sc3.9.1 build

README.md

@@ -1,7 +1,7 @@
 # supercolliderStandaloneRPI2
 Standalone for Raspberry Pi 2 or 3 with Raspbian Stretch including the full IDE.
 
-This is the audio synthesis program [SuperCollider](http://github.com/supercollider/supercollider) (3.9.0, commit 8567e7e, 16jan2018) + [sc3-plugins](https://github.com/supercollider/sc3-plugins) (master, commit aa606ed, 5nov2017) compiled for rpi2 and rpi3.
+This is the audio synthesis program [SuperCollider](http://github.com/supercollider/supercollider) (3.9.1, commit f15598c, 9feb2018) + [sc3-plugins](https://github.com/supercollider/sc3-plugins) (master, commit 9307b41, 2feb2018) compiled for rpi2 and rpi3.
 
 It was built using [this guide](http://supercollider.github.io/development/building-raspberrypi.html) on a **Raspberry Pi 3** under [2017-11-29-raspbian-stretch](http://raspberrypi.org/downloads/raspbian/) (Raspbian Stretch with Desktop). It also works on the **Raspberry Pi 2**.
 For **Raspberry Pi 1** and **Raspberry Pi Zero** use [this repository](https://github.com/redFrik/supercolliderStandaloneRPI1).
@@ -49,7 +49,7 @@ KNOWN ISSUES:
 jack
 --
 
-If you start SuperCollider without having Jack already running (like when autostarting or running headless), Jack will automatically launch when you boot sc server. the audio settings then used are found in the file...
+If you start SuperCollider without having Jack already running (like when autostarting or running headless), Jack will automatically launch when you boot sc server. The audio settings then used are found in the file...
 
 * `nano ~/.jackdrc`
 

share/system/HelpSource/Classes/UnitTest.schelp

@@ -26,17 +26,31 @@ code::
 UnitTest.gui
 ::
 
-METHOD:: runTest
-Run a single test for methodName.
+METHOD:: run
+Run all methods whose names begin with code::test_::.
 
-ARGUMENT:: methodName
-Should have the format code::"TestPolyPlayerPool:test_prepareChildrenToBundle"::
+code::
+TestUnitTest.new.run;
+::
+
+ARGUMENT:: reset
+If code::true::, first runs link::#*reset:: on the class.
+
+ARGUMENT:: report
+If code::true::, outputs the test results.
+
+METHOD:: runTest
+Run a single test method.
 
 code::
 UnitTest.reset;
 UnitTest.runTest("TestUnitTest:test_assert");
 ::
 
+ARGUMENT:: methodName
+A link::Classes/String:: referring to the class and test method
+within it, e.g. code::"TestPolyPlayerPool:test_prepareChildrenToBundle"::.
+
 
 METHOD:: runAll
 runs all subclasses of UnitTest
@@ -47,133 +61,135 @@ UnitTest.runAll;
 ::
 
 
-METHOD:: runTestClassForClass
-runs testclass for a class
-
-ARGUMENT:: class
-the class
-
-ARGUMENT:: reset
-reset UnitTest environment
-
-ARGUMENT:: report
-report on pass/fail
-
-
-
-
-
-
 INSTANCEMETHODS::
 
-METHOD:: run
-All method names that start with test_ are invoked.
-
-ARGUMENT:: reset
-(describe argument here)
-
-ARGUMENT:: report
-(describe argument here)
-
-returns:: (describe returnvalue here)
+METHOD:: runTestMethod
+Run a single test method of this class
 
 code::
-TestUnitTest.new.run;
+TestUnitTest.new.runTestMethod(TestUnitTest.findMethod(\test_assert));
 ::
 
+ARGUMENT:: method
+the method to run
 
+returns:: (describe returnvalue here)
 
+PRIVATE:: s, run, findTestMethods, setUp, tearDown
 
 
-METHOD:: runTestMethod
-Run a single test method of this class
+METHOD:: assert
 
-ARGUMENT:: method
-the method to run
+Asserts that an expression must be true.
 
 code::
-TestUnitTest.new.runTestMethod(TestUnitTest.findMethod(\test_assert));
+this.assert(2 == 2, "passes");
+this.assert(2 == 2.00001, "fails");
 ::
 
-returns:: (describe returnvalue here)
+argument:: boolean
+A boolean, where code::true:: means that the test is passed.
 
