langchain/libs/text-splitters/langchain_text_splitters/character.py
Nisarg Trivedi 1252ccce6f
text-splitters[minor]: Added Haskell support in langchain.text_splitter module (#16191)
- **Description:** Haskell language support added in text_splitter
module
  - **Dependencies:** No
  - **Twitter handle:** @nisargtr

If no one reviews your PR within a few days, please @-mention one of
@baskaryan, @eyurtsev, @hwchase17.

---------

Co-authored-by: Bagatur <baskaryan@gmail.com>
2024-03-29 20:17:50 +00:00

618 lines
20 KiB
Python

from __future__ import annotations
import re
from typing import Any, List, Optional
from langchain_text_splitters.base import Language, TextSplitter
class CharacterTextSplitter(TextSplitter):
"""Splitting text that looks at characters."""
def __init__(
self, separator: str = "\n\n", is_separator_regex: bool = False, **kwargs: Any
) -> None:
"""Create a new TextSplitter."""
super().__init__(**kwargs)
self._separator = separator
self._is_separator_regex = is_separator_regex
def split_text(self, text: str) -> List[str]:
"""Split incoming text and return chunks."""
# First we naively split the large input into a bunch of smaller ones.
separator = (
self._separator if self._is_separator_regex else re.escape(self._separator)
)
splits = _split_text_with_regex(text, separator, self._keep_separator)
_separator = "" if self._keep_separator else self._separator
return self._merge_splits(splits, _separator)
def _split_text_with_regex(
text: str, separator: str, keep_separator: bool
) -> List[str]:
# Now that we have the separator, split the text
if separator:
if keep_separator:
# The parentheses in the pattern keep the delimiters in the result.
_splits = re.split(f"({separator})", text)
splits = [_splits[i] + _splits[i + 1] for i in range(1, len(_splits), 2)]
if len(_splits) % 2 == 0:
splits += _splits[-1:]
splits = [_splits[0]] + splits
else:
splits = re.split(separator, text)
else:
splits = list(text)
return [s for s in splits if s != ""]
class RecursiveCharacterTextSplitter(TextSplitter):
"""Splitting text by recursively look at characters.
Recursively tries to split by different characters to find one
that works.
"""
def __init__(
self,
separators: Optional[List[str]] = None,
keep_separator: bool = True,
is_separator_regex: bool = False,
**kwargs: Any,
) -> None:
"""Create a new TextSplitter."""
super().__init__(keep_separator=keep_separator, **kwargs)
self._separators = separators or ["\n\n", "\n", " ", ""]
self._is_separator_regex = is_separator_regex
def _split_text(self, text: str, separators: List[str]) -> List[str]:
"""Split incoming text and return chunks."""
final_chunks = []
# Get appropriate separator to use
separator = separators[-1]
new_separators = []
for i, _s in enumerate(separators):
_separator = _s if self._is_separator_regex else re.escape(_s)
if _s == "":
separator = _s
break
if re.search(_separator, text):
separator = _s
new_separators = separators[i + 1 :]
break
_separator = separator if self._is_separator_regex else re.escape(separator)
splits = _split_text_with_regex(text, _separator, self._keep_separator)
# Now go merging things, recursively splitting longer texts.
_good_splits = []
_separator = "" if self._keep_separator else separator
for s in splits:
if self._length_function(s) < self._chunk_size:
_good_splits.append(s)
else:
if _good_splits:
merged_text = self._merge_splits(_good_splits, _separator)
final_chunks.extend(merged_text)
_good_splits = []
if not new_separators:
final_chunks.append(s)
else:
other_info = self._split_text(s, new_separators)
final_chunks.extend(other_info)
if _good_splits:
merged_text = self._merge_splits(_good_splits, _separator)
final_chunks.extend(merged_text)
return final_chunks
def split_text(self, text: str) -> List[str]:
return self._split_text(text, self._separators)
@classmethod
def from_language(
cls, language: Language, **kwargs: Any
) -> RecursiveCharacterTextSplitter:
separators = cls.get_separators_for_language(language)
return cls(separators=separators, is_separator_regex=True, **kwargs)
