Re: [PATCH v3 5/9] ref-filter: add option to match literal pattern

On Mon, Jul 20, 2015 at 4:01 AM, Christian Couder
[off-list ref] wrote:

On Mon, Jul 20, 2015 at 8:24 AM, Eric Sunshine [off-list ref] wrote:

quoted

On Sat, Jul 18, 2015 at 3:12 PM, Karthik Nayak [off-list ref] wrote:

quoted

+static int filter_pattern_match(struct ref_filter *filter, const char *refname)
+{
+       if (!*filter->name_patterns)
+               return 1;
+       if (filter->match_as_path)
+               return match_name_as_path(filter->name_patterns, refname);
+       return match_pattern(filter->name_patterns, refname);
+}

@@ -1034,7 +1057,7 @@ static int ref_filter_handler(const char *refname, const struct object_id *oid,
-       if (*filter->name_patterns && !match_name_as_path(filter->name_patterns, refname))
+       if (!filter_pattern_match(filter, refname))
                return 0;

I find it much more difficult to grok the new logic due to
'*filter->name_patterns' having moved into the called function and its
negation inside the function returning 1 which is then negated (again)
upon return here. This sort of twisty logic places a higher cognitive
load on the reader. Retaining the original logic makes the code far
simpler to understand:

    if (*filter->name_patterns &&
        !filter_pattern_match(filter, refname))
        return 0;

although it's a bit less nicely encapsulated, so I dunno...

I think a comment before filter_pattern_match() and perhaps also one
inside it might help. For example something like:

/* Return 1 if the refname matches one of the patterns, otherwise 0. */
static int filter_pattern_match(struct ref_filter *filter, const char *refname)
{
       if (!*filter->name_patterns)
               return 1; /* No pattern always matches */
       if (filter->match_as_path)
               return match_name_as_path(filter->name_patterns, refname);
       return match_pattern(filter->name_patterns, refname);
}

Yes, the comments do improve the situation and may be a reasonable compromise...