do subroutineCall

[jav4di]

The book says "we propose parsing do subroutineCall statements as if their syntax were do expression".

So I implemented compileDo as follows:

private void compileDo() {
    xmlWriter.printStartTag(JackXmlWriter.DO_STATEMENT_TAG_NAME);
    process(TokenType.KEYWORD); // 'do'
    compileExpression();
    process(TokenType.SYMBOL); // ';'
    xmlWriter.printEndTag(JackXmlWriter.DO_STATEMENT_TAG_NAME);
}

Which of course, given ExpressionLessSquare/Main.jack, outputs the following XML:

<keyword> do </keyword>
<expression>
<term>
<identifier> game </identifier>
<symbol> . </symbol>
<identifier> run </identifier>
<symbol> ( </symbol>
<expressionList>
</expressionList>
<symbol> ) </symbol>
</term>
</expression>
<symbol> ; </symbol>
</doStatement>

But the provided compare file contains:

<doStatement>
<keyword> do </keyword>
<identifier> game </identifier>
<symbol> . </symbol>
<identifier> run </identifier>
<symbol> ( </symbol>
<expressionList>
</expressionList>
<symbol> ) </symbol>
<symbol> ; </symbol>
</doStatement>

Have I misinterpreted the point in the book, or is the compare XML file out of sync with the book's advice?

[jav4di]

For now, I will use a printTags flag:

private void compileExpression(final boolean printTags) {
    if (printTags) {
        xmlWriter.printStartTag(JackXmlWriter.EXPRESSION_TAG_NAME);
    }
    // ...
    if (printTags) {
        xmlWriter.printEndTag(JackXmlWriter.EXPRESSION_TAG_NAME);
    }
}

Perhaps this is the solution intended by the authors.

[fjc]

according to the course slide of page number 133 in Coursera, it shows that:

 https://drive.google.com/file/d/1ujgcS7GoI-zu56FxhfkTAvEgZ6JT7Dxl/view