Skip to content

P
pdf-miner
Commit 46bcddf4 authored by myhloli's avatar myhloli
Browse files
Options

disable s3 test

parent d458b705
Hide whitespace changes
Inline Side-by-side
Showing with 25 additions and 25 deletions
tests/test_cli/test_cli.py
View file @ 46bcddf4
...@@ -19,32 +19,32 @@ class TestCli: ...@@ -19,32 +19,32 @@ class TestCli:
#common.count_folders_and_check_contents(pdf_res_path) #common.count_folders_and_check_contents(pdf_res_path)
def test_pdf_specify_jsonl(self): # def test_pdf_specify_jsonl(self):
""" # """
输入jsonl, 默认方式解析 # 输入jsonl, 默认方式解析
""" # """
cmd = "cd %s && export PYTHONPATH=. && python magic_pdf/cli/magicpdf.py json-command --json 's3://llm-process-pperf/ebook_index_textbook_40k/中高考&竞赛知识点/part-663f1ef5e7c1-009416.jsonl?bytes=0,1133972'" % (code_path) # cmd = "cd %s && export PYTHONPATH=. && python magic_pdf/cli/magicpdf.py json-command --json 's3://llm-process-pperf/ebook_index_textbook_40k/中高考&竞赛知识点/part-663f1ef5e7c1-009416.jsonl?bytes=0,1133972'" % (code_path)
logging.info(cmd) # logging.info(cmd)
common.check_shell(cmd) # common.check_shell(cmd)
#common.count_folders_and_check_contents(pdf_res_path) # #common.count_folders_and_check_contents(pdf_res_path)
def test_pdf_specify_jsonl_txt(self): # def test_pdf_specify_jsonl_txt(self):
""" # """
输入jsonl, txt方式解析 # 输入jsonl, txt方式解析
""" # """
cmd = "cd %s && export PYTHONPATH=. && python magic_pdf/cli/magicpdf.py json-command --json 's3://llm-process-pperf/ebook_index_textbook_40k/中高考&竞赛知识点/part-663f1ef5e7c1-009416.jsonl?bytes=0,1133972' --method txt" % (code_path) # cmd = "cd %s && export PYTHONPATH=. && python magic_pdf/cli/magicpdf.py json-command --json 's3://llm-process-pperf/ebook_index_textbook_40k/中高考&竞赛知识点/part-663f1ef5e7c1-009416.jsonl?bytes=0,1133972' --method txt" % (code_path)
logging.info(cmd) # logging.info(cmd)
common.check_shell(cmd) # common.check_shell(cmd)
#common.count_folders_and_check_contents(pdf_res_path) # #common.count_folders_and_check_contents(pdf_res_path)
#
def test_pdf_specify_jsonl_ocr(self): # def test_pdf_specify_jsonl_ocr(self):
""" # """
输入jsonl, ocr方式解析 # 输入jsonl, ocr方式解析
""" # """
cmd = "cd %s && export PYTHONPATH=. && python magic_pdf/cli/magicpdf.py json-command --json 's3://llm-process-pperf/ebook_index_textbook_40k/中高考&竞赛知识点/part-663f1ef5e7c1-009416.jsonl?bytes=0,1133972' --method ocr" % (code_path) # cmd = "cd %s && export PYTHONPATH=. && python magic_pdf/cli/magicpdf.py json-command --json 's3://llm-process-pperf/ebook_index_textbook_40k/中高考&竞赛知识点/part-663f1ef5e7c1-009416.jsonl?bytes=0,1133972' --method ocr" % (code_path)
logging.info(cmd) # logging.info(cmd)
common.check_shell(cmd) # common.check_shell(cmd)
#common.count_folders_and_check_contents(pdf_res_path) # #common.count_folders_and_check_contents(pdf_res_path)
if __name__ == "__main__": if __name__ == "__main__":
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment