[script.subtitles.zimukux] 0.3

script.subtitles.zimukux/addon.xml

@@ -1,5 +1,5 @@
 <?xml version="1.0" encoding="UTF-8" standalone="yes"?>
-<addon id="script.subtitles.zimukux" name="ZiMuKuX" version="0.2.1" provider-name="pizzamx">
+<addon id="script.subtitles.zimukux" name="ZiMuKuX" version="0.3b" provider-name="pizzamx">
     <requires>
         <import addon="xbmc.python" version="3.0.0"/>
         <import addon="script.module.beautifulsoup4" version="4.6.2"/>
@@ -20,8 +20,15 @@
             <fanart>resources/fanart.png</fanart>
         </assets>
         <news>
-v0.2.1(2023/12/11)
-- Merged https://github.com/pizzamx/zimuku_for_kodi/pull/23（thanks @jiangpengcheng for solving the cookie issue）
+v0.3(2025/06/30)
+- Hardcoded OCR mechanism, credits to @realaboo( and behind-the-scene Gemini Code Assist)
+
+v0.2c(2024/10/28)
+- Add new OCR vender (Tencent), ref: https://console.cloud.tencent.com/ocr/overview
+
+v0.2b(2024/4/6)
+- Need to resolve Captcha issue again
+- Do change OCR service URL to your own, ref: https://cloud.baidu.com/doc/OCR/s/dk3iqnq51
 
 v0.2.0(2023/5/30)
 - Merged https://github.com/pizzamx/zimuku_for_kodi/pull/18 (big thanks to lm317379829 for solving the captcha issue)

script.subtitles.zimukux/resources/language/resource.language.en_GB/strings.po

@@ -9,10 +9,6 @@ msgctxt "#30101"
 msgid "Site URL"
 msgstr ""
 
-msgctxt "#301011"
-msgid "OCR URL"
-msgstr "OCR API"
-
 msgctxt "#30200"
 msgid "Sub preference"
 msgstr ""

script.subtitles.zimukux/resources/language/resource.language.zh_CN/strings.po

@@ -9,10 +9,6 @@ msgctxt "#30101"
 msgid "Site URL"
 msgstr "网址"
 
-msgctxt "#301011"
-msgid "OCR URL"
-msgstr "OCR API 的地址（请勿随意修改）"
-
 msgctxt "#30200"
 msgid "Sub preference"
 msgstr "字幕下载偏好"

