git/list[1] front-page[2] threads[3] people[4] search[5] about
 

Re: [PATCH v2 03/27] parse_arg(): Really test that argument is properly terminated

From
Junio C Hamano <gitster@pobox.com>
Date
Mar 31, 2014, 21:36 UTC
Message-ID
<xmqqlhvq3vaj.fsf@gitster.dls.corp.google.com>
In-Reply-To
<1395683820-17304-4-git-send-email-mhagger@alum.mit.edu>
Michael Haggerty <mhagger@alum.mit.edu> writes:
Show 63 quoted lines
> Test that the argument is properly terminated by either whitespace or
> a NUL character, even if it is quoted, to be consistent with the
> non-quoted case.  Adjust the tests to expect the new error message.
> Add a docstring to the function, incorporating the comments that were
> formerly within the function plus some added information.
>
> Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
> ---
>  builtin/update-ref.c  | 20 +++++++++++++++-----
>  t/t1400-update-ref.sh |  4 ++--
>  2 files changed, 17 insertions(+), 7 deletions(-)
>
> diff --git a/builtin/update-ref.c b/builtin/update-ref.c
> index 1292cfe..02b5f95 100644
> --- a/builtin/update-ref.c
> +++ b/builtin/update-ref.c
> @@ -62,16 +62,26 @@ static void update_store_old_sha1(struct ref_update *update,
>  	update->have_old = *oldvalue || line_termination;
>  }
>  
> +/*
> + * Parse one whitespace- or NUL-terminated, possibly C-quoted argument
> + * and append the result to arg.  Return a pointer to the terminator.
> + * Die if there is an error in how the argument is C-quoted.  This
> + * function is only used if not -z.
> + */
>  static const char *parse_arg(const char *next, struct strbuf *arg)
>  {
> -	/* Parse SP-terminated, possibly C-quoted argument */
> -	if (*next != '"')
> +	if (*next == '"') {
> +		const char *orig = next;
> +
> +		if (unquote_c_style(arg, next, &next))
> +			die("badly quoted argument: %s", orig);
> +		if (*next && !isspace(*next))
> +			die("unexpected character after quoted argument: %s", orig);
> +	} else {
>  		while (*next && !isspace(*next))
>  			strbuf_addch(arg, *next++);
> -	else if (unquote_c_style(arg, next, &next))
> -		die("badly quoted argument: %s", next);
> +	}
>  
> -	/* Return position after the argument */
>  	return next;
>  }
>  
> diff --git a/t/t1400-update-ref.sh b/t/t1400-update-ref.sh
> index 29391c6..774f8c5 100755
> --- a/t/t1400-update-ref.sh
> +++ b/t/t1400-update-ref.sh
> @@ -356,10 +356,10 @@ test_expect_success 'stdin fails on badly quoted input' '
>  	grep "fatal: badly quoted argument: \\\"master" err
>  '
>  
> -test_expect_success 'stdin fails on arguments not separated by space' '
> +test_expect_success 'stdin fails on junk after quoted argument' '
>  	echo "create \"$a\"master" >stdin &&
>  	test_must_fail git update-ref --stdin <stdin 2>err &&
> -	grep "fatal: expected SP but got: master" err
> +	grep "fatal: unexpected character after quoted argument: \\\"$a\\\"master" err
>  '
Interesting.

I would have expected that "We used to check only one of the two codepaths and the other was loose, fix it" to be accompanied by "So here is an _addition_ to the test suite to validate the other case that used to be loose, now tightened", not "We update one existing case". The test before and after the patch is about a c-quoted string, so I am not sure if we are still testing the right thing.

The code in update-ref.c after the patch does look reasonable, though.

