Update the docstring of untokenize() to match the docs

[tomasr8]

Documentation

In #128031, it was deemed too risky to change untokenize() to more closely match the original input string. We should update the docstring of untokenize to reflect that and prevent confusion in the (rare) case when untokenize does not match the input.

The docs already make that clear so we can simply reuse that explanation:

cpython/Doc/library/tokenize.rst

Lines 94 to 98 in 2228e92

	The reconstructed script is returned as a single string. The result is
	guaranteed to tokenize back to match the input so that the conversion is
	lossless and round-trips are assured. The guarantee applies only to the
	token type and token string as the spacing between tokens (column
	positions) may change.

Linked PRs

• gh-128519: Update the docstring of untokenize() to match the docs #128521
• [3.13] gh-128519: Align the docstring of untokenize() to match the docs (GH-128521) #128531
• [3.12] gh-128519: Align the docstring of untokenize() to match the docs (GH-128521) #128532

[erlend-aasland]

BTW, it would have been perfectly fine to use #128031 for #128521, instead of creating a new issue.

[erlend-aasland]

Resolved by:

• gh-128519: Update the docstring of untokenize() to match the docs #128521
• [3.13] gh-128519: Align the docstring of untokenize() to match the docs (GH-128521) #128531
• [3.12] gh-128519: Align the docstring of untokenize() to match the docs (GH-128521) #128532

[tomasr8]

BTW, it would have been perfectly fine to use #128031 for #128521, instead of creating a new issue.

I thought since that issue was closed, it was better to make a new one.. Sorry for the noise!