Add checkEOL() to every command

[RFO-BASIC]

Most commands do not check to insure that there is no extraneous (non comment) characters after syntax processing on a command is done. This lack of checking can make debugging some types of coding errors difficult.

If you see a command that does not have a checkEOL(), put it in.....and TEST that change.

[jMarcS]

I've added a second checkEOL() that takes no arguments. It's exactly like checkEOL(false). In the same check-in, I changed every checkEOL(false) to checkEOL(). There is absolutely no other change in this check-in. That makes it pretty safe, but I tested it anyway -- well, almost. I tested 138 of the 139 occurrences. The one I did not check is in executeCLIENT_GETFILE. I'll get to BASIC! sockets soon, I promise.
The next step is going to be to changed every checkEOL(true) to checkEOL(false), but that will require some extra work almost everywhere it happens.

[jMarcS]

Round 2: the intent was just to remove checkEOL(boolean), leaving only the non-parameterized checkEOL(). That meant adjusting the places that used to rely on checkEOL(true) bumping the line number, distinguishing correct from incorrect ones. And that was a good excuse to touch a lot of code: added return false to some places that call RunTimeError() and removed some of the redundant SyntaxError() calls. I also reordered some code in getVar() and doUserFunction(), for readability and maybe a little performance. With a very few exceptions I thoroughly tested everything I touched.

One last-minute trivial change that affected dozens and dozens of lines I did not test: everywhere '};" was functionally identical to "}" I took out the semi-colon. I let the compiler test those.

This is a big, messy change, and I don't think there's anything critical in there, so I plan to hold off pushing this to RFO-BASIC until after v01.71 is out, Besides, that'll give me time to deal with the merge.

[jMarcS]

Round 2 is checked in.

[jMarcS]

Issue #41 gave me a close look at executeIF and its ilk. There was one checkEOL() missing (extra characters after IF with no THEN). That case still got a syntax error; all that was missing was the standard runtime error message.

But I wrapped in another change, as the commit comment says. A while back we got a big performance improvement by changing a lot of Integer objects to int primitives. This go-round I changed constants pushed onto the IfElseStack from int to Integer. The idea was to save a lot of boxing and unboxing, making all the conditions just Object reference compares. It didn't actually help any -- the benchmark results before and after are indistinguishable. Oh well, live and learn.

It did have one side-effect: switch cases can't be Objects, so the switch at the top of StatementExecuter is now an if-else tree.

Other minor changes:

• executeIf -- consolidated "return true" statements, eliminated SingleLine variable, readability tweaks
• SingleLineIf -- reduced String creation, eliminated SaveELB variable.
• executeELSE -- fixed a comment, combined parallel IfElseStack.push calls
• executeELSEIF -- fixed a comment, consolidated "return true", made it look more like executeIf

Testing: breakpoints showed f04_if_else.bas covers all the positive paths. I just needed to inject manual syntax errors to confirm the one added checkEOL works and all other syntax errors are still reported.

[jMarcS]

Covered most BT commands with fix for Issue 47 on 2012/12/12.
Covered most BYTE and TEXT file commands with fix for issue 51 on 2013/01/06.
Covered SQL commands with groundwork for issue 61 on 2013/01/27.
Covered SOCKET commands with groundwork for issues 1&62 on 2013/02/02.
Covered TIMER commands with groundwork for issue 72 on 2013/02/24.
Covered FN.DEF, MyPhoneNumber, SOUNDPOOL and ARRAY commands in commits in March, 2013.
Covered LIST commands 2013/04/01.