EiffelStudio 20.11 is now available
Eiffel Users
We are pleased to announce the availability of EiffelStudio 20.11.
Try it right now! All the download sites have been updated.
The highlights are:
- Supported once classes.
- A new WrapC wizard is available to assist the user wrapping C libraries.
- Autotest has been updated and is now fully compatible with void safety.
- And new wrappers for mpack, cairo, librsvg). They are available via https://iron.eiffel.com/.
- Plus numerous improvements across the libraries (new graphs, b-trees, union-find in unstable/library/base_extension, reCaptcha v3 support, fixed known web vulnerabilities in the CMS component.
Let's remind the main highlights of previous EiffelStudio 20.05 :
- The new support for multi-branch expressions that are similar to multi-branch instructions, but Then_parts and Else_part contain expressions instead of instruction sequences.
- Improved support for Unicode: in the search tools, in code completion to insert Unicode symbols.
See the release notes for details.
Happy Eiffeling,
-- The Eiffel Software Team
Larry Rix
Bertrand Meyer
The cited addition going back to 20.05
> Improved support for Unicode: in the search tools, in code completion to insert Unicode symbols
includes completion support in EiffelStudio making it easy to use Unicode quantifier expressions such as
if ∃ c: your_string ¦ c.is_upper then …
A blog post I just wrote, https://bertrandmeyer.com/2020/12/24/right-forms-expression/, gives more details.
-- Bertrand Meyer
--
You received this message because you are subscribed to the Google Groups "Eiffel Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to eiffel-users...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/eiffel-users/f12176b3-8d8f-4bb5-9485-8c12d12ae240n%40googlegroups.com.
Volkan Arslan
Just a comment, nothing earth-shattering but a (very) small detail:
I don't know if there are any technical or non-technical reasons, but it would be nice to support beside
the "broken bar" ¦
also
the "vertical line" | (Unicode Basic Latin, position: 0x007C, see https://unicode-table.com/en/007C/)
since most keyboards nowadays do have the "vertical line" | symbol too.
Regards
Volkan
On 24/12/2020 12:44, Bertrand Meyer wrote:
Bertrand Meyer
Yes, it’s a broken bar, because the normal bar would create a syntactical conflict since it is allowed in free operators, where it is actually used fairly often in practice. We did look for a way to allow it anyway, through ad hoc techniques for syntactical disambiguation, but did not find an obvious solution. The expectation is that most of the time the broken bar will be generated automatically through completion, so this shouldn’t be too much of a burden.
The issue here is backward compatibility == we do not want to break existing code.
-- BM
--
You received this message because you are subscribed to the Google Groups "Eiffel Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to eiffel-users...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/eiffel-users/c3f81d9d-a8fb-8504-92b4-a1c5945b5633%40gmx.net.
Philippe Laré
« The issue here is backward compatibility == we do not want to break existing code. »
To view this discussion on the web visit https://groups.google.com/d/msgid/eiffel-users/044701d6d9ff%249c679410%24d536bc30%24%40inf.ethz.ch.
Bertrand Meyer
Sorry, it was meant as two hyphens approximating an em-dash “—“, my finger slipped to the right.
-- BM
To view this discussion on the web visit https://groups.google.com/d/msgid/eiffel-users/CAN%3DG_amWC%3DEgRaoX1oT%2BhWbUdUhd7G0B%3Dtrg15%3DdLoaV%2BKbZ%3DQ%40mail.gmail.com.
Woland's Cat
Richard
Howard Thomson
--
You received this message because you are subscribed to the Google Groups "Eiffel Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to eiffel-users...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/eiffel-users/7f8c0ef2-4c6e-4955-acc2-2c365164f3aan%40googlegroups.com.
Richard
Richard
Richard
Louis M
Good day,
By looking at the 20.11 release of EiffelStudio, I just notice that the GPL LICENCE files of the 19.05 has been replaced by a proprietary EULA. Being a free software advocate, using Free Software is extremely important for me. So is someone can confirm that Eiffel Studio still has an Open-source license (like it is still indicated there: https://www.eiffel.com/eiffelstudio/licensing/) or is Eiffel now only behind a proprietary EULA?
Thanks,
Louis M
--
You received this message because you are subscribed to the Google Groups "Eiffel Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to eiffel-users...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/eiffel-users/d1450bb3-e183-4602-b448-f64bbda3be16n%40googlegroups.com.
Ulrich Windl
in
Nachricht <041f01d6d9ea$2c038090$840a81b0$@inf.ethz.ch>:
>
>
>
> > Improved support for Unicode: in the search tools, in
code
> completion to insert Unicode symbols
>
>
>
> includes completion support in EiffelStudio making it easy to use Unicode
> quantifier expressions such as
>
>
>
> if ∃ c: your_string | c.is_upper then …
Wondering: What operators to implement, which not?: In the past Betrand was
against cryptic operators (even though being a mathemacican), but the
preference seems to have changed.
