Activity for REDUCE

Francis Wright committed [r4240]

Website: revised the Obtaining REDUCE page, especially to add more about using Linux.

Posted by Francis Wright on 15 Oct 2017

Dream Scatter posted a comment on discussion Open Discussion

Greetings, I'd like to invite you to ponder the similarity between REDUCE and Panini's grammar: As in modern logic, considerable effort is expended distinguishing the ‘‘mention’’ of actual Sanskrit and the ‘‘use’’ of a formal grammar to describe and generate it. Pān: ini constructed a formal metalanguage for linguistic description by extending ordinary Sanskrit; this formalism then expresses the algorithms used to derive spoken expressions. For modern logicians the analogous task is organized through...

Posted by Dream Scatter on 13 Oct 2017

Arthur Norman committed [r4239]

CSL now uses a Mersenne Twister random number generator rather than the

Posted by Arthur Norman on 11 Oct 2017

Francis Wright committed [r4238]

Website: add a footer to all main pages that shows the date the page was last modified; use the SimplePie library locally to display RSS feeds.

Posted by Francis Wright on 8 Oct 2017

Rainer Schöpf committed [r4237]

PSL for 64bit Linux:

Posted by Rainer Schöpf on 7 Oct 2017

Dream Scatter posted a comment on discussion Developers

Aha, I see. Thanks for the clarification, I do not familiarity with your code but am trying to learn some things!

Posted by Dream Scatter on 4 Oct 2017

Rainer Schöpf posted a comment on discussion Developers

I believe you are mistaken: Eberhard's change was not to CSL or PSL, but to Reduce proper. Hence, the switch is available in both CSL and PSL Reduce. However, as I tried to explain in a previous post, PSL is not Unicode-ready, which means that printing of non-ASCII characters does not work as expected: Reduce (Free PSL version, revision 4209), 4-Oct-2017 ... 1: off nat; 2: on unicode_in_off_nat; 3: #alpha; + #beta;; α + β$ 4: ab + #beta;; β + ab$ 5: off unicode_in_off_nat; 6: ab + #beta;; !�!� +...

Posted by Rainer Schöpf on 4 Oct 2017

Dream Scatter posted a comment on discussion Developers

Thanks for your response. I believe Eberhard Schruefer already made the necessary modifications to CSL as explained in this post: https://sourceforge.net/p/reduce-algebra/discussion/899365/thread/5b0447af/#04ee This is the commit: https://sourceforge.net/p/reduce-algebra/code/4234/ However, it would be good to know if this switch will also be available in PSL

Posted by Dream Scatter on 4 Oct 2017

Francis Wright committed [r4236]

Website: fixed some broken links in the bibliography.

Posted by Francis Wright on 3 Oct 2017

Arthur Norman posted a comment on discussion Developers

greetings from longyearbyen where I am just back from 100% lack of connectivity. I can fix whatever is needed in CSL when I have unpacked some polar bear pics having got home later today. Arthur -------- Original message -------- From: Dream Scatter chakravala@users.sf.net Date: 29/09/2017 19:35 (GMT+01:00) To: "[reduce-algebra:discussion]" 899365@discussion.reduce-algebra.p.re.sf.net Subject: [reduce-algebra:discussion] Unicode or UTF8 support? That's great, thank you for the assistance. It is now...

Posted by Arthur Norman on 3 Oct 2017

Rainer Schöpf committed [r4235]

Updated list of known switches

Posted by Rainer Schöpf on 2 Oct 2017

Dream Scatter posted a comment on discussion Developers

Alright, in that case the name unicode_in_off_nat should be fine, since it is only used once. thanks

Posted by Dream Scatter on 30 Sep 2017

Eberhard Schruefer posted a comment on discussion Developers

The switch can be given any name as long as it doesn't clash with something already existing. Preferably, the name should be intuitive (off unat or utf_nat suggests to me that one would also turn off the unicode printing). At the moment the switch unicode_in_off_nat is not turned off when nat is turned on. It has no consequencies whatsoever when nat is turned on. So when you say 'on unicode_in_off_nat' you can toggle nat as many times as you want in your program and have always the unicode characters...

