[gimp] app: update action search to use glib API.
- From: Jehan Pagès <jehanp src gnome org>
- To: commits-list gnome org
- Cc:
- Subject: [gimp] app: update action search to use glib API.
- Date: Tue, 16 Sep 2014 02:06:15 +0000 (UTC)
commit b9a9169656cae021c57797e00cd33b87abfd1543
Author: Jehan <jehan girinstud io>
Date: Tue Sep 16 03:45:56 2014 +0200
app: update action search to use glib API.
I now normalize with g_str_tokenize_and_fold() which uses standard
Unicode normalization.
I don't use g_str_match_string() directly though, because I want to
run additional checks to order the results by relevance. For instance
I still want actions whose labels starts with the search string to be
at the top, and results with same order as search token before those
with a different order. Then results with match in the tooltip. Finally
I also returns results with partial match in the label, and the rest in
the tooltip, though at the bottom of the list.
Other than that, this returns the same results as g_str_match_string()
with a similar algorithm. In particular now we only match the start of
tokens (a substring in the middle of a token won't match anymore).
I kept the small 2-character trick matching the first letters of the
first 2 words of the label, but I got rid of the fuzzy search (that none
really found ever relevant anyway).
app/dialogs/action-search-dialog.c | 237 ++++++++++++++++++------------------
app/widgets/gimpaction-history.c | 2 +-
app/widgets/gimpaction-history.h | 2 +-
3 files changed, 119 insertions(+), 122 deletions(-)
---
diff --git a/app/dialogs/action-search-dialog.c b/app/dialogs/action-search-dialog.c
index 7d7cb4b..aa4ba75 100644
--- a/app/dialogs/action-search-dialog.c
+++ b/app/dialogs/action-search-dialog.c
@@ -94,13 +94,10 @@ static void action_search_add_to_results_list (GtkAction *a
static void action_search_run_selected (SearchDialog *private);
static void action_search_history_and_actions (const gchar *keyword,
SearchDialog *private);
-static gboolean action_fuzzy_match (gchar *string,
- gchar *key);
-static gchar * action_search_normalize_string (const gchar *str);
static gboolean action_search_match_keyword (GtkAction *action,
const gchar* keyword,
gint *section,
- gboolean match_fuzzy);
+ Gimp *gimp);
static void action_search_hide (SearchDialog *private);
@@ -583,7 +580,7 @@ action_search_history_and_actions (const gchar *keyword,
! GIMP_GUI_CONFIG (private->gimp->config)->search_show_unavailable)
continue;
- if (action_search_match_keyword (action, keyword, §ion, TRUE))
+ if (action_search_match_keyword (action, keyword, §ion, private->gimp))
{
GList *list3;
@@ -613,61 +610,16 @@ action_search_history_and_actions (const gchar *keyword,
g_list_free_full (history_actions, (GDestroyNotify) g_object_unref);
}
-/**
- * Fuzzy search matching.
- * Returns: TRUE if all the letters of `key` are found in `string`,
- * in the same order (even with intermediate letters).
- */
-static gboolean
-action_fuzzy_match (gchar *string,
- gchar *key)
-{
- gchar *remaining_string = string;
-
- if (strlen (key) == 0 )
- return TRUE;
-
- if ((remaining_string = strchr (string, key[0])) != NULL )
- return action_fuzzy_match (remaining_string + 1,
- key + 1);
- else
- return FALSE;
-}
-
-/*
- * Returns: a newly allocated lowercased string, which replaced any
- * spacing characters into a single space and stripped out any leading
- * and trailing space.
