OUDON · OUDON · Aug 30, 2022 · Aug 11, 2022 · Aug 11, 2022 · Aug 14, 2022
diff --git a/Makefile b/Makefile
@@ -1,3 +1,15 @@
+.PHONY: install
+install:
+	pip install -e ".[dev]"
+
+.PHONY: uninstall
+uninstall:
+	yes Y | pip uninstall rmqrcode
+
+.PHONY: test
+test:
+	python -m pytest
+
 .PHONY: lint
 lint:
 	flake8 src

diff --git a/README.md b/README.md
@@ -1,8 +1,8 @@
-# An rMQR Code Generator
+# Rectangular Micro QR Code (rMQR Code) Generator
 ![reop-url](https://user-images.githubusercontent.com/14174940/172978619-accbf9d0-9dd8-4b19-b47e-ad139a68dcc9.png)
 
 
-The rMQR Code is a rectangular two-dimensional barcode. This package can generate an rMQR Code image. This is implemented based on [ISO/IEC 23941: Rectangular Micro QR Code (rMQR) bar code symbology specification](https://www.iso.org/standard/77404.html).
+The rMQR Code is a rectangular two-dimensional barcode. This is easy to print in narrow space compared to conventional QR Code. This package can generate an rMQR Code image. This is implemented based on [ISO/IEC 23941: Rectangular Micro QR Code (rMQR) bar code symbology specification](https://www.iso.org/standard/77404.html).
 
 [![pytest](https://github.com/OUDON/rmqrcode-python/actions/workflows/python-app.yml/badge.svg?branch=main)](https://github.com/OUDON/rmqrcode-python/actions/workflows/python-app.yml)
 ![PyPI](https://img.shields.io/pypi/v/rmqrcode?color=blue)
@@ -48,7 +48,7 @@ optional arguments:
                         Strategy how to determine rMQR Code size.
 ```
 
-### Generate rMQR Code
+### Generate rMQR Code in scripts
 Alternatively, you can also use in python scripts:
 ```py
 from rmqrcode import rMQR
@@ -71,7 +71,7 @@ The `fit_strategy` parameter is enum value of `rmqrcode.FitStrategy` to specify
 - **`FitStrategy.MINIMIZE_HEIGHT`**: Try to minimize height.
 - **`FitStrategy.BALANCED`**: Try to keep balance of width and height.
 
-Here is an example of images genereated by each fit strategies for data `Test test test`:
+Here is an example of images generated by each fit strategies for data `Test test test`:
 ![Example of fit strategies](https://user-images.githubusercontent.com/14174940/175759120-7fb5ec71-c258-4646-9b91-6865b3eeac3f.png)
 
 ### Save as image
@@ -88,8 +88,8 @@ image.save("my_qr.png")
 ### Select rMQR Code size manually
 To select rMQR Code size manually, use `rMQR()` constructor.
 ```py
+from rmqrcode import rMQR, ErrorCorrectionLevel
 qr = rMQR('R11x139', ErrorCorrectionLevel.H)
-qr.make("https://oudon.xyz")
 ```
 
 `R11x139` means 11 rows and 139 columns. The following table shows available combinations.
@@ -103,22 +103,43 @@ qr.make("https://oudon.xyz")
 |R15|❌|✅|✅|✅|✅|✅|
 |R17|❌|✅|✅|✅|✅|✅|
 
+### Encoding Modes and Segments
 
-## 🛠️ Under the Hood
-### Encoding modes
+The rMQR Code has the four encoding modes Numeric, Alphanumeric, Byte and Kanji to convert data efficiently. We can select encoding mode for each data segment separately.
+The following example shows how to encode data "123Abc". The first segment is for "123" in the Numeric mode. The second segment is for "Abc" in the Byte mode.
+We can select an encoding mode by passing the `encoder_class` argument to the `rMQR#add_segment` method. In this example, the length of bits after encoding is 47 in the case combined with the Numeric mode and the Byte mode, which is shorter than 56 in the Byte mode only.
 
-The rMQR Code has the four encoding modes Numeric, Alphanumeric, Byte and Kanji to convert data efficiently. However, this package supoprts only Byte mode currently.
+```py
+from rmqrcode import rMQR, ErrorCorrectionLevel, encoder
+qr = rMQR('R7x43', ErrorCorrectionLevel.M)
+qr.add_segment("123", encoder_class=encoder.NumericEncoder)
+qr.add_segment("Abc", encoder_class=encoder.ByteEncoder)
+qr.make()
+```
+
+The value for `encoder_class` is listed in the below table.
+
+|Mode|Value of encoder_class|Characters|
+|-|-|-|
+|Numeric|NumericEncoder|0-9|
+|Alphanumeric|AlphanumericEncoder|0-9 A-Z \s $ % * + - . / :|
+|Byte|ByteEncoder|Any|
+|Kanji|KanjiEncoder|from 0x8140 to 0x9FFC, from 0xE040 to 0xEBBF in Shift JIS value|
+
+### Optimal Segmentation
+The `rMQR.fit` method mentioned above computes the optimal segmentation.
+For example, the data "123Abc" is divided into the following two segments.
 
-|Mode|Supported?|
-|-|:-:|
-|Numeric| |
-|Alphanumeric||
-|Byte|✅|
-|Kanji||
+|Segment No.|Data|Encoding Mode|
+|-|-|-|
+|Segment1|123|Numeric|
+|Segment2|Abc|Byte|
 
+In the case of other segmentation like "123A bc", the length of the bit string after
+encoding will be longer than the above optimal case.
 
-## 🤝 Contiributing
-Any suggestions are welcome! If you are interesting in contiributing, please read [CONTRIBUTING](https://github.com/OUDON/rmqrcode-python/blob/develop/CONTRIBUTING.md).
+## 🤝 Contributing
+Any suggestions are welcome! If you are interesting in contributing, please read [CONTRIBUTING](https://github.com/OUDON/rmqrcode-python/blob/develop/CONTRIBUTING.md).
 
 
 ## 📚 References

diff --git a/example.py b/example.py
@@ -2,6 +2,7 @@
 from rmqrcode import ErrorCorrectionLevel
 from rmqrcode import QRImage
 from rmqrcode import FitStrategy
+from rmqrcode import encoder
 
 import logging
 
@@ -24,9 +25,11 @@ def main():
     print(qr)
 
     # Determine rMQR version manually
-    # version = 'R13x99'
+    # version = 'R7x43'
     # qr = rMQR(version, error_correction_level)
-    # qr.make(data)
+    # qr.add_segment("123", encoder_class=encoder.NumericEncoder)
+    # qr.add_segment("Abc", encoder_class=encoder.ByteEncoder)
+    # qr.make()
     # print(qr)
 
     # Save as png

diff --git a/setup.cfg b/setup.cfg
@@ -1,6 +1,6 @@
 [metadata]
 name = rmqrcode
-version = 0.2.0
+version = 0.3.0
 author = Takahiro Tomita
 author_email = ttp8101@gmail.com
 description = An rMQR Code Generetor

diff --git a/src/rmqrcode/__init__.py b/src/rmqrcode/__init__.py
@@ -1,5 +1,21 @@
+from . import encoder
 from .format.error_correction_level import ErrorCorrectionLevel
 from .qr_image import QRImage
-from .rmqrcode import DataTooLongError, FitStrategy, IllegalVersionError, rMQR
+from .rmqrcode import (
+    DataTooLongError,
+    FitStrategy,
+    IllegalVersionError,
+    NoSegmentError,
+    rMQR,
+)
 
-__all__ = ("rMQR", "DataTooLongError", "FitStrategy", "IllegalVersionError", "QRImage", "ErrorCorrectionLevel")
+__all__ = (
+    "rMQR",
+    "DataTooLongError",
+    "FitStrategy",
+    "IllegalVersionError",
+    "NoSegmentError",
+    "QRImage",
+    "ErrorCorrectionLevel",
+    "encoder",
+)
diff --git a/src/rmqrcode/console.py b/src/rmqrcode/console.py
@@ -25,8 +25,8 @@ def _make_qr(data, ecc, version, fit_strategy):
             qr = rMQR(version, ecc)
         except IllegalVersionError:
             _show_error_and_exit("Error: Illegal version.")
-        qr.make(data)
-
+        qr.add_segment(data)
+        qr.make()
     return qr
 
 

diff --git a/src/rmqrcode/encoder/__init__.py b/src/rmqrcode/encoder/__init__.py
@@ -0,0 +1,7 @@
+from .alphanumeric_encoder import AlphanumericEncoder
+from .byte_encoder import ByteEncoder
+from .encoder_base import IllegalCharacterError
+from .kanji_encoder import KanjiEncoder
+from .numeric_encoder import NumericEncoder
+
+__all__ = ("ByteEncoder", "NumericEncoder", "IllegalCharacterError", "AlphanumericEncoder", "KanjiEncoder")
diff --git a/src/rmqrcode/encoder/alphanumeric_encoder.py b/src/rmqrcode/encoder/alphanumeric_encoder.py
@@ -0,0 +1,92 @@
+import re
+
+from .encoder_base import EncoderBase
+
+
+class AlphanumericEncoder(EncoderBase):
+    CHARACTER_MAP = {
+        "0": 0,
+        "1": 1,
+        "2": 2,
+        "3": 3,
+        "4": 4,
+        "5": 5,
+        "6": 6,
+        "7": 7,
+        "8": 8,
+        "9": 9,
+        "A": 10,
+        "B": 11,
+        "C": 12,
+        "D": 13,
+        "E": 14,
+        "F": 15,
+        "G": 16,
+        "H": 17,
+        "I": 18,
+        "J": 19,
+        "K": 20,
+        "L": 21,
+        "M": 22,
+        "N": 23,
+        "O": 24,
+        "P": 25,
+        "Q": 26,
+        "R": 27,
+        "S": 28,
+        "T": 29,
+        "U": 30,
+        "V": 31,
+        "W": 32,
+        "X": 33,
+        "Y": 34,
+        "Z": 35,
+        " ": 36,
+        "$": 37,
+        "%": 38,
+        "*": 39,
+        "+": 40,
+        "-": 41,
+        ".": 42,
+        "/": 43,
+        ":": 44,
+    }
+
+    @classmethod
+    def mode_indicator(cls):
+        return "010"
+
+    @classmethod
+    def _encoded_bits(cls, data):
+        res = ""
+        data_grouped = cls._group_by_2characters(data)
+        for s in data_grouped:
+            if len(s) == 2:
+                value = cls.CHARACTER_MAP[s[0]] * 45 + cls.CHARACTER_MAP[s[1]]
+                res += bin(value)[2:].zfill(11)
+            elif len(s) == 1:
+                value = cls.CHARACTER_MAP[s[0]]
+                res += bin(value)[2:].zfill(6)
+        return res
+
+    @classmethod
+    def _group_by_2characters(cls, data):
+        res = []
+        while data != "":
+            res.append(data[:2])
+            data = data[2:]
+        return res
+
+    @classmethod
+    def length(cls, data, character_count_indicator_length):
+        return (
+            len(cls.mode_indicator()) + character_count_indicator_length + 11 * (len(data) // 2) + 6 * (len(data) % 2)
+        )
+
+    @classmethod
+    def characters_num(cls, data):
+        return len(data)
+
+    @classmethod
+    def is_valid_characters(cls, data):
+        return bool(re.match(r"^[0-9A-Z\s\$\%\*\+\-\.\/\:]*$", data))
diff --git a/src/rmqrcode/encoder/byte_encoder.py b/src/rmqrcode/encoder/byte_encoder.py
@@ -1,21 +1,27 @@
-class ByteEncoder:
-    MODE_INDICATOR = "011"
+from .encoder_base import EncoderBase
 
-    @staticmethod
-    def _encoded_bits(s):
+
+class ByteEncoder(EncoderBase):
+    @classmethod
+    def mode_indicator(cls):
+        return "011"
+
+    @classmethod
+    def _encoded_bits(cls, s):
         res = ""
         encoded = s.encode("utf-8")
         for byte in encoded:
             res += bin(byte)[2:].zfill(8)
         return res
 
-    @staticmethod
-    def encode(data, character_count_length):
-        res = ByteEncoder.MODE_INDICATOR
-        res += bin(len(data))[2:].zfill(character_count_length)
-        res += ByteEncoder._encoded_bits(data)
-        return res
+    @classmethod
+    def length(cls, data, character_count_indicator_length):
+        return len(cls.mode_indicator()) + character_count_indicator_length + 8 * len(data.encode("utf-8"))
 
-    @staticmethod
-    def length(data):
+    @classmethod
+    def characters_num(cls, data):
         return len(data.encode("utf-8"))
+
+    @classmethod
+    def is_valid_characters(cls, data):
+        return True  # Any characters can encode in the Byte Mode