Posted by Eberhard Schruefer on 30 Sep 2017

Dream Scatter posted a comment on discussion Developers

Is it necessary for the unicode_in_off_nat switch to have such a long name? for a command like that, which might be used frequently, I would prefer to minimize the time it takes to stream the characters into the pipe could it be named to something like unat or utf_nat ? will unicode_in_off_nat be switched off when nat is turned on and off again or does it stay on the whole time? i.e. will I have to turn unicode_in_off_nat mode back on each time I enter and exit nat mode?

Posted by Dream Scatter on 30 Sep 2017

Dream Scatter posted a comment on discussion Developers

Thank you, I appreciate it!

Posted by Dream Scatter on 30 Sep 2017

Eberhard Schruefer posted a comment on discussion Developers

I just added a switch unicode_in_off_nat that prints unicode characters as such in 'off nat' mode. E.g. [eschruefer@herbrand ~]$ redcsl -w Reduce (Free CSL version, revision 4218), 30-Sep-17 ... 1: off nat; 2: on unicode_in_off_nat; 3: #alpha; + #beta;; α + β$ 4: ab + #beta;; ab + β$ 5: off unicode_in_off_nat; 6: ab + #beta;; ab + !#03b2;$ Eberhard

Posted by Eberhard Schruefer on 30 Sep 2017

Eberhard Schruefer committed [r4234]

Added switch unicode_in_off_nat to have unicode characters displayed as such when nat is off.

Posted by Eberhard Schruefer on 30 Sep 2017

Dream Scatter posted a comment on discussion Developers

Thanks for the explanation. There is one difference between the CSL and PSL that I can perceive though: with the "redcsl -w" mode I am able to input a character like Ω and it is recognized, while in PSL the character is not recognized as an input unless written out as #03a9; With CSL combined with my package I can apply Reduce to julia langauge expressions julia> :(Ω^2+2Ω+1) |> factor :((Ω + 1) ^ 2) internally, this command goes through a sequence like this: convert the julia expression object into...

Posted by Dream Scatter on 30 Sep 2017

Francis Wright committed [r4233]

Website: ensure that the text in the Download REDUCE from SourceForge box stays black.

Posted by Francis Wright on 30 Sep 2017

Rainer Schöpf posted a comment on discussion Developers

The exclamation mark is the escape character. See section 2.3 (Identifiers) of the Reduce manual. Currently, Unicode is not properly supported in Reduce: if you want to use non-ASCII characters in identifiers, you must escape them on input. You can use them in strings. There is no other way. CSL and PSL Reduce do not really differ in that respect. The Unicode characters are not altered: When the switch NAT is set to off, output is written so that it can be read by the Reduce reader without any changes....

Posted by Rainer Schöpf on 30 Sep 2017

Dream Scatter posted a comment on discussion Developers

Well, I did come up with a work around. When I want to parse a reduce expression into a julia expression, I just check if there is a "!#" and if there is, I evaluate it in nat mode first in order to convert the character to unicode. So I did work around the issue with the nat mode and solved it from the julia side by making extra calls to reduce in these cases. However, it would still help to know if my above question can be solved from the reduce side of pipe.

Posted by Dream Scatter on 29 Sep 2017

Dream Scatter posted a comment on discussion Developers

That's great, thank you for the assistance. It is now possible to use both CSL and PSL with my package, the option can be set using an environment variable in Julia, which I will explain in the upcoming documentation. However, there is one more issue to make the unicode characters in CSL fully compatible 1: off nat; Ω; !#03a9;$ For displaying in nat mode, there is no problem. however, when nat is turned off the unicode character is replaced with the !#xxxx; code This presents a problem, since now...

Posted by Dream Scatter on 29 Sep 2017

Francis Wright committed [r4232]

Website: added details of another book.

Posted by Francis Wright on 29 Sep 2017

Francis Wright committed [r4231]

Website: added a remark about running REDUCE on Windows under Cygwin.

