refactor: add itemized_carrier instead of carrier consists of ResourceHolder

This commit is contained in:
Junhan Chang
2025-09-29 20:36:45 +08:00
parent a5397ffe12
commit 10aabb7592
6 changed files with 373 additions and 121 deletions

View File

@@ -0,0 +1,322 @@
"""
自动化液体处理工作站物料类定义 - 简化版
Automated Liquid Handling Station Resource Classes - Simplified Version
"""
from __future__ import annotations
from typing import Dict, Optional
from pylabrobot.resources.coordinate import Coordinate
from pylabrobot.resources.container import Container
from pylabrobot.resources.resource_holder import ResourceHolder
from pylabrobot.resources import Resource as ResourcePLR
class Bottle(Container):
"""瓶子类 - 简化版,不追踪瓶盖"""
def __init__(
self,
name: str,
diameter: float,
height: float,
max_volume: float,
size_x: float = 0.0,
size_y: float = 0.0,
size_z: float = 0.0,
barcode: Optional[str] = "",
category: str = "container",
model: Optional[str] = None,
):
super().__init__(
name=name,
size_x=diameter,
size_y=diameter,
size_z=height,
max_volume=max_volume,
category=category,
model=model,
)
self.diameter = diameter
self.height = height
self.barcode = barcode
def serialize(self) -> dict:
return {
**super().serialize(),
"diameter": self.diameter,
"height": self.height,
"barcode": self.barcode,
}
from string import ascii_uppercase as LETTERS
from typing import Dict, List, Optional, Type, TypeVar, Union, Sequence, Tuple
import pylabrobot
from pylabrobot.resources.resource_holder import ResourceHolder
T = TypeVar("T", bound=ResourceHolder)
S = TypeVar("S", bound=ResourceHolder)
class ItemizedCarrier(ResourcePLR):
"""Base class for all carriers."""
def __init__(
self,
name: str,
size_x: float,
size_y: float,
size_z: float,
sites: Optional[Dict[Union[int, str], Optional[ResourcePLR]]] = None,
category: Optional[str] = "carrier",
model: Optional[str] = None,
):
super().__init__(
name=name,
size_x=size_x,
size_y=size_y,
size_z=size_z,
category=category,
model=model,
)
sites = sites or {}
self.sites: List[Optional[ResourcePLR]] = list(sites.values())
self._ordering = sites
self.num_items = len(self.sites)
self.child_locations: Dict[str, Coordinate] = {}
for spot, resource in sites.items():
if resource is not None and getattr(resource, "location", None) is None:
raise ValueError(f"resource {resource} has no location")
if resource is not None:
self.child_locations[spot] = resource.location
else:
self.child_locations[spot] = Coordinate.zero()
@property
def capacity(self):
"""The number of sites on this carrier."""
return len(self.sites)
def __len__(self) -> int:
"""Return the number of sites on this carrier."""
return len(self.sites)
def assign_child_resource(
self,
resource: ResourcePLR,
location: Optional[Coordinate],
reassign: bool = True,
spot: Optional[int] = None,
):
idx = spot if spot is not None else len(self.sites)
if not reassign and self.sites[idx] is not None:
raise ValueError(f"a site with index {idx} already exists")
super().assign_child_resource(resource, location=location, reassign=reassign)
self.sites[idx] = resource
def assign_resource_to_site(self, resource: ResourcePLR, spot: int):
if self.sites[spot] is not None and not isinstance(self.sites[spot], ResourceHolder):
raise ValueError(f"spot {spot} already has a resource, {resource}")
self.assign_child_resource(resource, location=self.child_locations.get(str(spot)), spot=spot)
def unassign_child_resource(self, resource: ResourcePLR):
found = False
for spot, res in enumerate(self.sites):
if res == resource:
self.sites[spot] = None
found = True
break
if not found:
raise ValueError(f"Resource {resource} is not assigned to this carrier")
if hasattr(resource, "unassign"):
resource.unassign()
def __getitem__(
self,
identifier: Union[str, int, Sequence[int], Sequence[str], slice, range],
) -> Union[List[T], T]:
"""Get the items with the given identifier.
This is a convenience method for getting the items with the given identifier. It is equivalent
to :meth:`get_items`, but adds support for slicing and supports single items in the same
functional call. Note that the return type will always be a list, even if a single item is
requested.
Examples:
Getting the items with identifiers "A1" through "E1":
>>> items["A1:E1"]
[<Item A1>, <Item B1>, <Item C1>, <Item D1>, <Item E1>]
Getting the items with identifiers 0 through 4 (note that this is the same as above):
>>> items[range(5)]
[<Item A1>, <Item B1>, <Item C1>, <Item D1>, <Item E1>]
Getting items with a slice (note that this is the same as above):
>>> items[0:5]
[<Item A1>, <Item B1>, <Item C1>, <Item D1>, <Item E1>]
Getting a single item:
>>> items[0]
[<Item A1>]
"""
if isinstance(identifier, str):
if ":" in identifier: # multiple # TODO: deprecate this, use `"A1":"E1"` instead (slice)
