forked from aniani/vim
patch 8.2.1893: fuzzy matching does not support multiple words
Problem: Fuzzy matching does not support multiple words.
Solution: Add support for matching white space separated words. (Yegappan
Lakshmanan, closes #7163)
This commit is contained in:
@@ -7367,8 +7367,15 @@ matchfuzzy({list}, {str} [, {dict}]) *matchfuzzy()*
|
|||||||
the strings in {list} that fuzzy match {str}. The strings in
|
the strings in {list} that fuzzy match {str}. The strings in
|
||||||
the returned list are sorted based on the matching score.
|
the returned list are sorted based on the matching score.
|
||||||
|
|
||||||
|
The optional {dict} argument always supports the following
|
||||||
|
items:
|
||||||
|
matchseq When this item is present and {str} contains
|
||||||
|
multiple words separated by white space, then
|
||||||
|
returns only matches that contain the words in
|
||||||
|
the given sequence.
|
||||||
|
|
||||||
If {list} is a list of dictionaries, then the optional {dict}
|
If {list} is a list of dictionaries, then the optional {dict}
|
||||||
argument supports the following items:
|
argument supports the following additional items:
|
||||||
key key of the item which is fuzzy matched against
|
key key of the item which is fuzzy matched against
|
||||||
{str}. The value of this item should be a
|
{str}. The value of this item should be a
|
||||||
string.
|
string.
|
||||||
@@ -7382,6 +7389,9 @@ matchfuzzy({list}, {str} [, {dict}]) *matchfuzzy()*
|
|||||||
matching is NOT supported. The maximum supported {str} length
|
matching is NOT supported. The maximum supported {str} length
|
||||||
is 256.
|
is 256.
|
||||||
|
|
||||||
|
When {str} has multiple words each separated by white space,
|
||||||
|
then the list of strings that have all the words is returned.
|
||||||
|
|
||||||
If there are no matching strings or there is an error, then an
|
If there are no matching strings or there is an error, then an
|
||||||
empty list is returned. If length of {str} is greater than
|
empty list is returned. If length of {str} is greater than
|
||||||
256, then returns an empty list.
|
256, then returns an empty list.
|
||||||
@@ -7401,7 +7411,12 @@ matchfuzzy({list}, {str} [, {dict}]) *matchfuzzy()*
|
|||||||
:echo v:oldfiles->matchfuzzy("test")
|
:echo v:oldfiles->matchfuzzy("test")
|
||||||
< results in a list of file names fuzzy matching "test". >
|
< results in a list of file names fuzzy matching "test". >
|
||||||
:let l = readfile("buffer.c")->matchfuzzy("str")
|
:let l = readfile("buffer.c")->matchfuzzy("str")
|
||||||
< results in a list of lines in "buffer.c" fuzzy matching "str".
|
< results in a list of lines in "buffer.c" fuzzy matching "str". >
|
||||||
|
:echo ['one two', 'two one']->matchfuzzy('two one')
|
||||||
|
< results in ['two one', 'one two']. >
|
||||||
|
:echo ['one two', 'two one']->matchfuzzy('two one',
|
||||||
|
\ {'matchseq': 1})
|
||||||
|
< results in ['two one'].
|
||||||
|
|
||||||
matchfuzzypos({list}, {str} [, {dict}]) *matchfuzzypos()*
|
matchfuzzypos({list}, {str} [, {dict}]) *matchfuzzypos()*
|
||||||
Same as |matchfuzzy()|, but returns the list of matched
|
Same as |matchfuzzy()|, but returns the list of matched
|
||||||
|
|||||||
165
src/search.c
165
src/search.c
@@ -4203,16 +4203,16 @@ the_end:
|
|||||||
* Ported from the lib_fts library authored by Forrest Smith.
|
* Ported from the lib_fts library authored by Forrest Smith.
|
||||||
* https://github.com/forrestthewoods/lib_fts/tree/master/code
|
* https://github.com/forrestthewoods/lib_fts/tree/master/code
|
||||||
*
|
*
|
||||||
* Blog describing the algorithm:
|
* The following blog describes the fuzzy matching algorithm:
|
||||||
* https://www.forrestthewoods.com/blog/reverse_engineering_sublime_texts_fuzzy_match/
|
* https://www.forrestthewoods.com/blog/reverse_engineering_sublime_texts_fuzzy_match/
|
||||||
*
|
*
|
||||||
* Each matching string is assigned a score. The following factors are checked:
|
* Each matching string is assigned a score. The following factors are checked:
|
||||||
* Matched letter
|
* - Matched letter
|
||||||
* Unmatched letter
|
* - Unmatched letter
|
||||||
* Consecutively matched letters
|
* - Consecutively matched letters
|
||||||
* Proximity to start
|
* - Proximity to start
|
||||||
* Letter following a separator (space, underscore)
|
* - Letter following a separator (space, underscore)
|
||||||
* Uppercase letter following lowercase (aka CamelCase)
|
* - Uppercase letter following lowercase (aka CamelCase)
|
||||||
*
|
*
|
||||||
* Matched letters are good. Unmatched letters are bad. Matching near the start
|
* Matched letters are good. Unmatched letters are bad. Matching near the start
|
||||||
* is good. Matching the first letter in the middle of a phrase is good.
|
* is good. Matching the first letter in the middle of a phrase is good.
|
||||||
@@ -4222,16 +4222,17 @@ the_end:
|
|||||||
* File paths are different from file names. File extensions may be ignorable.
|
* File paths are different from file names. File extensions may be ignorable.
|
||||||
* Single words care about consecutive matches but not separators or camel
|
* Single words care about consecutive matches but not separators or camel
|
||||||
* case.
|
* case.
|
||||||
* Score starts at 0
|
* Score starts at 100
|
||||||
* Matched letter: +0 points
|
* Matched letter: +0 points
|
||||||
* Unmatched letter: -1 point
|
* Unmatched letter: -1 point
|
||||||
* Consecutive match bonus: +5 points
|
* Consecutive match bonus: +15 points
|
||||||
* Separator bonus: +10 points
|
* First letter bonus: +15 points
|
||||||
* Camel case bonus: +10 points
|
* Separator bonus: +30 points
|
||||||
* Unmatched leading letter: -3 points (max: -9)
|
* Camel case bonus: +30 points
|
||||||
|
* Unmatched leading letter: -5 points (max: -15)
|
||||||
*
|
*
|
||||||
* There is some nuance to this. Scores don’t have an intrinsic meaning. The
|
* There is some nuance to this. Scores don’t have an intrinsic meaning. The
|
||||||
* score range isn’t 0 to 100. It’s roughly [-50, 50]. Longer words have a
|
* score range isn’t 0 to 100. It’s roughly [50, 150]. Longer words have a
|
||||||
* lower minimum score due to unmatched letter penalty. Longer search patterns
|
* lower minimum score due to unmatched letter penalty. Longer search patterns
|
||||||
* have a higher maximum score due to match bonuses.
|
* have a higher maximum score due to match bonuses.
|
||||||
*
|
*
|
||||||
@@ -4247,6 +4248,7 @@ the_end:
|
|||||||
*/
|
*/
|
||||||
typedef struct
|
typedef struct
|
||||||
{
|
{
|
||||||
|
int idx; // used for stable sort
|
||||||
listitem_T *item;
|
listitem_T *item;
|
||||||
int score;
|
int score;
|
||||||
list_T *lmatchpos;
|
list_T *lmatchpos;
|
||||||
@@ -4267,6 +4269,8 @@ typedef struct
|
|||||||
#define MAX_LEADING_LETTER_PENALTY -15
|
#define MAX_LEADING_LETTER_PENALTY -15
|
||||||
// penalty for every letter that doesn't match
|
// penalty for every letter that doesn't match
|
||||||
#define UNMATCHED_LETTER_PENALTY -1
|
#define UNMATCHED_LETTER_PENALTY -1
|
||||||
|
// penalty for gap in matching positions (-2 * k)
|
||||||
|
#define GAP_PENALTY -2
|
||||||
// Score for a string that doesn't fuzzy match the pattern
|
// Score for a string that doesn't fuzzy match the pattern
|
||||||
#define SCORE_NONE -9999
|
#define SCORE_NONE -9999
|
||||||
|
|
||||||
@@ -4319,6 +4323,8 @@ fuzzy_match_compute_score(
|
|||||||
// Sequential
|
// Sequential
|
||||||
if (currIdx == (prevIdx + 1))
|
if (currIdx == (prevIdx + 1))
|
||||||
score += SEQUENTIAL_BONUS;
|
score += SEQUENTIAL_BONUS;
|
||||||
|
else
|
||||||
|
score += GAP_PENALTY * (currIdx - prevIdx);
|
||||||
}
|
}
|
||||||
|
|
||||||
// Check for bonuses based on neighbor character value
|
// Check for bonuses based on neighbor character value
|
||||||
@@ -4334,7 +4340,7 @@ fuzzy_match_compute_score(
|
|||||||
while (sidx < currIdx)
|
while (sidx < currIdx)
|
||||||
{
|
{
|
||||||
neighbor = (*mb_ptr2char)(p);
|
neighbor = (*mb_ptr2char)(p);
|
||||||
(void)mb_ptr2char_adv(&p);
|
MB_PTR_ADV(p);
|
||||||
sidx++;
|
sidx++;
|
||||||
}
|
}
|
||||||
curr = (*mb_ptr2char)(p);
|
curr = (*mb_ptr2char)(p);
|
||||||
@@ -4362,6 +4368,10 @@ fuzzy_match_compute_score(
|
|||||||
return score;
|
return score;
|
||||||
}
|
}
|
||||||
|
|
||||||
|
/*
|
||||||
|
* Perform a recursive search for fuzzy matching 'fuzpat' in 'str'.
|
||||||
|
* Return the number of matching characters.
|
||||||
|
*/
|
||||||
static int
|
static int
|
||||||
fuzzy_match_recursive(
|
fuzzy_match_recursive(
|
||||||
char_u *fuzpat,
|
char_u *fuzpat,
|
||||||
@@ -4386,11 +4396,11 @@ fuzzy_match_recursive(
|
|||||||
// Count recursions
|
// Count recursions
|
||||||
++*recursionCount;
|
++*recursionCount;
|
||||||
if (*recursionCount >= FUZZY_MATCH_RECURSION_LIMIT)
|
if (*recursionCount >= FUZZY_MATCH_RECURSION_LIMIT)
|
||||||
return FALSE;
|
return 0;
|
||||||
|
|
||||||
// Detect end of strings
|
// Detect end of strings
|
||||||
if (*fuzpat == '\0' || *str == '\0')
|
if (*fuzpat == '\0' || *str == '\0')
|
||||||
return FALSE;
|
return 0;
|
||||||
|
|
||||||
// Loop through fuzpat and str looking for a match
|
// Loop through fuzpat and str looking for a match
|
||||||
first_match = TRUE;
|
first_match = TRUE;
|
||||||
@@ -4411,7 +4421,7 @@ fuzzy_match_recursive(
|
|||||||
|
|
||||||
// Supplied matches buffer was too short
|
// Supplied matches buffer was too short
|
||||||
if (nextMatch >= maxMatches)
|
if (nextMatch >= maxMatches)
|
||||||
return FALSE;
|
return 0;
|
||||||
|
|
||||||
// "Copy-on-Write" srcMatches into matches
|
// "Copy-on-Write" srcMatches into matches
|
||||||
if (first_match && srcMatches)
|
if (first_match && srcMatches)
|
||||||
@@ -4444,12 +4454,12 @@ fuzzy_match_recursive(
|
|||||||
// Advance
|
// Advance
|
||||||
matches[nextMatch++] = strIdx;
|
matches[nextMatch++] = strIdx;
|
||||||
if (has_mbyte)
|
if (has_mbyte)
|
||||||
(void)mb_ptr2char_adv(&fuzpat);
|
MB_PTR_ADV(fuzpat);
|
||||||
else
|
else
|
||||||
++fuzpat;
|
++fuzpat;
|
||||||
}
|
}
|
||||||
if (has_mbyte)
|
if (has_mbyte)
|
||||||
(void)mb_ptr2char_adv(&str);
|
MB_PTR_ADV(str);
|
||||||
else
|
else
|
||||||
++str;
|
++str;
|
||||||
strIdx++;
|
strIdx++;
|
||||||
@@ -4469,12 +4479,12 @@ fuzzy_match_recursive(
|
|||||||
// Recursive score is better than "this"
|
// Recursive score is better than "this"
|
||||||
memcpy(matches, bestRecursiveMatches, maxMatches * sizeof(matches[0]));
|
memcpy(matches, bestRecursiveMatches, maxMatches * sizeof(matches[0]));
|
||||||
*outScore = bestRecursiveScore;
|
*outScore = bestRecursiveScore;
|
||||||
return TRUE;
|
return nextMatch;
|
||||||
}
|
}
|
||||||
else if (matched)
|
else if (matched)
|
||||||
return TRUE; // "this" score is better than recursive
|
return nextMatch; // "this" score is better than recursive
|
||||||
|
|
||||||
return FALSE; // no match
|
return 0; // no match
|
||||||
}
|
}
|
||||||
|
|
||||||
/*
|
/*
|
||||||
@@ -4485,45 +4495,110 @@ fuzzy_match_recursive(
|
|||||||
* Scores values have no intrinsic meaning. Possible score range is not
|
* Scores values have no intrinsic meaning. Possible score range is not
|
||||||
* normalized and varies with pattern.
|
* normalized and varies with pattern.
|
||||||
* Recursion is limited internally (default=10) to prevent degenerate cases
|
* Recursion is limited internally (default=10) to prevent degenerate cases
|
||||||
* (fuzpat="aaaaaa" str="aaaaaaaaaaaaaaaaaaaaaaaaaaaaaa").
|
* (pat_arg="aaaaaa" str="aaaaaaaaaaaaaaaaaaaaaaaaaaaaaa").
|
||||||
* Uses char_u for match indices. Therefore patterns are limited to MAXMATCHES
|
* Uses char_u for match indices. Therefore patterns are limited to MAXMATCHES
|
||||||
* characters.
|
* characters.
|
||||||
*
|
*
|
||||||
* Returns TRUE if 'fuzpat' matches 'str'. Also returns the match score in
|
* Returns TRUE if 'pat_arg' matches 'str'. Also returns the match score in
|
||||||
* 'outScore' and the matching character positions in 'matches'.
|
* 'outScore' and the matching character positions in 'matches'.
|
||||||
*/
|
*/
|
||||||
static int
|
static int
|
||||||
fuzzy_match(
|
fuzzy_match(
|
||||||
char_u *str,
|
char_u *str,
|
||||||
char_u *fuzpat,
|
char_u *pat_arg,
|
||||||
|
int matchseq,
|
||||||
int *outScore,
|
int *outScore,
|
||||||
matchidx_T *matches,
|
matchidx_T *matches,
|
||||||
int maxMatches)
|
int maxMatches)
|
||||||
{
|
{
|
||||||
int recursionCount = 0;
|
int recursionCount = 0;
|
||||||
int len = MB_CHARLEN(str);
|
int len = MB_CHARLEN(str);
|
||||||
|
char_u *save_pat;
|
||||||
|
char_u *pat;
|
||||||
|
char_u *p;
|
||||||
|
int complete = FALSE;
|
||||||
|
int score = 0;
|
||||||
|
int numMatches = 0;
|
||||||
|
int matchCount;
|
||||||
|
|
||||||
*outScore = 0;
|
*outScore = 0;
|
||||||
|
|
||||||
return fuzzy_match_recursive(fuzpat, str, 0, outScore, str, len, NULL,
|
save_pat = vim_strsave(pat_arg);
|
||||||
matches, maxMatches, 0, &recursionCount);
|
if (save_pat == NULL)
|
||||||
|
return FALSE;
|
||||||
|
pat = save_pat;
|
||||||
|
p = pat;
|
||||||
|
|
||||||
|
// Try matching each word in 'pat_arg' in 'str'
|
||||||
|
while (TRUE)
|
||||||
|
{
|
||||||
|
if (matchseq)
|
||||||
|
complete = TRUE;
|
||||||
|
else
|
||||||
|
{
|
||||||
|
// Extract one word from the pattern (separated by space)
|
||||||
|
p = skipwhite(p);
|
||||||
|
if (*p == NUL)
|
||||||
|
break;
|
||||||
|
pat = p;
|
||||||
|
while (*p != NUL && !VIM_ISWHITE(PTR2CHAR(p)))
|
||||||
|
{
|
||||||
|
if (has_mbyte)
|
||||||
|
MB_PTR_ADV(p);
|
||||||
|
else
|
||||||
|
++p;
|
||||||
|
}
|
||||||
|
if (*p == NUL) // processed all the words
|
||||||
|
complete = TRUE;
|
||||||
|
*p = NUL;
|
||||||
|
}
|
||||||
|
|
||||||
|
score = 0;
|
||||||
|
recursionCount = 0;
|
||||||
|
matchCount = fuzzy_match_recursive(pat, str, 0, &score, str, len, NULL,
|
||||||
|
matches + numMatches, maxMatches - numMatches,
|
||||||
|
0, &recursionCount);
|
||||||
|
if (matchCount == 0)
|
||||||
|
{
|
||||||
|
numMatches = 0;
|
||||||
|
break;
|
||||||
|
}
|
||||||
|
|
||||||
|
// Accumulate the match score and the number of matches
|
||||||
|
*outScore += score;
|
||||||
|
numMatches += matchCount;
|
||||||
|
|
||||||
|
if (complete)
|
||||||
|
break;
|
||||||
|
|
||||||
|
// try matching the next word
|
||||||
|
++p;
|
||||||
|
}
|
||||||
|
|
||||||
|
vim_free(save_pat);
|
||||||
|
return numMatches != 0;
|
||||||
}
|
}
|
||||||
|
|
||||||
/*
|
/*
|
||||||
* Sort the fuzzy matches in the descending order of the match score.
