Minimal example: the current SVG 2 draft says that the equivalent path for:

<circle cx="0" cy="0" r="100" />

should be:

    M                100    0
    A 100 100 0 0 0   0   100
    A 100 100 0 0 0 -100    0
    A 100 100 0 0 0   0  -100
    A 100 100 0 0 0  100    0
    Z

(ignoring the segment-completing close path operation to make it a valid path data string)

See below how Chrome renders this path (left), compared to how Chrome renders the path if we change sweep-flag to 1 (right).

<?xml version="1.0" standalone="no"?>
<svg width="550px" height="300px" viewBox="0 0 550 300" xmlns="http://www.w3.org/2000/svg">
  <path transform="translate(150,150)" d="
    M                100    0
    A 100 100 0 0 0   0   100
    A 100 100 0 0 0 -100    0
    A 100 100 0 0 0   0  -100
    A 100 100 0 0 0  100    0
    Z" />
  <path transform="translate(400,150)" d = "
    M                100    0
    A 100 100 0 0 1    0  100
    A 100 100 0 0 1 -100    0
    A 100 100 0 0 1   0  -100
    A 100 100 0 0 1  100    0
    Z" />
</svg>

You can draw a circle in different ways using paths, so that's a bit tricky.

<?xml version="1.0" standalone="no"?>
<svg width="550px" height="300px" viewBox="0 0 550 300" xmlns="http://www.w3.org/2000/svg">
   <path transform="translate(150,150)" d="
     M                100    0
    A 100 100 0 0 0   0   -100
    A 100 100 0 0 0 -100    0
    A 100 100 0 0 0   0  100
    A 100 100 0 0 0  100   0       Z" />
</svg>

draws a circle, with a sweep-flag set to 0.
The example in the 2nd comment is not from the spec prose, so it's difficult to interpret this as a spec bug.

@caribouW3 I think you overlooked some parts of the spec prose: the coordinates of the equivalent arc commands of a <circle> are mandated by the spec, so you cannot change them like you have done in your example.

In other words, the following example (that I provided earlier):

<path transform="translate(150,150)" d="
    M                100    0
    A 100 100 0 0 0   0   100
    A 100 100 0 0 0 -100    0
    A 100 100 0 0 0   0  -100
    A 100 100 0 0 0  100    0
    Z" />

is exactly what the current spec mandates as equivalent path for <circle cx="0" cy="0" r="100" /> (no room for interpretation), while the example that you provide does not.

The current spec prose is very specific (but wrong) and currently mandates that the equivalent path for a <circle> element is the following:

• A move-to command to the point cx+r,cy;
• arc to cx,cy+r;
• arc to cx-r,cy;
• arc to cx,cy-r;
• arc with a segment-completing close path operation.

Where:

• cx, cy, and r refer to the parameters of the <circle> element
• All arc flags are set to zero
• The rx and ry parameters of the arc commands are equal to the r parameter of the <circle> element

So with this circle: <circle cx="0" cy="0" r="100" />, we have:

• Move to cx+r, cy = 0+100, 0
• Arc to cx,cy+r = 0, 0+100
• Arc to cx-r,cy = 0-100, 0
• Arc to cx,cy-r = 0, 0-100
• Segment-completing arc: 0+100, 0 (same coordinate as the move to command)

And the above is not a circle but the star shape on the top of this comment.

I have exhaustively implemented these equivalent paths in my SVG renderer and I am very confident that the current prose is incorrect, and that the sweep-flag should be mandated to be set to one, as corrected by #878.