Spaces:
Running
Running
File size: 5,140 Bytes
cfd3735 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 |
"""Test text splitting functionality."""
import pytest
from langchain.docstore.document import Document
from langchain.text_splitter import (
CharacterTextSplitter,
RecursiveCharacterTextSplitter,
)
def test_character_text_splitter() -> None:
"""Test splitting by character count."""
text = "foo bar baz 123"
splitter = CharacterTextSplitter(separator=" ", chunk_size=7, chunk_overlap=3)
output = splitter.split_text(text)
expected_output = ["foo bar", "bar baz", "baz 123"]
assert output == expected_output
def test_character_text_splitter_empty_doc() -> None:
"""Test splitting by character count doesn't create empty documents."""
text = "foo bar"
splitter = CharacterTextSplitter(separator=" ", chunk_size=2, chunk_overlap=0)
output = splitter.split_text(text)
expected_output = ["foo", "bar"]
assert output == expected_output
def test_character_text_splitter_separtor_empty_doc() -> None:
"""Test edge cases are separators."""
text = "f b"
splitter = CharacterTextSplitter(separator=" ", chunk_size=2, chunk_overlap=0)
output = splitter.split_text(text)
expected_output = ["f", "b"]
assert output == expected_output
def test_character_text_splitter_long() -> None:
"""Test splitting by character count on long words."""
text = "foo bar baz a a"
splitter = CharacterTextSplitter(separator=" ", chunk_size=3, chunk_overlap=1)
output = splitter.split_text(text)
expected_output = ["foo", "bar", "baz", "a a"]
assert output == expected_output
def test_character_text_splitter_short_words_first() -> None:
"""Test splitting by character count when shorter words are first."""
text = "a a foo bar baz"
splitter = CharacterTextSplitter(separator=" ", chunk_size=3, chunk_overlap=1)
output = splitter.split_text(text)
expected_output = ["a a", "foo", "bar", "baz"]
assert output == expected_output
def test_character_text_splitter_longer_words() -> None:
"""Test splitting by characters when splits not found easily."""
text = "foo bar baz 123"
splitter = CharacterTextSplitter(separator=" ", chunk_size=1, chunk_overlap=1)
output = splitter.split_text(text)
expected_output = ["foo", "bar", "baz", "123"]
assert output == expected_output
def test_character_text_splitting_args() -> None:
"""Test invalid arguments."""
with pytest.raises(ValueError):
CharacterTextSplitter(chunk_size=2, chunk_overlap=4)
def test_merge_splits() -> None:
"""Test merging splits with a given separator."""
splitter = CharacterTextSplitter(separator=" ", chunk_size=9, chunk_overlap=2)
splits = ["foo", "bar", "baz"]
expected_output = ["foo bar", "baz"]
output = splitter._merge_splits(splits, separator=" ")
assert output == expected_output
def test_create_documents() -> None:
"""Test create documents method."""
texts = ["foo bar", "baz"]
splitter = CharacterTextSplitter(separator=" ", chunk_size=3, chunk_overlap=0)
docs = splitter.create_documents(texts)
expected_docs = [
Document(page_content="foo"),
Document(page_content="bar"),
Document(page_content="baz"),
]
assert docs == expected_docs
def test_create_documents_with_metadata() -> None:
"""Test create documents with metadata method."""
texts = ["foo bar", "baz"]
splitter = CharacterTextSplitter(separator=" ", chunk_size=3, chunk_overlap=0)
docs = splitter.create_documents(texts, [{"source": "1"}, {"source": "2"}])
expected_docs = [
Document(page_content="foo", metadata={"source": "1"}),
Document(page_content="bar", metadata={"source": "1"}),
Document(page_content="baz", metadata={"source": "2"}),
]
assert docs == expected_docs
def test_metadata_not_shallow() -> None:
"""Test that metadatas are not shallow."""
texts = ["foo bar"]
splitter = CharacterTextSplitter(separator=" ", chunk_size=3, chunk_overlap=0)
docs = splitter.create_documents(texts, [{"source": "1"}])
expected_docs = [
Document(page_content="foo", metadata={"source": "1"}),
Document(page_content="bar", metadata={"source": "1"}),
]
assert docs == expected_docs
docs[0].metadata["foo"] = 1
assert docs[0].metadata == {"source": "1", "foo": 1}
assert docs[1].metadata == {"source": "1"}
def test_iterative_text_splitter() -> None:
"""Test iterative text splitter."""
text = """Hi.\n\nI'm Harrison.\n\nHow? Are? You?\nOkay then f f f f.
This is a weird text to write, but gotta test the splittingggg some how.
Bye!\n\n-H."""
splitter = RecursiveCharacterTextSplitter(chunk_size=10, chunk_overlap=1)
output = splitter.split_text(text)
expected_output = [
"Hi.",
"I'm",
"Harrison.",
"How? Are?",
"You?",
"Okay then",
"f f f f.",
"This is a",
"a weird",
"text to",
"write, but",
"gotta test",
"the",
"splittingg",
"ggg",
"some how.",
"Bye!\n\n-H.",
]
assert output == expected_output
|