Leon/docutranslate
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

AiTranslatorConfig添加skip_translate参数

Browse Source
This commit is contained in:
xunbu
2025-08-28 20:35:36 +08:00
parent 7e1fb345c6
commit 7d71e6382c
11 changed files with 237 additions and 150 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
docutranslate/static/i18nData.json
View File

@@ -52,7 +52,7 @@
"chunkSizeLabel": "分块大小", "chunkSizeLabel": "分块大小",
"resetBtn": "重置", "resetBtn": "重置",
"concurrentLabel": "并发数", "concurrentLabel": "并发数",
"glossaryGenTitle": "5. 术语表", "glossaryGenTitle": "术语表",
"glossaryLabel": "术语表 (可选)", "glossaryLabel": "术语表 (可选)",
"glossaryHelp": "选择一个或多个CSV文件。文件需包含'src'和'dst'两列标题，分别代表原文和译文。", "glossaryHelp": "选择一个或多个CSV文件。文件需包含'src'和'dst'两列标题，分别代表原文和译文。",
"viewGlossaryBtn": "查看术语表", "viewGlossaryBtn": "查看术语表",
@@ -206,7 +206,7 @@
"chunkSizeLabel": "Chunk Size", "chunkSizeLabel": "Chunk Size",
"resetBtn": "Reset", "resetBtn": "Reset",
"concurrentLabel": "Concurrency", "concurrentLabel": "Concurrency",
"glossaryGenTitle": "5. Glossary", "glossaryGenTitle": "Glossary",
"glossaryLabel": "Glossary (Optional)", "glossaryLabel": "Glossary (Optional)",
"glossaryHelp": "Select one or more CSV files. Files must have 'src' and 'dst' headers, representing source and destination text respectively.", "glossaryHelp": "Select one or more CSV files. Files must have 'src' and 'dst' headers, representing source and destination text respectively.",
"viewGlossaryBtn": "View Glossary", "viewGlossaryBtn": "View Glossary",

2
docutranslate/static/index.html
View File

File diff suppressed because one or more lines are too long

16
docutranslate/translator/ai_translator/base.py
View File

@@ -1,5 +1,5 @@
from abc import abstractmethod from abc import abstractmethod
from dataclasses import dataclass from dataclasses import dataclass, field
from typing import TypeVar from typing import TypeVar
from docutranslate.agents.agent import ThinkingMode from docutranslate.agents.agent import ThinkingMode
@@ -10,10 +10,10 @@ from docutranslate.translator.base import Translator, TranslatorConfig
@dataclass(kw_only=True) @dataclass(kw_only=True)
class AiTranslatorConfig(TranslatorConfig): class AiTranslatorConfig(TranslatorConfig):
base_url: str base_url: str | None = field(default=None,metadata={"description": "OpenAI兼容地址，当skip_translate为False时为必填项"})
api_key: str api_key: str | None = field(default=None,metadata={"description": "当skip_translate为False时为必填项"})
model_id: str model_id: str | None = field(default=None,metadata={"description": "当skip_translate为False时为必填项"})
to_lang: str to_lang: str = "简体中文"
custom_prompt: str | None = None custom_prompt: str | None = None
temperature: float = 0.7 temperature: float = 0.7
thinking: ThinkingMode = "default" thinking: ThinkingMode = "default"
@@ -23,6 +23,7 @@ class AiTranslatorConfig(TranslatorConfig):
glossary_dict: dict[str:str] | None = None glossary_dict: dict[str:str] | None = None
glossary_generate_enable: bool = False glossary_generate_enable: bool = False
glossary_agent_config: GlossaryAgentConfig | None = None glossary_agent_config: GlossaryAgentConfig | None = None
skip_translate: bool = False
T = TypeVar('T', bound=Document) T = TypeVar('T', bound=Document)
@@ -35,8 +36,12 @@ class AiTranslator(Translator[T]):
def __init__(self, config: AiTranslatorConfig): def __init__(self, config: AiTranslatorConfig):
super().__init__(config=config) super().__init__(config=config)
self.skip_translate = config.skip_translate
self.glossary_agent = None self.glossary_agent = None
self.glossary_dict_gen = None self.glossary_dict_gen = None
if not self.skip_translate and (config.base_url is None or config.api_key is None or config.model_id is None):
raise ValueError("skip_translate不为false时，base_url、api_key、model_id为必填项")
if config.glossary_generate_enable: if config.glossary_generate_enable:
if config.glossary_agent_config: if config.glossary_agent_config:
self.glossary_agent = GlossaryAgent(config.glossary_agent_config) self.glossary_agent = GlossaryAgent(config.glossary_agent_config)
@@ -53,6 +58,7 @@ class AiTranslator(Translator[T]):
logger=self.logger, logger=self.logger,
) )
self.glossary_agent = GlossaryAgent(glossary_agent_config) self.glossary_agent = GlossaryAgent(glossary_agent_config)
@abstractmethod @abstractmethod
def translate(self, document: T) -> Document: def translate(self, document: T) -> Document:
... ...