- */
-static gchar *
-action_search_normalize_string (const gchar *str)
-{
- GRegex *spaces_regex;
- gchar *normalized_str;
- gint i;
-
- spaces_regex = g_regex_new ("[ \n\t\r]+", 0, 0, NULL);
- normalized_str = g_regex_replace_literal (spaces_regex, str, -1, 0, " ", 0, NULL);
-
- g_regex_unref (spaces_regex);
-
- normalized_str = g_strstrip (normalized_str);
-
- for (i = 0 ; i < strlen (normalized_str); i++)
- normalized_str[i] = tolower (normalized_str[i]);
-
- return normalized_str;
-}
-
static gboolean
action_search_match_keyword (GtkAction *action,
const gchar *keyword,
gint *section,
- gboolean match_fuzzy)
+ Gimp *gimp)
{
- gboolean matched = FALSE;
- gchar *key;
- gchar *label;
+ gboolean matched = FALSE;
+ gchar **key_tokens;
+ gchar **label_tokens;
+ gchar **label_alternates = NULL;
gchar *tmp;
if (keyword == NULL)
@@ -682,108 +634,153 @@ action_search_match_keyword (GtkAction *action,
return TRUE;
}
- key = action_search_normalize_string (keyword);
- tmp = gimp_strip_uline (gtk_action_get_label (action));
- label = action_search_normalize_string (tmp);
+ key_tokens = g_str_tokenize_and_fold (keyword, gimp->config->language, NULL);
+ tmp = gimp_strip_uline (gtk_action_get_label (action));
+ label_tokens = g_str_tokenize_and_fold (tmp, gimp->config->language, &label_alternates);
g_free (tmp);
/* If keyword is two characters, then match them with first letters
* of first and second word in the labels. For instance 'gb' will
* list 'Gaussian Blur...'
*/
- if (strlen (key) == 2)
+ if (g_strv_length (key_tokens) == 1 && g_utf8_strlen (key_tokens[0], -1) == 2)
{
- gchar* space_pos;
-
- space_pos = strchr (label, ' ');
-
- if (space_pos != NULL)
+ gunichar c1 = g_utf8_get_char (key_tokens[0]);
+ gunichar c2 = g_utf8_get_char (g_utf8_find_next_char (key_tokens[0], NULL));
+
+ if ((g_strv_length (label_tokens) > 1 &&
+ g_utf8_get_char (label_tokens[0]) == c1 &&
+ g_utf8_get_char (label_tokens[1]) == c2) ||
+ (g_strv_length (label_alternates) > 1 &&
+ g_utf8_get_char (label_alternates[0]) == c1 &&
+ g_utf8_get_char (label_alternates[1]) == c2))
{
- space_pos++;
-
- if (key[0] == label[0] && key[1] == *space_pos)
+ matched = TRUE;
+ if (section)
{
- matched = TRUE;
- if (section)
- {
- *section = 1;
- }
+ *section = 1;
}
}
}
- if (! matched)
+ if (! matched && g_strv_length (label_tokens) > 0)
{
- gchar *substr;
-
- substr = strstr (label, key);
- if (substr)
+ gint previous_matched = -1;
+ gboolean match_start;
+ gboolean match_ordered;
+ gint i;
+
+ matched = TRUE;
+ match_start = TRUE;
+ match_ordered = TRUE;
+ for (i = 0; key_tokens[i] != NULL; i++)
{
- matched = TRUE;
- if (section)
+ gint j;
+ for (j = 0; label_tokens[j] != NULL; j++)
{
- /* If the substring is the label start, this is a nicer match. */
- *section = (substr == label) ? 1 : 2;
+ if (g_str_has_prefix (label_tokens[j], key_tokens[i]))
+ {
+ goto one_matched;
+ }
}
+ for (j = 0; label_alternates[j] != NULL; j++)
+ {
+ if (g_str_has_prefix (label_alternates[j], key_tokens[i]))
+ {
+ goto one_matched;
+ }
+ }
+ matched = FALSE;
+one_matched:
+ if (previous_matched > j)
+ match_ordered = FALSE;
+ previous_matched = j;
+
+ if (i != j)
+ match_start = FALSE;
+
+ continue;
}
- else if (strlen (key) > 2)
+
+ if (matched && section)
{
- gchar *tooltip = NULL;
+ /* If the key is the label start, this is a nicer match.
+ * Then if key tokens are found in the same order in the label.
