Merge pull request #6 from bxparks/develop

publish doxygen docs on GitHub Pages

Author: bxparks

CHANGELOG.md

@@ -1,18 +1,20 @@
 # Changelog
 
-* v0.3.1 (2018-03-20)
+* 0.3.2 (2018-03-22)
+    * Publish doxygen docs using GitHub Pages.
+* 0.3.1 (2018-03-20)
     * Add support for Verbosity::kTestExpired.
     * Clean up code in examples/ directory and add documentation in README.md.
-* v0.3.0 (2018-03-19)
+* 0.3.0 (2018-03-19)
     * Implement all remaining macros from ArduinoUnit:
       assertTestXxx(), checkTestXxx(), externTest(), externTesting().
-* v0.2.0 (2018-03-16)
+* 0.2.0 (2018-03-16)
     * TestRunner can time out long running tests. Default time out is 10
       seconds, but is configurable using TestRunner::setTimeout().
-* v0.1.1 (2018-03-15)
+* 0.1.1 (2018-03-15)
     * Fix small bug with Test::setPassOrFail() which caused assertXxx()
       macros which returned true to terminate the testing() test cases.
-* v0.1.0 (2018-03-15)
+* 0.1.0 (2018-03-15)
     * First merge into 'master' branch and tagged.
 * (2018-03-12)
     * Initial upload to GitHub.

README.md

@@ -1,6 +1,6 @@
 # AUnit - Unit Testing Framework for Arduino Platforms
 
-Version: 0.3.1 (2018-03-20)
+Version: 0.3.2 (2018-03-22)
 
 ## Summary
 

library.properties

@@ -1,5 +1,5 @@
 name=AUnit
-version=0.3.1
+version=0.3.2
 author=Brian T. Park
 maintainer=Brian T. Park <[email protected]>
 sentence=A unit testing framework for Arduino platforms inspired by ArduinoUnit.

src/AUnit.h

@@ -41,6 +41,6 @@ SOFTWARE.
 #include "aunit/MetaAssertion.h"
 
 // Version format: 010203 == "1.2.3"
-#define AUNIT_VERSION 000301
+#define AUNIT_VERSION 000302
 
 #endif

tests/README.md

@@ -0,0 +1,13 @@
+# Unit Tests
+
+It may seem that we have a circular referencing problem (or a tautology) by
+using AUnit to write unit tests for itself. It could be argued that we can't
+really be sure that the code is actually testing anything since the same bug
+could be validating itself. But I built `AUnitTest` incrementally in layers,
+first validating the first layer of primitive functions (e.g.
+`compareString()`), then using that to test the next layer (e.g.
+`compareEqual()`), then using that to test the next layer (e.g.
+`assertEqual()`), then using that to test meta-assertions (e.g.
+`assertTestEqual()`), and so on. In practice, I think these tests are better
+than nothing. At the very least, it verifies that the library compiles and runs
+on the various supported platforms.