47
docutranslate/translator/ai_translator/docx_translator.py
View File

@@ -37,20 +37,22 @@ class DocxTranslator(AiTranslator):
def __init__(self, config: DocxTranslatorConfig): def __init__(self, config: DocxTranslatorConfig):
super().__init__(config=config) super().__init__(config=config)
self.chunk_size = config.chunk_size self.chunk_size = config.chunk_size
agent_config = SegmentsTranslateAgentConfig( self.translate_agent = None
custom_prompt=config.custom_prompt, if not self.skip_translate:
to_lang=config.to_lang, agent_config = SegmentsTranslateAgentConfig(
baseurl=config.base_url, custom_prompt=config.custom_prompt,
key=config.api_key, to_lang=config.to_lang,
model_id=config.model_id, baseurl=config.base_url,
temperature=config.temperature, key=config.api_key,
thinking=config.thinking, model_id=config.model_id,
max_concurrent=config.concurrent, temperature=config.temperature,
timeout=config.timeout, thinking=config.thinking,
logger=self.logger, max_concurrent=config.concurrent,
glossary_dict=config.glossary_dict timeout=config.timeout,
) logger=self.logger,
self.translate_agent = SegmentsTranslateAgent(agent_config) glossary_dict=config.glossary_dict
)
self.translate_agent = SegmentsTranslateAgent(agent_config)
self.insert_mode = config.insert_mode self.insert_mode = config.insert_mode
self.separator = config.separator self.separator = config.separator
@@ -160,10 +162,14 @@ class DocxTranslator(AiTranslator):
if self.glossary_agent: if self.glossary_agent:
self.glossary_dict_gen = self.glossary_agent.send_segments(original_texts, self.chunk_size) self.glossary_dict_gen = self.glossary_agent.send_segments(original_texts, self.chunk_size)
self.translate_agent.update_glossary_dict(self.glossary_dict_gen) if self.translate_agent:
self.translate_agent.update_glossary_dict(self.glossary_dict_gen)
# 调用翻译 agent # 调用翻译 agent
translated_texts = self.translate_agent.send_segments(original_texts, self.chunk_size) if self.translate_agent:
translated_texts = self.translate_agent.send_segments(original_texts, self.chunk_size)
else:
translated_texts = original_texts
# 将翻译结果写回文档 # 将翻译结果写回文档
document.content = self._after_translate(doc, elements_to_translate, translated_texts, original_texts) document.content = self._after_translate(doc, elements_to_translate, translated_texts, original_texts)
@@ -184,11 +190,14 @@ class DocxTranslator(AiTranslator):
if self.glossary_agent: if self.glossary_agent:
self.glossary_dict_gen = await self.glossary_agent.send_segments_async(original_texts, self.chunk_size) self.glossary_dict_gen = await self.glossary_agent.send_segments_async(original_texts, self.chunk_size)
self.translate_agent.update_glossary_dict(self.glossary_dict_gen) if self.translate_agent:
self.translate_agent.update_glossary_dict(self.glossary_dict_gen)
# 异步调用翻译 agent # 异步调用翻译 agent
translated_texts = await self.translate_agent.send_segments_async(original_texts, self.chunk_size) if self.translate_agent:
translated_texts = await self.translate_agent.send_segments_async(original_texts, self.chunk_size)
else:
translated_texts = original_texts
# 将翻译结果写回文档 # 将翻译结果写回文档
document.content = await asyncio.to_thread(self._after_translate, doc, elements_to_translate, translated_texts, document.content = await asyncio.to_thread(self._after_translate, doc, elements_to_translate, translated_texts,
original_texts) original_texts)

