Report problems with Grapher, CAS and User Interface

[compsystems]

Hello, In this thread I will report problem situations with CAS, as well as with the Grapher and the User Interface.

I hope that there might be a working team that can solve these problems

Sorry, my language is Spanish, this is an AI translation.

=========================
P1_CAS: In the History view the hp-prime CAS operation is not synchronized with the system UI.


Entering the following algebraic expression works logically correctly, but when retrieving the input expression from the history to the input expression line, some arguments are not printed legibly because there is a vertical shift up of the arguments.

a*x^2+b*x+c | a = -6,b = -4,c = 2; [ENTER] returns -6*x^2-4*x+2

By math rule everything to the left of the symbolic operator | is considered the expression to be evaluated and everything to the right are the values of the substitution.

even if you use parentheses to say what the arguments are also fails.

a*x^2+b*x+c | (a = -6,b = -4,c = 2); [ENTER] returns -6*x^2-4*x+2

does not fail if rectangular brackets [] or {} are used, but it should work only with () curved brackets. In addition, the semicolon (; ) infers the end of the argument, symbol that is also vertically defaced

a*x^2+b*x+c | [a = -6,b = -4,c = 2]; [ENTER] returns -6*x^2-4*x+2

[compsystems]

P1_UI: The (-) sign when placed in the history it is not centered with respect to the horizontal bar of the division, it is always above it, by law of signs the print must be aligned with the horizontal bar of the division.

[KeithB]

Try the other operator. The prime has *2* negation operators. The "minus" and unary minus as selected by the +/- key.

The size and vertical justification are used to tell them apart.

[compsystems]

P2_UI: The following situation happens very rarely

When executing the following algebraic expression, the exact coefficient numbers are changed to decimal numbers, i.e. the UI adds a dot.

subst(a*x^2+b*x+c, [a = -6,b = -4,c = 2]); [ENTER] returns -6*x^2-4*x+2
subst(a*x^2+b*x+c, {a = -6,b = -4,c = 2}); [ENTER] returns -6*x^2-4*x+2

Run it a few times and you will most likely see the following:
-6. *x^2 -4. *x + 2.0

[compsystems]

P3_UI: When you copy symbols to the character input line, the bottom part is hidden. Configure with minifont



Another improvement would be to store the last position of the scanned character when the character table is reopened.

[roadrunner]

(04-02-2024 01:46 PM)compsystems Wrote:  1: When entering the following algebraic expression it works fine, but when retrieving the input expression in the input expressions line, certain arguments are not printed legibly because there is a vertical offset in the arguments.

(a*x^2+b*x+c) | (a = -6,b = -4,c = 2) [ENTER] returns -6*x^2-4*x+2

everything on the left side of the symbolic operator | It is considered the expression to be evaluated and everything on the right side is the values of the substitution.

The Where function requires a list for multiple variables. Use this:

a*x^2+b*x+c|({a = -6,b = -4,c = 2}

instead of this:

(a*x^2+b*x+c) | (a = -6,b = -4,c = 2)

and that behavior is not observed.

-road

[compsystems]

The following input statement is interpreted as everything to the left of the operator | is the input expression and everything to the right is the substitutions

a*x^2+b*x+c | a=-6,b=-4,c=2 [Enter] -6*x^2-4*x+2 ok

with curved parentheses should be enough

a*x^2+b*x+c | (a=-6,b=-4,c=2) [Enter] -6*x^2-4*x+2 ok

or

(a*x^2+b*x+c) | (a=-6,b=-4,c=2) [Enter] -6*x^2-4*x+2 ok

[jte]

(04-02-2024 01:46 PM)compsystems Wrote:  Hello, CAS operation on hp-prime is not synchronized with the system GUI. I will report these problematic situations in this thread.
⋮

Thanks! I thought I'd try to add some comments from this thread to the bug tracker I've set up.

Quote:⋮
2: The following situation happens very rarely

When executing the following algebraic expression, the exact coeficient numbers are changed to decimal numbers, that is, a point is added.

subst(a*x^2+b*x+c,[a = -6,b = -4,c = 2]) [ENTER] returns -6*x^2-4*x+2

Run it a few times and you will most likely see the following:

-6. *x^2 -4. *x + 2.0

I've tried to recreate this on both HP Prime Virtual Calculators and on a physical HP Prime, with a variety of different settings for Entry mode and Simplification, but could not reproduce it.

While writing this post (after writing the sentence just before this one), I did try the above subst, but with an "a" variable set to "-6.", and then (repeatedly) got the -6. in the output.