+ * Finally we show at the end if the key tokens are found with a different order. */
+ *section = match_ordered ? (match_start ? 1 : 2) : 3;
+ }
+ }
- if (gtk_action_get_tooltip (action)!= NULL)
- {
- tooltip = action_search_normalize_string (gtk_action_get_tooltip (action));
+ if (! matched && g_utf8_strlen (key_tokens[0], -1) > 2 &&
+ gtk_action_get_tooltip (action) != NULL)
+ {
+ gchar **tooltip_tokens;
+ gchar **tooltip_alternates = NULL;
+ gboolean mixed_match;
+ gint i;
- if (strstr (tooltip, key))
+ tooltip_tokens = g_str_tokenize_and_fold (gtk_action_get_tooltip (action),
+ gimp->config->language, &tooltip_alternates);
+
+ if (g_strv_length (tooltip_tokens) > 0)
+ {
+ matched = TRUE;
+ mixed_match = FALSE;
+
+ for (i = 0; key_tokens[i] != NULL; i++)
+ {
+ gint j;
+ for (j = 0; tooltip_tokens[j] != NULL; j++)
{
- matched = TRUE;
- if (section)
+ if (g_str_has_prefix (tooltip_tokens[j], key_tokens[i]))
{
- *section = 3;
+ goto one_tooltip_matched;
}
}
- }
-
- if (! matched && strchr (key, ' '))
- {
- gchar **words;
- gchar **word;
-
- matched = TRUE;
- if (section)
+ for (j = 0; tooltip_alternates[j] != NULL; j++)
{
- *section = 4;
+ if (g_str_has_prefix (tooltip_alternates[j], key_tokens[i]))
+ {
+ goto one_tooltip_matched;
+ }
}
-
- words = g_strsplit (key, " ", 0);
- for (word = &words[0]; *word; ++word)
+ for (j = 0; label_tokens[j] != NULL; j++)
{
- if (! strstr (label, *word) &&
- (! tooltip || ! strstr (tooltip, *word)))
+ if (g_str_has_prefix (label_tokens[j], key_tokens[i]))
{
- matched = FALSE;
- break;
+ mixed_match = TRUE;
+ goto one_tooltip_matched;
}
}
-
- g_strfreev (words);
+ for (j = 0; label_alternates[j] != NULL; j++)
+ {
+ if (g_str_has_prefix (label_alternates[j], key_tokens[i]))
+ {
+ mixed_match = TRUE;
+ goto one_tooltip_matched;
+ }
+ }
+ matched = FALSE;
+one_tooltip_matched:
+ continue;
}
-
- g_free (tooltip);
- }
-
- if (! matched && match_fuzzy && action_fuzzy_match (label, key))
- {
- matched = TRUE;
- if (section)
+ if (matched && section)
{
- *section = 5;
+ /* Matching the tooltip is section 4. We don't go looking
+ * for start of string or token order for tooltip match.
+ * But if the match is mixed on tooltip and label (there are
+ * no match for *only* label or *only* tooltip), this is
+ * section 5. */
+ *section = mixed_match ? 5 : 4;
}
}
+ g_strfreev (tooltip_tokens);
+ g_strfreev (tooltip_alternates);
}
- g_free (label);
- g_free (key);
+ g_strfreev (key_tokens);
+ g_strfreev (label_tokens);
+ g_strfreev (label_alternates);
return matched;
}
diff --git a/app/widgets/gimpaction-history.c b/app/widgets/gimpaction-history.c
index b317628..3a503d2 100644
--- a/app/widgets/gimpaction-history.c
+++ b/app/widgets/gimpaction-history.c
@@ -279,7 +279,7 @@ gimp_action_history_search (Gimp *gimp,
! config->search_show_unavailable)
continue;
- if (match_func (action, keyword, NULL, FALSE))
+ if (match_func (action, keyword, NULL, gimp))
result = g_list_prepend (result, g_object_ref (action));
}
diff --git a/app/widgets/gimpaction-history.h b/app/widgets/gimpaction-history.h
index 1f0f402..2ac4464 100644
--- a/app/widgets/gimpaction-history.h
+++ b/app/widgets/gimpaction-history.h
@@ -25,7 +25,7 @@
typedef gboolean (* GimpActionMatchFunc) (GtkAction *action,
const gchar *keyword,
gint *section,
- gboolean match_fuzzy);
+ Gimp *gimp);
void gimp_action_history_init (Gimp *gimp);
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]
