Since Bazel does not provide a way to map labels to strings, I am wondering how to work around this via Skylark.
Following my partial horrible "workaround".
First the statics:
_INDEX_COUNT = 50
def _build_label_mapping():
lmap = {}
for i in range(_INDEX_COUNT):
lmap ["map_name%s" % i] = attr.string()
lmap ["map_label%s" % i] = attr.label(allow_files = True)
return lmap
_LABEL_MAPPING = _build_label_mapping()
And in the implementation:
item_pairs = {}
for i in range(_INDEX_COUNT):
id = getattr(ctx.attr, "map_name%s" % i)
if not id:
continue
mapl = getattr(ctx.attr, "map_label%s" % i)
if len(mapl.files):
item_pairs[id] = list(mapl.files)[0].path
else:
item_pairs[id] = ""
if item_pairs:
arguments += [
"--map", str(item_pairs), # Pass json data
]
And then the rule:
_foo = rule(
implementation = _impl,
attrs = dict({
"srcs": attr.label_list(allow_files = True, mandatory = True),
}.items() + _LABEL_MAPPING.items()),
Which needs to be wrapped like:
def foo(map={}, **kwargs):
map_args = {}
# TODO: Check whether order of items is defined
for i, item in enumerate(textures.items()):
key, value = item
map_args["map_name%s" % i] = key
map_args["map_label%s" % i] = value
return _foo(
**dict(map_args.items() + kwargs.items())
)
Is there a better way of doing that in Skylark?
To rephrase your question, you want to create a rule attribute mapping from string to label?
This is currently not supported (see list of attributes), but you can file a feature request for this.
Do you think using "label_keyed_string_dict" is a reasonable workaround? (it won't work if you have duplicated keys)
Related
How do I get the full path to the required value in the table? I want to track changes in another table through a proxy table.
I understand that I need to use metatables and __index in it. But I haven't been able to come up with a tracker yet.
Sample table structure:
Objects = {
Panel = { layer = 1, x = 600, y = 328, w = 331, h = 491;
objects = {
label = { layer = 1, x = 0, y = 0, text = 'header' };
Window = { layer = 2, x = 400, y = 100, w = 100, h = 100;
objects = {
label = { layer = 1, x = 0, y = 0, text = 'lorem ipsum dorem' };
};
};
};
};
};
Path: Objects.Panel.objects.Window.objects.label.text
I tried to create a metatable for each of the tables and collect the result of each call to __index into a table in order to roughly understand which key and value were retrieved or changed in order to synchronize these values with other tables.
This will prove itself to be horrendously slow and memory inefficient. Anyway, you were right on the track: proxy and handle __index and __newindex metamethods to your liking. This being said you also need to track the state of the proxy somehow.
You can try to hide it with some closures and upvalues but the easy way is to store the information directly in the proxy tables:
function make_tracker (o, name)
local mt = {}
mt.__index = function (proxy, key)
local path = {unpack(rawget(proxy, "__path"))} -- Stupid shallow copy
local object = rawget(proxy, "__to")
table.insert(path, key)
if type(object[key]) == "table" then
return setmetatable({__to = object[key], __path = path}, mt)
else
return table.concat(path, ".") .. " = " .. tostring(object[key])
end
end
return setmetatable({__to = o, __path = {name}}, mt)
end
__to fields indicates what proxy should point to and __path is there to cover fields we have trespassed so far. It does a shallow copy, so that one can use subproxies with local variables. name parameter is there to initialize the name of the first table, as you just simply can't know that. You use it like this:
local tObjects = make_tracker(Objects, "Objects")
local subproxy = tObjects.Panel.objects.Window
print(subproxy.objects.label.text)
print(tObjects.Panel.objects.label.text)
print(subproxy.x)
-- prints:
-- Objects.Panel.objects.Window.objects.label.text = lorem ipsum dorem
-- Objects.Panel.objects.label.text = header
-- Objects.Panel.objects.Window.x = 400
Of course, I doubt that appending the path to the original value is what you want. Modify insides of else block:
return table.concat(path, ".") .. " = " .. tostring(object[key])
according to your needs, e.g:
register_tracked_path(table.concat(path, "."))
return object[key]
If you want to handle modification of values you need to extend the metatable with similar __newindex.
I've made a table like this
TableAlpha = {
Alpha = 3648,
Beta = 6593,
Charlie = 2358,
Delta = 6483,
Echo = 4736
}
Im wondering how can i pull out 3 keys with lowest values inside the table?
local TableAlpha = {
Alpha = 3648,
Beta = 6593,
Charlie = 2358,
Delta = 6483,
Echo = 4736
}
--- Returns the keys of tab sorted numerically by their values
local function ascending(tab)
local list = {}
for key, integer in pairs(tab) do
table.insert(list, {integer, key})
end
table.sort(list, function(left, right) return left[1] < right[1] end)
for i, tuple in ipairs(list) do
list[i] = tuple[2]
end
return list
end
local unpack = unpack or table.unpack or error("Could not find an unpack function!")
print(unpack(ascending(TableAlpha), 1, 3))
I'm trying to figure out how to dynamically build a series of sub-tables inside a lua table. For example
function BuildsubTable()
local oTable = {}
local name = {"P1","P2"}
for i = 1, 2 do
oTable[i] = {name = name[i], "x" = i + 2, "y" = i + 1}
end
return oTable
end
expected output:
oTable = {
{name = "P1", "x"=3, "y"=2},
{name = "P2", "x"=4, "y"=3}
}
Which obviously doesn't work, but you get the idea of what I'm trying to do. This is a rather simple task but in LUA 5.3 this is proving to be difficult. I cannot find a good example of building a table in this manner. Any solutions or other ideas would be appreciated. In Python I would use a class or a simple dictionary.
