Bug (or wrong use?) of 'At()' - Inconsistent results - Сообщения
#1 Опубликовано: 12.03.2019 06:54:06
Hi,
I starting with smath and still try easy examples.
Is found the results of at() to be inconstistent,
but it may be an bad understandig. Could someone review
my appended file?
Many thanks, Johannes
Bug in at.sm
I starting with smath and still try easy examples.
Is found the results of at() to be inconstistent,
but it may be an bad understandig. Could someone review
my appended file?
Many thanks, Johannes
Bug in at.sm
#2 Опубликовано: 12.03.2019 07:08:03
It is by design. If you give multiple replacement rules for the same variable ( x=a then x=b ) only the first is used (tbf there's no filter on how many times you can enter the same variable name on the LHS, thus the function tries even to evaluate x in b, but x doesn't exists anymore since is already evaluated in a)
If you like my plugins please consider to support the program buying a license; for personal contributions to me: paypal.me/dcprojects
#3 Опубликовано: 12.03.2019 07:12:20
But why is it different for the two definitions of L?
Or does it look equal for you? (Than it may only occur for me.)
Or does it look equal for you? (Than it may only occur for me.)
#4 Опубликовано: 12.03.2019 07:23:08
Sorry, I'm on a linux machine right now and I missed a part of the document, my comment was about the first yellow region without maxima. 
If you like my plugins please consider to support the program buying a license; for personal contributions to me: paypal.me/dcprojects
#5 Опубликовано: 12.03.2019 08:35:59
Here comes an image. This may show up differences with smath or maxima versions.
#6 Опубликовано: 12.03.2019 08:51:00
L is differently defined in the two versions (use the dynamic assistant by mouse hovering over the definitions to see the difference)
in the second case, L isnt't stored in evaluated shape, thus vectorize() doesn't recognise the expression as vector and executes the expression only once.
in the second case, L isnt't stored in evaluated shape, thus vectorize() doesn't recognise the expression as vector and executes the expression only once.
Martin Kraska
Pre-configured portable distribution of SMath Studio: https://en.smath.info/wiki/SMath%20with%20Plugins.ashx
#7 Опубликовано: 12.03.2019 09:24:44
OK. I see.
But how can i get L stored in evaluated shape?
In your example:
it works nice.
If i try the same
it does not. Did i miss an configuration?
But how can i get L stored in evaluated shape?
In your example:
it works nice.
If i try the same
it does not. Did i miss an configuration?
#8 Опубликовано: 12.03.2019 10:19:30
it does not. Did i miss an configuration?
... plot first, solve(,,,,) native Smath.
#9 Опубликовано: 12.03.2019 10:35:02
This is the numerical solution.
My question is about assignin symbolic results.
In the example you see mkraska using L:=Solve(,), and assigning the results.
In my try, you see the result is not assigned as expected,
last line shows
x.1= x=result
but it should be
x.1=result.
My question is about assignin symbolic results.
In the example you see mkraska using L:=Solve(,), and assigning the results.
In my try, you see the result is not assigned as expected,
last line shows
x.1= x=result
but it should be
x.1=result.
#10 Опубликовано: 12.03.2019 11:40:18
I found the difference between Program Versions.
I tried V0.99.6884 and all works as expected.
L is now assigned identically for both definitions.
The bug may be not located in at().
I tried V0.99.6884 and all works as expected.
L is now assigned identically for both definitions.
The bug may be not located in at().
#11 Опубликовано: 12.03.2019 12:21:16
It seems that the evaluation logic in the plug-in interface has changed.
The function Solve() indeed shows a different behaviour (not evaluated/simplified when used in definitions, as if it was behind a line operator)
Other functions like int() don't seem to be affected by the problem.
Blatt39.sm
The function Solve() indeed shows a different behaviour (not evaluated/simplified when used in definitions, as if it was behind a line operator)
Other functions like int() don't seem to be affected by the problem.
Blatt39.sm
Martin Kraska
Pre-configured portable distribution of SMath Studio: https://en.smath.info/wiki/SMath%20with%20Plugins.ashx
#12 Опубликовано: 12.03.2019 16:53:29
This is the numerical solution.
My question is about assignin symbolic results.
In the example you see mkraska using L:=Solve(,), and assigning the results.
In my try, you see the result is not assigned as expected,
Users keep complaining about Maxima from SS versions/versions.
Maple/Smath are more stable companions.
#13 Опубликовано: 12.03.2019 18:02:12
Users keep complaining about Maxima from SS versions/versions.
Maple/Smath are more stable companions.
Jean, I know that many users don't care, but once you vote for maple, can you tell me under what license it is distributed?
Is it legally safe to use it in public service like university? Can it be recommended under the light of respect for IP?
Martin Kraska
Pre-configured portable distribution of SMath Studio: https://en.smath.info/wiki/SMath%20with%20Plugins.ashx
#14 Опубликовано: 12.03.2019 21:32:35
Jean, I know that many users don't care, but once you vote for maple, can you tell me under what license it is distributed?
Is it legally safe to use it in public service like university? Can it be recommended under the light of respect for IP?
Martin, you have several specific questions that only Andrey can answer.
Maxima was formerly MacSyma, originally piloted by the MIT, conjointly
MIT/University of Toronto. MIT pedaled too long in butter, so that this
branch of research founded itself "Maple Waterloo".
Most of the stuff was borrowed from book.
My understanding is that: part of Maple is courtesy to Smath whereas
Smath is freeware. Viacheslav is the guru Maple/Smath, is it ?
Probably same as the Matlab "mwXXX" ... odesolvers ... etc.
Jean
#15 Опубликовано: 12.03.2019 23:45:34
Hi. The line statement seems to correct all issues, except the vectorization one.
Best regards.
Alvaro.
Best regards.
Alvaro.
-
Новые сообщения
-
Нет новых сообщений