-PRIVATE:: assert, assertArrayFloatEquals, asynchAssert, tearDown, setUp, assertEquals, failed, passed, assertFloatEquals, s, ifAsserts, findTestMethods, wait, bootServer
+argument:: message
+A message describing the purpose of the assertion, e.g. code::"foo
+should be less than bar"::.
+Posted if code::report:: is true.
 
+argument:: report
+Reports the result of the test if code::true::.
 
-SECTION:: Writing tests by subclassing UnitTest
+argument:: onFailure
+If not code::nil::, a failure stops the tests and evaluates this function.
 
-code::
-YourClass.test
-::
-Runs the test class for code::YourClass::, which is assumed to be named code::TestYourClass::.
+METHOD:: assertEquals
 
-If no test class if found it politely declines.
+Asserts that an expression code::a:: is equal to the value code::b::,
+where equality is determined according to code::a::'s implementation
+of code::==::.
 
-note::
-UnitTests for the Common library classes are kept in the CommonTests quark. 	This enables you to easily install and uninstall these tests.
+code::
+this.assertEquals(2 + 2, 4, "passes");
+this.assertEquals(2 + 2, 5, "fails");
 ::
 
+argument:: a
+the experimentally produced value
 
-Make sure to implement the methods code::setUp:: and code::tearDown::. They will be called before and after the test.
-
-METHOD:: assert
-
-argument:: test
-Make sure that this returns a boolean, where code::true:: means that the test is passed.
+argument:: b
+the expected value
 
 argument:: message
-posted if report is true.
+A message describing the purpose of the assertion, e.g.
+code::"Adding two numbers should return their sum."::.
+Posted if code::report:: is true.
+
+argument:: report
+Reports the result of the test if code::true:: (default is code::true::)
 
 argument:: onFailure
 If not code::nil::, a failure stops the tests and evaluates this function.
 
-code::
-UnitTest.new.assert(2 == 2, "Two does equal two.");
-UnitTest.new.assert(2 == 2.00001, "Two does equal two.");
-::
-
 METHOD:: assertFloatEquals
 
-Make sure that the two floats code::a:: and code::b:: equal within a given range (code::within::).
+Asserts that an expression code::a:: returning a float is within a
+given range (code::within::) of being equal to code::b::.
 
+code::
+this.assertFloatEquals(2, 2.0001, "Pass since 2 is close enough to 2.0001.", 0.001);
+this.assertFloatEquals(2, 2.0001, "Fail since 2 isn't close enough to 2.0001.", 0.0001);
+::
 
 argument:: a
-a float
+the experimentally produced float value
 
 argument:: b
-a float
+the expected float value
 
 argument:: message
-posted if report is true.
+A message describing the purpose of the assertion, e.g.
+code::"Adding two numbers should return their sum."::.
+Posted if code::report:: is true.
+
+argument:: within
+The range within which code::a:: must be of code::b:: in
+order for the test to pass.
+
+argument:: report
+Reports the result of the test if code::true:: (default is code::true::)
 
 argument:: onFailure
 If not code::nil::, a failure stops the tests and evaluates this function.
 