Posted by Francis Wright on 29 Sep 2017

Rainer Schöpf posted a comment on discussion Developers

Proper support for multibyte characters in PSL is not an easy task, as the assumption of 1 character = 1 byte is implicit in PSL. What will probably be implemented is a way to print and read multibyte UTF-8 characters in a way that they appear to be single character objects. Rainer

Posted by Rainer Schöpf on 29 Sep 2017

Rainer Schöpf posted a comment on discussion Developers

int2id is guaranteed to work only for an argument in the range 0 .. 127. For the general case use int2wideid. Its argument is the Unicode codepoint (which is independent of the encoding). Since you can enter any Unicode character via its numeric codepoint, as either hexnumber #Xxxxx; or decimal number #Unnnn; you can input ASCII EOT (aka ctrl-D) as #U4; : symbolic write '!#U4; ; Rainer

Posted by Rainer Schöpf on 29 Sep 2017

Eberhard Schruefer posted a comment on discussion Developers

One possibility would be to go write int2id 4; Eberhard

Posted by Eberhard Schruefer on 29 Sep 2017

Dream Scatter posted a comment on discussion Developers

Also, how can I do something like 1: symbolic write(string 4); +++ Error undefined function (1 arg): string how to do it in CSL? this works fine in PSL. I need this as a delimiter for knowing when to stop reading the pipe.

Posted by Dream Scatter on 29 Sep 2017

Dream Scatter modified a comment on discussion Developers

Great! that also works for me with no problems. I was using the plain redcsl before. Is there any way of doing that with redpsl?

Posted by Dream Scatter on 29 Sep 2017

Dream Scatter posted a comment on discussion Developers

Great! that also works for me with no problems. I was using the plain redcsl before.

Posted by Dream Scatter on 29 Sep 2017

Eberhard Schruefer posted a comment on discussion Developers

Well, it works here for me with CSL. I can cut and pate and also send utf-8 characters over a pipe to Reduce. [eschruefer@herbrand ~]$ redcsl -w Reduce (Free CSL version, revision 4218), 20-Sep-17 ... 1: Φ; Φ 2: #Omega;; Ω The first is just the \Phi cut and pasted from your post into the terminal. The second is just to show that you get the correct symbol. How are you invoking CSL Reduce? Do you have the '-w' option as shown above on the command line? Eberhard

Posted by Eberhard Schruefer on 28 Sep 2017

Dream Scatter posted a comment on discussion Developers

No, it does not work for me. For example, in CSL it yields the wrong character n: #Omega;; Φ while in PSL it yields the correct character. This is with revision 4015 of reduce. Perhaps if I recompile with new version it will work. If I try to directly paste the character into CSL prompt, then it pastes as "\u03a9" So it seems that the UTF8 support is a bit quirky. For my purposes, it would be useful if I could send PSL an input with UTF8 characters in it via the pipe. This does not seem to be currently...

Posted by Dream Scatter on 28 Sep 2017

Eberhard Schruefer posted a comment on discussion Developers

That should work. Please try it with CSL to see if it makes a difference. Eberhard

Posted by Eberhard Schruefer on 28 Sep 2017

Eberhard Schruefer posted a comment on discussion Developers

Yes, it should also work with PSL. A unicode character in Reduce is represented as something that starts with # followed by a name or the numerical representation of an entity and is terminated by a semicolon.

Posted by Eberhard Schruefer on 28 Sep 2017

Dream Scatter posted a comment on discussion Developers

So it seems that outputting these characters is no problem for reduce, but when I try to input one of those characters I get 1: Ω;; Declare � operator ? (Y or N) is there any way to interpret an input that has Ω in it?

Posted by Dream Scatter on 28 Sep 2017

Dream Scatter posted a comment on discussion Developers

It also works in PSL too then? Why does it require 2 semicolons?

Posted by Dream Scatter on 27 Sep 2017

Dream Scatter posted a comment on discussion Developers

Oh I see. However, my program uses PSL but that is useful to know about the CSL.

Posted by Dream Scatter on 27 Sep 2017

Eberhard Schruefer posted a comment on discussion Developers

In CSL based Reduce you can do for example 2: #Omega;; Ω 3: #03a9;; Ω Note the semicolon after the name. Here is another example 6: #alpha; + #beta;; α + β Eberhard

Posted by Eberhard Schruefer on 27 Sep 2017

Dream Scatter posted a comment on discussion Developers

Greetings, Is there any support for unicode or utf8 characters planned? It could be useful for the Reduce.jl package, since it currently requires an extra text-substituation layer to translate some greek characters from a julia expression to a reduce expression, as julia expressions natively support it but reduce does not. This is not currently a huge issue, but having full character support would be very nice. Regards,

Posted by Dream Scatter on 27 Sep 2017

Dream Scatter posted a comment on discussion Help

Thanks, that is indeed what I wanted. I figured there must be some sort of hold command, but I wasn't sure how to find the syntax for it. !*hold(e**(i*pi))$ ws where i => im; im*pi e

Posted by Dream Scatter on 27 Sep 2017

Eberhard Schruefer posted a comment on discussion Help

I don't know if this is what you want, but you can use the hold operator to prevent evalauation. 1: !*hold(e**(i*pi)); i*pi e Eberhard

Posted by Eberhard Schruefer on 26 Sep 2017

Dream Scatter modified a comment on discussion Help

Well, I kind of solved the issue from the previous post by adjusting my parser, but it would still be interesting to know if that is possible or if it is possible to send reduce a statement like e*(ipi) without evaluating, but simply returning the statement unevaluated as "e*(ipi)"

Posted by Dream Scatter on 25 Sep 2017

Dream Scatter posted a comment on discussion Help

Well, I kind of solved the issue from the previous post by adjusting my parser, but it would still be interesting to know if that is possible or if it is possible to send reduce a statement like e*(ipi) without evaluating, but simply returning the statement unevaluated as "e*(ipi)"

Posted by Dream Scatter on 25 Sep 2017

Dream Scatter posted a comment on discussion Help

Is there a way to evaluate expression$ ws where i => im; without the "expression" part being evaluated? For example, consider the reduce expression "x:= i+1", and I want to convert that expression into "x:=im+1", is there any way to apply the substitution on some reduce expression without evaluating it? This is could be convenient, if possible.

Posted by Dream Scatter on 25 Sep 2017

Dream Scatter modified a comment on discussion Help

Thanks for the response. Yes, that seems to do the job for me! Also, now I see that "im" does not give "i" in redpsl. This is because I had a command after the initialization of redpsl that sets im:=i.

Posted by Dream Scatter on 25 Sep 2017

Dream Scatter posted a comment on discussion Help

Thanks for the response. Yes, that seems to do the job for me! Also, now I see that "im" does not give "i" in redpsl. This is an unintended side-effect of the "reduce> " repl when using Reduce.jl in julia. I will have to look into the bug.

Posted by Dream Scatter on 25 Sep 2017

Francis Wright posted a comment on discussion Help

A somewhat inelegant way to have REDUCE output i as im would be instead of evaluating expression; to evaluate expression$ ws where i => im; I don't think that has any unexpected side-effects. Maybe you could hide this input change within the Julia interface. REDUCE doesn't interpret im as i on input by default as far as I can see. (And I wouldn't expect it to.) If I evaluate im^2 then it stays as im^2, whereas i^2 evaluates to -1. However, you should be able to use the where operator on input in...

Posted by Francis Wright on 25 Sep 2017

Raffaele Vitolo committed [r4230]

new module for evolutionary vector fields and their bracket

Posted by Raffaele Vitolo on 24 Sep 2017

Dream Scatter posted a comment on discussion Help

Greetings, I am developing the Reduce.jl interface for the julia language: https://github.com/chakravala/Reduce.jl It would be of great convenience for my application if the imaginary output of reduce could be represented using "im" instead of "i", since I need to convert this value between the two languages. For most things I am able to find a work around, but for this particular value it is very difficult to do so. It seems that Reduce already is able to interpret "im" as the "i" value. Is there...

