On 30.04.2025 00:25, ichthyo wrote:
So the point in question is rather: what we store through the
UnifiedPresets and thus also what we place into the Clipboard
is marked as Zyn-compatible.
Is this intentional? Do we want/need it to be this way?
On Wed, 30 Apr 2025 Kristian Amlie <krist...@amlie.name> wrote:
Hmm, copy-pasting between Yoshi and Zyn.. I wonder if anybody does that.
I suppose it has some value, theoretically, but I can't imagine it's
common.
Maybe the best rule to follow is "if it ain't broke, don't fix it", and
keep this one working..?
On 30.04.25 10:43, Will Godfrey wrote:
Some of this is relevant to backward and forward compatibility's *within*
Yoshimi versions.
Although we try to retain compatibility with older versions of both Yoshimi
and Zyn, there's a practical limit to what we can do, but we should be able
to maintain full forward compatibility.
Files stored in the clipboard could have a different extension
(or even none), but why make that a different (extra) bit of coding?
Agreed, there would be no need for such a distinction
The extension of the file does not matter for this question, because
Yoshimi just reads in the files it produces here. Even the "clipboard"
is in fact a file in a temporary directory.
Compatibility between different versions of Yoshimi is indeed important,
but as far as I can see, it is handled directly in the code reading from
the xml tree.
What I don't understand: why do we use a Zyn-doctype and root-Element
on files we put in the clipboard and which are stored as presets?
Was this done intentionally so (for the UnifiedPresets)?
Sure, Yoshimi can read both doctypes.
Zyn can presumably only read its own doctype.
Furthermore, I have not found any code which actually *acts* based on the
version number found in the headers. All we do is to warn the user when
loading an old version or a newer Zyn version.
But the actual loading code is always fault tolerant. It always tries to
load both a Zyn root element or alternatively a Yoshimi root element.
And the code reading in actual settings is also fault tolerant and
e.g. tries first to read in the extended precision settings from yoshimi,
even on files marked as "Zyn compatible"
So the file extension, the doctype and root element, and the actual
data loading are really completely independent from each other.
As said: for all other kinds of files, we set the Yoshimi Doctype.
With the notable exception of *.xiz (Instrument) files, where we deliberately
produce a Zyn doctype. In that case, it is clear why we do it: because the
user wants it this way (and configured yoshimi accordingly)
Instrument files (*.xiz vs *.xiy) are also the only case where we actually
look for different XML content, and at different places in the tree.
_______________________________________________
Yoshimi-devel mailing list
Yoshimi-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/yoshimi-devel
