增加html工作流

This commit is contained in:
xunbu
2025-08-18 18:44:08 +08:00
parent 943c540edc
commit fd061960f0
11 changed files with 383 additions and 13 deletions

View File

@@ -0,0 +1,55 @@
from dataclasses import dataclass
from pathlib import Path
from typing import Self
from docutranslate.exporter.base import ExporterConfig
from docutranslate.exporter.html.html2html_exporter import Html2HtmlExporter
from docutranslate.ir.document import Document
from docutranslate.translator.ai_translator.html_translator import HtmlTranslatorConfig, HtmlTranslator
from docutranslate.workflow.base import Workflow, WorkflowConfig
from docutranslate.workflow.interfaces import HTMLExportable
@dataclass(kw_only=True)
class HtmlWorkflowConfig(WorkflowConfig):
translator_config: HtmlTranslatorConfig
class HtmlWorkflow(Workflow[HtmlWorkflowConfig, Document, Document], HTMLExportable):
def __init__(self, config: HtmlWorkflowConfig):
super().__init__(config=config)
if config.logger:
for sub_config in [self.config.translator_config]:
if sub_config:
sub_config.logger = config.logger
def _pre_translate(self, document_original: Document):
document = document_original.copy()
translate_config = self.config.translator_config
translator = HtmlTranslator(translate_config)
return document, translator
def translate(self) -> Self:
document, translator = self._pre_translate(self.document_original)
translator.translate(document)
self.document_translated = document
return self
async def translate_async(self) -> Self:
document, translator = self._pre_translate(self.document_original)
await translator.translate_async(document)
self.document_translated = document
return self
def export_to_html(self, _: ExporterConfig = None) -> str:
docu = self._export(Html2HtmlExporter())
return docu.content.decode()
def save_as_html(self, name: str = None, output_dir: Path | str = "./output",
_: ExporterConfig | None = None) -> Self:
self._save(exporter=Html2HtmlExporter(), name=name, output_dir=output_dir)
return self