After browsing the Unicode tables a bit, I wonder about these operators:
Should they all be implemented in the appropriate classes?:
∃∀∄∈∉∋∌∌⊂⊃⊄⊅⊆⊇⊈⊉⊊⊋
(Apologies to those not having Unicode capabilities to read)
Regards,
Ulrich
>
>
>
> A blog post I just wrote,
> https://bertrandmeyer.com/2020/12/24/right-forms-expression/, gives more
> details.
>
>
>
> -- Bertrand Meyer
>
>
>
> From: eiffel...@googlegroups.com [mailto:eiffel...@googlegroups.com]
> On Behalf Of Larry Rix
> Sent: Thursday, December 24, 2020 00:26
> To: Eiffel Users <eiffel...@googlegroups.com>
> Subject: [eiffel-users] Re: EiffelStudio 20.11 is now available
>
>
>
> YAY!!!
>
> On Tuesday, December 22, 2020 at 8:33:50 AM UTC-5 Eiffel Users wrote:
>
> We are pleased to announce the availability of EiffelStudio 20.11.
>
> <https://www.eiffel.org/downloads> have been updated.
>
> The highlights are:
>
> * Supported once classes
> <https://www.eiffel.org/blog/Alexander%20Kogtenkov/2020/12/once-classes> .
> * A new WrapC wizard is available to assist the user wrapping C
libraries.
> * Autotest has been updated and is now fully compatible with void
safety.
> * And new wrappers for mpack, cairo, librsvg). They are available via
> https://iron.eiffel.com/.
> * Plus numerous improvements across the libraries (new graphs, b-trees,
> known web vulnerabilities in the CMS component.
>
> Let's remind the main highlights of previous EiffelStudio 20.05 :
>
> instead of instruction sequences.
> insert Unicode symbols.
>
> See the release notes
>
> for details.
>
> Happy Eiffeling,
>
> -- The Eiffel Software Team
>
> Link: https://www.eiffel.org/blog/eiffelstudio_20_11
>
> --
> You received this message because you are subscribed to the Google Groups
> "Eiffel Users" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to eiffel-users...@googlegroups.com
>
https://groups.google.com/d/msgid/eiffel-users/f12176b3-8d8f-4bb5-9485-8c12d1
> 2ae240n%40googlegroups.com
>
> 12ae240n%40googlegroups.com?utm_medium=email&utm_source=footer> .
> --
> You received this message because you are subscribed to the Google Groups
> "Eiffel Users" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to eiffel-users...@googlegroups.com.
> To view this discussion on the web visit
>
> 81b0%24%40inf.ethz.ch.
Ulrich Windl
Nachricht <c3f81d9d-a8fb-8504...@gmx.net>:
> they really look quite impressive.
>
> Just a comment, nothing earth-shattering but a (very) small detail:
>
> I don't know if there are any technical or non-technical reasons, but it
> would be nice to support beside
> the "broken bar" |
> also
> the "vertical line" | (Unicode Basic Latin, position: 0x007C,
> see https://unicode-table.com/en/007C/)
> since most keyboards nowadays do have the "vertical line" | symbol too.
"vertical line" is "󠁼", which is not "vertical bar" (ASCII code). I
doubt a keyboard will have a key other than for the ASCII vertical bar.
Regards,
Ulrich
>
> Regards
> Volkan
>
>
> On 24/12/2020 12:44, Bertrand Meyer wrote:
>>
>> The cited addition going back to 20.05
>>
>> > Improved support for Unicode: in the search tools, in code
>> completion to insert Unicode symbols
>>
>> includes completion support in EiffelStudio making it easy to use
>> Unicode quantifier expressions such as
>>
>> <https://bertrandmeyer.com/2020/12/24/right-forms-expression/>, gives
>>
>> -- Bertrand Meyer
>>
>
> You received this message because you are subscribed to the Google Groups
> "Eiffel Users" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to eiffel-users...@googlegroups.com.
> To view this discussion on the web visit
>
> 5b5633%40gmx.net.
Ulrich Windl
in
Nachricht <04b001d6dab9$bd727c50$385774f0$@inf.ethz.ch>:
> slipped to the right.
>
>
>
> -- BM
>
>
>
> From: eiffel...@googlegroups.com [mailto:eiffel...@googlegroups.com]
> On Behalf Of Philippe Laré
> Sent: Friday, December 25, 2020 13:25
> To: eiffel...@googlegroups.com
> Subject: Re: [eiffel-users] Re: EiffelStudio 20.11 is now available
>
>
>
>
> « The issue here is backward compatibility == we do not want to break
> existing code. »
>
>
> == ?
>
> ��
>
> Ph.