script.subtitles.zimukux/resources/lib/ocr.py

@@ -0,0 +1,207 @@
+# -*- coding: utf-8 -*-
+"""
+Subtitle add-on for Kodi 19+ derived from https://github.com/taxigps/xbmc-addons-chinese/tree/master/service.subtitles.zimuku
+Copyright (C) <2021>  <[email protected]>
+
+This program is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2 of the License, or
+(at your option) any later version.
+
+This program is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License along
+with this program; if not, write to the Free Software Foundation, Inc.,
+51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
+"""
+
+"""
+A simple OCR script to recognize 5 digits from a Base64 encoded BMP image,
+using only standard Python libraries.
+
+This script is designed to work in restricted environments like Kodi add-ons.
+"""
+import base64
+import struct
+import sys
+from typing import Tuple, List
+
+
+class BmpOcr:
+    """
+    Performs OCR on a specific 100x27 24-bit BMP image of 5 digits.
+
+    The method uses template matching based on a few sample pixels
+    for each character.
+    """
+    # Image properties
+    IMG_WIDTH = 100
+    IMG_HEIGHT = 27
+    CHAR_WIDTH = 20
+    NUM_CHARS = 5
+
+    # BMP header constants
+    PIXEL_DATA_OFFSET = 54  # For 24-bit BMP without a color palette
+
+    # OCR sampling points relative to the top-left of a 20x27 character box.
+    # These points are chosen to effectively distinguish between digits 0-9.
+    # (x, y)
+    SAMPLE_POINTS = [
+        (10, 7),   # P0: Top-center
+        (7, 8),    # P1: Top-left
+        (12, 8),   # P2: Top-right
+        (10, 13),  # P3: Center
+        (7, 19),   # P4: Bottom-left
+        (12, 19),  # P5: Bottom-right
+        (10, 20),  # P6: Bottom-center
+        (6, 13),   # P7: Middle-left
+        (14, 13)   # P8: Middle-right
+    ]
+
+    # Pre-defined feature vectors for digits 0-9.
+    # 1 represents a foreground (dark) pixel, 0 represents a background (light) pixel.
+    # The vectors for '1', '2', '6', '7', '8' are derived from the sample image.
+    # The rest are crafted based on a standard digital font shape.
+    DIGIT_TEMPLATES = {
+        '0': [1, 1, 1, 1, 1, 1, 1, 1, 0],#
+        '1': [0, 1, 0, 0, 0, 0, 1, 0, 0],#
+        '2': [1, 0, 1, 0, 1, 0, 1, 0, 0],#
+        '3': [1, 0, 1, 1, 0, 1, 1, 0, 0],#
+        '4': [0, 0, 1, 0, 0, 1, 0, 0, 0],#
+        '5': [1, 1, 0, 0, 0, 1, 1, 0, 0],#
+        '6': [1, 0, 1, 1, 1, 1, 1, 1, 0],#
+        '7': [1, 0, 1, 0, 0, 0, 0, 0, 0],#
+        '8': [1, 1, 1, 1, 1, 1, 1, 0, 0],#
+        '9': [1, 1, 1, 0, 1, 0, 1, 0, 0],#
+    }
+
+    def __init__(self, b64_string: str):
+        """
+        Initializes the OCR with a Base64 encoded BMP string.
+        """
+        try:
+            self.image_data = base64.b64decode(b64_string)
+        except (ValueError, TypeError):
+            raise ValueError("Invalid Base64 string provided.")
+
+        # Basic validation of the BMP header
+        if len(self.image_data) < self.PIXEL_DATA_OFFSET or self.image_data[0:2] != b'BM':
+            raise ValueError("Data is not a valid BMP.")
+
+        width = struct.unpack_from('<i', self.image_data, 18)[0]
+        height = struct.unpack_from('<i', self.image_data, 22)[0]
+
+        if width != self.IMG_WIDTH or height != self.IMG_HEIGHT:
+            raise ValueError(f"Expected image dimensions {self.IMG_WIDTH}x{self.IMG_HEIGHT}, "
+                             f"but got {width}x{height}.")
+
+        # Calculate row size with padding (stride)
+        self.row_stride = (self.IMG_WIDTH * 3 + 3) & ~3
+
+    def _get_pixel(self, x: int, y: int) -> Tuple[int, int, int]:
+        """
+        Gets the (B, G, R) color tuple for a pixel at (x, y).
+        Handles the bottom-up row order of BMP files.
+        """
+        # BMP rows are stored bottom-up
+        bmp_y = self.IMG_HEIGHT - 1 - y
+        offset = self.PIXEL_DATA_OFFSET + (bmp_y * self.row_stride) + (x * 3)
+        b, g, r = self.image_data[offset:offset + 3]
+        return b, g, r
+
+    def _is_foreground(self, x: int, y: int, threshold: int = 70) -> bool:
+        """
+        Determines if a pixel is part of the foreground (a digit).
+        The digits are dark gray/black, background is light gray.
+        """
+        b, g, r = self._get_pixel(x, y)
+        # Use average brightness to determine if it's foreground
+        return (r + g + b) / 3 < threshold
+
+    def _match_digit(self, feature_vector: List[int]) -> str:
+        """
+        Finds the best matching digit for a given feature vector.
+        It calculates the Hamming distance between the input vector and each template.
+        """
+        min_diff = float('inf')
+        found_digit = '?'
+
+        for digit_char, template_vector in self.DIGIT_TEMPLATES.items():
+            # Calculate Hamming distance (number of differing bits)
+            diff = sum(v1 != v2 for v1, v2 in zip(feature_vector, template_vector))
+
+            if diff < min_diff:
+                min_diff = diff
+                found_digit = digit_char
+
+            # Perfect match, no need to check further
+            if min_diff == 0:
+                break
+
+        return found_digit
+
+    def recognize(self) -> str:
+        """
+        Recognizes all 5 digits in the image and returns them as a string.
+        """
+        result = []
+        one_offset = 0
+        for i in range(self.NUM_CHARS):
+            char_x_offset = i * self.CHAR_WIDTH
+
+            # Generate the feature vector for the current character
+            feature_vector = [
+                1 if self._is_foreground(char_x_offset + px - one_offset, py) else 0
+                for px, py in self.SAMPLE_POINTS
+            ]
+            print(feature_vector)
+
+            # Find the best match for the vector
+            recognized_char = self._match_digit(feature_vector)
+            if recognized_char == '1':
+                one_offset += 1
+            elif recognized_char == '4':
+                one_offset -= 1
+
+            result.append(recognized_char)
+
+        return "".join(result)
+
+
+def main():
+    """
+    Main function to run the OCR process.
+    """
+
+    # parse file path from first argument
+    if len(sys.argv) < 2:
+        b64_file_path = 'base64.txt'
+    else:
+        b64_file_path = sys.argv[1]
+
+    try:
+        with open(b64_file_path, 'r') as f:
+            b64_data = f.read().strip()
+    except FileNotFoundError:
+        print(f"Error: The file '{b64_file_path}' was not found.", file=sys.stderr)
+        sys.exit(1)
+
+    if not b64_data:
+        print(f"Error: The file '{b64_file_path}' is empty.", file=sys.stderr)
+        sys.exit(1)
+
+    try:
+        ocr = BmpOcr(b64_data)
+        result = ocr.recognize()
+        print(f"Input {b64_file_path}")
+        print(f"Recognized digits: {result}")
+    except ValueError as e:
+        print(f"An error occurred: {e}", file=sys.stderr)
+        sys.exit(1)
+
+
+if __name__ == "__main__":
+    main()

script.subtitles.zimukux/resources/lib/sub_provider_service.py

@@ -229,11 +229,10 @@ def run():
                  else __addon__.getSetting("proxy_server"))
         os.environ["HTTP_PROXY"] = os.environ["HTTPS_PROXY"] = proxy
 
-    ocrUrl= __addon__.getSetting("ocr_url")
 
     # 查询
     agent = zmkagnt.Zimuku_Agent(zimuku_base_url, __temp__, logger, Unpacker(),
-                                 {'subtype': tpe, 'sublang': lang}, ocrUrl)
+                                 {'subtype': tpe, 'sublang': lang})
 
     handle_params(params)
     xbmcplugin.endOfDirectory(int(sys.argv[1]))