upgrade to v0.2: code maintaining

.gitignore

@@ -0,0 +1,7 @@
+build/
+dist/
+.venv/
+.idea/
+.pytest_cache/
+*/**/__pycache__/
+*.egg-info/

dev_requirements.txt

@@ -0,0 +1 @@
+twine

dev_scripts/prase.py

@@ -0,0 +1,23 @@
+import pickle
+import pathlib
+
+WORKSPACE_DIR = pathlib.Path(__file__).parent.parent
+
+data = {}
+
+for i in [
+    WORKSPACE_DIR / "chaizi/chaizi-ft.txt",
+    WORKSPACE_DIR / "chaizi/chaizi-jt.txt",
+]:
+    with open(i, "rt") as fd:
+        for line in fd:
+            item_list = line.strip().split("\t")
+            key = item_list[0]
+            value = [i.strip().split() for i in item_list[1:]]
+
+            data[key] = value
+
+output_file = WORKSPACE_DIR / "hanzi_chaizi/data/data.pkl"
+
+with open(output_file, "wb") as fd:
+    pickle.dump(data, fd)

example_code.py → example_code/example_code.py

@@ -1,6 +1,6 @@
 from hanzi_chaizi import HanziChaizi
 
 hc = HanziChaizi()
-result = hc.query('名')
+result = hc.query("名")
 
-print(result)
+print(result)

hanzi_chaizi/__init__.py

@@ -1,22 +1 @@
-import pickle
-
-import pkg_resources
-
-
-class HanziChaizi(object):
-    def __init__(self):
-        data_file = pkg_resources.resource_filename(__name__, "data/data.pkl")
-
-        with open(data_file, 'rb') as fd:
-            self.data = pickle.load(fd)
-
-    def query(self, input_char, default=None):
-        result = self.data.get(input_char, default)
-        return result[0]
-
-
-if __name__ == "__main__":
-    hc = HanziChaizi()
-    result = hc.query('名')
-
-    print(result)
+from .hanzi_chaizi import HanziChaizi

hanzi_chaizi/hanzi_chaizi.py

@@ -0,0 +1,15 @@
+import pickle
+
+import pkg_resources
+
+
+class HanziChaizi(object):
+    def __init__(self):
+        data_file = pkg_resources.resource_filename(__name__, "data/data.pkl")
+
+        with open(data_file, "rb") as fd:
+            self.data = pickle.load(fd)
+
+    def query(self, input_char, default=None):
+        result = self.data.get(input_char, default)
+        return result[0]

makefile

@@ -0,0 +1,8 @@
+pkg_build:
+	python setup.py bdist_wheel
+pkg_upload:
+	twine upload dist/*
+pkg_clean:
+	rm -rf build dist *.egg-info
+test:
+	pytest

pytest.ini

@@ -0,0 +1,2 @@
+[pytest]
+pythonpath = .

setup.py

@@ -1,13 +1,15 @@
 from setuptools import setup
 
 setup(
-    name='hanzi_chaizi',
-    version='0.1',
-    packages=['hanzi_chaizi'],
-    url='',
-    license='',
-    author='Xiaoquan Kong',
-    author_email='[email protected]',
-    description='',
-    include_package_data=True
+    name="hanzi_chaizi",
+    version="0.2",
+    packages=["hanzi_chaizi"],
+    url="https://github.com/howl-anderson/hanzi_chaizi",
+    license="",
+    author="Xiaoquan Kong",
+    author_email="[email protected]",
+    description="""
+The hanzi_chaizi package provides tools for breaking down Chinese characters into basic structural units, such as strokes and components. This decomposition reveals structural similarities between characters with similar shapes. These features can be used in deep learning models to capture character-based attributes, particularly shape-related ones, for applications like natural language processing or character recognition.
+    """.strip(),
+    include_package_data=True,
 )

test_requirements.txt

@@ -0,0 +1 @@
+pytest

tests/test_hanzi_chaizi.py

@@ -0,0 +1,8 @@
+from hanzi_chaizi import HanziChaizi
+
+
+def test_query():
+    hc = HanziChaizi()
+    result = hc.query("名")
+
+    assert result == ["夕", "口"]