Posted by Dream Scatter on 24 Sep 2017

Rainer Schöpf committed [r4229]

Restore accidentally overwritten pslmem.sh script

Posted by Rainer Schöpf on 24 Sep 2017

Raffaele Vitolo committed [r4228]

corrected a misprint

Posted by Raffaele Vitolo on 23 Sep 2017

Raffaele Vitolo committed [r4227]

This is an update of CDE to version 2.1. Main changes are:

Posted by Raffaele Vitolo on 23 Sep 2017

Raffaele Vitolo committed [r4226]

fixed one example program

Posted by Raffaele Vitolo on 23 Sep 2017

Francis Wright committed [r4225]

Website: added more details for installing and running REDUCE on Windows.

Posted by Francis Wright on 23 Sep 2017

REDUCE released /snapshot_2017-09-22/Reduce-svn4224-src.tar.gz

Posted by REDUCE on 22 Sep 2017

REDUCE released /snapshot_2017-09-22/linux-tar/reduce-csl_2017-09-22_i386.tgz

Posted by REDUCE on 22 Sep 2017

REDUCE released /snapshot_2017-09-22/linux-rpm/reduce-psl-2017_09-22.x86_64.rpm

Posted by REDUCE on 22 Sep 2017

REDUCE released /snapshot_2017-09-22/linux-deb/reduce-complete_2017-09-22_i386.deb

Posted by REDUCE on 22 Sep 2017

REDUCE released /snapshot_2017-09-22/linux-deb/reduce-psl_2017-09-22_i386.deb

Posted by REDUCE on 22 Sep 2017

REDUCE released /snapshot_2017-09-22/linux-rpm/reduce-complete-2017_09-22.x86_64.rpm

Posted by REDUCE on 22 Sep 2017

REDUCE released /snapshot_2017-09-22/Reduce-Setup_2017-09-22.exe

Posted by REDUCE on 22 Sep 2017

REDUCE released /snapshot_2017-09-22/Reduce-snapshot_2017-09-22.dmg

Posted by REDUCE on 22 Sep 2017

REDUCE released /snapshot_2017-09-22/linux-deb/reduce-addons_2017-09-22_amd64.deb

Posted by REDUCE on 22 Sep 2017

REDUCE released /snapshot_2017-09-22/linux-tar/reduce-common_2017-09-22_all.tgz

Posted by REDUCE on 22 Sep 2017

REDUCE released /snapshot_2017-09-22/linux-deb/reduce-csl_2017-09-22_amd64.deb

Posted by REDUCE on 22 Sep 2017

REDUCE released /snapshot_2017-09-22/linux-tar/reduce-freestanding_2017-09-22_x86_64.tar.bz2

Posted by REDUCE on 22 Sep 2017

REDUCE released /snapshot_2017-09-22/linux-rpm/reduce-common-2017_09-22.noarch.rpm

Posted by REDUCE on 22 Sep 2017

REDUCE released /snapshot_2017-09-22/linux-rpm/reduce-psl-2017_09-22.i386.rpm

Posted by REDUCE on 22 Sep 2017

REDUCE released /snapshot_2017-09-22/linux-tar/reduce-psl_2017-09-22_amd64.tgz

Posted by REDUCE on 22 Sep 2017

REDUCE released /snapshot_2017-09-22/linux-tar/reduce-csl_2017-09-22_amd64.tgz

Posted by REDUCE on 22 Sep 2017

REDUCE released /snapshot_2017-09-22/linux-rpm/reduce-addons-2017_09-22.i386.rpm

Posted by REDUCE on 22 Sep 2017

REDUCE released /snapshot_2017-09-22/linux-rpm/reduce-csl-2017_09-22.x86_64.rpm

Posted by REDUCE on 22 Sep 2017

REDUCE released /snapshot_2017-09-22/linux-deb/reduce-addons_2017-09-22_i386.deb

Posted by REDUCE on 22 Sep 2017

REDUCE released /snapshot_2017-09-22/linux-tar/reduce-addons_2017-09-22_amd64.tgz

Posted by REDUCE on 22 Sep 2017

REDUCE released /snapshot_2017-09-22/linux-tar/reduce-freestanding_2017-09-22_i686.tar.bz2

