Sets of labels for a node

[GrimPixel]

There is a possibility of optimisation: aggregating formulae in the same structure.
For example, the node （Ａ＋Ｂ）Ｃ／２, or if following the Python 2's naming rule opApBcpCs2, which represents $X = (A+B) C/2$:

• area of a trapezoid: $A = (a+b) h/2$
• displacement in an uniformly accelerated linear motion $x = (v_0+v_t) t/2$

I hope to find a way to circumvent that naming restriction, so the node names are easily readable.

They can be aggregated to one node with two sets of labels in different languages:

labels = [
    ['area of a trapezoid', 'upper base', 'lower base', 'height'], # eng-Latn-US
    ['梯形面积', '上底', '下底', '高'], # cmn-Hans-CN
    ['A,S', 'a', 'b', 'h'], # zxx-Zsym-ZZ

    ['displacement of a uniformly accelerated linear motion', 'start velocity', 'end velocity', 'time'], # eng-Latn-US
    ['匀加速直线运动位移', '初速度', '末速度', '时间'],  # cmn-Hans-CN
    ['x,s', 'v_0', 'v_t', 't'], # zxx-Zsym-ZZ
]

Labels can be searched as well.

It saves plenty of work if implemented.

[GrimPixel]

There is an alternative solution: allow functions to have aliases. They will look like this:
X_area_of_a_trapezoid__A_upper_base__B_lower_base__C_height
X_梯形面积__A_上底__B_下底__C_高
This still requires lifting the restriction on function names and is less intuitive.

[GrimPixel]

It is great to know that you are hooked, too. Let me summerise all proposals.

each file contains one unique equation

node names are the output and inputs

folders for files:

• area_of_a_trapezoid___upper_base__lower_base__height
• displacement_of_a_uniformly_accelerated_linear_motion___start_velocity__end_velocity__time

file contents:

def area_of_a_trapezoid___upper_base__lower_base__height(a:float=0.0, b:float=0.0, h:float=0.0) -> [{"name": "A", "type": float}]
	return (a+b)*h/2

def displacement_of_a_uniformly_accelerated_linear_motion___start_velocity__end_velocity__time(v_0:float=0.0, v_t:float=0.0, t:float=0.0) -> [{"name": "x", "type": float}]
	return (v_0+v_t)*t/2

node names are outputs with numbers

folders for files:

• area_of_a_trapezoid_1
• displacement_of_a_uniformly_accelerated_linear_motion_1

file contents:

def area_of_a_trapezoid_1(a:float=0.0, b:float=0.0, h:float=0.0) -> [{"name": "A", "type": float}]
	return (a+b)*h/2

def displacement_of_a_uniformly_accelerated_linear_motion_1(v_0:float=0.0, v_t:float=0.0, t:float=0.0) -> [{"name": "x", "type": float}]
	return (v_0+v_t)*t/2

node names are the output and inputs, represented by symbols

folders for files:

• X_area_of_a_trapezoid___A_upper_base__B_lower_base__C_height
• X_displacement_of_a_uniformly_accelerated_linear_motion___A_start_velocity__B_end_velocity__C_time

file contents:

def area_of_a_trapezoid___upper_base__lower_base__height(A:float=0.0, B:float=0.0, C:float=0.0) -> [{"name": "X", "type": float}]
	return (A+B)*C/2

def displacement_of_a_uniformly_accelerated_linear_motion___start_velocity__end_velocity__time(A:float=0.0, B:float=0.0, C:float=0.0) -> [{"name": "X", "type": float}]
	return (A+B)*C/2

each file contains one unique equation calculation

node names are fullwidth characters of parameters (invalid)

folders for files:

• （Ａ＋Ｂ）Ｃ／２

file contents:

def （Ａ＋Ｂ）Ｃ／２(A:float=0.0, B:float=0.0, C:float=0.0) -> [{"name": "X", "type": float}]
	return (A+B)*C/2

node names are the calculation with text with escape characters (here XML/HTML)

folders for files:

• lpar_A_plus_B_rpar_C_sol_2

file contents:

def lpar_A_plus_B_rpar_C_sol_2(A:float=0.0, B:float=0.0, C:float=0.0) -> [{"name": "X", "type": float}]
	return (A+B)*C/2

node names are texts in a natural or constructed language (here: modified Lojban)

folders for files:

• vei_ay0_suhi_by0_veho_cy0_fihu_re

file contents:

def vei_ay0_suhi_by0_veho_cy0_fihu_re(A:float=0.0, B:float=0.0, C:float=0.0) -> [{"name": "X", "type": float}]
	return (A+B)*C/2

node names are the hash value with an underscore (here: MD5)

folders for files:

• _4fe252e3443e2f8e71b0a96ba938959d

file contents:

def _4fe252e3443e2f8e71b0a96ba938959d(A:float=0.0, B:float=0.0, C:float=0.0) -> [{"name": "X", "type": float}]
	return (A+B)*C/2

each derived file contains one unique equation, imported from a base file

folders for files:

• _4fe252e3443e2f8e71b0a96ba938959d
• area_of_a_trapezoid___upper_base__lower_base__height
• displacement_of_a_uniformly_accelerated_linear_motion___start_velocity__end_velocity__time

file contents:

def _4fe252e3443e2f8e71b0a96ba938959d(A:float=0.0, B:float=0.0, C:float=0.0) -> [{"name": "X", "type": float}]
	return (A+B)*C/2

import _4fe252e3443e2f8e71b0a96ba938959d._main_._4fe252e3443e2f8e71b0a96ba938959d
def area_of_a_trapezoid___upper_base__lower_base__height(a:float=0.0, b:float=0.0, h:float=0.0, b:float=0.0) -> [{"name": "A", "type": float}]
	return _4fe252e3443e2f8e71b0a96ba938959d(a, b, h)

import _4fe252e3443e2f8e71b0a96ba938959d._main_._4fe252e3443e2f8e71b0a96ba938959d
def displacement_of_a_uniformly_accelerated_linear_motion___start_velocity__end_velocity__time(v_0:float=0.0, v_t:float=0.0, t:float=0.0) -> [{"name": "x", "type": float}]
	return _4fe252e3443e2f8e71b0a96ba938959d(v_0, v_t, t)

each file contains forms of similar formulae

folders for files:

• area_of_a_trapezoid__upper_base__lower_base__height
• displacement_of_a_uniformly_accelerated_linear_motion__start_velocity__end_velocity__time

file contents:

def area_of_a_trapezoid__upper_base__lower_base__height(A:float=0.0, a:float=0.0, b:float=0.0, h:float=0.0)
	if A is null and null not in [a, b, h]:
		return (a+b)*h/2
	elif a is null and null not in [A, b, h]:
		return 2*A/h-b
	elif b is null and null not in [A, a, h]:
		return 2*A/h-a
	elif h is null and null not in [A, a, b]:
		return 2*A/(a+b)

def displacement_of_a_uniformly_accelerated_linear_motion__start_velocity__end_velocity__time(x:float=0.0, v_0:float=0.0, v_t:float=0.0, t:float=0.0)
	if x is null and null not in [v_0, v_t, t]:
		return (v_0+v_t)*t/2
	elif v_0 is null and null not in [x, v_t, t]:
		return 2*x/t-v_t
	elif v_t is null and null not in [x, v_0, t]:
		return 2*x/t-v_0
	elif t is null and null not in [x, v_0, v_t]:
		return 2*x/(v_0+v_t)

each file contains forms of similar formulae calculation, imported from a base file, with a name of hash value (here: MD5)

folders for files:

• _7e0bb96c9462f6a2e4d921dc491fb0b7
• area_of_a_trapezoid__upper_base__lower_base__height
• displacement_of_a_uniformly_accelerated_linear_motion__start_velocity__end_velocity__time

file contents:

def _7e0bb96c9462f6a2e4d921dc491fb0b7(A:float=0.0, B:float=0.0, C:float=0.0, D:float=0.0) -> [{"name": "X", "type": float}]
    if A is null and null not in [B, C, D]:
		return (B+C)*D/2
	elif B is null and null not in [A, C, D]:
		return 2*A/D-C
	elif C is null and null not in [A, B, D]:
		return 2*A/D-B
	elif D is null and null not in [A, B, C]:
		return 2*A/(B+C)

import _7e0bb96c9462f6a2e4d921dc491fb0b7._main_._7e0bb96c9462f6a2e4d921dc491fb0b7
def area_of_a_trapezoid__upper_base__lower_base__height(A:float=0.0, a:float=0.0, b:float=0.0, h:float=0.0)
    return _7e0bb96c9462f6a2e4d921dc491fb0b7(A, a, b, h)

import _7e0bb96c9462f6a2e4d921dc491fb0b7._main_._7e0bb96c9462f6a2e4d921dc491fb0b7
def displacement_of_a_uniformly_accelerated_linear_motion__start_velocity__end_velocity__time(x:float=0.0, v_0:float=0.0, v_t:float=0.0, t:float=0.0)
    return _7e0bb96c9462f6a2e4d921dc491fb0b7(x, v_0, v_t, t)

[GrimPixel]

There are dimensions:

naming

1. output with numbering
2. output and input with text in a natural or constructed language
3. input with fullwidth characters
4. input with text with escape characters
5. input with a hash value with an underscore
6. the other side of 0 of a formula with text with escape characters
7. the other side of 0 of a formula with a hash value with an underscore

meaning of a node

1. a unique formula
2. a unique calculation for similar formulae
3. all forms of a unique formula
4. all forms of a unique calculation for similar formulae

dependency

1. files are independent from each other
2. some files are the base and others depend on them

labeling for node names, output and input names

1. not applied
2. applied

I tend to choose the combination that represents the nature of calculation to my understanding, so I prefer the combination of “input with text with escape characters (XML/HTML)”, “a unique calculation for similar formulae”, “files are independent from each other”, “applied”.

[GrimPixel]

It is working now.

The font doesn't include tab and non-Japanese Han characters.

I think it serves the educational purpose in a flashcard-way: check if your memory of a formula is correct.

Update: Node names are Python + XML/HTML now. What you see is what you get.

Heron's formula

[KennedyRichard]

Thank you for the detailed feedback. I appreciate it a lot. This shows a lot of dedication.

When it is the time to implement some of the agreed features or study them better to determine what should be implemented I'll certainly revisit this content and take it into consideration, as well as link other people/discussions to it so others can give feedback as well.

[GrimPixel]

I am thinking of moving the labels from the .py file to a .yaml or a .tsv file instead, so the .py files are clean as they should be. This saves the memory when loading nodes, too. In addition, it saves the efforts of typing quotation marks and commas.

Since it's your wish to avoid external packages, I will do with TSV. The first line of the file is the text of input/output on the node, the corresponding fields are the labels.

pi_ast_A_ast_B.tsv:

X	A	B	
椭圆面积	半长轴长	半短轴长	cmn-Hans-CN
area of an ellipse	semi-major axis	semi-minor axis	eng-Latn-US
A,S	a	b	zxx-Zsym-ZZ

[KennedyRichard]

Also, although Python recommends ascii/English characters for names in Python, it is up to the users whether they decide using them or not. As long as the names are valid Python identifiers, it doesn't matter if they use non-ascii characters.

Since Python itself doesn't restrict the characters used, I see no problem implementing this possibility in Nodezator as well (instead of just letting people write non-english characters in widgets and comments, like I said before on another discussion). However, this is something that will take time until I implement, because Nodezator has other priority features to be implemented before. But I'll keep all of this in mind and I do think people must have the option to choose whether they want to write their code in English or their mother tongue.

[GrimPixel]

I will try another approach: build a database and generate node files from it.

[KennedyRichard]

Interesting. Although this is a workaround, it might be worth considering what you are trying to accomplish.

I've been working on Nodezator's priority features, but I didn't forget about your suggestions/requests. As I said, I'll consider them for implementation once other basic missing features are implemented, although it will take time.

The automated tests that I'm implementing right now are an important step towards this, since it will allow me to develop faster and much safer (with less chance of breaking existing stuff).

Have a nice weekend.

Peace.