45
docutranslate/translator/ai_translator/epub_translator.py
View File

@@ -28,14 +28,22 @@ class EpubTranslator(AiTranslator):
def __init__(self, config: EpubTranslatorConfig): def __init__(self, config: EpubTranslatorConfig):
super().__init__(config=config) super().__init__(config=config)
self.chunk_size = config.chunk_size self.chunk_size = config.chunk_size
agent_config = SegmentsTranslateAgentConfig( self.translate_agent = None
custom_prompt=config.custom_prompt, to_lang=config.to_lang, if not self.skip_translate:
baseurl=config.base_url, key=config.api_key, model_id=config.model_id, agent_config = SegmentsTranslateAgentConfig(
temperature=config.temperature, thinking=config.thinking, custom_prompt=config.custom_prompt,
max_concurrent=config.concurrent, timeout=config.timeout, logger=self.logger, to_lang=config.to_lang,
glossary_dict=config.glossary_dict baseurl=config.base_url,
) key=config.api_key,
self.translate_agent = SegmentsTranslateAgent(agent_config) model_id=config.model_id,
temperature=config.temperature,
thinking=config.thinking,
max_concurrent=config.concurrent,
timeout=config.timeout,
logger=self.logger,
glossary_dict=config.glossary_dict
)
self.translate_agent = SegmentsTranslateAgent(agent_config)
self.insert_mode = config.insert_mode self.insert_mode = config.insert_mode
self.separator = config.separator self.separator = config.separator
@@ -175,8 +183,12 @@ class EpubTranslator(AiTranslator):
return self return self
if self.glossary_agent: if self.glossary_agent:
self.glossary_dict_gen = self.glossary_agent.send_segments(original_texts, self.chunk_size) self.glossary_dict_gen = self.glossary_agent.send_segments(original_texts, self.chunk_size)
self.translate_agent.update_glossary_dict(self.glossary_dict_gen) if self.translate_agent:
translated_texts = self.translate_agent.send_segments(original_texts, self.chunk_size) self.translate_agent.update_glossary_dict(self.glossary_dict_gen)
if self.translate_agent:
translated_texts = self.translate_agent.send_segments(original_texts, self.chunk_size)
else:
translated_texts = original_texts
document.content = self._after_translate( document.content = self._after_translate(
all_files, items_to_translate, translated_texts, original_texts all_files, items_to_translate, translated_texts, original_texts
) )
@@ -195,11 +207,14 @@ class EpubTranslator(AiTranslator):
if self.glossary_agent: if self.glossary_agent:
self.glossary_dict_gen = await self.glossary_agent.send_segments_async(original_texts, self.chunk_size) self.glossary_dict_gen = await self.glossary_agent.send_segments_async(original_texts, self.chunk_size)
self.translate_agent.update_glossary_dict(self.glossary_dict_gen) if self.translate_agent:
self.translate_agent.update_glossary_dict(self.glossary_dict_gen)
translated_texts = await self.translate_agent.send_segments_async( if self.translate_agent:
original_texts, self.chunk_size translated_texts = await self.translate_agent.send_segments_async(
) original_texts, self.chunk_size
)
else:
translated_texts = original_texts
document.content = await asyncio.to_thread( document.content = await asyncio.to_thread(
self._after_translate, all_files, items_to_translate, translated_texts, original_texts self._after_translate, all_files, items_to_translate, translated_texts, original_texts
) )

48
docutranslate/translator/ai_translator/html_translator.py
View File

@@ -85,20 +85,22 @@ class HtmlTranslator(AiTranslator):
def __init__(self, config: HtmlTranslatorConfig): def __init__(self, config: HtmlTranslatorConfig):
super().__init__(config=config) super().__init__(config=config)
self.chunk_size = config.chunk_size self.chunk_size = config.chunk_size
agent_config = SegmentsTranslateAgentConfig( self.translate_agent = None
custom_prompt=config.custom_prompt, if not self.skip_translate:
to_lang=config.to_lang, agent_config = SegmentsTranslateAgentConfig(
baseurl=config.base_url, custom_prompt=config.custom_prompt,
key=config.api_key, to_lang=config.to_lang,
model_id=config.model_id, baseurl=config.base_url,
temperature=config.temperature, key=config.api_key,
thinking=config.thinking, model_id=config.model_id,
max_concurrent=config.concurrent, temperature=config.temperature,
timeout=config.timeout, thinking=config.thinking,
logger=self.logger, max_concurrent=config.concurrent,
glossary_dict=config.glossary_dict timeout=config.timeout,
) logger=self.logger,
self.translate_agent = SegmentsTranslateAgent(agent_config) glossary_dict=config.glossary_dict
)
self.translate_agent = SegmentsTranslateAgent(agent_config)
self.insert_mode = config.insert_mode self.insert_mode = config.insert_mode
self.separator = config.separator self.separator = config.separator
@@ -199,9 +201,12 @@ class HtmlTranslator(AiTranslator):
if self.glossary_agent: if self.glossary_agent:
self.glossary_dict_gen = self.glossary_agent.send_segments(original_texts, self.chunk_size) self.glossary_dict_gen = self.glossary_agent.send_segments(original_texts, self.chunk_size)
self.translate_agent.update_glossary_dict(self.glossary_dict_gen) if self.translate_agent:
self.translate_agent.update_glossary_dict(self.glossary_dict_gen)
translated_texts = self.translate_agent.send_segments(original_texts, self.chunk_size) if self.translate_agent:
translated_texts = self.translate_agent.send_segments(original_texts, self.chunk_size)
else:
translated_texts = original_texts
document.content = self._after_translate(soup, translatable_items, translated_texts, original_texts) document.content = self._after_translate(soup, translatable_items, translated_texts, original_texts)
return self return self
@@ -218,9 +223,12 @@ class HtmlTranslator(AiTranslator):
if self.glossary_agent: if self.glossary_agent:
self.glossary_dict_gen = await self.glossary_agent.send_segments_async(original_texts, self.chunk_size) self.glossary_dict_gen = await self.glossary_agent.send_segments_async(original_texts, self.chunk_size)
self.translate_agent.update_glossary_dict(self.glossary_dict_gen) if self.translate_agent:
self.translate_agent.update_glossary_dict(self.glossary_dict_gen)
translated_texts = await self.translate_agent.send_segments_async(original_texts, self.chunk_size) if self.translate_agent:
translated_texts = await self.translate_agent.send_segments_async(original_texts, self.chunk_size)
else:
translated_texts = original_texts
document.content = await asyncio.to_thread( document.content = await asyncio.to_thread(
self._after_translate, soup, translatable_items, translated_texts, original_texts self._after_translate, soup, translatable_items, translated_texts, original_texts
) )

48
docutranslate/translator/ai_translator/json_translator.py
View File

@@ -18,18 +18,22 @@ class JsonTranslator(AiTranslator):
def __init__(self, config: JsonTranslatorConfig): def __init__(self, config: JsonTranslatorConfig):
super().__init__(config=config) super().__init__(config=config)
self.chunk_size = config.chunk_size self.chunk_size = config.chunk_size
agent_config = SegmentsTranslateAgentConfig(custom_prompt=config.custom_prompt, self.translate_agent = None
to_lang=config.to_lang, if not self.skip_translate:
baseurl=config.base_url, agent_config = SegmentsTranslateAgentConfig(
key=config.api_key, custom_prompt=config.custom_prompt,
model_id=config.model_id, to_lang=config.to_lang,
temperature=config.temperature, baseurl=config.base_url,
thinking=config.thinking, key=config.api_key,
max_concurrent=config.concurrent, model_id=config.model_id,
timeout=config.timeout, temperature=config.temperature,
logger=self.logger, thinking=config.thinking,
glossary_dict=config.glossary_dict) max_concurrent=config.concurrent,
self.translate_agent = SegmentsTranslateAgent(agent_config) timeout=config.timeout,
logger=self.logger,
glossary_dict=config.glossary_dict
)
self.translate_agent = SegmentsTranslateAgent(agent_config)
self.jsonpaths = config.json_paths self.jsonpaths = config.json_paths
def _extract_matches(self, content: dict) -> list[Any]: def _extract_matches(self, content: dict) -> list[Any]:
@@ -76,10 +80,13 @@ class JsonTranslator(AiTranslator):
original_texts = [match.value for match in all_matches] original_texts = [match.value for match in all_matches]
if self.glossary_agent: if self.glossary_agent:
self.glossary_dict_gen = self.glossary_agent.send_segments(original_texts, self.chunk_size) self.glossary_dict_gen = self.glossary_agent.send_segments(original_texts, self.chunk_size)
self.translate_agent.update_glossary_dict(self.glossary_dict_gen) if self.translate_agent:
self.translate_agent.update_glossary_dict(self.glossary_dict_gen)
# 步骤 2: 批量翻译提取出的文本 # 步骤 2: 批量翻译提取出的文本
translated_texts = self.translate_agent.send_segments(original_texts, self.chunk_size) if self.translate_agent:
translated_texts = self.translate_agent.send_segments(original_texts, self.chunk_size)
else:
translated_texts = original_texts
# 健壮性检查：确保翻译回来的项目数量与发送的一致 # 健壮性检查：确保翻译回来的项目数量与发送的一致
if len(original_texts) != len(translated_texts): if len(original_texts) != len(translated_texts):
raise ValueError("翻译服务返回的项目数量与发送的数量不匹配。") raise ValueError("翻译服务返回的项目数量与发送的数量不匹配。")
@@ -107,11 +114,14 @@ class JsonTranslator(AiTranslator):
if self.glossary_agent: if self.glossary_agent:
self.glossary_dict_gen = await self.glossary_agent.send_segments_async(original_texts, self.chunk_size) self.glossary_dict_gen = await self.glossary_agent.send_segments_async(original_texts, self.chunk_size)
self.translate_agent.update_glossary_dict(self.glossary_dict_gen) if self.translate_agent:
self.translate_agent.update_glossary_dict(self.glossary_dict_gen)
# 步骤 2: 批量翻译提取出的文本
translated_texts = await self.translate_agent.send_segments_async(original_texts, self.chunk_size)
# 步骤 2: 批量翻译提取出的文本
if self.translate_agent:
translated_texts = await self.translate_agent.send_segments_async(original_texts, self.chunk_size)
else:
translated_texts = original_texts
# 健壮性检查：确保翻译回来的项目数量与发送的一致 # 健壮性检查：确保翻译回来的项目数量与发送的一致
if len(original_texts) != len(translated_texts): if len(original_texts) != len(translated_texts):
raise ValueError("翻译服务返回的项目数量与发送的数量不匹配。") raise ValueError("翻译服务返回的项目数量与发送的数量不匹配。")

42
docutranslate/translator/ai_translator/md_translator.py
View File

@@ -19,18 +19,20 @@ class MDTranslator(AiTranslator):
def __init__(self, config: MDTranslatorConfig): def __init__(self, config: MDTranslatorConfig):
super().__init__(config=config) super().__init__(config=config)
self.chunk_size = config.chunk_size self.chunk_size = config.chunk_size
agent_config = MDTranslateAgentConfig(custom_prompt=config.custom_prompt, self.translate_agent = None
to_lang=config.to_lang, if not self.skip_translate:
baseurl=config.base_url, agent_config = MDTranslateAgentConfig(custom_prompt=config.custom_prompt,
key=config.api_key, to_lang=config.to_lang,
model_id=config.model_id, baseurl=config.base_url,
temperature=config.temperature, key=config.api_key,
thinking=config.thinking, model_id=config.model_id,
max_concurrent=config.concurrent, temperature=config.temperature,
timeout=config.timeout, thinking=config.thinking,
logger=self.logger, max_concurrent=config.concurrent,
glossary_dict=config.glossary_dict) timeout=config.timeout,
self.translate_agent = MDTranslateAgent(agent_config) logger=self.logger,
glossary_dict=config.glossary_dict)
self.translate_agent = MDTranslateAgent(agent_config)
def translate(self, document: MarkdownDocument) -> Self: def translate(self, document: MarkdownDocument) -> Self:
self.logger.info("正在翻译markdown") self.logger.info("正在翻译markdown")
@@ -38,9 +40,13 @@ class MDTranslator(AiTranslator):
chunks: list[str] = split_markdown_text(document.content.decode(), self.chunk_size) chunks: list[str] = split_markdown_text(document.content.decode(), self.chunk_size)
if self.glossary_agent: if self.glossary_agent:
self.glossary_dict_gen = self.glossary_agent.send_segments(chunks, self.chunk_size) self.glossary_dict_gen = self.glossary_agent.send_segments(chunks, self.chunk_size)
self.translate_agent.update_glossary_dict(self.glossary_dict_gen) if self.translate_agent:
self.translate_agent.update_glossary_dict(self.glossary_dict_gen)
self.logger.info(f"markdown分为{len(chunks)}块") self.logger.info(f"markdown分为{len(chunks)}块")
result: list[str] = self.translate_agent.send_chunks(chunks) if self.translate_agent:
result: list[str] = self.translate_agent.send_chunks(chunks)
else:
result = chunks
content = join_markdown_texts(result) content = join_markdown_texts(result)
# 做一些加强鲁棒性的操作 # 做一些加强鲁棒性的操作
content = content.replace(r'\（', r'\(') content = content.replace(r'\（', r'\(')
@@ -57,10 +63,14 @@ class MDTranslator(AiTranslator):
if self.glossary_agent: if self.glossary_agent:
self.glossary_dict_gen = await self.glossary_agent.send_segments_async(chunks, self.chunk_size) self.glossary_dict_gen = await self.glossary_agent.send_segments_async(chunks, self.chunk_size)
self.translate_agent.update_glossary_dict(self.glossary_dict_gen) if self.translate_agent:
self.translate_agent.update_glossary_dict(self.glossary_dict_gen)
self.logger.info(f"markdown分为{len(chunks)}块") self.logger.info(f"markdown分为{len(chunks)}块")
result: list[str] = await self.translate_agent.send_chunks_async(chunks) if self.translate_agent:
result: list[str] = await self.translate_agent.send_chunks_async(chunks)
else:
result = chunks
def run(): def run():
content = join_markdown_texts(result) content = join_markdown_texts(result)

