Usage of NestedFilter and changes in mapping - elasticsearch

Question

I'm trying to create a filter on the basis of some product attributes which I've indexed. Earlier, they were indexed as follows -:
"attributes": {
"name1": "value1",
"name2": "value2",
....
}
The filter that I used earlier was generated according to URL query parameters as follows -:
"/search?q=product&color=black&color=blue&size=xl"
would lead to -
>>> and_filter = ANDFilter(
[
ORFilter(
[
TermFilter('color', 'blue'),
TermFilter('color', 'black')
]
),
ORFilter(
[
TermFilter('size', 'xl')
]
)
]
)
>>> main_filter = BoolFilter().add_must(and_filter)
Due to some changes in the backend, the mapping had to be changed to a nested one.
New Mapping -:
"attributes":
{
"type": "nested",
"properties": {
"name": {"type": "string"},
"display_name": {"type": "string"},
"type": {"type": "string"},
"value": {"type": "string"}
}
}
I thought of the new filter as so -:
>>> and_filter = ANDFilter(
[
ORFilter(
[
ANDFilter(
[
TermFilter("attributes.name", "color"),
TermFilter("attributes.value", "blue"),
]
),
ANDFilter(
[
TermFilter("attributes.name", "color"),
TermFilter("attributes.value", "black"),
]
)
]
),
ORFilter(
[
ANDFilter(
[
TermFilter("attributes.name", "size"),
TermFilter("attributes.value", "xl"),
]
)
]
)
]
)
>>> nested_filter = NestedFilter("attributes", BoolFilter().add_must(and_filter))
However, this doesn't seem to be the right way to do it. When I try to generate facets over products with this filter applied, all the counts come out to be zero everytime.
Additionally, trying to search for products does not yield the expected results when the filter is applied.
I would appreciate some pointers as to how to design the filter correctly.
EDIT:
Old Filter -:
{
"bool": {
"must": [
{
"and": [
{
"or": [
{
"term": {
"color": "blue"
}
},
{
"term": {
"color": "black"
}
}
]
},
{
"or": [
{
"term": {
"size": "xl"
}
}
]
}
]
}
]
}
}
New Filter -:
{
"nested": {
"filter": {
"bool": {
"must": [
{
"and": [
{
"or": [
{
"and": [
{
"term": {
"attributes.name": "color"
}
},
{
"term": {
"attributes.value": "blue"
}
}
]
},
{
"and": [
{
"term": {
"attributes.name": "color"
}
},
{
"term": {
"attributes.value": "black"
}
}
]
}
]
},
{
"or": [
{
"and": [
{
"term": {
"attributes.name": "size"
}
},
{
"term": {
"attributes.value": "xl"
}
}
]
}
]
}
]
}
]
}
},
"path": "attributes"
}
}