This is a <path> element without path data 'd', but with (not applicable) attributes originating from a <circle> element (same applies to the other instances of <path> elements not rendered due to the same mismatch of element type and element attributes).
I do not know under which circumstances Inkscape >= 0.91 could possibly write such objects by itself [1], or whether some (third-party) SVG optimization (on save, or externally run) is in play.
[1] There was a known issue when support for writing and editing basic SVG shapes <ellipse> and <circle> was initially introduced, which at the time exposed itself in the XML editor mostly - IIRC the information in the saved SVG file was correct nevertheless: https://sourceforge.net/p/inkscape/mailman/message/31489984/
Mc wrote:
> (since the affected objects are no longer present)
The objects seem to be still present (about 448 of them) - look e.g. at object with id 'circle4515-8-6' in the example file:
<path
style=" opacity: 1;fill: #ffffff; fill-opacity: 1;stroke: none;stroke- width:1; stroke- linecap: butt;stroke- linejoin: miter;stroke- miterlimit: 4;stroke- dasharray: none;stroke- dashoffset: 9.60000038; stroke- opacity: 1"
id="circle4515- 8-6"
cy="636. 75836"
cx="-613. 87347" />
r="3"
This is a <path> element without path data 'd', but with (not applicable) attributes originating from a <circle> element (same applies to the other instances of <path> elements not rendered due to the same mismatch of element type and element attributes).
I do not know under which circumstances Inkscape >= 0.91 could possibly write such objects by itself [1], or whether some (third-party) SVG optimization (on save, or externally run) is in play.
[1] There was a known issue when support for writing and editing basic SVG shapes <ellipse> and <circle> was initially introduced, which at the time exposed itself in the XML editor mostly - IIRC the information in the saved SVG file was correct nevertheless: /sourceforge. net/p/inkscape/ mailman/ message/ 31489984/
https:/