Some more information might help. Are you seeing this with behaviour with the latest public release (2.1.14732 (2023 04 21))? (Does it occur even with a completely fresh / blank calculator?)

[dg1969]

(04-18-2024 01:02 AM)jte Wrote:  Are you seeing this with behaviour with the latest public release (2.1.14732 (2023 04 21))?

Hi,
I haven't seen this version before! Did I miss something? My physical HP G2 displays:
2.1.14730 (2023 04 13)

[parisse]

subst does not quote it's arguments, therefore if a is assigned say to 2
subst(a*x^2+b*x+c,[a = -6,b = -4,c = 2])
will eval a*x^2+b*x+c to 2*x^2+b*x+c, as well as a=-6 to 2=-6, and a will not be replaced by -6.
A warning is displayed in the terminal.

[compsystems]

P4_UI: Another problem is that when there is an expression in the input line, and if you later go to the [help] menu and search for something, when you return the expression in the input line is cut off.

[komame]

(04-08-2024 08:25 PM)compsystems Wrote:  When you copy symbols to the character input line, the bottom part is hidden.



Another improvement would be to store the last position of the scanned character when the character table is reopened.

I managed to reproduce this issue. It's related to the font size settings in Home Settings on page 2. This issue affects every edit field (anywhere, in any view/application where edit fields are present) when the font is set to "Small Font". For the other font sizes, the problem does not occur.

(04-19-2024 12:51 AM)compsystems Wrote:  In HP-Prime, sometimes the outputs, integers, have a dot added.

Another problem is that when there is an expression in the input line, and if you later go to the [help] menu and search for something, when you return the expression in the input line is cut off.

I managed to reproduce this bug as well. I've filed a ticket for this issue in the bug tracker.

[komame]

(04-20-2024 06:29 AM)komame Wrote:  [...] I've filed a ticket for this issue in the bug tracker.

..and the bug got fixed!

Thank you, Jeff.

[compsystems]

P5_UI: When typing '(x^m)^n^p' and executing the input, and then retrieving it on the input line of the expression, when trying to step down from the exponent 'n' to include a parenthesis, the cursor location jumps to the end of the expression, it must be located before the variable 'p' not after
'(x^m)^n|^p' => '(x^m)^n^p|'

[compsystems]

P6_UI:

Typing (x^m)^n:; and executing the input, then retrieving it on the input line of the expression and executing it again does not interpret the nodisp() function

The nodisp() function is synonymous with ( :; ), that is, it does not display the result of the input.

[Albert Chan]

(05-02-2024 11:57 AM)compsystems Wrote:  Typing (x^m)^n:; and executing the input, then retrieving it on the input line of the expression and executing it again does not interpret the nodisp() function

Solution: add a user-defined nodisp function.

> nodisp := z->"Done"

[compsystems]

This is not the only function that is not well interpreted, there are several similar ones.

[jte]

(05-02-2024 11:51 AM)compsystems Wrote:  Hello
When typing '(x^m)^n^p' and executing the input, and then retrieving it on the input line of the expression, when trying to step down from the exponent 'n' to include a parenthesis, the cursor location jumps to the end of the expression, it must be located before the variable 'p' not after
'(x^m)^n|^p' => '(x^m)^n^p|'

A similar sort of issue arises if one enters 1^2^3 and then tries to select the 1^2 portion. It is fairly typical for equation editors to apply / enforce / create / assume some structure to the input that is not visibly shown (e.g., with 1^2^3 there could be a tree structure in memory; the top node being "^" with a left child node of "1" and a right child node of "2^3"). I’m glad the the HP Prime editor does less of this than some editors I’ve experienced.

If one instead enters (1^2)^3 and then removes the parentheses, one can then reinsert them around the 1^2 (the top "^" node has "1^2" as the left child and "3" as the right child). (With this structure, selecting and copying the visible 2^3 ends up copying "^2^3".) Note that this equation editor structure does not override evaluation rules (one can try 3^3^3 with both editing structure — direct entry and by entering (3^3)^3 and then removing the parentheses).

Playing around with this just now, I do see something I’ll add to the bug tracker. (Enter "(1^2)" and then delete just the "1".)

[compsystems]

Sometimes the TRUE/FALSE logic output is set to a numeric value, creating confusion.

[compsystems]

Graph_01: An improvement in the graphic engine:

The cursor symbol (+) should change color not only in the intersection, but also in all its shape, because it is not well displayed (it is almost hidden).