Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Add a new API to get the frequency of phrases #216

Open
wants to merge 2 commits into
base: master
Choose a base branch
from
Open

Add a new API to get the frequency of phrases #216

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
1f5ff65
Add new API to get the frequency of phrases
mangokingTW Mar 31, 2016
41a01f1
Add a new test case for chewing_userphrase_freq()
mangokingTW Apr 17, 2016
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 4 additions & 0 deletions include/chewingio.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -529,6 +529,10 @@ CHEWING_API int chewing_userphrase_get(ChewingContext *ctx,
char *phrase_buf, unsigned int phrase_len,
char *bopomofo_buf, unsigned int bopomofo_len);

CHEWING_API int chewing_userphrase_get_freq(ChewingContext *ctx,
const char phrase_buf[], const char bopomofo_buf[],
unsigned int *orig_freq, unsigned int *max_freq, unsigned int *user_freq, unsigned int *time);
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Split unsigned int *user_freq, unsigned int *time to new line for pretty code.


CHEWING_API int chewing_userphrase_add(ChewingContext *ctx, const char *phrase_buf, const char *bopomofo_buf);

CHEWING_API int chewing_userphrase_remove(ChewingContext *ctx, const char *phrase_buf, const char *bopomofo_buf);
Expand Down
122 changes: 122 additions & 0 deletions src/chewingio.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1979,6 +1979,128 @@ CHEWING_API int chewing_userphrase_get(ChewingContext *ctx,
#endif
}

CHEWING_API int chewing_userphrase_get_freq(ChewingContext *ctx,
const char phrase_buf[], const char bopomofo_buf[],
unsigned int *orig_freq, unsigned int *max_freq, unsigned int *user_freq, unsigned int *time)
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Ditto. Split unsigned int *user_freq, unsigned int *time).

{
ChewingData *pgdata = NULL;

int ret;
int phone_len;
int phrase_len;
uint16_t phone_buf[MAX_PHONE_SEQ_LEN];
#ifdef WITH_SQLITE3
int i;
#else
const TreeType *tree_pos;
Phrase phrase;
UserPhraseData *uphrase;
#endif

if (!ctx || !phrase_buf || !bopomofo_buf) {
return -1;
}

pgdata = ctx->data;
phrase_len = ueStrLen(phrase_buf);
phone_len = UintArrayFromBopomofo(NULL, 0, bopomofo_buf);

ret = UintArrayFromBopomofo(phone_buf, phone_len + 1, bopomofo_buf);
if (ret == -1) {
return 0;
}

if (phone_len != phrase_len) {
LOG_WARN("Do not update userphrase because phone_buf length %d != phrase_buf length %d", phone_len, phrase_len);
return USER_UPDATE_FAIL;
}

if (phrase_len > MAX_PHRASE_LEN) {
LOG_WARN("phrase_buf length %d > MAX_PHRASE_LEN (%d)", phrase_len, MAX_PHRASE_LEN);
return -1;
}

#if WITH_SQLITE3
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Can you avoid manipulating SQLite here? Conceptually, we wish to maintain SQLite-specific functions in file userphrase-sql.c rather than every piece of libchewing source files.

Copy link
Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Does it mean that I should add some functions to manipulate SQLite for my API in file userphrase-sql.c?

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Beware of the configurations of libchewing contain one kind of combinations where SQLite3 is disabled.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

And, we expect developers can focus on the functionality of specific operations without digging every source files.

ret = sqlite3_reset(pgdata->static_data.stmt_userphrase[STMT_USERPHRASE_SELECT_BY_PHONE_PHRASE]);
if (ret != SQLITE_OK) {
LOG_ERROR("sqlite3_reset returns %d.", ret);
return ret;
}

ret = sqlite3_bind_int(pgdata->static_data.stmt_userphrase[STMT_USERPHRASE_SELECT_BY_PHONE_PHRASE], BIND_USERPHRASE_LENGTH, phone_len);
if (ret != SQLITE_OK) {
LOG_ERROR("sqlite3_bind_int returns %d.", ret);
return ret;
}

