Skip to content

Commit

Permalink
chore: remove unused code and class in text splitter (langgenius#4864)
Browse files Browse the repository at this point in the history
  • Loading branch information
@bowenliang123
bowenliang123 authored Jun 3, 2024
1 parent b98a1a3 commit 5d15aca
Showing 1 changed file with 0 additions and 369 deletions.
369 changes: 0 additions & 369 deletions api/core/splitter/text_splitter.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -6,7 +6,6 @@
from abc import ABC, abstractmethod
from collections.abc import Callable, Collection, Iterable, Sequence, Set
from dataclasses import dataclass
from enum import Enum
from typing import (
Any,
Literal,
Expand Down Expand Up @@ -477,27 +476,6 @@ def _encode(_text: str) -> list[int]:
return split_text_on_tokens(text=text, tokenizer=tokenizer)


class Language(str, Enum):
"""Enum of the programming languages."""

CPP = "cpp"
GO = "go"
JAVA = "java"
JS = "js"
PHP = "php"
PROTO = "proto"
PYTHON = "python"
RST = "rst"
RUBY = "ruby"
RUST = "rust"
SCALA = "scala"
SWIFT = "swift"
MARKDOWN = "markdown"
LATEX = "latex"
HTML = "html"
SOL = "sol"


class RecursiveCharacterTextSplitter(TextSplitter):
"""Splitting text by recursively look at characters.
Expand Down Expand Up @@ -554,350 +532,3 @@ def _split_text(self, text: str, separators: list[str]) -> list[str]:

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, **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.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.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.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",
" ",
"",
]
else:
raise ValueError(
f"Language {language} is not supported! "
f"Please choose from {list(Language)}"
)

0 comments on commit 5d15aca

Please sign in to comment.