threads / patch / 34803

patchcheck-ignore: Add option to ignore index contents

Subject: [PATCH] check-ignore: Add option to ignore index contents

## tl;dr

6 messages between Aug 29, 2013 and Aug 30, 2013. Diffs are folded; open one to read it.

replies: 5people: 5as markdown or json

Dave Williams· Aug 29, 2013, 22:46 UTC · lore

check-ignore currently shows how .gitignore rules would treat untracked paths. Tracked paths do not generate useful output. This prevents debugging of why a path became tracked unexpectedly unless that path is first removed from the index with git rm --cached <path>

This option (-i, --ignore-index) simply by-passes the check for the path being in the index and hence allows tracked path to be checked too.

Whilst this behaviour deviates from the characteristics of git add and git status its use case is unlikely to cause any user confusion.

Signed-off-by: Dave Williams <dave@opensourcesolutions.co.uk>
---
 Documentation/git-check-ignore.txt |  6 ++++++
 builtin/check-ignore.c             | 10 +++++++---
 2 files changed, 13 insertions(+), 3 deletions(-)
Show changes to 2 files +13 −3

Documentation/git-check-ignore.txt, builtin/check-ignore.c

diff --git a/Documentation/git-check-ignore.txt b/Documentation/git-check-ignore.txt
index d2df487..bb878ff 100644
--- a/Documentation/git-check-ignore.txt
+++ b/Documentation/git-check-ignore.txt
@@ -45,6 +45,12 @@ OPTIONS
 	not be possible to distinguish between paths which match a
 	pattern and those which don't.
 
+-i, --ignore-index::
+	Don't look in the index when undertaking the checks. This means
+	the results deviate from those seen by git add and git status
+	but is useful when understanding why a path became tracked by
+	e.g. git add . and was not ignored as expected by the rules.
+
 OUTPUT
 ------
 
diff --git a/builtin/check-ignore.c b/builtin/check-ignore.c
index 4a8fc70..c8f6ae1 100644
--- a/builtin/check-ignore.c
+++ b/builtin/check-ignore.c
@@ -5,7 +5,7 @@
 #include "pathspec.h"
 #include "parse-options.h"
 
-static int quiet, verbose, stdin_paths, show_non_matching;
+static int quiet, verbose, stdin_paths, show_non_matching, ignore_index;
 static const char * const check_ignore_usage[] = {
 "git check-ignore [options] pathname...",
 "git check-ignore [options] --stdin < <list-of-paths>",
@@ -24,6 +24,8 @@ static const struct option check_ignore_options[] = {
 		    N_("input paths are terminated by a null character")),
 	OPT_BOOLEAN('n', "non-matching", &show_non_matching,
 		    N_("show non-matching input paths")),
+	OPT_BOOLEAN('i', "ignore-index", &ignore_index,
+		    N_("ignore index when checking")),
 	OPT_END()
 };
 
@@ -82,7 +84,9 @@ static int check_ignore(struct dir_struct *dir,
 	 * should not be ignored, in order to be consistent with
 	 * 'git status', 'git add' etc.
 	 */
-	seen = find_pathspecs_matching_against_index(pathspec);
+	if (!ignore_index) {
+		seen = find_pathspecs_matching_against_index(pathspec);
+	}
 	for (i = 0; pathspec[i]; i++) {
 		path = pathspec[i];
 		full_path = prefix_path(prefix, prefix
@@ -90,7 +94,7 @@ static int check_ignore(struct dir_struct *dir,
 		full_path = check_path_for_gitlink(full_path);
 		die_if_path_beyond_symlink(full_path, prefix);
 		exclude = NULL;
-		if (!seen[i]) {
+		if (ignore_index || !seen[i]) {
 			exclude = last_exclude_matching(dir, full_path, &dtype);
 		}
 		if (!quiet && (exclude || show_non_matching))