@staticmethod
def get_separators_for_language(language: Language) -> List[str]:
if language == Language.CPP:
return [
# Split along class definitions
"\nclass ",
# Split along function definitions
"\nvoid ",
"\nint ",
"\nfloat ",
"\ndouble ",
# Split along control flow statements
"\nif ",
"\nfor ",
"\nwhile ",
"\nswitch ",
"\ncase ",
# Split by the normal type of lines
"\n\n",
"\n",
" ",
"",
]
elif language == Language.GO:
return [
# Split along function definitions
"\nfunc ",
"\nvar ",
"\nconst ",
"\ntype ",
# Split along control flow statements
"\nif ",
"\nfor ",
"\nswitch ",
"\ncase ",
# Split by the normal type of lines
"\n\n",
"\n",
" ",
"",
]
elif language == Language.JAVA:
return [
# Split along class definitions
"\nclass ",
# Split along method definitions
"\npublic ",
"\nprotected ",
"\nprivate ",
"\nstatic ",
# Split along control flow statements
"\nif ",
"\nfor ",
"\nwhile ",
"\nswitch ",
"\ncase ",
# Split by the normal type of lines
"\n\n",
"\n",
" ",
"",
]
elif language == Language.KOTLIN:
return [
# Split along class definitions
"\nclass ",
# Split along method definitions
"\npublic ",
"\nprotected ",
"\nprivate ",
"\ninternal ",
"\ncompanion ",
"\nfun ",
"\nval ",
"\nvar ",
# Split along control flow statements
"\nif ",
"\nfor ",
"\nwhile ",
"\nwhen ",
"\ncase ",
"\nelse ",
# Split by the normal type of lines
"\n\n",
"\n",
" ",
"",
]
elif language == Language.JS:
return [
# Split along function definitions
"\nfunction ",
"\nconst ",
"\nlet ",
"\nvar ",
"\nclass ",
# Split along control flow statements
"\nif ",
"\nfor ",
"\nwhile ",
"\nswitch ",
"\ncase ",
"\ndefault ",
# Split by the normal type of lines
"\n\n",
"\n",
" ",
"",
]
elif language == Language.TS:
return [
"\nenum ",
"\ninterface ",
"\nnamespace ",
"\ntype ",
# Split along class definitions
"\nclass ",
# Split along function definitions
"\nfunction ",
"\nconst ",
"\nlet ",
"\nvar ",
# Split along control flow statements
"\nif ",
"\nfor ",
"\nwhile ",
"\nswitch ",
"\ncase ",
"\ndefault ",
# Split by the normal type of lines
"\n\n",
"\n",
" ",
"",
]
elif language == Language.PHP:
return [
# Split along function definitions
"\nfunction ",
# Split along class definitions
"\nclass ",
# Split along control flow statements
"\nif ",
"\nforeach ",
"\nwhile ",
"\ndo ",
"\nswitch ",
"\ncase ",
# Split by the normal type of lines
"\n\n",
"\n",
" ",
"",
]
elif language == Language.PROTO:
return [
# Split along message definitions
"\nmessage ",
# Split along service definitions
"\nservice ",
# Split along enum definitions
"\nenum ",
# Split along option definitions
"\noption ",
# Split along import statements
"\nimport ",
# Split along syntax declarations
"\nsyntax ",
# Split by the normal type of lines
"\n\n",
"\n",
" ",
"",
]
elif language == Language.PYTHON:
return [
# First, try to split along class definitions
"\nclass ",
"\ndef ",
"\n\tdef ",
# Now split by the normal type of lines
"\n\n",
"\n",
" ",
"",
]
elif language == Language.RST:
return [
# Split along section titles
"\n=+\n",
"\n-+\n",
"\n\\*+\n",
# Split along directive markers
"\n\n.. *\n\n",
# Split by the normal type of lines
"\n\n",
"\n",
" ",
"",
]
elif language == Language.RUBY:
return [
# Split along method definitions
"\ndef ",
"\nclass ",
# Split along control flow statements
"\nif ",
"\nunless ",
"\nwhile ",
"\nfor ",
"\ndo ",
"\nbegin ",
"\nrescue ",
# Split by the normal type of lines
"\n\n",
"\n",
" ",
"",
]
elif language == Language.RUST:
return [
# Split along function definitions
"\nfn ",
"\nconst ",
"\nlet ",
# Split along control flow statements
"\nif ",
"\nwhile ",
"\nfor ",
"\nloop ",
"\nmatch ",
"\nconst ",
# Split by the normal type of lines
"\n\n",
"\n",
" ",
"",
]
elif language == Language.SCALA:
return [
# Split along class definitions
"\nclass ",
"\nobject ",
# Split along method definitions
"\ndef ",
"\nval ",
"\nvar ",
# Split along control flow statements
"\nif ",
"\nfor ",
"\nwhile ",
"\nmatch ",
"\ncase ",
# Split by the normal type of lines
"\n\n",
"\n",
" ",
"",
]
