Merge pull request #33 from codingpot/feat/verup-0.10.3

feat: add splitSentence

Author: mrchypark

.devcontainer/Dockerfile

@@ -9,21 +9,11 @@ ARG NODE_VERSION="none"
 RUN if [ "${NODE_VERSION}" != "none" ]; then su vscode -c "umask 0002 && . /usr/local/share/nvm/nvm.sh && nvm install ${NODE_VERSION} 2>&1"; fi
 
 # [Optional] Uncomment this section to install additional OS packages.
-RUN apt-get update && export DEBIAN_FRONTEND=noninteractive \
-    && apt-get -y install --no-install-recommends cmake \
-    && git clone https://github.com/bab2min/Kiwi.git \
-    && cd Kiwi \
-    && git submodule sync \
-    && git submodule update --init --recursive \
-    && mkdir build \
-    && cd build \
-    && cmake -DCMAKE_BUILD_TYPE=Release ../ \
-    && make && make install \
-    && ldconfig \
-    && cd ../.. && rm -rf Kiwi \
-    && apt-get --purge -y remove cmake \
-    && apt autoremove -y \
-    && rm -rf /var/lib/apt/lists/*
+#RUN apt-get update && export DEBIAN_FRONTEND=noninteractive \
+#    && apt-get -y install --no-install-recommends cmake \
+#    && apt-get --purge -y remove cmake \
+#    && apt autoremove -y \
+#    && rm -rf /var/lib/apt/lists/*
 
 
 RUN sh -c "$(curl -fsSL https://starship.rs/install.sh)" -- --yes

Makefile

@@ -1,4 +1,4 @@
-KIWI_VERSION := "v0.10.2"
+KIWI_VERSION := "v0.10.3"
 
 .PHONY: test
 test: ModelGenerator/default.dict

kiwi.go

@@ -122,6 +122,34 @@ func (k *Kiwi) Analyze(text string, topN int, options AnalyzeOption) ([]TokenRes
 	return res, nil
 }
 
+// SplitResult returns the Sentences.
+type SplitResult struct {
+	Text  string
+	Begin int
+	End   int
+}
+
+// SplitSentence returns the line of sentences.
+func (k *Kiwi) SplitSentence(text string, options AnalyzeOption) ([]SplitResult, error) {
+	kiwiSsH := C.kiwi_split_into_sents(k.handler, C.CString(text), C.int(options), nil)
+	defer C.kiwi_ss_close(kiwiSsH)
+
+	resSize := int(C.kiwi_ss_size(kiwiSsH))
+	res := make([]SplitResult, resSize)
+
+	for i := 0; i < resSize; i++ {
+		begin := int(C.kiwi_ss_begin_position(kiwiSsH, C.int(i)))
+		end := int(C.kiwi_ss_end_position(kiwiSsH, C.int(i)))
+		res[i] = SplitResult{
+			Text:  text[begin:end],
+			Begin: begin,
+			End:   end,
+		}
+	}
+
+	return res, nil
+}
+
 // Close frees the resource allocated for Kiwi and returns the exit status.
 // This must be called after New.
 // Returns 0 if successful.

kiwi_test.go

@@ -9,7 +9,7 @@ import (
 )
 
 func TestKiwiVersion(t *testing.T) {
-	assert.Equal(t, KiwiVersion(), "0.10.2")
+	assert.Equal(t, KiwiVersion(), "0.10.3")
 }
 
 func TestAnalyze(t *testing.T) {
@@ -63,6 +63,27 @@ func TestAnalyze(t *testing.T) {
 	assert.Equal(t, 0, kiwi.Close())
 }
 
+func TestSplitSentence(t *testing.T) {
+	kiwi := New("./ModelGenerator", 1, KIWI_BUILD_DEFAULT)
+	res, _ := kiwi.SplitSentence("여러 문장으로 구성된 텍스트네 이걸 분리해줘", KIWI_MATCH_ALL)
+
+	expected := []SplitResult{
+		{
+			Text:  "여러 문장으로 구성된 텍스트네",
+			Begin: 0,
+			End:   42,
+		},
+		{
+			Text:  "이걸 분리해줘",
+			Begin: 43,
+			End:   62,
+		},
+	}
+
+	assert.Equal(t, expected, res)
+	assert.Equal(t, 0, kiwi.Close())
+}
+
 func TestAddWordFail(t *testing.T) {
 	kb := NewBuilder("./ModelGenerator", 1, KIWI_BUILD_INTEGRATE_ALLOMORPH)
 	add := kb.AddWord("아버지가", "SKO", 0)