-- 
1.8.4.rc3
Adam Spiers· Aug 30, 2013, 00:10 UTC · re: Dave Williams · lore

Re: [PATCH] check-ignore: Add option to ignore index contents

On Thu, Aug 29, 2013 at 11:46:52PM +0100, Dave Williams wrote:
Show 12 quoted lines
> check-ignore currently shows how .gitignore rules would treat untracked
> paths. Tracked paths do not generate useful output.  This prevents
> debugging of why a path became tracked unexpectedly unless that path is
> first removed from the index with git rm --cached <path>
> 
> This option (-i, --ignore-index) simply by-passes the check for the path
> being in the index and hence allows tracked path to be checked too.
> 
> Whilst this behaviour deviates from the characteristics of git add and
> git status its use case is unlikely to cause any user confusion.
> 
> Signed-off-by: Dave Williams <dave@opensourcesolutions.co.uk>
[snipped]

Sounds like a great idea to me, and the code changes look reasonable; however you forgot to add corresponding tests :-) Please can you expand the patch to include them?

But I'm in favour in principle - thanks a lot for working on this!
Duy Nguyen· Aug 30, 2013, 00:51 UTC · re: Dave Williams · lore

Re: [PATCH] check-ignore: Add option to ignore index contents

On Fri, Aug 30, 2013 at 5:46 AM, Dave Williams <dave@opensourcesolutions.co.uk> wrote:

Show 7 quoted lines
> check-ignore currently shows how .gitignore rules would treat untracked
> paths. Tracked paths do not generate useful output.  This prevents
> debugging of why a path became tracked unexpectedly unless that path is
> first removed from the index with git rm --cached <path>
>
> This option (-i, --ignore-index) simply by-passes the check for the path
> being in the index and hence allows tracked path to be checked too.
Maybe --no-index, already used by git-diff and git-grep.
> Whilst this behaviour deviates from the characteristics of git add and
> git status its use case is unlikely to cause any user confusion.
-- 
Duy
Junio C Hamano· Aug 30, 2013, 04:13 UTC · re: Duy Nguyen · lore

Re: [PATCH] check-ignore: Add option to ignore index contents

Duy Nguyen <pclouds@gmail.com> writes:
Show 11 quoted lines
> On Fri, Aug 30, 2013 at 5:46 AM, Dave Williams
> <dave@opensourcesolutions.co.uk> wrote:
>> check-ignore currently shows how .gitignore rules would treat untracked
>> paths. Tracked paths do not generate useful output.  This prevents
>> debugging of why a path became tracked unexpectedly unless that path is
>> first removed from the index with git rm --cached <path>
>>
>> This option (-i, --ignore-index) simply by-passes the check for the path
>> being in the index and hence allows tracked path to be checked too.
>
> Maybe --no-index, already used by git-diff and git-grep.

I am somewhat ashamed that I didn't think of it myself when I saw the patch, but I agree that would be a good change for consistency.

>
>> Whilst this behaviour deviates from the characteristics of git add and
>> git status its use case is unlikely to cause any user confusion.
Junio C Hamano· Aug 30, 2013, 00:57 UTC · re: Dave Williams · lore

Re: [PATCH] check-ignore: Add option to ignore index contents

Dave Williams <dave@opensourcesolutions.co.uk> writes:
Show 30 quoted lines
> check-ignore currently shows how .gitignore rules would treat untracked
> paths. Tracked paths do not generate useful output.  This prevents
> debugging of why a path became tracked unexpectedly unless that path is
> first removed from the index with git rm --cached <path>
>
> This option (-i, --ignore-index) simply by-passes the check for the path
> being in the index and hence allows tracked path to be checked too.
>
> Whilst this behaviour deviates from the characteristics of git add and
> git status its use case is unlikely to cause any user confusion.
>
> Signed-off-by: Dave Williams <dave@opensourcesolutions.co.uk>
> ---
>  Documentation/git-check-ignore.txt |  6 ++++++
>  builtin/check-ignore.c             | 10 +++++++---
>  2 files changed, 13 insertions(+), 3 deletions(-)
>
> diff --git a/Documentation/git-check-ignore.txt b/Documentation/git-check-ignore.txt
> index d2df487..bb878ff 100644
> --- a/Documentation/git-check-ignore.txt
> +++ b/Documentation/git-check-ignore.txt
> @@ -45,6 +45,12 @@ OPTIONS
>  	not be possible to distinguish between paths which match a
>  	pattern and those which don't.
>  
> +-i, --ignore-index::
> +	Don't look in the index when undertaking the checks. This means
> +	the results deviate from those seen by git add and git status
> +	but is useful when understanding why a path became tracked by
> +	e.g. git add . and was not ignored as expected by the rules.

