Tables and Vectors

SKILL provides two data structures called table and vector. These are mapped similar to remote objects which means that the full content of these data structures is never send to the python side unless requested.

The following examples assume that the skillbridge is running and a workspace was opened like this:

from skillbridge import Workspace, Symbol
ws = Workspace.open()

Tables

You can create a new table

t = ws.make_table('MyTable')
# translates to makeTable("MyTable")

Or access an existing table (e.g. through a global variable)

t = ws.__.existing_table

Now t behaves like a python dictionary.

>>> t['x'] = 1
>>> t[2] = 3
>>> t[Symbol('x')] = 4
>>> t['x']
1
>>> t[2]
3
>>> t[Symbol('x')]
4
>>> t['missing']
Traceback (most recent call last):
    ...
KeyError: 'missing'
>>> t.get('missing')
None
>>> dict(t)
{'x': 1, 2: 3, Symbol('x'): 4}

It is possible to provide a default value like this:

t = ws.make_table('MyDefaultTable', None)
# translates to makeTable("MyDefaultTable" nil)

>>> t['missing']
None

Warning

The default value is evaluated on the SKILL side. That means only values that can be safely send to SKILL are possible. Empty containers and False are converted to None as usual.

>>> ws.make_table('nil', [])['missing']  # translates to makeTable("nil" nil)
None

As a convenience the attribute access to tables is an alias for item access with Symbol keys.

t = ws.make_table('Table')
t.snake_case = 10
print(t[Symbol('snakeCase')])  # prints 10
t[Symbol('snakeCase')] = 20
print(t.snake_case)  # prints 20

Vectors

Vectors behave like python sequences with a somewhat fixed length. Note, that vectors without a default value behave different from the usual python lists.

v = ws.make_vector(10)
# translates to makeVector(10)

>>> len(v)
10
>>> v[0]
Traceback (most recent call last):
    ...
IndexError: 0
>>> list(v)
[]

You have to fill the “empty” slots of the vector before you can use them

>>> v[0] = 1
>>> v[2] = 3
>>> list(v)
[1]
>>> v[1] = 2
>>> list(v)
[1, 2, 3]

Vectors with a default value behave more like python sequences

v = ws.make_vector(5, 0)
# translates to makeVector(5 0)

>>> list(v)
[0, 0, 0, 0, 0]
>>> v[0]
0
>>> v[0] = 10
>>> list(v)
[10, 0, 0, 0, 0]