I don't have time to write web-crawlers.
laturk
@laturk
@Oldes That is good to know. Thanks!
dsunanda
@dsunanda
It's not just with web resources - Red cannot read this file (written with R2)
write %test.txt to-string to-char 163  ;; valid "£" character
Vladimir Vasilyev
@9214
>> write %test.txt to string! to char! 163
>> read %test.txt
== "£"
Oldes Huhuman
@Oldes
@laturk you should also know, that Rebol2 has serious issues in the call implementation and it may fail.
dsunanda
@dsunanda
Did you write it from an R2 session? This is what I see:
read %test.txt
*** Access Error: invalid UTF-8 encoding: #{A3000000}
*** Where: read
laturk
@laturk
@Oldes Noted. Thanks.
Vladimir Vasilyev
@9214
@dsunanda well, maybe it's R2 to blame? R3 works fine.
Oldes Huhuman
@Oldes

@dsunanda Rebol2 does not save it in UTF-8!

>>  to-binary to-string to-char 163
== #{A3}

In Red or Rebol3:

>> to-binary to-string to-char 163
== #{C2A3}
dsunanda
@dsunanda
It's more that Red cannot read many plain ASCII files - ie those containing characters with the high bit set. Not sure blaming that on R2 is helpful :) It's an interoperability issue between Red and systems that write plain ASCII files.
Oldes Huhuman
@Oldes
I agree, that it is Red's limitation. Rebol3 can load such an ascii text:
>> to-string #{A3}
== "£"
>> to-string #{C2A3}
== "£"
Gregg Irwin
@greggirwin
Some Latin1 support was being worked on in Red, and read has an /as refinement for future extensions in this area.
On system words, there is also the what command.
I've never liked that name. It was carried over from R2.
Oldes Huhuman
@Oldes
@greggirwin do you have better name?
Gregg Irwin
@greggirwin
Not yet.
Gabriele Santilli
@giesse
@dsunanda To be pedantic, that's Latin1 not ASCII. Red can do ASCII. :) But, in general, there's not much Red can do magically, you need to tell it how the text is encoded. You could port my R2 code to Red if this is an issue for you. :P https://giesse.github.io/rebol-power-mezz/#section-5.21
dsunanda
@dsunanda

Thanks Gabriele - it's good to be precise.

The Red documentation on the differences between READ in Rebol and Red does not highlight the limitations in reading Latin1 - so the issue may be a bug.

Though it does note that the current "simple" functionality may change - so this may be addressed in later iterations.

Boleslav Březovský
@rebolek
@9214 it happens not only with Google, but for example with one message in Gitter too and of course with lot of Czech pages, which are using CP-1250 for ages and don't bother to update.
I have a simple function to auto-convert such pages, when using send-request https://github.com/rebolek/red-tools/blob/master/http-tools.red#L378
Gabriele Santilli
@giesse
@dsunanda R2 doesn't really care about the text encoding, aside from CR and LF when not using /binary (so, it could mess up text encoded in something not ASCII-compatible). Red and R3 will try to decode the text, so if it's not UTF-8 you get problems. It might make sense to relax the decoding and just replace anything invalid with a ? or something like that, instead of throwing an error... perhaps with a /strict refinement to enforce the current behavior... but in general if you don't know the encoding of some text then it would be best to treat it as binary instead.
GiuseppeChillemi
@GiuseppeChillemi
I Need to create and app that takes a geotagged picture and send It to a server together with email and phone id.
Could I make It in Red and publish on the play store ? Do we have everything ?
Gregg Irwin
@greggirwin
@GiuseppeChillemi, the Android branch is experimental, and we're still Alpha. I'll say Red isn't ready for that, but the camera interface works, there is no email support, but maybe an HTTP service could be used. We haven't looked at Play Store publishing at all yet, so that's something to be researched.
GiuseppeChillemi
@GiuseppeChillemi
Ok, I Need to find and alternate multiplatform language (ios android ) which Is Quick to learn
Semseddin Moldibi
@endo64
@GiuseppeChillemi or use a hybrid solution like ionic
Boleslav Březovský
@rebolek
Is it somehow possible for a function to access its context, something like self in object!?
Boleslav Březovský
@rebolek
context? , of course.
thanks @rebolek
GiuseppeChillemi
@GiuseppeChillemi
@rebolek You are thanking yourself, I am concerned about you, how many hours your are in front of your PC without taking a pause ?
Boleslav Březovský
@rebolek
:smile:
BeardPower
@BeardPower
It's all just in his head ;-)
GiuseppeChillemi
@GiuseppeChillemi
@BeardPower he referenced his inner self object
BeardPower
@BeardPower
@GiuseppeChillemi :-)
Gregg Irwin
@greggirwin
:^)
Jose Luis
@planetsizecpu
@rebolek @BeardPower @GiuseppeChillemi and we must not forget that everything is data, we included, we are only a block like [GTACATCG ...] ;-)
GiuseppeChillemi
@GiuseppeChillemi
At the end, all of us will be bound to the global context.
BeardPower
@BeardPower
@planetsizecpu Or is data everything? ;-)
Jose Luis
@planetsizecpu
@BeardPower :-) I'm not sure, but think about this: when our REPL loop finishes evaluating us, there will be an actor from the local environment that will move us to the GC of the global context hahaha (I think it's time to move to the chit-chat room or someone will throw us)
Boleslav Březovský
@rebolek
It's fine. Red is leaking memory like there's no tomorrow, so there's a good chance GC will never catch you.
Max
@moliad

It's fine. Red is leaking memory like there's no tomorrow

Yeah, we noticed that too. After christmas we'll have to work on the block lexer cause it seems like its using 100x the memory it should (As per my quick mental calculations)

Boleslav Březovský
@rebolek
The good thing is that this kind of bugs is very simple to fix.
Toomas Vooglaid
@toomasv
:clap: :clap: :exclamation:
Dave Andersen
@dander
:clap: Great job team, and keep up the good work!
I totally missed the console highlighting. Any reason that's not on by default? I'm so excited for IO/async :smile:
Vladimir Vasilyev
@9214

Any reason that's not on by default?

I think because it's static and can be extremely confusing. And also limited only to a handful of datatypes.

Gregg Irwin
@greggirwin
@dander, what @9214 said, and it's also more experimental than a polished feature. We can enhance the settings dialog to allow customization, and are happy to have people play with theme ideas. The current color choices have not been run by Ethan Schoonover.