<[] is similar to [] :: operators>

1. How does this work?
Read the Top First

It basically sees if a string is similar to another string

2. Is there a workaround
NO

fdreerf wrote:

What does “similar” even mean in the context of this block? It could mean basically anything.

UmbreonFX wrote:

By is similar to, I mean that the two strings have similar meanings or the strings contain similar text

-Accio- wrote:

What do you mean by similar? This seems very ambiguous. For example, would it return true for Scratch and Scrathc? What about for 11 and 12?

DJUmbreon2 wrote:

fdreerf wrote:

What does “similar” even mean in the context of this block? It could mean basically anything.

Welp, Here we go

UmbreonFX wrote:

By is similar to, I mean that the two strings have similar meanings or the strings contain similar text

DJUmbreon2 wrote:

fdreerf wrote:

What does “similar” even mean in the context of this block? It could mean basically anything.

Welp, Here we go

UmbreonFX wrote:

By is similar to, I mean that the two strings have similar meanings or the strings contain similar text

fdreerf wrote:

DJUmbreon2 wrote:

fdreerf wrote:

What does “similar” even mean in the context of this block? It could mean basically anything.

Welp, Here we go

UmbreonFX wrote:

By is similar to, I mean that the two strings have similar meanings or the strings contain similar text

What does that exactly mean? A computer can't decide what's similar or not without a rigorous definition.

DJUmbreon2 wrote:

fdreerf wrote:

DJUmbreon2 wrote:

fdreerf wrote:

What does “similar” even mean in the context of this block? It could mean basically anything.

Welp, Here we go

UmbreonFX wrote:

By is similar to, I mean that the two strings have similar meanings or the strings contain similar text

What does that exactly mean? A computer can't decide what's similar or not without a rigorous definition.

Oh god

<[test] is more than (50)% similar to [te5t]::operators> // true
<[test] is more than (50)% similar to [7e57]::operators> // false

(similarity of [test] and [te5t]::operators) // outputs 75 because 75% of characters match.

awesome-llama wrote:

The solution to this is to define what similar is exactly, which I can tell from this topic that it could mean how many characters of two strings are the same. Also something else to consider is what is done with letters which are swapped around, e.g. test and tset. I ignored this for this example as it is easier to understand without. But perhaps there is a special weighting to this situation, like it only counts as 1 dissimilar letter or something

awesome-llama wrote:

The solution to this is to define what similar is exactly, which I can tell from this topic that it could mean how many characters of two strings are the same. Also something else to consider is what is done with letters which are swapped around, e.g. test and tset. I ignored this for this example as it is easier to understand without. But perhaps there is a special weighting to this situation, like it only counts as 1 dissimilar letter or something.

Here is a modification to the original to make it work:

<[test] is more than (50)% similar to [te5t]::operators> // true
<[test] is more than (50)% similar to [7e57]::operators> // false

From this, it is possible to make a reporter block version:

(similarity of [test] and [te5t]::operators) // outputs 75 because 75% of characters match.

awesome-llama wrote:

-snip-
Here is a modification to the original to make it work:

<[test] is more than (50)% similar to [te5t]::operators> // true
<[test] is more than (50)% similar to [7e57]::operators> // false

-snip-

<[test] is [more than v] (50)% similar to [te5t] :: operators>

DarthVader4Life wrote:

awesome-llama wrote:

-snip-
Here is a modification to the original to make it work:

<[test] is more than (50)% similar to [te5t]::operators> // true
<[test] is more than (50)% similar to [7e57]::operators> // false

-snip-

For the boolean blocks, I think that it's best if we do this:

<[test] is [more than v] (50)% similar to [te5t] :: operators>

The dropdown would contain “more than”, “less than”, and “equal to”.

awesome-llama wrote:

The solution to this is to define what similar is exactly, which I can tell from this topic that it could mean how many characters of two strings are the same. Also something else to consider is what is done with letters which are swapped around, e.g. test and tset. I ignored this for this example as it is easier to understand without. But perhaps there is a special weighting to this situation, like it only counts as 1 dissimilar letter or something.

Here is a modification to the original to make it work:

<[test] is more than (50)% similar to [te5t]::operators> // true
<[test] is more than (50)% similar to [7e57]::operators> // false

From this, it is possible to make a reporter block version:

(similarity of [test] and [te5t]::operators) // outputs 75 because 75% of characters match.

fdreerf wrote:

DarthVader4Life wrote:

-snip-

The boolean blocks are unnecessary since you can just plug the reporter into the preexisting boolean comparison blocks. Frankly, this procedure isn't used very often and the workaround is a matter of going through all of the letters and see if they match or not.

(Similarity of [this] and [that] :: sensing)

cruncherBones wrote:

awesome-llama wrote:

The solution to this is to define what similar is exactly, which I can tell from this topic that it could mean how many characters of two strings are the same. Also something else to consider is what is done with letters which are swapped around, e.g. test and tset. I ignored this for this example as it is easier to understand without. But perhaps there is a special weighting to this situation, like it only counts as 1 dissimilar letter or something.

Here is a modification to the original to make it work:

<[test] is more than (50)% similar to [te5t]::operators> // true
<[test] is more than (50)% similar to [7e57]::operators> // false

From this, it is possible to make a reporter block version:

(similarity of [test] and [te5t]::operators) // outputs 75 because 75% of characters match.

How would percent be calculated?

DarthVader4Life wrote:

cruncherBones wrote:

awesome-llama wrote:

The solution to this is to define what similar is exactly, which I can tell from this topic that it could mean how many characters of two strings are the same. Also something else to consider is what is done with letters which are swapped around, e.g. test and tset. I ignored this for this example as it is easier to understand without. But perhaps there is a special weighting to this situation, like it only counts as 1 dissimilar letter or something.

Here is a modification to the original to make it work:

<[test] is more than (50)% similar to [te5t]::operators> // true
<[test] is more than (50)% similar to [7e57]::operators> // false

From this, it is possible to make a reporter block version:

(similarity of [test] and [te5t]::operators) // outputs 75 because 75% of characters match.

How would percent be calculated?

I'm pretty sure it'd be how you would normally calculate a precentage, which is dividing a number by a number, then multiplying by 100 and taking the first 2 numbers that come up. Ex. 10/100 = .1 if you multiply that by 100, you get 10. Therefore, 10 is 10 percent of 100.

For what this block does, it would count each character in the first text input that is in the same place as the same character in the second text input and count each character in the second text input. Then, it would divide the first text input's total number of same characters in the same place as the second text input by the total number of characters in the second text input. After this, it'll multiply by 100 and take the first 2 numbers.

Here's some math to spare you.

Text input 1: ti1
Text input 2: ti2

Text input 1's simularities: 2
Text input 2's character count: 3

2/3 = .666667
.666667 * 100 = 66.6667

66.6667's first two numbers are 66.
The percentage is 66.

6d66yh wrote:

DarthVader4Life wrote:

cruncherBones wrote:

awesome-llama wrote:

The solution to this is to define what similar is exactly, which I can tell from this topic that it could mean how many characters of two strings are the same. Also something else to consider is what is done with letters which are swapped around, e.g. test and tset. I ignored this for this example as it is easier to understand without. But perhaps there is a special weighting to this situation, like it only counts as 1 dissimilar letter or something.

Here is a modification to the original to make it work:

<[test] is more than (50)% similar to [te5t]::operators> // true
<[test] is more than (50)% similar to [7e57]::operators> // false

From this, it is possible to make a reporter block version:

(similarity of [test] and [te5t]::operators) // outputs 75 because 75% of characters match.

How would percent be calculated?

I'm pretty sure it'd be how you would normally calculate a precentage, which is dividing a number by a number, then multiplying by 100 and taking the first 2 numbers that come up. Ex. 10/100 = .1 if you multiply that by 100, you get 10. Therefore, 10 is 10 percent of 100.

For what this block does, it would count each character in the first text input that is in the same place as the same character in the second text input and count each character in the second text input. Then, it would divide the first text input's total number of same characters in the same place as the second text input by the total number of characters in the second text input. After this, it'll multiply by 100 and take the first 2 numbers.

Here's some math to spare you.

Text input 1: ti1
Text input 2: ti2

Text input 1's simularities: 2
Text input 2's character count: 3

2/3 = .666667
.666667 * 100 = 66.6667

66.6667's first two numbers are 66.
The percentage is 66.

So the percentage of 1e-9 is 17 (the first two numbers of 1e-7). How does that make any sense?