Your problem is that you quote the string indices. The generic syntax for declaring a table key inside the table constructor is [<key>] = <value>, for example, [20] = 20 or ["x"] = i + 2.
A shorthand for ["<key>"] = <value>, that is, for string indices that are valid variable names, you can write <key> = <value>, for example x = i + 2.
In your code you use a mix of both and write { ..., "x" = i + 2, ... }. A quick google search shows me that in Python, which you mention, you quote string keys in dictionaries, so you probably mixed that up with Lua?
EDIT: I noticed this a bit late, but you can also use ipairs to iterate the first table and table.insert to insert values:
function BuildsubTable()
local oTable = {}
local name = {"P1","P2"}
for i,name in ipairs(name) do
table.insert(oTable, {name = name, "x" = i + 2, "y" = i + 1})
end
return oTable
end
Use
oTable[i] = {name = name[i], x = i + 2, y = i + 1}
DarkWiiPlayers & lhf's answers are the proper way.
But here is how you can fix your current code if you intend to use a string as a key
function BuildsubTable()
local oTable = {}
local name = {"P1","P2"}
for i = 1, 2 do
oTable[i] = {name = name[i], ["x"] = i + 2, ["y"] = i + 1}
end
return oTable
end
Output
{
[1] = { ['name'] = 'P1', ['x'] = 3, ['y'] = 2},
[2] = { ['name'] = 'P2', ['x'] = 4, ['y'] = 3}
}
I am not sure if this is the right place to post this, but I have been trying to code up a simple decision tree class for a while and am getting lost at various points.
Specifically, I'm not sure what kind of data structure would represent a recursive tree that uses (feature, value) as nodes.
class DecisionTree():
def entropy(self, data):
# if there's nothing in this region, entropy is 1
if len(data) <= 1:
return 1
target_col = data.ix[:,-1]
size = float(len(target_col))
classes = Counter(target_col)
# if there's only one class, entropy is 1
if len(classes) == 1:
return 1
else:
probs = [i / size for i in classes.values()]
entropy = np.sum([-probs[i]*np.log(probs[i]) for i in range(len(probs))])
return entropy
def what_to_split_on(self, data):
split_feature = -1
best_entropy = 0.0
base_entropy = self.entropy(data)
for f, feature in enumerate(data.T):
unique_vals = list(set(feature))
for val in unique_vals:
left, right = self.split(f, val)
prop_left = float(len(left)) / (len(left) + len(right))
prop_right = 1 - prop_left
e_1 = prop_left * self.entropy(left)
e_2 = prop_right * self.entropy(right)
entropy_change = base_entropy - e_1 - e_2
if entropy_change > best_entropy:
best_entropy = entropy_change
split_feature = f; split_val = val
if split_feature != -1:
return split_feature, split_val
def split(self, data, f, val):
left = np.array([row for row in data if row[f] == val])
right = np.array([row for row in data if row[f] != val])
return left, right
def create_tree(self, data):
if self.entropy(data) == 1:
return
feature, value = self.what_to_split_on(data)
dt = Tree(feature, value)
left_child = np.array([row for row in data if row[feature] == value])
right_child = np.array([row for row in data if row[feature] == value])
feature, value = self.what_to_split_on(left_child)
sub_left = create_tree(left_child)
dt.insert_left(sub_left)
feature, value = self.what_to_split_on(right_child)
sub_right = create_tree(right_child)
dt.insert_right(sub_right)
return dt
I'm making a serialization library for Lua, and I'm using LPeg to parse the string. I've got K/V pairs working (with the key explicitly named), but now I'm going to add auto-indexing.
It'll work like so:
#"value"
#"value2"
Will evaluate to
{
[1] = "value"
[2] = "value2"
}
I've already got the value matching working (strings, tables, numbers, and Booleans all work perfectly), so I don't need help with that; what I'm looking for is the indexing. For each match of #[value pattern], it should capture the number of #[value pattern]'s found - in other words, I can match a sequence of values ("#"value1" #"value2") but I don't know how to assign them indexes according to the number of matches. If that's not clear enough, just comment and I'll attempt to explain it better.
Here's something of what my current pattern looks like (using compressed notation):
local process = {} -- Process a captured value
process.number = tonumber
process.string = function(s) return s:sub(2, -2) end -- Strip of opening and closing tags
process.boolean = function(s) if s == "true" then return true else return false end
number = [decimal number, scientific notation] / process.number
string = [double or single quoted string, supports escaped quotation characters] / process.string
boolean = P("true") + "false" / process.boolean
table = [balanced brackets] / [parse the table]
type = number + string + boolean + table
at_notation = (P("#") * whitespace * type) / [creates a table that includes the key and value]
As you can see in the last line of code, I've got a function that does this:
k,v matched in the pattern
-- turns into --
{k, v}
-- which is then added into an "entry table" (I loop through it and add it into the return table)
Based on what you've described so far, you should be able to accomplish this using a simple capture and table capture.
Here's a simplified example I knocked up to illustrate:
lpeg = require 'lpeg'
l = lpeg.locale(lpeg)
whitesp = l.space ^ 0
bool_val = (l.P "true" + "false") / function (s) return s == "true" end
num_val = l.digit ^ 1 / tonumber
string_val = '"' * l.C(l.alnum ^ 1) * '"'
val = bool_val + num_val + string_val
at_notation = l.Ct( (l.P "#" * whitesp * val * whitesp) ^ 0 )
local testdata = [[
#"value1"
#42
# "value2"
#true
]]
local res = l.match(at_notation, testdata)
The match returns a table containing the contents:
{
[1] = "value1",
[2] = 42,
[3] = "value2",
[4] = true
}