|
* Sort the fuzzy matches in the descending order of the match score.
|
||||||
|
* For items with same score, retain the order using the index (stable sort)
|
||||||
*/
|
*/
|
||||||
static int
|
static int
|
||||||
fuzzy_item_compare(const void *s1, const void *s2)
|
fuzzy_match_item_compare(const void *s1, const void *s2)
|
||||||
{
|
{
|
||||||
int v1 = ((fuzzyItem_T *)s1)->score;
|
int v1 = ((fuzzyItem_T *)s1)->score;
|
||||||
int v2 = ((fuzzyItem_T *)s2)->score;
|
int v2 = ((fuzzyItem_T *)s2)->score;
|
||||||
|
int idx1 = ((fuzzyItem_T *)s1)->idx;
|
||||||
|
int idx2 = ((fuzzyItem_T *)s2)->idx;
|
||||||
|
|
||||||
return v1 == v2 ? 0 : v1 > v2 ? -1 : 1;
|
return v1 == v2 ? (idx1 - idx2) : v1 > v2 ? -1 : 1;
|
||||||
}
|
}
|
||||||
|
|
||||||
/*
|
/*
|
||||||
* Fuzzy search the string 'str' in a list of 'items' and return the matching
|
* Fuzzy search the string 'str' in a list of 'items' and return the matching
|
||||||
* strings in 'fmatchlist'.
|
* strings in 'fmatchlist'.
|
||||||
|
* If 'matchseq' is TRUE, then for multi-word search strings, match all the
|
||||||
|
* words in sequence.
|
||||||
* If 'items' is a list of strings, then search for 'str' in the list.
|
* If 'items' is a list of strings, then search for 'str' in the list.
|
||||||
* If 'items' is a list of dicts, then either use 'key' to lookup the string
|
* If 'items' is a list of dicts, then either use 'key' to lookup the string
|
||||||
* for each item or use 'item_cb' Funcref function to get the string.
|
* for each item or use 'item_cb' Funcref function to get the string.
|
||||||
@@ -4531,9 +4606,10 @@ fuzzy_item_compare(const void *s1, const void *s2)
|
|||||||
* matches for each item.
|
* matches for each item.
|
||||||
*/
|
*/
|
||||||
static void
|
static void
|
||||||
match_fuzzy(
|
fuzzy_match_in_list(
|
||||||
list_T *items,
|
list_T *items,
|
||||||
char_u *str,
|
char_u *str,
|
||||||
|
int matchseq,
|
||||||
char_u *key,
|
char_u *key,
|
||||||
callback_T *item_cb,
|
callback_T *item_cb,
|
||||||
int retmatchpos,
|
int retmatchpos,
|
||||||
@@ -4561,6 +4637,7 @@ match_fuzzy(
|
|||||||
char_u *itemstr;
|
char_u *itemstr;
|
||||||
typval_T rettv;
|
typval_T rettv;
|
||||||
|
|
||||||
|
ptrs[i].idx = i;
|
||||||
ptrs[i].item = li;
|
ptrs[i].item = li;
|
||||||
ptrs[i].score = SCORE_NONE;
|
ptrs[i].score = SCORE_NONE;
|
||||||
itemstr = NULL;
|
itemstr = NULL;
|
||||||
@@ -4593,25 +4670,34 @@ match_fuzzy(
|
|||||||
}
|
}
|
||||||
|
|
||||||
if (itemstr != NULL
|
if (itemstr != NULL
|
||||||
&& fuzzy_match(itemstr, str, &score, matches,
|
&& fuzzy_match(itemstr, str, matchseq, &score, matches,
|
||||||
sizeof(matches) / sizeof(matches[0])))
|
sizeof(matches) / sizeof(matches[0])))
|
||||||
{
|
{
|
||||||
// Copy the list of matching positions in itemstr to a list, if
|
// Copy the list of matching positions in itemstr to a list, if
|
||||||
// 'retmatchpos' is set.
|
// 'retmatchpos' is set.
|
||||||
if (retmatchpos)
|
if (retmatchpos)
|
||||||
{
|
{
|
||||||
int j;
|
int j = 0;
|
||||||
int strsz;
|
char_u *p;
|
||||||
|
|
||||||
ptrs[i].lmatchpos = list_alloc();
|
ptrs[i].lmatchpos = list_alloc();
|
||||||
if (ptrs[i].lmatchpos == NULL)
|
if (ptrs[i].lmatchpos == NULL)
|
||||||
goto done;
|
goto done;
|
||||||
strsz = MB_CHARLEN(str);
|
|
||||||
for (j = 0; j < strsz; j++)
|
p = str;
|
||||||
|
while (*p != NUL)
|
||||||
|
{
|
||||||
|
if (!VIM_ISWHITE(PTR2CHAR(p)))
|
||||||
{
|
{
|
||||||
if (list_append_number(ptrs[i].lmatchpos,
|
if (list_append_number(ptrs[i].lmatchpos,
|
||||||
matches[j]) == FAIL)
|
matches[j]) == FAIL)
|
||||||
goto done;
|
goto done;
|
||||||
|
j++;
|
||||||
|
}
|
||||||
|
if (has_mbyte)
|
||||||
|
MB_PTR_ADV(p);
|
||||||
|
else
|
||||||
|
++p;
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
ptrs[i].score = score;
|
ptrs[i].score = score;
|
||||||
@@ -4627,7 +4713,7 @@ match_fuzzy(
|
|||||||
|
|
||||||
// Sort the list by the descending order of the match score
|
// Sort the list by the descending order of the match score
|
||||||
qsort((void *)ptrs, (size_t)len, sizeof(fuzzyItem_T),
|
qsort((void *)ptrs, (size_t)len, sizeof(fuzzyItem_T),
|
||||||
fuzzy_item_compare);
|
fuzzy_match_item_compare);
|
||||||
|
|
||||||
// For matchfuzzy(), return a list of matched strings.