ret = sqlite3_bind_text(pgdata->static_data.stmt_userphrase[STMT_USERPHRASE_SELECT_BY_PHONE_PHRASE], BIND_USERPHRASE_PHRASE, phrase_buf, -1, SQLITE_STATIC);
if (ret != SQLITE_OK) {
LOG_ERROR("sqlite3_bind_text returns %d", ret);
return ret;
}

for (i = 0; i < phone_len; ++i) {
ret = sqlite3_bind_int(pgdata->static_data.stmt_userphrase[STMT_USERPHRASE_SELECT_BY_PHONE_PHRASE], BIND_USERPHRASE_PHONE_0 + i, phone_buf[i]);
if (ret != SQLITE_OK) {
LOG_ERROR("sqlite3_bind_int returns %d", ret);
return ret;
}
}

for (i = phone_len; i < MAX_PHRASE_LEN; ++i) {
ret = sqlite3_bind_int(pgdata->static_data.stmt_userphrase[STMT_USERPHRASE_SELECT_BY_PHONE_PHRASE], BIND_USERPHRASE_PHONE_0 + i, 0);
if (ret != SQLITE_OK) {
LOG_ERROR("sqlite3_bind_int returns %d", ret);
return ret;
}
}

ret = sqlite3_step(pgdata->static_data.stmt_userphrase[STMT_USERPHRASE_SELECT_BY_PHONE_PHRASE]);

if (ret != SQLITE_ROW) return -1;

*orig_freq = sqlite3_column_int(pgdata->static_data.stmt_userphrase[STMT_USERPHRASE_SELECT_BY_PHONE_PHRASE],
SQL_STMT_USERPHRASE[STMT_USERPHRASE_SELECT_BY_PHONE_PHRASE].column
[COLUMN_USERPHRASE_ORIG_FREQ]);

*max_freq = sqlite3_column_int(pgdata->static_data.stmt_userphrase[STMT_USERPHRASE_SELECT_BY_PHONE_PHRASE],
SQL_STMT_USERPHRASE[STMT_USERPHRASE_SELECT_BY_PHONE_PHRASE].column
[COLUMN_USERPHRASE_MAX_FREQ]);

*user_freq = sqlite3_column_int(pgdata->static_data.stmt_userphrase[STMT_USERPHRASE_SELECT_BY_PHONE_PHRASE],
SQL_STMT_USERPHRASE[STMT_USERPHRASE_SELECT_BY_PHONE_PHRASE].column
[COLUMN_USERPHRASE_USER_FREQ]);

*time = sqlite3_column_int(pgdata->static_data.stmt_userphrase[STMT_USERPHRASE_SELECT_BY_PHONE_PHRASE],
SQL_STMT_USERPHRASE[STMT_USERPHRASE_SELECT_BY_PHONE_PHRASE].column
[COLUMN_USERPHRASE_TIME]);
#else
tree_pos = TreeFindPhrase(pgdata, 0, phone_len - 1, phone_buf);
if (tree_pos) {
GetPhraseFirst(pgdata, &phrase, tree_pos);
do {
if (!strcmp(phrase.phrase, phrase_buf)) {
*orig_freq = phrase.freq;
*max_freq = phrase.freq;
break;
}
} while (GetVocabNext(pgdata, &phrase));
}

uphrase = UserGetPhraseFirst(pgdata, phone_buf);
while (uphrase) {
if (!strcmp(uphrase->wordSeq, phrase_buf)) {
*user_freq = uphrase->userfreq;
break;
}
uphrase = UserGetPhraseNext(pgdata, phone_buf);
}

#endif

return 0;
}

CHEWING_API int chewing_userphrase_add(ChewingContext *ctx, const char *phrase_buf, const char *bopomofo_buf)
{
ChewingData *pgdata;
Expand Down
67 changes: 67 additions & 0 deletions test/test-userphrase.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -447,11 +447,78 @@ void test_userphrase_enumerate_rewind()
chewing_delete(ctx);
}

void test_userphrase_enumerate_frequency()
{
ChewingContext *ctx;
int ret;
unsigned int expect_len;

const char phrase[] = "\xE6\xB8\xAC\xE8\xA9\xA6" /* 測試 */ ;
char phrase_buf[50];
unsigned int phrase_len;

const char bopomofo[] = "\xE3\x84\x98\xE3\x84\x9C\xCB\x8B \xE3\x84\x95\xCB\x8B"; /* ㄘㄜˋ ㄕˋ */
char bopomofo_buf[50];
unsigned int bopomofo_len;

unsigned int orig_freq = 0, orig_freq_b = 0;
unsigned int max_freq = 0, max_freq_b = 0;
unsigned int user_freq = 0, user_freq_b = 0;
unsigned int time = 0, time_b = 0;

int i;

clean_userphrase();

ctx = chewing_new();
start_testcase(ctx, fd);

ret = chewing_userphrase_add(ctx, phrase, bopomofo);
ok(ret == 1, "chewing_userphrase_add() return value `%d' shall be `%d'", ret, 1);
ret = chewing_userphrase_lookup(ctx, phrase, bopomofo);
ok(ret == 1, "chewing_userphrase_lookup() return value `%d' shall be `%d'", ret, 1);

ret = chewing_userphrase_enumerate(ctx);
ok(ret == 0, "chewing_userphrase_enumerate() return value `%d' shall be `%d'", ret, 0);

ret = chewing_userphrase_has_next(ctx, &phrase_len, &bopomofo_len);
ok(ret == 1, "chewing_userphrase_has_next() return value `%d' shall be `%d'", ret, 1);
expect_len = strlen(phrase) + 1;
ok(phrase_len >= expect_len, "chewing_userphrase_has_next() shall set phrase_len `%d' >= `%d'", phrase_len,
expect_len);
expect_len = strlen(bopomofo) + 1;
ok(bopomofo_len >= expect_len, "chewing_userphrase_has_next() shall set bopomofo_len `%d' >= `%d'", bopomofo_len,
expect_len);
ret = chewing_userphrase_get(ctx, phrase_buf, sizeof(phrase_buf), bopomofo_buf, sizeof(bopomofo_buf));
ok(ret == 0, "chewing_userphrase_get() return value `%d' shall be `%d'", ret, 0);
ok(strcmp(phrase_buf, phrase) == 0, "chewing_userphrase_get() shall set phrase_buf `%s' to `%s'", phrase_buf,
phrase);
ok(strcmp(bopomofo_buf, bopomofo) == 0, "chewing_userphrase_get() shall set bopomofo_buf `%s' to `%s'",
bopomofo_buf, bopomofo);

ret = chewing_userphrase_get_freq(ctx, phrase_buf, bopomofo_buf, &orig_freq, &max_freq, &user_freq, &time);
ok(ret == 0, "chewing_userphrase_get_freq() return value `%d' shall be `%d'", ret, 0);
for( i=0 ; i < 32 ; i++ ){ // frequently type 測試 to increase the frequency of it.
type_keystroke_by_string(ctx, "hk4g4<SL><SL><E>");
}
ret = chewing_userphrase_get_freq(ctx, phrase_buf, bopomofo_buf, &orig_freq_b, &max_freq_b, &user_freq_b, &time_b);
ok(ret == 0, "chewing_userphrase_get_freq() return value `%d' shall be `%d'", ret, 0);
ok(orig_freq_b == orig_freq, "chewing_userphrase_get_freq() orig_freq `%d' shall be `%d'", orig_freq_b, orig_freq);
ok(max_freq_b == max_freq, "chewing_userphrase_get_freq() max_freq `%d' shall be `%d'", max_freq_b, max_freq);
ok(user_freq_b > user_freq, "chewing_userphrase_get_freq() shall set user_freq `%d' > `%d'", user_freq_b, user_freq);
ok(time_b >= time, "chewing_userphrase_get_freq() shall set time `%d' >= `%d'", time_b, time);
printf("ret=%d. orig=%d. max=%d. user=%d. time=%d.\n",ret,orig_freq,max_freq,user_freq,time);
Copy link
Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

forgot to delete this line


chewing_delete(ctx);

}

void test_userphrase_enumerate()
{
test_userphrase_enumerate_normal();
test_userphrase_enumerate_empty();
test_userphrase_enumerate_rewind();
test_userphrase_enumerate_frequency();
}

void test_userphrase_manipulate_normal()
Expand Down