elif language == Language.SWIFT:
return [
# Split along function definitions
"\nfunc ",
# Split along class definitions
"\nclass ",
"\nstruct ",
"\nenum ",
# Split along control flow statements
"\nif ",
"\nfor ",
"\nwhile ",
"\ndo ",
"\nswitch ",
"\ncase ",
# Split by the normal type of lines
"\n\n",
"\n",
" ",
"",
]
elif language == Language.MARKDOWN:
return [
# First, try to split along Markdown headings (starting with level 2)
"\n#{1,6} ",
# Note the alternative syntax for headings (below) is not handled here
# Heading level 2
# ---------------
# End of code block
"```\n",
# Horizontal lines
"\n\\*\\*\\*+\n",
"\n---+\n",
"\n___+\n",
# Note that this splitter doesn't handle horizontal lines defined
# by *three or more* of ***, ---, or ___, but this is not handled
"\n\n",
"\n",
" ",
"",
]
elif language == Language.LATEX:
return [
# First, try to split along Latex sections
"\n\\\\chapter{",
"\n\\\\section{",
"\n\\\\subsection{",
"\n\\\\subsubsection{",
# Now split by environments
"\n\\\\begin{enumerate}",
"\n\\\\begin{itemize}",
"\n\\\\begin{description}",
"\n\\\\begin{list}",
"\n\\\\begin{quote}",
"\n\\\\begin{quotation}",
"\n\\\\begin{verse}",
"\n\\\\begin{verbatim}",
# Now split by math environments
"\n\\\begin{align}",
"$$",
"$",
# Now split by the normal type of lines
" ",
"",
]
elif language == Language.HTML:
return [
# First, try to split along HTML tags
"<body",
"<div",
"<p",
"<br",
"<li",
"<h1",
"<h2",
"<h3",
"<h4",
"<h5",
"<h6",
"<span",
"<table",
"<tr",
"<td",
"<th",
"<ul",
"<ol",
"<header",
"<footer",
"<nav",
# Head
"<head",
"<style",
"<script",
"<meta",
"<title",
"",
]
elif language == Language.CSHARP:
return [
"\ninterface ",
"\nenum ",
"\nimplements ",
"\ndelegate ",
"\nevent ",
# Split along class definitions
"\nclass ",
"\nabstract ",
# Split along method definitions
"\npublic ",
"\nprotected ",
"\nprivate ",
"\nstatic ",
"\nreturn ",
# Split along control flow statements
"\nif ",
"\ncontinue ",
"\nfor ",
"\nforeach ",
"\nwhile ",
"\nswitch ",
"\nbreak ",
"\ncase ",
"\nelse ",
# Split by exceptions
"\ntry ",
"\nthrow ",
"\nfinally ",
"\ncatch ",
# Split by the normal type of lines
"\n\n",
"\n",
" ",
"",
]
elif language == Language.SOL:
return [
# Split along compiler information definitions
"\npragma ",
"\nusing ",
# Split along contract definitions
"\ncontract ",
"\ninterface ",
"\nlibrary ",
# Split along method definitions
"\nconstructor ",
"\ntype ",
"\nfunction ",
"\nevent ",
"\nmodifier ",
"\nerror ",
"\nstruct ",
"\nenum ",
# Split along control flow statements
"\nif ",
"\nfor ",
"\nwhile ",
"\ndo while ",
"\nassembly ",
# Split by the normal type of lines
"\n\n",
"\n",
" ",
"",
]
elif language == Language.COBOL:
return [
# Split along divisions
"\nIDENTIFICATION DIVISION.",
"\nENVIRONMENT DIVISION.",
"\nDATA DIVISION.",
"\nPROCEDURE DIVISION.",
# Split along sections within DATA DIVISION
"\nWORKING-STORAGE SECTION.",
"\nLINKAGE SECTION.",
"\nFILE SECTION.",
# Split along sections within PROCEDURE DIVISION
"\nINPUT-OUTPUT SECTION.",
# Split along paragraphs and common statements
"\nOPEN ",
"\nCLOSE ",
"\nREAD ",
"\nWRITE ",
"\nIF ",
"\nELSE ",
"\nMOVE ",
"\nPERFORM ",
"\nUNTIL ",
"\nVARYING ",
"\nACCEPT ",
"\nDISPLAY ",
"\nSTOP RUN.",
# Split by the normal type of lines
"\n",
" ",
"",
]
elif language == Language.HASKELL:
return [
# Split along function definitions
"\nmain :: ",
"\nmain = ",
"\nlet ",
"\nin ",
"\ndo ",
"\nwhere ",
"\n:: ",
"\n= ",
# Split along type declarations
"\ndata ",
"\nnewtype ",
"\ntype ",
"\n:: ",
# Split along module declarations
"\nmodule ",
# Split along import statements
"\nimport ",
"\nqualified ",
"\nimport qualified ",
# Split along typeclass declarations
"\nclass ",
"\ninstance ",
# Split along case expressions
"\ncase ",
# Split along guards in function definitions
"\n| ",
# Split along record field declarations
"\ndata ",
"\n= {",
"\n, ",
# Split by the normal type of lines
"\n\n",
"\n",
" ",
"",
]
else:
raise ValueError(
f"Language {language} is not supported! "
f"Please choose from {list(Language)}"
)