Bikey-Wagon Learnings #14
Conversation
mawildoer
force-pushed
the
mawildoer/rough-edges-2
branch
from
27bbefd
to
ee10ce3
Compare
mawildoer
force-pushed
the
mawildoer/rough-edges-2
branch
2 times, most recently
from
bd7c815
to
408c46d
Compare
mawildoer
force-pushed
the
mawildoer/rough-edges-2
branch
from
408c46d
to
755aca5
Compare
Move FuncSet and FuncDict hashable collections to libs/util. Make colours for links and nodes unique.
mawildoer
force-pushed
the
mawildoer/rough-edges-2
branch
from
9b27698
to
e263abb
Compare
|
Ready to merge |
| interface_type: type[T] = Electrical, | ||
| ) -> None: | ||
| super().__init__() | ||
|
|
There was a problem hiding this comment.
init in fab ll is purely for argument passing
split into init and preinit
def __init__(
self,
width: Size,
interface_type: type[T] = Electrical,
) -> None:
super().__init__()
self._interface_type = interface_type
self._width = width
def __preinit__(self): ...There was a problem hiding this comment.
Wait though, if __preinit__ is called before __init__ then how can I access _width during my static configuration?
There was a problem hiding this comment.
The order is init, annotations, d_fields, rt_fields, preinit, postinit
I can see where the confusion comes from.
The pre/post refers to pre/post base class construction
| self.add(F.can_attach_to_footprint_symmetrically()) | ||
| self.add(F.has_designator_prefix_defined("H")) | ||
| # TODO: "removed" isn't really true, but seems to work | ||
| self.add(has_part_picked_remove()) | ||
|
|
||
| # TODO: generate the kicad footprint instead of loading it | ||
| self.add( | ||
| F.has_footprint_defined( | ||
| KicadFootprint( | ||
| f"NetTie:NetTie-2_SMD_Pad{width_mm:.1f}mm", pin_names=["1", "2"] | ||
| ) | ||
| ) | ||
| ) |
There was a problem hiding this comment.
static traits should be in the class body
attach_to_footprint: F.can_attach_to_footprint_symmetrically
designator_prefix = L.f_field(F.has_designator_prefix_defined)("H")
@L.rt_field
def footprint(self):
width_mm = NetTie.Size(self._width).value
return F.has_footprint_defined(
KicadFootprint(
f"NetTie:NetTie-2_SMD_Pad{width_mm:.1f}mm", pin_names=["1", "2"]
)
)| super().__init__() | ||
|
|
||
| # dynamically construct the interfaces | ||
| self.unnamed = self.add([interface_type(), interface_type()], "unnamed") |
There was a problem hiding this comment.
No need to do the self.unnamed =, it's done by self.add automatically
There was a problem hiding this comment.
Use times:
self.add(times(2, self._interface_type), "unnamed")| from faebryk.library.Electrical import Electrical | ||
| from faebryk.library.KicadFootprint import KicadFootprint |
There was a problem hiding this comment.
Avoid direct imports from faebryk.library, use F.
There was a problem hiding this comment.
I'm pulling this into my project for now and separating the PRs.
Here's another draft for it specifically: #39
|
|
||
| def __init__( | ||
| self, | ||
| width: Size, |
There was a problem hiding this comment.
Shouldn't the width be a parameter instead of constructor argument?
There was a problem hiding this comment.
Ideally, but then I'd need to defer construction until later to make the footprint trait work.
Until that's all working this seems simpler and clearer
There was a problem hiding this comment.
You don't need delayed construction, because you are not constructing anything. Adding a trait is always possible.
You can use is_implemented in the trait to check whether the width is 'ready' for making a footprint.
There was a problem hiding this comment.
width: TBD[Quantity]
...
@L.rt_field
def footprint(self):
class _(F.has_footprint):
@staticmethod
def get_footprint():
value = self.width.get_most_narrow()
assert isinstance(value, Constant)
width_mm = value ...
return F.KicadFootprint(f"NetTie:NetTie-2_SMD_Pad{width_mm:.1f}mm", pin_names=["1", "2"])
@staticmethod
def is_implemented(_self):
return isinstance(self.width.get_most_narrow(), Constant)
return _()| """ | ||
|
|
||
| def __init__( | ||
| self, data: Sequence[T] | None = None, hasher: Callable[[T], Hashable] = id |
| self._deref = {hasher(d[0]): d[0] for d in data} if data else {} | ||
| self._data = {hasher(d[0]): d[1] for d in data} if data else {} |
There was a problem hiding this comment.
see comment in FuncSet about multidict / hash uniqueness
| def __contains__(self, item: T): | ||
| return self._hasher(item) in self._deref | ||
|
|
||
| def __iter__(self) -> Iterable[T]: |
There was a problem hiding this comment.
I think iter returns Iterator[T], not sure tho
test/libs/util_func_collections.py
Outdated
There was a problem hiding this comment.
Nice thank you added tests!
I'd add one for non default hash func that has collisions: e.g `lambda x: id(x) % 10'
| offer_missing_install("dash_cytoscape") | ||
| offer_missing_install("dash") |
There was a problem hiding this comment.
See comment in visualize.
Having a poetry dep group for this and then check if its there and if not print: install via poetry ... might be cleaner
There was a problem hiding this comment.
Yeah, this one's a little weird. Not sure I love the offer_missing pattern for the deps at all.
Separately, we should shoot for a vanilla pip/pipx install working by default, not poetry. Let's detangle that where when we come by it
There was a problem hiding this comment.
Afaik currently everything works with a pip only setup
Commit per change
Checklist
Please read and execute the following:
Code of Conduct
By submitting this issue, you agree to follow our Code of Conduct: