Re: [PATCH 3/3] fast-export, fast-import: implement signed-commits

[Taylor Blau <ttaylorr@github.com>]

On Mon, Apr 19, 2021 at 04:54:41PM -0600, Luke Shumaker wrote:
> From: Luke Shumaker <lukeshu@datawire.io>
>
> fast-export has an existing --signed-tags= flag that controls how to
> handle tag signatures.  However, there is no equivalent for commit
> signatures; it just silently strips the signature out of the commit
> (analogously to --signed-tags=strip).
>
> While signatures are generally problematic for fast-export/fast-import
> (because hashes are likely to change), if they're going to support tag
> signatures, there's no reason to not also support commit signatures.
>
> So, implement signed-commits.
>
> On the fast-export side, try to be as much like signed-tags as possible,
> in both implementation and in user-interface; with the exception that
> the default should be `--signed-commits=strip` (compared to the default
> `--signed-tags=abort`), in order to continue defaulting to the
> historical behavior.  Only bother implementing "gpgsig", not
> "gpgsig-sha256"; the existing signed-tag support doesn't implement
> "gpgsig-sha256" either.
>
> On the fast-import side, I'm not entirely sure that I got the ordering
> correct between "gpgsig" and "encoding" when generating the commit
> object.
>
> Signed-off-by: Luke Shumaker <lukeshu@datawire.io>
> ---
>  Documentation/git-fast-export.txt | 12 +++++
>  Documentation/git-fast-import.txt |  7 +++
>  builtin/fast-export.c             | 86 +++++++++++++++++++++++++------
>  builtin/fast-import.c             | 15 ++++++
>  t/t9350-fast-export.sh            | 70 +++++++++++++++++++++++++
>  5 files changed, 174 insertions(+), 16 deletions(-)
>
> diff --git a/Documentation/git-fast-export.txt b/Documentation/git-fast-export.txt
> index d4a2bfe037..6fdb678b54 100644
> --- a/Documentation/git-fast-export.txt
> +++ b/Documentation/git-fast-export.txt
> @@ -39,6 +39,18 @@ warning will be displayed, with 'verbatim', they will be silently
>  exported and with 'warn-verbatim', they will be exported, but you will
>  see a warning.
>
> +--signed-commits=(verbatim|warn-verbatim|warn-strip|strip|abort)::
> +	Specify how to handle signed commits.  Since any transformation
> +	after the export can change the commit (which can also happen
> +	when excluding revisions) the signatures will not match.
> ++
> +When asking to 'abort', this program will die when encountering a
> +signed commit.  With 'strip' (which is the default), the commits will
> +silently be made unsigned, with 'warn-strip' they will be made
> +unsigned but a warning will be displayed, with 'verbatim', they will
> +be silently exported and with 'warn-verbatim', they will be exported,
> +but you will see a warning.
> +

OK, this all seems normal to me. But it may be worth shortening it to
say "behaves exactly as --signed-tags, but for commits", or something.

>  --tag-of-filtered-object=(abort|drop|rewrite)::
>  	Specify how to handle tags whose tagged object is filtered out.
>  	Since revisions and files to export can be limited by path,
> diff --git a/Documentation/git-fast-import.txt b/Documentation/git-fast-import.txt
> index 458af0a2d6..3d0c5dbf7d 100644
> --- a/Documentation/git-fast-import.txt
> +++ b/Documentation/git-fast-import.txt
> @@ -437,6 +437,7 @@ change to the project.
>  	original-oid?
>  	('author' (SP <name>)? SP LT <email> GT SP <when> LF)?
>  	'committer' (SP <name>)? SP LT <email> GT SP <when> LF
> +	('gpgsig' LF data)?

Is this missing a LF after data?

> +static const char *find_signature(const char *begin, const char *end)
> +{
> +	const char *needle = "\ngpgsig ";
> +	char *bod, *eod, *eol;
> +
> +	bod = memmem(begin, end ? end - begin : strlen(begin),
> +		     needle, strlen(needle));
> +	if (!bod)
> +		return NULL;
> +	bod += strlen(needle);
> +	eod = strchrnul(bod, '\n');
> +	while (eod[0] == '\n' && eod[1] == ' ') {
> +		eod = strchrnul(eod+1, '\n');
> +	}
> +	*eod = '\0';
> +
> +	while ((eol = strstr(bod, "\n ")))
> +		memmove(eol+1, eol+2, strlen(eol+1));

Hmm. I'm not quite sure I follow these last two lines. Perhaps a comment
would help? The rest of this patch looks reasonable to me.

Thanks,
Taylor