Previous: Michael HaggertyNext: Michael Haggerty
Message 8 of 65 in “Clean up update-refs --stdin and implement ref_transaction”
  1. 00/27 Clean up update-refs --stdin and implement ref_transactionMichael Haggerty, Mar 24, 2014
  2. 01/27 t1400: Fix name and expected result of one testMichael Haggerty, Mar 24, 2014
  3. Junio C HamanoMar 31, 2014
  4. Michael HaggertyMar 31, 2014
  5. 02/27 t1400: Provide more usual input to the commandMichael Haggerty, Mar 24, 2014
  6. Junio C HamanoMar 31, 2014
  7. 03/27 parse_arg(): Really test that argument is properly terminatedMichael Haggerty, Mar 24, 2014
  8. Junio C HamanoMar 31, 2014
  9. Michael HaggertyMar 31, 2014
  10. 04/27 t1400: Add some more tests involving quoted argumentsMichael Haggerty, Mar 24, 2014
  11. 05/27 refs.h: Rename the action_on_err constantsMichael Haggerty, Mar 24, 2014
  12. 06/27 update_refs(): Fix constnessMichael Haggerty, Mar 24, 2014
  13. Junio C HamanoMar 31, 2014
  14. Michael HaggertyMar 31, 2014
  15. Junio C HamanoMar 31, 2014
  16. 07/27 update-ref --stdin: Read the whole input at onceMichael Haggerty, Mar 24, 2014
  17. 08/27 parse_cmd_verify(): Copy old_sha1 instead of evaluating <oldvalue> twiceMichael Haggerty, Mar 24, 2014
  18. 09/27 update-ref.c: Extract a new function, parse_refname()Michael Haggerty, Mar 24, 2014
  19. 10/27 update-ref --stdin: Improve error messages for invalid valuesMichael Haggerty, Mar 24, 2014
  20. 11/27 update-ref --stdin: Make error messages more consistentMichael Haggerty, Mar 24, 2014
  21. 12/27 update-ref --stdin: Simplify error messages for missing oldvaluesMichael Haggerty, Mar 24, 2014
  22. 13/27 t1400: Test that stdin -z update treats empty <newvalue> as zerosMichael Haggerty, Mar 24, 2014
  23. Junio C HamanoMar 31, 2014
  24. Michael HaggertyMar 31, 2014
  25. 14/27 update-ref.c: Extract a new function, parse_next_sha1()Michael Haggerty, Mar 24, 2014
  26. Brad KingMar 26, 2014
  27. 15/27 update-ref --stdin -z: Deprecate interpreting the empty string as zerosMichael Haggerty, Mar 24, 2014
  28. Junio C HamanoMar 31, 2014
  29. 16/27 t1400: Test one mistake at a timeMichael Haggerty, Mar 24, 2014
  30. Brad KingMar 26, 2014
  31. Junio C HamanoMar 31, 2014
  32. Michael HaggertyMar 31, 2014
  33. 17/27 update-ref --stdin: Improve the error message for unexpected EOFMichael Haggerty, Mar 24, 2014
  34. 18/27 update-ref --stdin: Harmonize error messagesMichael Haggerty, Mar 24, 2014
  35. Junio C HamanoMar 31, 2014
  36. Michael HaggertyMar 31, 2014
  37. Michael HaggertyApr 1, 2014
  38. Junio C HamanoApr 2, 2014
  39. 19/27 refs: Add a concept of a reference transactionMichael Haggerty, Mar 24, 2014
  40. Brad KingMar 26, 2014
  41. Michael HaggertyMar 26, 2014
  42. Junio C HamanoApr 1, 2014
  43. Michael HaggertyApr 2, 2014
  44. 20/27 update-ref --stdin: Reimplement using reference transactionsMichael Haggerty, Mar 24, 2014
  45. Junio C HamanoApr 1, 2014
  46. Michael HaggertyApr 2, 2014
  47. Junio C HamanoApr 3, 2014
  48. Michael HaggertyApr 4, 2014
  49. 21/27 refs: Remove API function update_refs()Michael Haggerty, Mar 24, 2014
  50. Junio C HamanoApr 1, 2014
  51. 22/27 struct ref_update: Rename field "ref_name" to "refname"Michael Haggerty, Mar 24, 2014
  52. Junio C HamanoApr 1, 2014
  53. Michael HaggertyApr 2, 2014
  54. 23/27 struct ref_update: Store refname as a FLEX_ARRAY.Michael Haggerty, Mar 24, 2014
  55. Junio C HamanoApr 1, 2014
  56. 24/27 ref_transaction_commit(): Introduce temporary variablesMichael Haggerty, Mar 24, 2014
  57. Junio C HamanoApr 1, 2014
  58. 25/27 struct ref_update: Add a lock memberMichael Haggerty, Mar 24, 2014
  59. 26/27 struct ref_update: Add type fieldMichael Haggerty, Mar 24, 2014
  60. Junio C HamanoApr 1, 2014
  61. Michael HaggertyApr 2, 2014
  62. Junio C HamanoApr 2, 2014
  63. 27/27 ref_transaction_commit(): Work with transaction->updates in placeMichael Haggerty, Mar 24, 2014
  64. Brad KingMar 26, 2014
  65. Michael HaggertyMar 26, 2014

Read the whole thread, see it on lore, or plain text.

$ cat FOOTERMessages come from the public archive at lore.kernel.org/git, fetched every hour. The front page is chosen and written each morning by an AI editor and can be wrong; the threads themselves are the record. About and API. For agents: an MCP server at https://gitlist.dev/mcp, and any thread, story or person page as Markdown by adding .md to its URL (or sending Accept: text/markdown). Details in /llms.txt.