|
// For matchfuzzy(), return a list of matched strings.
|
||||||
// ['str1', 'str2', 'str3']
|
// ['str1', 'str2', 'str3']
|
||||||
@@ -4687,6 +4773,7 @@ do_fuzzymatch(typval_T *argvars, typval_T *rettv, int retmatchpos)
|
|||||||
callback_T cb;
|
callback_T cb;
|
||||||
char_u *key = NULL;
|
char_u *key = NULL;
|
||||||
int ret;
|
int ret;
|
||||||
|
int matchseq = FALSE;
|
||||||
|
|
||||||
CLEAR_POINTER(&cb);
|
CLEAR_POINTER(&cb);
|
||||||
|
|
||||||
@@ -4737,6 +4824,8 @@ do_fuzzymatch(typval_T *argvars, typval_T *rettv, int retmatchpos)
|
|||||||
return;
|
return;
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
if ((di = dict_find(d, (char_u *)"matchseq", -1)) != NULL)
|
||||||
|
matchseq = TRUE;
|
||||||
}
|
}
|
||||||
|
|
||||||
// get the fuzzy matches
|
// get the fuzzy matches
|
||||||
@@ -4762,8 +4851,8 @@ do_fuzzymatch(typval_T *argvars, typval_T *rettv, int retmatchpos)
|
|||||||
goto done;
|
goto done;
|
||||||
}
|
}
|
||||||
|
|
||||||
match_fuzzy(argvars[0].vval.v_list, tv_get_string(&argvars[1]), key,
|
fuzzy_match_in_list(argvars[0].vval.v_list, tv_get_string(&argvars[1]),
|
||||||
&cb, retmatchpos, rettv->vval.v_list);
|
matchseq, key, &cb, retmatchpos, rettv->vval.v_list);
|
||||||
|
|
||||||
done:
|
done:
|
||||||
free_callback(&cb);
|
free_callback(&cb);
|
||||||
|
|||||||
@@ -22,16 +22,15 @@ func Test_matchfuzzy()
|
|||||||
call assert_equal(['aaaaaaaaaaaaaaaaaaaaaaaaaaaaaa'], matchfuzzy(['aaaaaaaaaaaaaaaaaaaaaaaaaaaaaa'], 'aa'))
|
call assert_equal(['aaaaaaaaaaaaaaaaaaaaaaaaaaaaaa'], matchfuzzy(['aaaaaaaaaaaaaaaaaaaaaaaaaaaaaa'], 'aa'))
|
||||||
call assert_equal(256, matchfuzzy([repeat('a', 256)], repeat('a', 256))[0]->len())
|
call assert_equal(256, matchfuzzy([repeat('a', 256)], repeat('a', 256))[0]->len())
|
||||||
call assert_equal([], matchfuzzy([repeat('a', 300)], repeat('a', 257)))
|
call assert_equal([], matchfuzzy([repeat('a', 300)], repeat('a', 257)))
|
||||||
|
" matches with same score should not be reordered
|
||||||
|
let l = ['abc1', 'abc2', 'abc3']
|
||||||
|
call assert_equal(l, l->matchfuzzy('abc'))
|
||||||
|
|
||||||
" Tests for match preferences
|
" Tests for match preferences
|
||||||
" preference for camel case match
|
" preference for camel case match
|
||||||
call assert_equal(['oneTwo', 'onetwo'], ['onetwo', 'oneTwo']->matchfuzzy('onetwo'))
|
call assert_equal(['oneTwo', 'onetwo'], ['onetwo', 'oneTwo']->matchfuzzy('onetwo'))
|
||||||
" preference for match after a separator (_ or space)
|
" preference for match after a separator (_ or space)
|
||||||
if has("win32")
|
|
||||||
call assert_equal(['onetwo', 'one two', 'one_two'], ['onetwo', 'one_two', 'one two']->matchfuzzy('onetwo'))
|
|
||||||
else
|
|
||||||
call assert_equal(['onetwo', 'one_two', 'one two'], ['onetwo', 'one_two', 'one two']->matchfuzzy('onetwo'))
|
call assert_equal(['onetwo', 'one_two', 'one two'], ['onetwo', 'one_two', 'one two']->matchfuzzy('onetwo'))
|
||||||
endif
|
|
||||||
" preference for leading letter match
|
" preference for leading letter match
|
||||||
call assert_equal(['onetwo', 'xonetwo'], ['xonetwo', 'onetwo']->matchfuzzy('onetwo'))
|
call assert_equal(['onetwo', 'xonetwo'], ['xonetwo', 'onetwo']->matchfuzzy('onetwo'))
|
||||||
" preference for sequential match
|
" preference for sequential match
|
||||||
@@ -42,6 +41,17 @@ func Test_matchfuzzy()
|
|||||||
call assert_equal(['one', 'onex', 'onexx'], ['onexx', 'one', 'onex']->matchfuzzy('one'))
|
call assert_equal(['one', 'onex', 'onexx'], ['onexx', 'one', 'onex']->matchfuzzy('one'))
|
||||||
" prefer complete matches over separator matches
|
" prefer complete matches over separator matches
|
||||||
call assert_equal(['.vim/vimrc', '.vim/vimrc_colors', '.vim/v_i_m_r_c'], ['.vim/vimrc', '.vim/vimrc_colors', '.vim/v_i_m_r_c']->matchfuzzy('vimrc'))
|
call assert_equal(['.vim/vimrc', '.vim/vimrc_colors', '.vim/v_i_m_r_c'], ['.vim/vimrc', '.vim/vimrc_colors', '.vim/v_i_m_r_c']->matchfuzzy('vimrc'))
|
||||||
|
" gap penalty
|
||||||
|
call assert_equal(['xxayybxxxx', 'xxayyybxxx', 'xxayyyybxx'], ['xxayyyybxx', 'xxayyybxxx', 'xxayybxxxx']->matchfuzzy('ab'))
|
||||||
|
|
||||||
|
" match multiple words (separated by space)
|
||||||
|
call assert_equal(['foo bar baz'], ['foo bar baz', 'foo', 'foo bar', 'baz bar']->matchfuzzy('baz foo'))
|
||||||
|
call assert_equal([], ['foo bar baz', 'foo', 'foo bar', 'baz bar']->matchfuzzy('one two'))
|
||||||
|
call assert_equal([], ['foo bar']->matchfuzzy(" \t "))
|
||||||
|
|
||||||
|
" test for matching a sequence of words
|
||||||
|
call assert_equal(['bar foo'], ['foo bar', 'bar foo', 'foobar', 'barfoo']->matchfuzzy('bar foo', {'matchseq' : 1}))
|
||||||
|
call assert_equal([#{text: 'two one'}], [#{text: 'one two'}, #{text: 'two one'}]->matchfuzzy('two one', #{key: 'text', matchseq: v:true}))
|
||||||
|
|
||||||
%bw!
|
%bw!
|
||||||
eval ['somebuf', 'anotherone', 'needle', 'yetanotherone']->map({_, v -> bufadd(v) + bufload(v)})
|
eval ['somebuf', 'anotherone', 'needle', 'yetanotherone']->map({_, v -> bufadd(v) + bufload(v)})
|
||||||
@@ -49,6 +59,7 @@ func Test_matchfuzzy()
|
|||||||
call assert_equal(1, len(l))
|
call assert_equal(1, len(l))
|
||||||
call assert_match('needle', l[0])
|
call assert_match('needle', l[0])
|
||||||
|
|
||||||
|
" Test for fuzzy matching dicts
|
||||||
let l = [{'id' : 5, 'val' : 'crayon'}, {'id' : 6, 'val' : 'camera'}]
|
let l = [{'id' : 5, 'val' : 'crayon'}, {'id' : 6, 'val' : 'camera'}]
|
||||||
call assert_equal([{'id' : 6, 'val' : 'camera'}], matchfuzzy(l, 'cam', {'text_cb' : {v -> v.val}}))
|
call assert_equal([{'id' : 6, 'val' : 'camera'}], matchfuzzy(l, 'cam', {'text_cb' : {v -> v.val}}))
|
||||||
call assert_equal([{'id' : 6, 'val' : 'camera'}], matchfuzzy(l, 'cam', {'key' : 'val'}))
|
call assert_equal([{'id' : 6, 'val' : 'camera'}], matchfuzzy(l, 'cam', {'key' : 'val'}))
|
||||||
@@ -64,6 +75,9 @@ func Test_matchfuzzy()
|
|||||||
call assert_fails("let x = matchfuzzy(l, 'cam', test_null_dict())", 'E715:')
|
call assert_fails("let x = matchfuzzy(l, 'cam', test_null_dict())", 'E715:')
|
||||||
call assert_fails("let x = matchfuzzy(l, 'foo', {'key' : test_null_string()})", 'E475:')
|
call assert_fails("let x = matchfuzzy(l, 'foo', {'key' : test_null_string()})", 'E475:')
|
||||||
call assert_fails("let x = matchfuzzy(l, 'foo', {'text_cb' : test_null_function()})", 'E475:')
|
call assert_fails("let x = matchfuzzy(l, 'foo', {'text_cb' : test_null_function()})", 'E475:')
|
||||||
|
" matches with same score should not be reordered
|
||||||
|
let l = [#{text: 'abc', id: 1}, #{text: 'abc', id: 2}, #{text: 'abc', id: 3}]
|
||||||
|
call assert_equal(l, l->matchfuzzy('abc', #{key: 'text'}))
|
||||||
|
|
||||||
let l = [{'id' : 5, 'name' : 'foo'}, {'id' : 6, 'name' : []}, {'id' : 7}]
|
let l = [{'id' : 5, 'name' : 'foo'}, {'id' : 6, 'name' : []}, {'id' : 7}]
|
||||||
call assert_fails("let x = matchfuzzy(l, 'foo', {'key' : 'name'})", 'E730:')
|
call assert_fails("let x = matchfuzzy(l, 'foo', {'key' : 'name'})", 'E730:')
|
||||||
@@ -75,7 +89,7 @@ func Test_matchfuzzy()
|
|||||||
let &encoding = save_enc
|
let &encoding = save_enc
|
||||||
endfunc
|
endfunc
|
||||||
|
|
||||||
" Test for the fuzzymatchpos() function
|
" Test for the matchfuzzypos() function
|
||||||
func Test_matchfuzzypos()
|
func Test_matchfuzzypos()
|
||||||
call assert_equal([['curl', 'world'], [[2,3], [2,3]]], matchfuzzypos(['world', 'curl'], 'rl'))
|
call assert_equal([['curl', 'world'], [[2,3], [2,3]]], matchfuzzypos(['world', 'curl'], 'rl'))
|
||||||
call assert_equal([['curl', 'world'], [[2,3], [2,3]]], matchfuzzypos(['world', 'one', 'curl'], 'rl'))
|
call assert_equal([['curl', 'world'], [[2,3], [2,3]]], matchfuzzypos(['world', 'one', 'curl'], 'rl'))
|
||||||
@@ -83,6 +97,10 @@ func Test_matchfuzzypos()
|
|||||||
\ [[0, 1, 2, 3, 4], [0, 1, 2, 3, 4]]],
|
\ [[0, 1, 2, 3, 4], [0, 1, 2, 3, 4]]],
|
||||||
\ matchfuzzypos(['hello world hello world', 'hello', 'world'], 'hello'))
|
\ matchfuzzypos(['hello world hello world', 'hello', 'world'], 'hello'))
|
||||||
call assert_equal([['aaaaaaa'], [[0, 1, 2]]], matchfuzzypos(['aaaaaaa'], 'aaa'))
|
call assert_equal([['aaaaaaa'], [[0, 1, 2]]], matchfuzzypos(['aaaaaaa'], 'aaa'))
|
||||||
|