>
>
>
> Le jeu. 24 déc. 2020 à 15:18, Bertrand Meyer <Bertran...@inf.ethz.ch
> Yes, it’s a broken bar, because the normal bar would create a syntactical
> conflict since it is allowed in free operators, where it is actually used
> fairly often in practice. We did look for a way to allow it anyway, through
> ad hoc techniques for syntactical disambiguation, but did not find an
obvious
> solution. The expectation is that most of the time the broken bar will be
> generated automatically through completion, so this shouldn’t be too much of
> a burden.
>
>
>
> The issue here is backward compatibility == we do not want to break existing
> code.
>
>
>
> -- BM
>
>
>
> From: eiffel...@googlegroups.com <mailto:eiffel...@googlegroups.com>
> On Behalf Of Volkan Arslan
> Sent: Thursday, December 24, 2020 14:57
>
https://groups.google.com/d/msgid/eiffel-users/c3f81d9d-a8fb-8504-92b4-a1c594
> 5b5633%40gmx.net
>
> 45b5633%40gmx.net?utm_medium=email&utm_source=footer> .
> --
> You received this message because you are subscribed to the Google Groups
> "Eiffel Users" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to eiffel-users...@googlegroups.com
>
> bc30%24%40inf.ethz.ch
>
<https://groups.google.com/d/msgid/eiffel-users/044701d6d9ff%249c679410%24d53
> 6bc30%24%40inf.ethz.ch?utm_medium=email&utm_source=footer> .
> --
> You received this message because you are subscribed to the Google Groups
> "Eiffel Users" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to eiffel-users...@googlegroups.com
>
> UdUhd7G0B%3Dtrg15%3DdLoaV%2BKbZ%3DQ%40mail.gmail.com
>
<https://groups.google.com/d/msgid/eiffel-users/CAN%3DG_amWC%3DEgRaoX1oT%2BhW
>
bUdUhd7G0B%3Dtrg15%3DdLoaV%2BKbZ%3DQ%40mail.gmail.com?utm_medium=email&utm_sou
> rce=footer> .
> --
> You received this message because you are subscribed to the Google Groups
> "Eiffel Users" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to eiffel-users...@googlegroups.com.
> To view this discussion on the web visit
>
> 74f0%24%40inf.ethz.ch.
Bertrand Meyer
-- BM
Eiffel Users
Louis M
Thanks. This is reassuring. I was wondering because in the past, both version was released at the same times.
Louis M
To view this discussion on the web visit https://groups.google.com/d/msgid/eiffel-users/3709c2f2-5973-4b95-82c9-bbb168690616n%40googlegroups.com.
Bertrand Meyer
Dear Ulrich,
> After browsing the Unicode tables a bit, I wonder about these operators:
> Should they all be implemented in the appropriate classes?:
∃∀∄∈∉∋∌∌⊂⊃⊄⊅⊆⊇⊈⊉⊊⊋
Indeed, class SUBSET of EiffelBase now has the basic membership and subset/superset operators as operator aliases. Not all of the above but the most common ones.
Also membership, in SET, but only ∋, as an operator alias for `has'. At the moment ∈ is not possible (see a discussion started by Jonathan Ostroff some time ago).
(The problem with ∈ is that in the simplest approach it would have to be in ANY, but that class should not depend on non-kernel classes such as SET.)
Union and intersection symbols È and Ç are not present because the corresponding features in SUBSET are procedures (used as e.g. a.merge (b), meaning a := a È b), I guess for performance reasons. Actually the function versions should be added (if only for purposes of writing contracts, but also for people preferring a functional style of programming), with the appropriate Unicode aliases.
∃ and ∀ are not operators (in the sense of denoting features of some class), but part of the syntax.
With best regards,
-- Bertrand Meyer
To view this discussion on the web visit https://groups.google.com/d/msgid/eiffel-users/5FF30946020000A10003DEF5%40gwsmtp.uni-regensburg.de.
Ulrich Windl
in
Nachricht <098701d6e2f2$03292740$097b75c0$@inf.ethz.ch>:
Dear Bertrand,
> ∃ and ∀ are not operators (in the sense of denoting features of some class),
> but part of the syntax.
parameters, a list of items, and a function accepting one item and returning a
Boolean, why aren't those operators (Obviously the function is applied to the
list until the final result had been determined)? I mean you need some Boolean
property and an iterable collection.
In case it was confusing, here is a sample quck and dirty implementation (in
Perl, because I was lazy):
~> cat /tmp/forall.pl
#!/usr/bin/perl
my @list = (1, 2, 3);
sub forall($@)
{
my ($fn, @list) = @_;
foreach (@list) {
return undef
unless ($fn->($_));
}
return 1;
}
print forall(sub ($) { $_[0] > 1 }, @list) ? 'true' : 'false', "\n";
print forall(sub ($) { $_[0] >= 1 }, @list) ? 'true' : 'false', "\n";
~> perl /tmp/forall.pl
false
true
Regards,
Ulrich