Posted by REDUCE on 22 Sep 2017

REDUCE released /snapshot_2017-09-22/linux-tar/reduce-complete_2017-09-22_i386.tgz

Posted by REDUCE on 22 Sep 2017

REDUCE released /snapshot_2017-09-22/linux-tar/reduce-addons_2017-09-22_i386.tgz

Posted by REDUCE on 22 Sep 2017

REDUCE released /snapshot_2017-09-22/linux-rpm/reduce-complete-2017_09-22.i386.rpm

Posted by REDUCE on 22 Sep 2017

REDUCE released /snapshot_2017-09-22/linux-deb/reduce-psl_2017-09-22_amd64.deb

Posted by REDUCE on 22 Sep 2017

REDUCE released /snapshot_2017-09-22/linux-deb/reduce-common_2017-09-22_all.deb

Posted by REDUCE on 22 Sep 2017

REDUCE released /snapshot_2017-09-22/linux-deb/reduce-csl_2017-09-22_i386.deb

Posted by REDUCE on 22 Sep 2017

REDUCE released /snapshot_2017-09-22/linux-rpm/reduce-csl-2017_09-22.i386.rpm

Posted by REDUCE on 22 Sep 2017

REDUCE released /snapshot_2017-09-22/linux-tar/reduce-psl_2017-09-22_i386.tgz

Posted by REDUCE on 22 Sep 2017

REDUCE released /snapshot_2017-09-22/linux-rpm/reduce-addons-2017_09-22.x86_64.rpm

Posted by REDUCE on 22 Sep 2017

REDUCE released /snapshot_2017-09-22/linux-deb/reduce-complete_2017-09-22_amd64.deb

Posted by REDUCE on 22 Sep 2017

REDUCE released /snapshot_2017-09-22/linux-tar/reduce-complete_2017-09-22_amd64.tgz

Posted by REDUCE on 22 Sep 2017

Rainer Schöpf committed [r4224]

Building PSL REDUCE on Windows: make sure that the 64bit version is built

Posted by Rainer Schöpf on 22 Sep 2017

Arthur Norman committed [r4223]

Another place where CSL had not kept a value safe from the garbage collector.

Posted by Arthur Norman on 20 Sep 2017

Arthur Norman committed [r4222]

Fix (I hope) a bug in the bytecode interpreter where over some tail-calls a

Posted by Arthur Norman on 19 Sep 2017

Francis Wright committed [r4221]

Website: renamed reduce-ide.php to index.php to correct a potentially broken link.

Posted by Francis Wright on 19 Sep 2017

Arthur Norman committed [r4220]

Change associated with when "make c-code" gets activated. This may be important

Posted by Arthur Norman on 19 Sep 2017

Arthur Norman committed [r4219]

I have re-worked the CSL code that translated from Lisp into C++ so that the

Posted by Arthur Norman on 18 Sep 2017

Francis Wright committed [r4218]

Website styling changes: slight drop shadow and animation in navigation panel, serif font for major headings, better heading levels on About page, removed small caps on Help Wanted page, fonts and animation on Books page.

Posted by Francis Wright on 18 Sep 2017

Arthur Norman committed [r4217]

Slight improvement so that if CSL or PSL has not been configured

Posted by Arthur Norman on 18 Sep 2017

Rainer Schöpf committed [r4216]

Improve README for Linux build process: describe which packages are required.

Posted by Rainer Schöpf on 18 Sep 2017

Arthur Norman committed [r4215]

CPPFLAGS was sometimes restored from OLDLDFLAGS rather then OLDCPPFLAGS

Posted by Arthur Norman on 17 Sep 2017

Eberhard Schruefer committed [r4214]

Fix a typo.

Posted by Eberhard Schruefer on 16 Sep 2017

Arthur Norman committed [r4213]

Small changes...

Posted by Arthur Norman on 16 Sep 2017

Arthur Norman committed [r4212]

A typo in this script, whihc could mislead people but ought not to

Posted by Arthur Norman on 16 Sep 2017

Arthur Norman committed [r4211]

I had included twice, whihc was silly.

Posted by Arthur Norman on 16 Sep 2017