uncurl - Convert curl to python-requests

Common errors

• Curl command options like -u , -X etc are rejected.

  cause The `uncurl.parse()` function can be less robust with certain `curl` command-line options when trying to directly generate a `requests` string, leading to incomplete or incorrect output.
  fix Instead of `uncurl.parse()`, use `uncurl.parse_context()` which returns a structured object containing URL, headers, method, data, cookies, and authentication. You can then use these components to build a `requests` call more reliably. Alternatively, simplify the `curl` command before passing it to `uncurl.parse()`.

• ModuleNotFoundError: No module named 'requests'

  cause `uncurl` converts `curl` commands into Python code that typically uses the `requests` library. However, `requests` is not a direct runtime dependency of `uncurl` itself and is not automatically installed when you install `uncurl`.
  fix You need to explicitly install the `requests` library in your environment: `pip install requests`.

• uncurl generated requests code not working as expected

  cause The translation from a `curl` command to `requests` code might not always be perfect, especially for non-standard headers, complex POST data, or specific `curl` flags (like `--compressed` which `requests` handles automatically, or certificate validation with `-k`/`--insecure`).
  fix Examine the original `curl` command closely. Use `uncurl.parse_context()` to get detailed components and then debug the `requests` call by inspecting `context.headers`, `context.data`, `context.cookies`, `context.auth`, and the `verify` parameter. Ensure `requests` is handling compression (often automatic) and certificate validation as intended.

Warnings

• gotcha The `uncurl` library's last release was in March 2021. This suggests a low level of active maintenance, which might affect its compatibility with very recent `curl` versions or advanced, newer `curl` features and syntax.
  Fix: For cutting-edge `curl` features or complex, evolving web environments, consider manually verifying `uncurl`'s output or exploring alternatives/forks like `uncurlx` for `httpx`.
• gotcha Complex `curl` commands, especially those with intricate authentication mechanisms (e.g., NTLM, Kerberos), multi-part form data, or highly specific header encoding, might not be fully or accurately translated by `uncurl`.
  Fix: Always inspect the output of `uncurl.parse_context()` and the resulting `requests` call. Manual adjustments to headers, data, or the `auth` parameter may be necessary. For problematic `curl` options like `-u` or `-X`, consider stripping them or handling them explicitly after parsing.
• gotcha The `uncurl` command-line tool's clipboard integration (e.g., `pbpaste | uncurl`) is primarily demonstrated and might work best for macOS. On other operating systems, you may need to manually pipe the `curl` command to `uncurl` or install the `pyperclip` library for seamless clipboard access.
  Fix: On Linux, use `xclip -o | uncurl` (with `xclip` installed) or manually paste the `curl` command string as an argument. For cross-platform clipboard support, install `pip install pyperclip` and integrate it into your script.

Install

• pip install uncurl Install `uncurl`

Imports

• parse
  wrong:

  import uncurl.parse

  correct:

  from uncurl import parse

  The `parse` function is directly imported from the `uncurl` package.
• parse_context
  wrong:

  import uncurl.parse_context

  correct:

  from uncurl import parse_context

  The `parse_context` function is directly imported and provides structured data.

Quickstart

This quickstart demonstrates how to use `uncurl.parse_context` to extract components from a `curl` command and then build and execute a `requests` call. This method is generally more robust than `uncurl.parse` for complex commands. It also shows how to get the `requests` code as a string directly.

import uncurl
import requests

curl_command = "curl 'https://pypi.python.org/pypi/uncurl' -H 'Accept-Encoding: gzip,deflate,sdch' -H 'User-Agent: Mozilla/5.0' --compressed"

# Get structured components of the curl command
context = uncurl.parse_context(curl_command)

# Construct and execute the requests call
response = requests.request(
    context.method.upper(),
    context.url,
    headers=context.headers,
    data=context.data,
    cookies=context.cookies,
    auth=context.auth,
    verify=(not context.insecure) # handle --insecure curl flag
)

print(f"Status Code: {response.status_code}")
# print(response.text)

# Alternatively, get the requests code as a string
# requests_code_string = uncurl.parse(curl_command)
# print(requests_code_string)