Merge pull request #216 from SergeStinckwich/master

Start to do experiments with Microdown

Author: SergeStinckwich

src/Math-AutomaticDifferenciation/PMDualNumber.class.st

@@ -1,11 +1,14 @@
 "
-In linear algebra, dual numbers extend the real numbers by adjoining one new element ε with the property ε^2 = 0 (ε is nilpotent).
-See here: https://en.wikipedia.org/wiki/Dual_number^
+In linear algebra, dual numbers extend the real numbers by adjoining one new element ε with the property ε^2 = 0 (ε is nilpotent). See here: [https://en.wikipedia.org/wiki/Dual_number](https://en.wikipedia.org/wiki/Dual_number)
 
-PMDualNumbers can be used to calculate the first derivative if one  creates them this way:
-	PMDualNumber value: aNumber eps:  derivativeOfaNumber (1 if the derivative with respect to aNumber is calculated, 0 otherwise)
+`PMDualNumber`s can be used to calculate the first derivative if one  creates them this way:
 
-PMDualNumbers can be mixed with Numbers.
+```
+PMDualNumber value: aNumber eps: derivativeOfANumber
+```
+ (1 if the derivative with respect to aNumber is calculated, 0 otherwise)
+
+`PMDualNumber`s can be mixed with Numbers.
 "
 Class {
 	#name : #PMDualNumber,

src/Math-AutomaticDifferenciation/PMHyperDualNumber.class.st

@@ -1,6 +1,6 @@
 "
-PMHyperDualNumbers can be used to additionally calculate second order derivatives.
-They can be mixed with Numbers, not with PMDualNumbers.
+`PMHyperDualNumber` can be used to additionally calculate second order derivatives.
+They can be mixed with `Number`s, not with `PMDualNumber`s.
 "
 Class {
 	#name : #PMHyperDualNumber,

src/Math-Core-Distribution/PMErfApproximation.class.st

@@ -1,9 +1,11 @@
 "
-A PMErfApproximation is a five term approximation to erf(x). This is a singleton class encapsulating the approximation. 
+A `PMErfApproximation` is a five term approximation to erf(x). This is a singleton class encapsulating the approximation. 
 
 Use with 
 
+```
 PMErfApproximation new value: aNumber
+```
 
 to receive erf(x). The result of the error function lies between zero and one.
 

src/Math-Core-Distribution/PMMitchellMooreGenerator.class.st

@@ -9,7 +9,7 @@ Class {
 	#classVars : [
 		'UniqueInstance'
 	],
-	#category : 'Math-Core-Distribution'
+	#category : #'Math-Core-Distribution'
 }
 
 { #category : #creation }
@@ -70,11 +70,10 @@ PMMitchellMooreGenerator >> floatValue [
 
 { #category : #initialization }
 PMMitchellMooreGenerator >> initialize: anArray lowIndex: anInteger [
-	
+
 	randoms := anArray.
 	lowIndex := anInteger.
-	highIndex := randoms size.
-	^self
+	highIndex := randoms size
 ]
 
 { #category : #information }

src/Math-Core-Distribution/PMProbabilityDensity.class.st

@@ -4,7 +4,7 @@ Class {
 	#instVars : [
 		'flatGenerator'
 	],
-	#category : 'Math-Core-Distribution'
+	#category : #'Math-Core-Distribution'
 }
 
 { #category : #information }
@@ -108,10 +108,11 @@ PMProbabilityDensity >> kurtosis [
 
 { #category : #information }
 PMProbabilityDensity >> parameters [
+
 	"Returns an Array containing the parameters of the distribution. 
 	It is used to print out the distribution and for fitting."
 
-	^self subclassResponsibility
+	self subclassResponsibility
 ]
 
 { #category : #display }

src/Math-Core-Process/PMFixpoint.class.st

@@ -1,10 +1,13 @@
 "
-A Fixpoint is just a little utility. it calculates the fixpoint of a block with one variable. a starting value for the variable is necessary. the variable does not need to be numerical, it can be anything the block can eat and spit out.
-example:
-a:=Fixpoint block: [:x| 1/ (1+x )] value: 20.0.
-a evaluate. 
-""-->0.6180339887498948""
+`PMFixpoint` is just a little utility. It calculates the fixpoint of a block with one variable. A starting value for the variable is necessary. The variable does not need to be numerical, it can be anything the block can eat and spit out.
 
+Example:
+
+```
+| a |
+a := PMFixpoint block: [:x| 1/(1+x)] value: 20.0.
+a evaluate
+```
 "
 Class {
 	#name : #PMFixpoint,
@@ -18,7 +21,7 @@ Class {
 		'verbose',
 		'equalityTest'
 	],
-	#category : 'Math-Core-Process'
+	#category : #'Math-Core-Process'
 }
 
 { #category : #'instance-creation' }

src/Math-Core-Process/PMIterativeProcess.class.st

@@ -1,14 +1,14 @@
 "
-A PMIterativeProcess class is an abstract base class for processes which follow an iterative pattern. 
+A `PMIterativeProcess` class is an abstract base class for processes which follow an iterative pattern. 
 
-Subclasses of PMIterativeProcess will redefine
-	initializeIterations
-	evaluateIteration
-	finalizeIterations
+Subclasses of `PMIterativeProcess` will redefine:
+- `initializeIterations`
+- `evaluateIteration`
+- `finalizeIterations`
 
-The instance variable result is used to store the most recent/best result, and is accessible through the result selector.
+The instance variable `result` is used to store the most recent/best result, and is accessible through the result selector.
 
-The maximumIterations: method allows control of the amount of work this method is allowed to do. Each evaluation of the iteration increments the instance variable iterations. When this number exceeds maximumIterations, the evaluate method will stop the process, and answer result.
+The maximumIterations: method allows control of the amount of work this method is allowed to do. Each evaluation of the iteration increments the instance variable `iterations`. When this number exceeds `maximumIterations,` the evaluate method will stop the process, and answer `result`.
 
 "
 Class {
@@ -21,7 +21,7 @@ Class {
 		'result',
 		'iterations'
 	],
-	#category : 'Math-Core-Process'
+	#category : #'Math-Core-Process'
 }
 
 { #category : #default }

src/Math-Core/PMFloatingPointMachine.class.st

@@ -1,28 +1,16 @@
 "
-A PMFloatingPointMachine represents the numerical precision of this system.
-
-Instance Variables
-
-	defaultNumericalPrecision:		The relative 
-numerical precision that can be expected for a general numerical computation. One should consider to numbers a and b equal if the relative difference between them is less than the default machine precision.
-
-	largestExponentArgument:		natural logarithm of largest number
-
-	largestNumber:		The largest positive number that can be represented in the machine
-
-	machinePrecision:		r^{-(n+1)}, with the largest n such that (1 + r^-n) - 1 != 0
-
-	negativeMachinePrecision:		r^{-(n+1)}, with the largest n such that (1 - r^-n) - 1 != 0
-
-	radix:		The radix of the floating point representation. This is often 2.
-
-	smallNumber:		A number that can be added to some value without noticeably changing the result of the computation
-
-	smallestNumber:		The smallest positive number different from 0.
-
-
-largestExponentArgument
-	- xxxxx
+A `PMFloatingPointMachine` represents the numerical precision of this system.
+
+##Instance Variables
+
+- `defaultNumericalPrecision`	The relative numerical precision that can be expected for a general numerical computation. One should consider to numbers a and b equal if the relative difference between them is less than the default machine precision,
+- `largestExponentArgument` Natural logarithm of largest number,
+- `largestNumber` The largest positive number that can be represented in the machine,
+- `machinePrecision` $r^{-(n+1)}$, with the largest n such that $(1 + r^{-n}) - 1$ != 0,
+- `negativeMachinePrecision` $r^{-(n+1)}$, with the largest n such that $(1 - r^{-n}) - 1$ != 0,
+- `radix` The radix of the floating point representation. This is often 2,
+- `smallNumber` A number that can be added to some value without noticeably changing the result of the computation,
+- `smallestNumber` The smallest positive number different from 0.
 
 This class is detailed in Object Oriented Implementation of Numerical Methods, Section 1.4.1 and 1.4.2.
 

src/Math-Core/PMWeightedPoint.class.st

@@ -1,7 +1,5 @@
 "
 I'm a simple point (two values with a weight and an error).
-
-(c) Copyrights Didier BESSET, 1999.
 "
 Class {
 	#name : #PMWeightedPoint,
@@ -12,7 +10,7 @@ Class {
 		'weight',
 		'error'
 	],
-	#category : 'Math-Core'
+	#category : #'Math-Core'
 }
 
 { #category : #creation }

src/Math-KDTree/PMIndexedKDTree.class.st

@@ -1,13 +1,13 @@
 "
-IndexedKDTree returns the indices of the nearest neighbours instead of the nearest neighbours itself. additionally it returns the squared distances between the vector and its neighbours, if more than one neighbour is searched.
+`PMIndexedKDTree` returns the indices of the nearest neighbours instead of the nearest neighbours itself. additionally it returns the squared distances between the vector and its neighbours, if more than one neighbour is searched.
 "
 Class {
 	#name : #PMIndexedKDTree,
 	#superclass : #PMKDTree,
 	#instVars : [
 		'index'
 	],
-	#category : 'Math-KDTree'
+	#category : #'Math-KDTree'
 }
 
 { #category : #'instance creation' }