apunkt/MetaGPT
1
0
Fork 0
mirror of https://github.com/FoundationAgents/MetaGPT.git synced 2026-05-27 14:25:20 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions

cr修改，单测完善

Browse source
This commit is contained in:
liuminhui 2024-07-22 17:10:29 +08:00
parent 79334de5a4
commit 758acf8ba6
18 changed files with 371 additions and 178 deletions
Download patch file Download diff file Expand all files Collapse all files

28
examples/rag/omniparse_client.py → examples/rag/omniparse.py
View file

@ -1,18 +1,16 @@
import asyncio
from llama_parse import ResultType
from metagpt.config2 import config
from metagpt.logs import logger
from metagpt.rag.parser.omniparse.client import OmniParseClient
from metagpt.rag.parser.omniparse.parse import OmniParse
from metagpt.rag.schema import OmniParseOptions, OmniParseType
from metagpt.const import EXAMPLE_DATA_PATH
from metagpt.logs import logger
from metagpt.rag.parser import OmniParse
from metagpt.rag.schema import OmniParseOptions, OmniParseType, ParseResultType
from metagpt.utils.omniparse_client import OmniParseClient
TEST_DOCX = EXAMPLE_DATA_PATH / "parse/test01.docx"
TEST_PDF = EXAMPLE_DATA_PATH / "parse/test02.pdf"
TEST_VIDEO = EXAMPLE_DATA_PATH / "parse/test03.mp4"
TEST_AUDIO = EXAMPLE_DATA_PATH / "parse/test04.mp3"
TEST_DOCX = EXAMPLE_DATA_PATH / "omniparse/test01.docx"
TEST_PDF = EXAMPLE_DATA_PATH / "omniparse/test02.pdf"
TEST_VIDEO = EXAMPLE_DATA_PATH / "omniparse/test03.mp4"
TEST_AUDIO = EXAMPLE_DATA_PATH / "omniparse/test04.mp3"
TEST_WEBSITE_URL = "https://github.com/geekan/MetaGPT"
@ -37,10 +35,6 @@ async def omniparse_client_example():
audio_parse_ret = await client.parse_audio(filelike=TEST_AUDIO)
logger.info(audio_parse_ret)
# website fixme:omniparse官方api还存在问题
# website_parse_ret = await client.parse_website(url=TEST_WEBSITE_URL)
# logger.info(website_parse_ret)
async def omniparse_example():
parser = OmniParse(
@ -48,10 +42,10 @@ async def omniparse_example():
base_url=config.omniparse.base_url,
parse_options=OmniParseOptions(
parse_type=OmniParseType.PDF,
result_type=ResultType.MD,
result_type=ParseResultType.MD,
max_timeout=120,
num_workers=3,
)
),
)
ret = parser.load_data(file_path=TEST_PDF)
logger.info(ret)
@ -67,5 +61,5 @@ async def main():
await omniparse_example()
if __name__ == '__main__':
if __name__ == "__main__":
asyncio.run(main())