48
docutranslate/translator/ai_translator/srt_translator.py
View File

@@ -24,20 +24,22 @@ class SrtTranslator(AiTranslator):
def __init__(self, config: SrtTranslatorConfig): def __init__(self, config: SrtTranslatorConfig):
super().__init__(config=config) super().__init__(config=config)
self.chunk_size = config.chunk_size self.chunk_size = config.chunk_size
agent_config = SegmentsTranslateAgentConfig( self.translate_agent = None
custom_prompt=config.custom_prompt, if not self.skip_translate:
to_lang=config.to_lang, agent_config = SegmentsTranslateAgentConfig(
baseurl=config.base_url, custom_prompt=config.custom_prompt,
key=config.api_key, to_lang=config.to_lang,
model_id=config.model_id, baseurl=config.base_url,
temperature=config.temperature, key=config.api_key,
thinking=config.thinking, model_id=config.model_id,
max_concurrent=config.concurrent, temperature=config.temperature,
timeout=config.timeout, thinking=config.thinking,
logger=self.logger, max_concurrent=config.concurrent,
glossary_dict=config.glossary_dict timeout=config.timeout,
) logger=self.logger,
self.translate_agent = SegmentsTranslateAgent(agent_config) glossary_dict=config.glossary_dict
)
self.translate_agent = SegmentsTranslateAgent(agent_config)
self.insert_mode = config.insert_mode self.insert_mode = config.insert_mode
self.separator = config.separator self.separator = config.separator
@@ -109,10 +111,13 @@ class SrtTranslator(AiTranslator):
return self return self
if self.glossary_agent: if self.glossary_agent:
self.glossary_dict_gen = self.glossary_agent.send_segments(original_texts, self.chunk_size) self.glossary_dict_gen = self.glossary_agent.send_segments(original_texts, self.chunk_size)
self.translate_agent.update_glossary_dict(self.glossary_dict_gen) if self.translate_agent:
self.translate_agent.update_glossary_dict(self.glossary_dict_gen)
# --- 步骤 2: 调用翻译Agent --- # --- 步骤 2: 调用翻译Agent ---
translated_texts = self.translate_agent.send_segments(original_texts, self.chunk_size) if self.translate_agent:
translated_texts = self.translate_agent.send_segments(original_texts, self.chunk_size)
else:
translated_texts = original_texts
# --- 步骤 3: 后处理并更新文档内容 --- # --- 步骤 3: 后处理并更新文档内容 ---
document.content = self._after_translate(subtitles, translated_texts, original_texts) document.content = self._after_translate(subtitles, translated_texts, original_texts)
return self return self
@@ -130,11 +135,14 @@ class SrtTranslator(AiTranslator):
if self.glossary_agent: if self.glossary_agent:
self.glossary_dict_gen = await self.glossary_agent.send_segments_async(original_texts, self.chunk_size) self.glossary_dict_gen = await self.glossary_agent.send_segments_async(original_texts, self.chunk_size)
self.translate_agent.update_glossary_dict(self.glossary_dict_gen) if self.translate_agent:
self.translate_agent.update_glossary_dict(self.glossary_dict_gen)
# --- 步骤 2: 调用翻译Agent (异步) --- # --- 步骤 2: 调用翻译Agent (异步) ---
translated_texts = await self.translate_agent.send_segments_async(original_texts, self.chunk_size) if self.translate_agent:
translated_texts = await self.translate_agent.send_segments_async(original_texts, self.chunk_size)
else:
translated_texts = original_texts
# --- 步骤 3: 后处理并更新文档内容 (I/O密集型) --- # --- 步骤 3: 后处理并更新文档内容 (I/O密集型) ---
document.content = await asyncio.to_thread( document.content = await asyncio.to_thread(
self._after_translate, subtitles, translated_texts, original_texts self._after_translate, subtitles, translated_texts, original_texts