 
+METHOD:: assertArrayFloatEquals
+Make sure that the two arrays of floats code::a:: and code::b:: equal within a given range (code::within::).
 
 code::
-UnitTest.new.assertFloatEquals(2, 2.00001, "Two does equal two.", 0.00001);
-UnitTest.new.assertFloatEquals(2, 2.001, "Two does equal two.", 0.0001);
+this.assertArrayFloatEquals([2, 3] + 0.0001, [2, 3], "Pass since pairs of floats are close enough", 0.001);
+this.assertArrayFloatEquals([2, 3.0001], [2, 3], "Fail since pairs of floats aren't both close enough", 0.0001);
 ::
 
-method:: assertArrayFloatEquals
-Make sure that the two arrays of floats code::a:: and code::b:: equal within a given range (code::within::).
-
 argument:: a
-an code::Array:: of floats
+the experimentally produced value, which is an code::Array:: of floats
 
 argument:: b
-an code::Array:: of floats
+the code::Array:: of float values expected
 
 argument:: message
-posted if report is true.
+A message describing the purpose of the assertion, e.g.
+code::"Arrays foo and bar should be equal."::.
+Posted if code::report:: is true.
+
+argument:: within
+The range within which each item of code::a:: must be of the
+corresponding item in code::b:: in order for the test to pass.
+
+argument:: report
+Reports the result of the test if code::true:: (default is code::true::)
 
 argument:: onFailure
 If not code::nil::, a failure stops the tests and evaluates this function.
 
-code::
-UnitTest.new.assertArrayFloatEquals([2, 3], [2, 3] + 0.00001, "Same Floats", 0.000001);
-::
 
 METHOD:: ifAsserts
 Make a further assertion only if it passed, or only if it failed.
@@ -210,14 +226,31 @@ code::
 )
 ::
 
+METHOD:: passed
+Register a passed test.
+
+code::
+this.passed(message: "this passed");
+::
+
 METHOD:: failed
-call failure directly.
+Register a test failure.
 
 code::
-UnitTest.new.passed(message: "this passed");
+this.failed(message: "this failed");
 ::
 
-SUBSECTION:: UnitTest (sub)class  example.
+EXAMPLES::
+
+Write tests by subclassing UnitTest, and defining methods whose names
+begins code::test_::.  Each test method will be called from a fresh
+instance of the subclass.
+
+If you implement the methods code::setUp:: and code::tearDown:: in
+your test class, they will be called before and after every test.
+This can help to eliminate repetitive code, and makes it easier to
+maintain your tests by ensuring that they all run under the same set
+of conditions.
 
 code::
 TestYourClass : UnitTest {
@@ -229,9 +262,7 @@ TestYourClass : UnitTest {
 	}
 
 	test_yourMethod {
-
-		// every method named test_
-		// will be run
+		// every method whose name begins with "test_" will be run
 
 		this.assert( 6 == 6, "6 should equal 6");
 

share/system/HelpSource/Classes/Unpack1FFT.schelp

@@ -34,10 +34,11 @@ code::
 (
 {
 	var sig, chain, stream, windowStarts, demand;
+	var fftsize = 1024;
 
 	sig = SinOsc.ar(LFDNoise3.kr(LFNoise0.kr(1) * 40 + 60) * 700 + 800);
 
-	chain = FFT(LocalBuf(1024), sig);
+	chain = FFT(LocalBuf(fftsize), sig);
 	stream = Unpack1FFT(chain, bufsize: fftsize, binindex: 0, whichmeasure: 0);
 
 	// each time an FFT window starts, the unpacker returns a new value

share/system/HelpSource/Reference/Synth-Definition-File-Format.schelp

@@ -16,13 +16,13 @@ list::
 This document describes SynthDef2's format. See link::#Original SynthDef format:: for the differences between SynthDef and SynthDef2.
 
 section:: Basic types
-All data is stored big endian. All data is packed, not padded or aligned.
+All data is stored big endian. All integers, unless otherwise noted, are signed. All data is packed, not padded or aligned.
 definitionlist::
 ## int32 || a 32 bit integer.
 ## int16 || a 16 bit integer.
 ## int8 || an 8 bit integer.
 ## float32 || a 32 bit IEEE floating point number.
-## pstring || a pascal format string: a byte giving the length followed by that many bytes.
+## pstring || a pascal format string: a byte (an strong::unsigned:: int8) giving the length followed by that many bytes.
 ::
 
 section:: File Format