Description
Transform a normal covector by inverse transpose without unit renormalization; singular linear maps reject.
A planar affine transform combines a linear map and translation. This family stores [a,b,c,d,e,f], mapping [x,y] to [a×x+c×y+e,b×x+d×y+f]. Positions include translation; displacements use only the linear block, and normal covectors use its inverse transpose. Composition order changes the result.
Transform a normal covector by inverse transpose without unit renormalization; singular linear maps reject.
When to use this tool
Use this tool to transform a normal covector by inverse transpose without unit renormalization; singular linear maps reject.
How to use the tool and how it works
- Enter matrix, normal using the input definitions below.
- Run the operation. Resolve malformed-input errors before interpreting the result.
- Read result.normal. Interpret signs, units, ordering and null values using the rules below.
Invert the supplied affine matrix, rejecting a zero/nonfinite determinant. Apply the inverse linear block’s transpose to the supplied normal, leaving out translation.1
| Input field | Meaning and units | Accepted input |
|---|---|---|
| matrix | Six finite coefficients [a,b,c,d,e,f] in the mapping defined above. | Dense list of at most 128 finite numbers, subject to the stated integer/shape rules. |
| normal | Two-component normal covector, transformed by inverse transpose without unit normalization. | Dense list of at most 128 finite numbers, subject to the stated integer/shape rules. |
| Output field | Interpretation |
|---|---|
| result | Object containing the fields below. |
| result.normal | Inverse-transpose transformed normal without unit normalization. |
{
"input": {
"matrix": [
2,
0,
0,
3,
4,
5
],
"normal": [
2,
3
]
},
"output": {
"result": {
"normal": [
1,
1
]
}
}
}Limitations and assumptions
- The output normal is not renormalized. Inversion includes finite-output checks, and near-singular matrices can magnify roundoff.
- Matrices have six finite coefficients. Point/vector inputs have two components within ±1,000,000,000; rectangle bounds are likewise bounded and ordered. Coordinate units and axis directions are caller-defined. No conditioning threshold is inferred for inversion.
- Every declared top-level input is required and unknown fields are rejected. Ordinary strings are limited to 8,192 UTF-16 code units and lists to 128 entries, with tighter operation-specific bounds where stated. JSON text has bounded nesting and node counts.
- Numeric outputs must be finite and within the absolute safe-integer magnitude; exact larger integers use decimal text. Source positions are zero-based UTF-16 offsets and span ends are exclusive unless a different convention is explicitly stated.
Alternative or Complementary methods
Check coefficient layout and application order when exchanging matrices. Use point, displacement or normal operations according to the quantity being transformed, and a separate conditioning analysis for numerically sensitive inverses.
References
Similar or alternative tools
- Affine Matrix About Anchor Transform
Move the linear transform’s origin to an explicit anchor, preserving existing translation.
- Affine Displacement Transform
Transform a displacement using only the linear block; translation is intentionally omitted.