search

princeton-nlp / SWE-bench

[ICLR 2024] SWE-Bench: Can Language Models Resolve Real-world Github Issues?
https://www.swebench.com
MIT License
1.47k stars 241 forks source link

Use the more portable . instead of source #85

Closed kevinlu1248 closed 3 months ago

kevinlu1248 commented 3 months ago

This avoids errors such as /bin/sh: 1: source: not found

$ python harness/run_evaluation.py     --predictions_path example-predictions.json  --swe_bench_tasks example.json     --log_dir example-log/     --testbed example-testbed/
Error: Command 'source /home/jgross/Documents/GitHub/SWE-bench/example-testbed/fakemodel/sqlfluff__sqlfluff/1.4/tmpdx4urtao/miniconda3/bin/activate sqlfluff__sqlfluff__1.4 && echo 'activate successful' && pip install -r /home/jgross/Documents/GitHub/SWE-bench/example-testbed/fakemodel/sqlfluff__sqlfluff/1.4/tmpfnim3stm/requirements.txt' returned non-zero exit status 127.
Error stdout:
Error stderr: /bin/sh: 1: source: not found

Error traceback: Traceback (most recent call last):
  File "/home/jgross/Documents/GitHub/SWE-bench/harness/context_manager.py", line 50, in __call__
    output = subprocess.run(cmd, **combined_args)
  File "/home/jgross/.local64/mambaforge/envs/swe-bench/lib/python3.9/subprocess.py", line 528, in run
    raise CalledProcessError(retcode, process.args,
subprocess.CalledProcessError: Command 'source /home/jgross/Documents/GitHub/SWE-bench/example-testbed/fakemodel/sqlfluff__sqlfluff/1.4/tmpdx4urtao/miniconda3/bin/activate sqlfluff__sqlfluff__1.4 && echo 'activate successful' && pip install -r /home/jgross/Documents/GitHub/SWE-bench/example-testbed/fakemodel/sqlfluff__sqlfluff/1.4/tmpfnim3stm/requirements.txt' returned non-zero exit status 127.

Traceback (most recent call last):
  File "/home/jgross/Documents/GitHub/SWE-bench/harness/run_evaluation.py", line 186, in <module>
    main(**vars(args))
  File "/home/jgross/Documents/GitHub/SWE-bench/harness/run_evaluation.py", line 162, in main
    eval_engine(args)
  File "/home/jgross/Documents/GitHub/SWE-bench/harness/engine_evaluation.py", line 164, in main
    setup_testbed(data_groups[0])
  File "/home/jgross/Documents/GitHub/SWE-bench/harness/engine_validation.py", line 89, in setup_testbed
    with TestbedContextManager(
  File "/home/jgross/Documents/GitHub/SWE-bench/harness/context_manager.py", line 288, in __enter__
    self.exec(cmd, shell=True)
  File "/home/jgross/Documents/GitHub/SWE-bench/harness/context_manager.py", line 58, in __call__
    raise e
  File "/home/jgross/Documents/GitHub/SWE-bench/harness/context_manager.py", line 50, in __call__
    output = subprocess.run(cmd, **combined_args)
  File "/home/jgross/.local64/mambaforge/envs/swe-bench/lib/python3.9/subprocess.py", line 528, in run
    raise CalledProcessError(retcode, process.args,
subprocess.CalledProcessError: Command 'source /home/jgross/Documents/GitHub/SWE-bench/example-testbed/fakemodel/sqlfluff__sqlfluff/1.4/tmpdx4urtao/miniconda3/bin/activate sqlfluff__sqlfluff__1.4 && echo 'activate successful' && pip install -r /home/jgross/Documents/GitHub/SWE-bench/example-testbed/fakemodel/sqlfluff__sqlfluff/1.4/tmpfnim3stm/requirements.txt' returned non-zero exit status 127.