42
docutranslate/translator/ai_translator/txt_translator.py
View File

@@ -16,27 +16,33 @@ class TXTTranslator(AiTranslator):
def __init__(self, config: TXTTranslatorConfig): def __init__(self, config: TXTTranslatorConfig):
super().__init__(config=config) super().__init__(config=config)
self.chunk_size = config.chunk_size self.chunk_size = config.chunk_size
agent_config = TXTTranslateAgentConfig(custom_prompt=config.custom_prompt, self.translate_agent =None
to_lang=config.to_lang, if not self.skip_translate:
baseurl=config.base_url, agent_config = TXTTranslateAgentConfig(custom_prompt=config.custom_prompt,
key=config.api_key, to_lang=config.to_lang,
model_id=config.model_id, baseurl=config.base_url,
temperature=config.temperature, key=config.api_key,
thinking=config.thinking, model_id=config.model_id,
max_concurrent=config.concurrent, temperature=config.temperature,
timeout=config.timeout, thinking=config.thinking,
logger=self.logger, max_concurrent=config.concurrent,
glossary_dict=config.glossary_dict) timeout=config.timeout,
self.translate_agent = TXTTranslateAgent(agent_config) logger=self.logger,
glossary_dict=config.glossary_dict)
self.translate_agent = TXTTranslateAgent(agent_config)
def translate(self, document: Document) -> Self: def translate(self, document: Document) -> Self:
self.logger.info("正在翻译txt") self.logger.info("正在翻译txt")
chunks: list[str] = split_markdown_text(document.content.decode(), max_block_size=self.chunk_size) chunks: list[str] = split_markdown_text(document.content.decode(), max_block_size=self.chunk_size)
if self.glossary_agent: if self.glossary_agent:
self.glossary_dict_gen = self.glossary_agent.send_segments(chunks, self.chunk_size) self.glossary_dict_gen = self.glossary_agent.send_segments(chunks, self.chunk_size)
self.translate_agent.update_glossary_dict(self.glossary_dict_gen) if self.translate_agent:
self.translate_agent.update_glossary_dict(self.glossary_dict_gen)
self.logger.info(f"txt分为{len(chunks)}块") self.logger.info(f"txt分为{len(chunks)}块")
result: list[str] = self.translate_agent.send_chunks(chunks) if self.translate_agent:
result: list[str] = self.translate_agent.send_chunks(chunks)
else:
result=chunks
content = "\n".join(result) content = "\n".join(result)
document.content = content.encode() document.content = content.encode()
self.logger.info("翻译完成") self.logger.info("翻译完成")
@@ -48,10 +54,14 @@ class TXTTranslator(AiTranslator):
if self.glossary_agent: if self.glossary_agent:
self.glossary_dict_gen = await self.glossary_agent.send_segments_async(chunks, self.chunk_size) self.glossary_dict_gen = await self.glossary_agent.send_segments_async(chunks, self.chunk_size)
self.translate_agent.update_glossary_dict(self.glossary_dict_gen) if self.translate_agent:
self.translate_agent.update_glossary_dict(self.glossary_dict_gen)
self.logger.info(f"txt分为{len(chunks)}块") self.logger.info(f"txt分为{len(chunks)}块")
result: list[str] = await self.translate_agent.send_chunks_async(chunks) if self.translate_agent:
result: list[str] = await self.translate_agent.send_chunks_async(chunks)
else:
result=chunks
content = "\n".join(result) content = "\n".join(result)
document.content = content.encode() document.content = content.encode()
self.logger.info("翻译完成") self.logger.info("翻译完成")

