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

Re: [PATCH 2/4] show: integrate with the sparse index

From
Derrick Stolee <derrickstolee@github.com>
Date
Apr 18, 2022, 12:28 UTC
Message-ID
<4c999e1f-e6bb-72e0-a27d-e072ce9487f8@github.com>
In-Reply-To
<Ylhs3t9nFS/IeO2Y@google.com>
On 4/14/2022 2:50 PM, Josh Steadmon wrote:
> On 2022.04.07 16:37, Derrick Stolee via GitGitGadget wrote:
>> From: Derrick Stolee <dstolee@microsoft.com>
Show 10 quoted lines
>>  	# Asking "git show" for directories in the index
>> -	# does not work as implemented. The error message is
>> -	# different for a full checkout and a sparse checkout
>> -	# when the directory is outside of the cone.
>> +	# changes depending on the existence of a sparse index.
> 
> The wording here seems awkward after these changes are applied. Without
> other context, it makes it sound to me like the command(s) used to show
> a directory change depending on the existence of a sparse index, rather
> than the fact that the behavior of `git show` changes.
I can see that.
Show 15 quoted lines
>> +	# The output of "git show" includes the way we referenced the
>> +	# objects, so strip that out.
>> +	test_line_count = 4 actual &&
>> +	tail -n 2 actual >actual-trunc &&
>> +	tail -n 2 expect >expect-trunc &&
>> +	test_cmp expect-trunc actual-trunc
>>  '
> 
> It's not specific to this commit, but in general I think the series of
> changes to this test would be easier to follow if we used hard-coded
> strings to compare against, rather than matching parts of files against
> each other. It makes it more clear to the reader exactly which behavior
> is changing, and can make it more obvious why certain output is
> undesirable. However, it would make the test more brittle to future
> changes.

The tests here are designed with this approach in mind: demonstrate success by comparing to existing behavior. We don't want to be coupled to the exact behavior of these commands, but we _do_ want to demonstrate that the sparse-checkout or sparse-index features do not change from the full-checkout behavior (unless we are demonstrating an expected difference).

In particular, using comparisons like this are also robust against changes in the test repository data shape, which has been necessary to update as bugs are found.

Thanks, -Stolee

Previous: Josh SteadmonNext: Derrick Stolee via GitGitGadget
Message 7 of 26 in “Sparse index integration with 'git show'”
  1. 0/4 Sparse index integration with 'git show'Derrick Stolee via GitGitGadget, Apr 7, 2022
  2. 1/4 t1092: add compatibility tests for 'git show'Derrick Stolee via GitGitGadget, Apr 7, 2022
  3. Josh SteadmonApr 14, 2022
  4. Derrick StoleeApr 18, 2022
  5. 2/4 show: integrate with the sparse indexDerrick Stolee via GitGitGadget, Apr 7, 2022
  6. Josh SteadmonApr 14, 2022
  7. Derrick StoleeApr 18, 2022
  8. 3/4 object-name: reject trees found in the indexDerrick Stolee via GitGitGadget, Apr 7, 2022
  9. Josh SteadmonApr 14, 2022
  10. Derrick StoleeApr 18, 2022
  11. 4/4 object-name: diagnose trees in index properlyDerrick Stolee via GitGitGadget, Apr 7, 2022
  12. Philip OakleyApr 7, 2022
  13. Junio C HamanoApr 12, 2022
  14. Derrick StoleeApr 12, 2022
  15. Junio C HamanoApr 12, 2022
  16. Josh SteadmonApr 14, 2022
  17. Junio C HamanoApr 14, 2022
  18. Derrick StoleeApr 18, 2022
  19. 0/5 Sparse index integration with 'git show'Derrick Stolee via GitGitGadget, Apr 26, 2022
  20. 1/5 t1092: add compatibility tests for 'git show'Derrick Stolee via GitGitGadget, Apr 26, 2022
  21. 2/5 show: integrate with the sparse indexDerrick Stolee via GitGitGadget, Apr 26, 2022
  22. 3/5 object-name: reject trees found in the indexDerrick Stolee via GitGitGadget, Apr 26, 2022
  23. 4/5 object-name: diagnose trees in index properlyDerrick Stolee via GitGitGadget, Apr 26, 2022
  24. 5/5 rev-parse: integrate with sparse indexDerrick Stolee via GitGitGadget, Apr 26, 2022
  25. Junio C HamanoApr 26, 2022
  26. Derrick StoleeApr 27, 2022

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.