Please think twice after you write "This means", "What this means is", etc. to see if you can get rid of sentence before that, which needed such clarification. If the sentence can stand on its own, and what follows "This means" is also useful information, then it is a sign that "What this means is" is better rephrased as "Note that".

Also, I think `git add .` should be quoted to make it clear where the sample command begins and ends. Using a pair of `` would be the most appropriate here.

Show 45 quoted lines
>  OUTPUT
>  ------
>  
> diff --git a/builtin/check-ignore.c b/builtin/check-ignore.c
> index 4a8fc70..c8f6ae1 100644
> --- a/builtin/check-ignore.c
> +++ b/builtin/check-ignore.c
> @@ -5,7 +5,7 @@
>  #include "pathspec.h"
>  #include "parse-options.h"
>  
> -static int quiet, verbose, stdin_paths, show_non_matching;
> +static int quiet, verbose, stdin_paths, show_non_matching, ignore_index;
>  static const char * const check_ignore_usage[] = {
>  "git check-ignore [options] pathname...",
>  "git check-ignore [options] --stdin < <list-of-paths>",
> @@ -24,6 +24,8 @@ static const struct option check_ignore_options[] = {
>  		    N_("input paths are terminated by a null character")),
>  	OPT_BOOLEAN('n', "non-matching", &show_non_matching,
>  		    N_("show non-matching input paths")),
> +	OPT_BOOLEAN('i', "ignore-index", &ignore_index,
> +		    N_("ignore index when checking")),
>  	OPT_END()
>  };
>  
> @@ -82,7 +84,9 @@ static int check_ignore(struct dir_struct *dir,
>  	 * should not be ignored, in order to be consistent with
>  	 * 'git status', 'git add' etc.
>  	 */
> -	seen = find_pathspecs_matching_against_index(pathspec);
> +	if (!ignore_index) {
> +		seen = find_pathspecs_matching_against_index(pathspec);
> +	}
>  	for (i = 0; pathspec[i]; i++) {
>  		path = pathspec[i];
>  		full_path = prefix_path(prefix, prefix
> @@ -90,7 +94,7 @@ static int check_ignore(struct dir_struct *dir,
>  		full_path = check_path_for_gitlink(full_path);
>  		die_if_path_beyond_symlink(full_path, prefix);
>  		exclude = NULL;
> -		if (!seen[i]) {
> +		if (ignore_index || !seen[i]) {
>  			exclude = last_exclude_matching(dir, full_path, &dtype);
>  		}
>  		if (!quiet && (exclude || show_non_matching))

Hmph, if ignoring what is in the index is what the new feature wants to do, why does it even need to touch check_ignore() function at all and add a global variable ignore_index to keep track of that state?

Isn't it sufficient to skip the call to read_cache() when that option is given?

You would need to add tests for this, too.
Thanks.
Johannes Sixt· Aug 30, 2013, 06:51 UTC · re: Dave Williams · lore

Re: [PATCH] check-ignore: Add option to ignore index contents

Am 8/30/2013 0:46, schrieb Dave Williams:
Show 5 quoted lines
> +-i, --ignore-index::
> +	Don't look in the index when undertaking the checks. This means
> +	the results deviate from those seen by git add and git status
> +	but is useful when understanding why a path became tracked by
> +	e.g. git add . and was not ignored as expected by the rules.

Did you mean to say "was not ignored as expected by the user"? Because the "the rules" cannot "expect" something, they only "are applied".

-- Hannes

← back to recent threads