return self.get_items(identifier)
return self.get_item(identifier) # single
if isinstance(identifier, int):
return self.get_item(identifier)
if isinstance(identifier, (slice, range)):
start, stop = identifier.start, identifier.stop
if isinstance(identifier.start, str):
start = list(self._ordering.keys()).index(identifier.start)
elif identifier.start is None:
start = 0
if isinstance(identifier.stop, str):
stop = list(self._ordering.keys()).index(identifier.stop)
elif identifier.stop is None:
stop = self.num_items
identifier = list(range(start, stop, identifier.step or 1))
return self.get_items(identifier)
if isinstance(identifier, (list, tuple)):
return self.get_items(identifier)
raise TypeError(f"Invalid identifier type: {type(identifier)}")
def get_item(self, identifier: Union[str, int, Tuple[int, int]]) -> T:
"""Get the item with the given identifier.
Args:
identifier: The identifier of the item. Either a string, an integer, or a tuple. If an
integer, it is the index of the item in the list of items (counted from 0, top to bottom, left
to right). If a string, it uses transposed MS Excel style notation, e.g. "A1" for the first
item, "B1" for the item below that, etc. If a tuple, it is (row, column).
Raises:
IndexError: If the identifier is out of range. The range is 0 to self.num_items-1 (inclusive).
"""
if isinstance(identifier, tuple):
row, column = identifier
identifier = LETTERS[row] + str(column + 1) # standard transposed-Excel style notation
if isinstance(identifier, str):
try:
identifier = list(self._ordering.keys()).index(identifier)
except ValueError as e:
raise IndexError(
f"Item with identifier '{identifier}' does not exist on " f"resource '{self.name}'."
) from e
if not 0 <= identifier < self.capacity:
raise IndexError(
f"Item with identifier '{identifier}' does not exist on " f"resource '{self.name}'."
)
# Cast child to item type. Children will always be `T`, but the type checker doesn't know that.
return self.sites[identifier]
def get_items(self, identifiers: Union[str, Sequence[int], Sequence[str]]) -> List[T]:
"""Get the items with the given identifier.
Args:
identifier: Deprecated. Use `identifiers` instead. # TODO(deprecate-ordered-items)
identifiers: The identifiers of the items. Either a string range or a list of integers. If a
string, it uses transposed MS Excel style notation. Regions of items can be specified using
a colon, e.g. "A1:H1" for the first column. If a list of integers, it is the indices of the
items in the list of items (counted from 0, top to bottom, left to right).
Examples:
Getting the items with identifiers "A1" through "E1":
>>> items.get_items("A1:E1")
[<Item A1>, <Item B1>, <Item C1>, <Item D1>, <Item E1>]
Getting the items with identifiers 0 through 4:
>>> items.get_items(range(5))
[<Item A1>, <Item B1>, <Item C1>, <Item D1>, <Item E1>]
"""
if isinstance(identifiers, str):
identifiers = pylabrobot.utils.expand_string_range(identifiers)
return [self.get_item(i) for i in identifiers]
def __setitem__(self, idx: Union[int, str], resource: Optional[ResourcePLR]):
"""Assign a resource to this carrier."""
if resource is None: # setting to None
assigned_resource = self[idx]
if assigned_resource is not None:
self.unassign_child_resource(assigned_resource)
else:
idx = list(self._ordering.keys()).index(idx) if isinstance(idx, str) else idx
self.assign_resource_to_site(resource, spot=idx)
def __delitem__(self, idx: int):
"""Unassign a resource from this carrier."""
assigned_resource = self[idx]
if assigned_resource is not None:
self.unassign_child_resource(assigned_resource)
def get_resources(self) -> List[ResourcePLR]:
"""Get all resources assigned to this carrier."""
return [resource for resource in self.sites.values() if resource is not None]
def __eq__(self, other):
return super().__eq__(other) and self.sites == other.sites
def get_free_sites(self) -> List[int]:
return [spot for spot, resource in self.sites.items() if resource is None]
def serialize(self):
return {
**super().serialize(),
"slots": [{
"label": str(identifier),
"visible": True if self[identifier] is not None else False,
"position": {"x": location.x, "y": location.y, "z": location.z},
"size": {"width": self._size_x, "height": self._size_y, "depth": self._size_z},
"content_type": ["bottle", "container", "tube", "bottle_carrier", "tip_rack"]
} for identifier, location in self.child_locations.items()]
}
class BottleCarrier(ItemizedCarrier):
"""瓶载架 - 直接继承自 TubeCarrier"""
def __init__(
self,
name: str,
size_x: float,
size_y: float,
size_z: float,
sites: Optional[Dict[Union[int, str], ResourceHolder]] = None,
category: str = "bottle_carrier",
model: Optional[str] = None,
):
super().__init__(
name=name,
size_x=size_x,
size_y=size_y,
size_z=size_z,
sites=sites,
category=category,
model=model,
)