Using multiple arguments for FilterRetriever

[hammerdirt-analyst]

Describe the bug
Hi, I am using the FilterRetriever and having problems with applying more than one filter.

This from a Chroma vectorstore. WITH ONE FILTER it works, I can not fiugre out how to put multiple filters.

the pseudocode:
retriever = FilterRetriever(context_document_store)

report_filters = {
'meta-key-a': param-a,
'meta-key-b'': param-b
}

this does not work:
result = retriever.run(report_filters)

this does work:
result = retriever.run({ 'meta-key-a': param-a})

In other words i can only pass one argument to the filter. Is this a problem with Chroma or whay am i missing here.

To Reproduce
context_document_store = ChromaDocumentStore()

document_pipeline = Pipeline()
document_pipeline.add_component("converter", MarkdownToDocument())
document_pipeline.add_component("cleaner", DocumentCleaner())
document_pipeline.add_component("splitter", DocumentSplitter(split_by="sentence", split_length=10))
document_pipeline.add_component("writer", DocumentWriter(document_store=context_document_store))
document_pipeline.connect("converter", "cleaner")
document_pipeline.connect("cleaner", "splitter")
document_pipeline.connect("splitter", "writer")

document_pipeline.run({"converter": {
"sources": ["vaud_report_results.md", "report_results.md", "lac_leman_g70.md"],
'meta':[
{'doc-id': 'Vaud 2015-11-15 2021-12-31', 'topic':'reports'},
{'doc-id': 'Bern 2015-11-15 2021-12-31', 'topic':'reports'},
{'doc-id': 'Lac Léman 2015-11-15 2021-12-31', 'topic': 'reports'}
]
}})

pipeline = Pipeline()
pipeline.add_component("converter", PyPDFToDocument())
pipeline.add_component("cleaner", DocumentCleaner())
pipeline.add_component("splitter", DocumentSplitter(split_by="page", split_length=1))
pipeline.add_component("writer", DocumentWriter(document_store=context_document_store))
pipeline.connect("converter", "cleaner")
pipeline.connect("cleaner", "splitter")
pipeline.connect("splitter", "writer")

file_names = [
'resources/brief_history_marine_litter.pdf',
'resources/coastline_litter_threshold_value_report_14_9_2020_final.pdf',
'resources/revealing_the_role_of_landuse.pdf',
'resources/Walvoort-ea-2021-Modelling-Forecasting-Beach-Litter-Assessment-Values-1.pdf',
'resources/eu-guide-marine-litter-2023.pdf',
'resources/land-use-marine-litter-malaysia.pdf'
]
metas = [
{'topic': 'history of research, methods of research'},
{'topic': 'threshold values, methods of calculation'},
{'topic': 'geospatial analysis, land use, feature evaluation'},
{'topic': 'threshold values, methods of calculation'},
{'topic': 'sampling protocols, methods of research'},
{'topic': 'geospatial analysis, land use, feature evaluation'}
]

pipeline.run({"converter": {"sources": file_names, "meta": metas}})

filters = { "operator": "AND",
"conditions": [
{"field": "meta.topic", "operator": "in", "value":['threshold values, methods of calculation']},
{"field": "meta.doc-id", "operator": "in", "value":['Lac Léman 2015-11-15 2021-12-31']},
]
}
retriever = FilterRetriever(context_document_store)
result = retriever.run(filters)
Describe your environment (please complete the following information):

• OS: Pop!_OS 22.04
• Haystack version: haystack-ai 2.31 (from conda list)
• Integration version: