Spaces:
Sleeping
Sleeping
File size: 8,100 Bytes
b3b0ccd | 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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 | import os
import glob
import config
from pathlib import Path
from langchain_text_splitters import MarkdownHeaderTextSplitter, RecursiveCharacterTextSplitter
class DocumentChunker:
def __init__(self):
if config.MIN_PARENT_SIZE <= 0 or config.MAX_PARENT_SIZE < config.MIN_PARENT_SIZE:
raise ValueError("Parent chunk sizes must be positive and MIN_PARENT_SIZE <= MAX_PARENT_SIZE.")
if not 0 <= config.CHILD_CHUNK_OVERLAP < config.CHILD_CHUNK_SIZE:
raise ValueError("CHILD_CHUNK_OVERLAP must be smaller than CHILD_CHUNK_SIZE.")
if config.CHILD_CHUNK_OVERLAP >= config.MAX_PARENT_SIZE:
raise ValueError("CHILD_CHUNK_OVERLAP must be smaller than MAX_PARENT_SIZE.")
self.__parent_splitter = MarkdownHeaderTextSplitter(
headers_to_split_on=config.HEADERS_TO_SPLIT_ON,
strip_headers=False
)
self.__child_splitter = RecursiveCharacterTextSplitter(
chunk_size=config.CHILD_CHUNK_SIZE,
chunk_overlap=config.CHILD_CHUNK_OVERLAP
)
self.__min_parent_size = config.MIN_PARENT_SIZE
self.__max_parent_size = config.MAX_PARENT_SIZE
@staticmethod
def __merge_metadata(target, source, prepend=False):
for key, value in source.items():
if key not in target:
target[key] = value
else:
first, second = (value, target[key]) if prepend else (target[key], value)
values = [
item.strip()
for raw in (first, second)
for item in str(raw).split(" -> ")
if item.strip()
]
target[key] = " -> ".join(dict.fromkeys(values))
def create_chunks(self, path_dir=config.MARKDOWN_DIR):
all_parent_chunks, all_child_chunks = [], []
for doc_path_str in sorted(glob.glob(os.path.join(path_dir, "*.md"))):
doc_path = Path(doc_path_str)
parent_chunks, child_chunks = self.create_chunks_single(doc_path)
all_parent_chunks.extend(parent_chunks)
all_child_chunks.extend(child_chunks)
return all_parent_chunks, all_child_chunks
def create_chunks_single(self, md_path, source_name=None):
doc_path = Path(md_path)
source_name = source_name or f"{doc_path.stem}.pdf"
with open(doc_path, "r", encoding="utf-8") as f:
parent_chunks = self.__parent_splitter.split_text(f.read())
merged_parents = self.__merge_small_parents(parent_chunks)
split_parents = self.__split_large_parents(merged_parents)
cleaned_parents = self.__clean_small_chunks(split_parents)
if any(len(chunk.page_content) > self.__max_parent_size for chunk in cleaned_parents):
raise ValueError("Parent chunking produced a chunk larger than MAX_PARENT_SIZE.")
all_parent_chunks, all_child_chunks = [], []
self.__create_child_chunks(
all_parent_chunks,
all_child_chunks,
cleaned_parents,
doc_path,
source_name,
)
return all_parent_chunks, all_child_chunks
def __merge_small_parents(self, chunks):
if not chunks:
return []
merged, current = [], None
for chunk in chunks:
if current is None:
current = chunk
else:
current.page_content += "\n\n" + chunk.page_content
self.__merge_metadata(current.metadata, chunk.metadata)
if len(current.page_content) >= self.__min_parent_size:
merged.append(current)
current = None
if current:
if merged:
merged[-1].page_content += "\n\n" + current.page_content
self.__merge_metadata(merged[-1].metadata, current.metadata)
else:
merged.append(current)
return merged
def __split_large_parents(self, chunks):
split_chunks = []
for chunk in chunks:
if len(chunk.page_content) <= self.__max_parent_size:
split_chunks.append(chunk)
else:
splitter = RecursiveCharacterTextSplitter(
chunk_size=self.__max_parent_size,
chunk_overlap=config.CHILD_CHUNK_OVERLAP
)
sub_chunks = splitter.split_documents([chunk])
split_chunks.extend(sub_chunks)
return split_chunks
def __rebalance_pair(self, first, second):
combined = first.page_content.rstrip() + "\n\n" + second.page_content.lstrip()
lower = max(1, len(combined) - self.__max_parent_size)
upper = min(self.__max_parent_size, len(combined) - 1)
if len(combined) >= 2 * self.__min_parent_size:
lower = max(lower, self.__min_parent_size)
upper = min(upper, len(combined) - self.__min_parent_size)
preferred = min(max(len(combined) // 2, lower), upper)
split_at = preferred
for separator in ("\n\n", "\n", " "):
before = combined.rfind(separator, lower, preferred + 1)
after = combined.find(separator, preferred, upper + 1)
if before >= lower:
split_at = before
break
if after != -1:
split_at = after
break
left_text = combined[:split_at].rstrip()
right_text = combined[split_at:].lstrip()
if len(combined) >= 2 * self.__min_parent_size and (
len(left_text) < self.__min_parent_size
or len(right_text) < self.__min_parent_size
):
split_at = preferred
left_text, right_text = combined[:split_at], combined[split_at:]
if not left_text or not right_text:
return first, second
metadata = dict(first.metadata)
self.__merge_metadata(metadata, second.metadata)
first.page_content, first.metadata = left_text, dict(metadata)
second.page_content, second.metadata = right_text, dict(metadata)
return first, second
def __clean_small_chunks(self, chunks):
cleaned = []
for i, chunk in enumerate(chunks):
if len(chunk.page_content) < self.__min_parent_size:
if cleaned and len(cleaned[-1].page_content) + 2 + len(chunk.page_content) <= self.__max_parent_size:
cleaned[-1].page_content += "\n\n" + chunk.page_content
self.__merge_metadata(cleaned[-1].metadata, chunk.metadata)
elif (
i < len(chunks) - 1
and len(chunk.page_content) + 2 + len(chunks[i + 1].page_content) <= self.__max_parent_size
):
chunks[i + 1].page_content = chunk.page_content + "\n\n" + chunks[i + 1].page_content
self.__merge_metadata(chunks[i + 1].metadata, chunk.metadata, prepend=True)
else:
cleaned.append(chunk)
else:
cleaned.append(chunk)
for i, chunk in enumerate(cleaned):
if len(chunk.page_content) >= self.__min_parent_size or len(cleaned) == 1:
continue
if i < len(cleaned) - 1:
cleaned[i], cleaned[i + 1] = self.__rebalance_pair(chunk, cleaned[i + 1])
else:
cleaned[i - 1], cleaned[i] = self.__rebalance_pair(cleaned[i - 1], chunk)
return cleaned
def __create_child_chunks(self, all_parent_pairs, all_child_chunks, parent_chunks, doc_path, source_name):
for i, p_chunk in enumerate(parent_chunks):
parent_id = f"{doc_path.stem}_p{i}"
p_chunk.metadata.update({"source": source_name, "parent_id": parent_id})
all_parent_pairs.append((parent_id, p_chunk))
all_child_chunks.extend(self.__child_splitter.split_documents([p_chunk]))
|