From 9670fe024b6ac484e863e93d970c445ce090a1df Mon Sep 17 00:00:00 2001
From: alalalalaki <harlan.zhu@gmail.com>
Date: Wed, 20 Jan 2021 03:30:39 +0900
Subject: [PATCH] Add z2h. Publish 0.0.4.

---
 jpstat/estat/core.py |  4 ++++
 jpstat/util/z2h.py   | 47 ++++++++++++++++++++++++++++++++++++++++++++
 jpstat/version.py    |  2 +-
 pyproject.toml       |  2 +-
 4 files changed, 53 insertions(+), 2 deletions(-)
 create mode 100644 jpstat/util/z2h.py

diff --git a/jpstat/estat/core.py b/jpstat/estat/core.py
index bc3fbcc..04e29a6 100644
--- a/jpstat/estat/core.py
+++ b/jpstat/estat/core.py
@@ -11,6 +11,7 @@
 from .. import config
 from .api import API
 from .util.clean import clean_dict_cols
+from ..util.z2h import str_z2h
 
 
 def get_list(statsCode=None, searchWord=None, outputRaw=False, key=None, lang=None, **kwargs):
@@ -23,6 +24,7 @@ def get_list(statsCode=None, searchWord=None, outputRaw=False, key=None, lang=No
                    'STATISTICS_NAME', 'TITLE',
                    'SURVEY_DATE', 'OPEN_DATE', 'OVERALL_TOTAL_NUMBER']
     df = df[cols_simple].pipe(clean_dict_cols, ['STAT_NAME', 'GOV_ORG', 'TITLE'])
+    df = df.applymap(str_z2h)
     return df
 
 
@@ -31,6 +33,7 @@ def get_stat(key=None, lang=None,):
     data = api.get_list(statsNameList="Y")
     df = pd.DataFrame(data['DATALIST_INF']['LIST_INF'])
     df = df.pipe(clean_dict_cols, ['STAT_NAME', 'GOV_ORG'])
+    df = df.applymap(str_z2h)
     return df
 
 
@@ -61,6 +64,7 @@ def get_data(statsDataId, return_note=True, key=None, lang=None,  **kwargs):
         df.drop(col_name, axis=1, inplace=True)
     df['Value'] = df['$']
     df.drop('$', axis=1, inplace=True)
+    df = df.applymap(str_z2h)
     if return_note:
         try:
             note = pd.DataFrame(data['STATISTICAL_DATA']['DATA_INF']['NOTE'])
diff --git a/jpstat/util/z2h.py b/jpstat/util/z2h.py
new file mode 100644
index 0000000..1eeab76
--- /dev/null
+++ b/jpstat/util/z2h.py
@@ -0,0 +1,47 @@
+"""
+This code is generally borrown from under-maintained pkg `japandas`
+and used to turn 「全角」strings to 「半角」strings.
+"""
+
+__version__ = 0.1
+
+from unicodedata import normalize
+
+# soundmarks require special handlings
+_ZALPHA = ('ＡＢＣＤＥＦＧＨＩＪＫＬＭＮＯＰＱＲＳＴＵＶＷＸＹＺ'
+           'ａｂｃｄｅｆｇｈｉｊｋｌｍｎｏｐｑｒｓｔｕｖｗｘｙｚ')
+_ZSYMBOL = '！＂＃＄％＆＇（）＊＋，－．／：；＜＝＞？＠［＼］＾＿｀｛｜｝～　'
+_ZDIGIT = '０１２３４５６７８９'
+
+# mapping from full-width to half-width
+_ALPHA_MAPPER = {c: normalize('NFKC', c) for c in _ZALPHA}
+_DIGIT_MAPPER = {c: normalize('NFKC', c) for c in _ZDIGIT}
+_SYMBOL_MAPPER = {c: normalize('NFKC', c) for c in _ZSYMBOL}
+
+# adding symbols that un-normalizable
+# https://www.utf8-chartable.de/unicode-utf8-table.pl?start=12224&names=-&utf8=string-literal
+_ZSYMBOL_MAPPER = {"〜": "~", }
+_SYMBOL_MAPPER.update(_ZSYMBOL_MAPPER)
+
+
+def _ord_dict(dict):
+    return {ord(k): v for k, v in dict.items()}
+
+
+# for unicode.translate
+_Z2H_ALPHA = _ord_dict(_ALPHA_MAPPER)
+_Z2H_DIGIT = _ord_dict(_DIGIT_MAPPER)
+_Z2H_SYMBOL = _ord_dict(_SYMBOL_MAPPER)
+
+mapper = dict()
+mapper.update(_Z2H_ALPHA)
+mapper.update(_Z2H_DIGIT)
+mapper.update(_Z2H_SYMBOL)
+
+
+def str_z2h(string: str, ):
+    try:
+        res = string.translate(mapper)
+        return res
+    except AttributeError:
+        return string
diff --git a/jpstat/version.py b/jpstat/version.py
index c301537..156d6f9 100644
--- a/jpstat/version.py
+++ b/jpstat/version.py
@@ -1 +1 @@
-__version__ = '0.0.3.2'
+__version__ = '0.0.4'
diff --git a/pyproject.toml b/pyproject.toml
index 6c3092a..7e64e85 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "jpstat"
-version = "0.0.3.2"
+version = "0.0.4"
 description = "A python library for accessing official statistics of Japan."
 authors = ["Xuanli Zhu <akaguro.koyomi@gmail.com>"]
 license = "MIT"