45
docutranslate/translator/ai_translator/xlsx_translator.py
View File

@@ -26,18 +26,22 @@ class XlsxTranslator(AiTranslator):
def __init__(self, config: XlsxTranslatorConfig): def __init__(self, config: XlsxTranslatorConfig):
super().__init__(config=config) super().__init__(config=config)
self.chunk_size = config.chunk_size self.chunk_size = config.chunk_size
agent_config = SegmentsTranslateAgentConfig(custom_prompt=config.custom_prompt, self.translate_agent = None
to_lang=config.to_lang, if not self.skip_translate:
baseurl=config.base_url, agent_config = SegmentsTranslateAgentConfig(
key=config.api_key, custom_prompt=config.custom_prompt,
model_id=config.model_id, to_lang=config.to_lang,
temperature=config.temperature, baseurl=config.base_url,
thinking=config.thinking, key=config.api_key,
max_concurrent=config.concurrent, model_id=config.model_id,
timeout=config.timeout, temperature=config.temperature,
logger=self.logger, thinking=config.thinking,
glossary_dict=config.glossary_dict) max_concurrent=config.concurrent,
self.translate_agent = SegmentsTranslateAgent(agent_config) timeout=config.timeout,
logger=self.logger,
glossary_dict=config.glossary_dict
)
self.translate_agent = SegmentsTranslateAgent(agent_config)
self.insert_mode = config.insert_mode self.insert_mode = config.insert_mode
self.separator = config.separator self.separator = config.separator
# --- 新增功能 --- # --- 新增功能 ---
@@ -156,9 +160,13 @@ class XlsxTranslator(AiTranslator):
return self return self
if self.glossary_agent: if self.glossary_agent:
self.glossary_dict_gen = self.glossary_agent.send_segments(original_texts, self.chunk_size) self.glossary_dict_gen = self.glossary_agent.send_segments(original_texts, self.chunk_size)
self.translate_agent.update_glossary_dict(self.glossary_dict_gen) if self.translate_agent:
self.translate_agent.update_glossary_dict(self.glossary_dict_gen)
# --- 步骤 2: 调用翻译函数 --- # --- 步骤 2: 调用翻译函数 ---
translated_texts = self.translate_agent.send_segments(original_texts, self.chunk_size) if self.translate_agent:
translated_texts = self.translate_agent.send_segments(original_texts, self.chunk_size)
else:
translated_texts = original_texts
document.content = self._after_translate(workbook, cells_to_translate, translated_texts, original_texts) document.content = self._after_translate(workbook, cells_to_translate, translated_texts, original_texts)
return self return self
@@ -173,11 +181,14 @@ class XlsxTranslator(AiTranslator):
if self.glossary_agent: if self.glossary_agent:
self.glossary_dict_gen = await self.glossary_agent.send_segments_async(original_texts, self.chunk_size) self.glossary_dict_gen = await self.glossary_agent.send_segments_async(original_texts, self.chunk_size)
self.translate_agent.update_glossary_dict(self.glossary_dict_gen) if self.translate_agent:
self.translate_agent.update_glossary_dict(self.glossary_dict_gen)
# --- 步骤 2: 调用翻译函数 --- # --- 步骤 2: 调用翻译函数 ---
translated_texts = await self.translate_agent.send_segments_async(original_texts, self.chunk_size) if self.translate_agent:
translated_texts = await self.translate_agent.send_segments_async(original_texts, self.chunk_size)
else:
translated_texts = original_texts
document.content = await asyncio.to_thread(self._after_translate, workbook, cells_to_translate, document.content = await asyncio.to_thread(self._after_translate, workbook, cells_to_translate,
translated_texts, original_texts) translated_texts, original_texts)
return self return self