call assert_equal([['a b'], [[0, 3]]], matchfuzzypos(['a b'], 'a b'))
|
||||||
|
call assert_equal([['a b'], [[0, 3]]], matchfuzzypos(['a b'], 'a b'))
|
||||||
|
call assert_equal([['a b'], [[0]]], matchfuzzypos(['a b'], ' a '))
|
||||||
|
call assert_equal([[], []], matchfuzzypos(['a b'], ' '))
|
||||||
call assert_equal([[], []], matchfuzzypos(['world', 'curl'], 'ab'))
|
call assert_equal([[], []], matchfuzzypos(['world', 'curl'], 'ab'))
|
||||||
let x = matchfuzzypos([repeat('a', 256)], repeat('a', 256))
|
let x = matchfuzzypos([repeat('a', 256)], repeat('a', 256))
|
||||||
call assert_equal(range(256), x[1][0])
|
call assert_equal(range(256), x[1][0])
|
||||||
@@ -104,6 +122,12 @@ func Test_matchfuzzypos()
|
|||||||
" best recursive match
|
" best recursive match
|
||||||
call assert_equal([['xoone'], [[2, 3, 4]]], matchfuzzypos(['xoone'], 'one'))
|
call assert_equal([['xoone'], [[2, 3, 4]]], matchfuzzypos(['xoone'], 'one'))
|
||||||
|
|
||||||
|
" match multiple words (separated by space)
|
||||||
|
call assert_equal([['foo bar baz'], [[8, 9, 10, 0, 1, 2]]], ['foo bar baz', 'foo', 'foo bar', 'baz bar']->matchfuzzypos('baz foo'))
|
||||||
|
call assert_equal([[], []], ['foo bar baz', 'foo', 'foo bar', 'baz bar']->matchfuzzypos('one two'))
|
||||||
|
call assert_equal([[], []], ['foo bar']->matchfuzzypos(" \t "))
|
||||||
|
call assert_equal([['grace'], [[1, 2, 3, 4, 2, 3, 4, 0, 1, 2, 3, 4]]], ['grace']->matchfuzzypos('race ace grace'))
|
||||||
|
|
||||||
let l = [{'id' : 5, 'val' : 'crayon'}, {'id' : 6, 'val' : 'camera'}]
|
let l = [{'id' : 5, 'val' : 'crayon'}, {'id' : 6, 'val' : 'camera'}]
|
||||||
call assert_equal([[{'id' : 6, 'val' : 'camera'}], [[0, 1, 2]]],
|
call assert_equal([[{'id' : 6, 'val' : 'camera'}], [[0, 1, 2]]],
|
||||||
\ matchfuzzypos(l, 'cam', {'text_cb' : {v -> v.val}}))
|
\ matchfuzzypos(l, 'cam', {'text_cb' : {v -> v.val}}))
|
||||||
@@ -126,6 +150,7 @@ func Test_matchfuzzypos()
|
|||||||
call assert_fails("let x = matchfuzzypos(l, 'foo', {'key' : 'name'})", 'E730:')
|
call assert_fails("let x = matchfuzzypos(l, 'foo', {'key' : 'name'})", 'E730:')
|
||||||
endfunc
|
endfunc
|
||||||
|
|
||||||
|
" Test for matchfuzzy() with multibyte characters
|
||||||
func Test_matchfuzzy_mbyte()
|
func Test_matchfuzzy_mbyte()
|
||||||
CheckFeature multi_lang
|
CheckFeature multi_lang
|
||||||
call assert_equal(['ンヹㄇヺヴ'], matchfuzzy(['ンヹㄇヺヴ'], 'ヹヺ'))
|
call assert_equal(['ンヹㄇヺヴ'], matchfuzzy(['ンヹㄇヺヴ'], 'ヹヺ'))
|
||||||
@@ -136,19 +161,19 @@ func Test_matchfuzzy_mbyte()
|
|||||||
call assert_equal(['ππbbππ', 'πππbbbπππ', 'ππππbbbbππππ', 'πbπ'],
|
call assert_equal(['ππbbππ', 'πππbbbπππ', 'ππππbbbbππππ', 'πbπ'],
|
||||||
\ matchfuzzy(['πbπ', 'ππbbππ', 'πππbbbπππ', 'ππππbbbbππππ'], 'ππ'))
|
\ matchfuzzy(['πbπ', 'ππbbππ', 'πππbbbπππ', 'ππππbbbbππππ'], 'ππ'))
|
||||||
|
|
||||||
|
" match multiple words (separated by space)
|
||||||
|
call assert_equal(['세 마리의 작은 돼지'], ['세 마리의 작은 돼지', '마리의', '마리의 작은', '작은 돼지']->matchfuzzy('돼지 마리의'))
|
||||||
|
call assert_equal([], ['세 마리의 작은 돼지', '마리의', '마리의 작은', '작은 돼지']->matchfuzzy('파란 하늘'))
|
||||||
|
|
||||||
" preference for camel case match
|
" preference for camel case match
|
||||||
call assert_equal(['oneĄwo', 'oneąwo'],
|
call assert_equal(['oneĄwo', 'oneąwo'],
|
||||||
\ ['oneąwo', 'oneĄwo']->matchfuzzy('oneąwo'))
|
\ ['oneąwo', 'oneĄwo']->matchfuzzy('oneąwo'))
|
||||||
" preference for complete match then match after separator (_ or space)
|
" preference for complete match then match after separator (_ or space)
|
||||||
if has("win32")
|
|
||||||
" order is different between Windows and Unix :(
|
|
||||||
" It's important that the complete match is first
|
|
||||||
call assert_equal(['ⅠⅡabㄟㄠ', 'ⅠⅡa bㄟㄠ', 'ⅠⅡa_bㄟㄠ'],
|
|
||||||
\ ['ⅠⅡabㄟㄠ', 'ⅠⅡa_bㄟㄠ', 'ⅠⅡa bㄟㄠ']->matchfuzzy('ⅠⅡabㄟㄠ'))
|
|
||||||
else
|
|
||||||
call assert_equal(['ⅠⅡabㄟㄠ'] + sort(['ⅠⅡa_bㄟㄠ', 'ⅠⅡa bㄟㄠ']),
|
call assert_equal(['ⅠⅡabㄟㄠ'] + sort(['ⅠⅡa_bㄟㄠ', 'ⅠⅡa bㄟㄠ']),
|
||||||
\ ['ⅠⅡabㄟㄠ', 'ⅠⅡa bㄟㄠ', 'ⅠⅡa_bㄟㄠ']->matchfuzzy('ⅠⅡabㄟㄠ'))
|
\ ['ⅠⅡabㄟㄠ', 'ⅠⅡa bㄟㄠ', 'ⅠⅡa_bㄟㄠ']->matchfuzzy('ⅠⅡabㄟㄠ'))
|
||||||
endif
|
" preference for match after a separator (_ or space)
|
||||||
|
call assert_equal(['ㄓㄔabㄟㄠ', 'ㄓㄔa_bㄟㄠ', 'ㄓㄔa bㄟㄠ'],
|
||||||
|
\ ['ㄓㄔa_bㄟㄠ', 'ㄓㄔa bㄟㄠ', 'ㄓㄔabㄟㄠ']->matchfuzzy('ㄓㄔabㄟㄠ'))
|
||||||
" preference for leading letter match
|
" preference for leading letter match
|
||||||
call assert_equal(['ŗŝţũŵż', 'xŗŝţũŵż'],
|
call assert_equal(['ŗŝţũŵż', 'xŗŝţũŵż'],
|
||||||
\ ['xŗŝţũŵż', 'ŗŝţũŵż']->matchfuzzy('ŗŝţũŵż'))
|
\ ['xŗŝţũŵż', 'ŗŝţũŵż']->matchfuzzy('ŗŝţũŵż'))
|
||||||
@@ -163,6 +188,7 @@ func Test_matchfuzzy_mbyte()
|
|||||||
\ ['ŗŝţxx', 'ŗŝţ', 'ŗŝţx']->matchfuzzy('ŗŝţ'))
|
\ ['ŗŝţxx', 'ŗŝţ', 'ŗŝţx']->matchfuzzy('ŗŝţ'))
|
||||||
endfunc
|
endfunc
|
||||||
|
|
||||||
|
" Test for matchfuzzypos() with multibyte characters
|
||||||
func Test_matchfuzzypos_mbyte()
|
func Test_matchfuzzypos_mbyte()
|
||||||
CheckFeature multi_lang
|
CheckFeature multi_lang
|
||||||
call assert_equal([['こんにちは世界'], [[0, 1, 2, 3, 4]]],
|
call assert_equal([['こんにちは世界'], [[0, 1, 2, 3, 4]]],
|
||||||
@@ -183,9 +209,13 @@ func Test_matchfuzzypos_mbyte()
|
|||||||
call assert_equal(range(256), x[1][0])
|
call assert_equal(range(256), x[1][0])
|
||||||
call assert_equal([[], []], matchfuzzypos([repeat('✓', 300)], repeat('✓', 257)))
|
call assert_equal([[], []], matchfuzzypos([repeat('✓', 300)], repeat('✓', 257)))
|
||||||
|
|
||||||
|
" match multiple words (separated by space)
|
||||||
|
call assert_equal([['세 마리의 작은 돼지'], [[9, 10, 2, 3, 4]]], ['세 마리의 작은 돼지', '마리의', '마리의 작은', '작은 돼지']->matchfuzzypos('돼지 마리의'))
|
||||||
|
call assert_equal([[], []], ['세 마리의 작은 돼지', '마리의', '마리의 작은', '작은 돼지']->matchfuzzypos('파란 하늘'))
|
||||||
|
|
||||||
" match in a long string
|
" match in a long string
|
||||||
call assert_equal([[repeat('♪', 300) .. '✗✗✗'], [[300, 301, 302]]],
|
call assert_equal([[repeat('ぶ', 300) .. 'ẼẼẼ'], [[300, 301, 302]]],
|
||||||
\ matchfuzzypos([repeat('♪', 300) .. '✗✗✗'], '✗✗✗'))
|
\ matchfuzzypos([repeat('ぶ', 300) .. 'ẼẼẼ'], 'ẼẼẼ'))
|
||||||
" preference for camel case match
|
" preference for camel case match
|
||||||
call assert_equal([['xѳѵҁxxѳѴҁ'], [[6, 7, 8]]], matchfuzzypos(['xѳѵҁxxѳѴҁ'], 'ѳѵҁ'))
|
call assert_equal([['xѳѵҁxxѳѴҁ'], [[6, 7, 8]]], matchfuzzypos(['xѳѵҁxxѳѴҁ'], 'ѳѵҁ'))
|
||||||
" preference for match after a separator (_ or space)
|
" preference for match after a separator (_ or space)
|
||||||
|
|||||||
@@ -750,6 +750,8 @@ static char *(features[]) =
|
|||||||
|
|
||||||
static int included_patches[] =
|
static int included_patches[] =
|
||||||
{ /* Add new patch number below this line */
|
{ /* Add new patch number below this line */
|
||||||
|
/**/
|
||||||
|
1893,
|
||||||
/**/
|
/**/
|
||||||
1892,
|
1892,
|
||||||
/**/
|
/**/
|
||||||
|
|||||||
Reference in New Issue
Block a user