Simpler install (#16)

* Cleanup tesseract files, no more need to install tess

* Update readme+add ico

* correction creation

* icon tweaking

README.md

@@ -14,20 +14,13 @@ It's called Utsushi's charm because I thought it would be funny to make a comple
 
 ## Requirements
 - A computer (Windows)
-  - Linux and Mac might work too, you wont be able to run the EXE and will have to run from source in a terminal window.
+  - Linux and Mac might work too, you wont be able to run the EXE and will have to run from source in a terminal window. Refer to [Running from source](#Running-from-source)
 - A USB cable to connect your switch to transfer files
 - This latest version of this downloaded to your computer (Utsushis-Charm_**vx_x**.zip)
   - You can find it [here](https://github.com/chpoit/utsushis-charm/releases/latest)
-- **Google Tesseract** installed and in path
-  - A copy of the version 4 is bundled with the release. Just run it, no extra packages needed
-    - Built by UB-Mannheim [License (Apache 2.0)](https://github.com/tesseract-ocr/tesseract/blob/master/LICENSE)
-    - Alternatively, download the same version here: [Installer here](https://digi.bib.uni-mannheim.de/tesseract/tesseract-ocr-w64-setup-v4.1.0.20190314.exe)
-    - Other Versions available on the [UB-Mannheim Github](https://github.com/UB-Mannheim/tesseract/wiki) page
 - Some knowledge of how to type things in the terminal
-  - AKA: Knowing how to type in a hacker box
 - Being able to read
 
-
 ## Steps
 
 0. Unequip all jewels. You will create "fake" charms otherwise.
@@ -156,19 +149,41 @@ In all seriousness, the work is done in a few broad steps:
 
 Sometimes windows will lock some files for a while and there is nothing you can do about it other than wait.
 
-# Running from source on a mac
+# Running from source
+Common requirements:
+- Python3 installed and in path
+- Set up a virtual environment (optional)
+- Install pip packages (in virtualenv if you use it.)
+
+Normal instructions apply once the application starts.
 
+## MacOS
 - Requirements:
   - have python3 and tesseract installed via brew (or some other way)
-  - setup a virtual environment
-    - `pip3 install virtualenv` to install virtualenv
-    - create a virtual env at the root of the repository `virtualenv -p python3 env`
+  - Virtual env on mac (optional):  `virtualenv -p python3 env`
 - Running:
   - switch to the virtualenv `source env/bin/activate` (run at the root of the repository)
   - set TESSDATA_PREFIX: `export TESSDATA_PREFIX=/usr/local/Cellar/tesseract/<version>/share/tessdata`
   - install the project dependencies: `pip3 install .`
   - run with `python3 main.py`
 
+## Linux
+- Requirements
+  - You will need to install Google tesseract with your package manager of choice.
+- Running
+  - `python3 main.py`
+  - 
+
+## Windows
+- Requirements
+  - **Google Tesseract** installed and in path
+    - A copy of the version 4 is bundled with the release. Just run it, no extra packages needed
+      - You can download the installer here version here: [Installer here](https://digi.bib.uni-mannheim.de/tesseract/tesseract-ocr-w64-setup-v4.1.0.20190314.exe)
+    - Make sure it's in path
+  - Python3 installed and in path
+- Running:
+  - `source env/bin/activate` (if you use a virtual env)
+
 # Extra command line options
 If you run from source, or call the executable from the terminal you can make use of the following flags/arguments to achieve different functionality
 

scripts/build_release.bat

@@ -17,11 +17,4 @@ cmd /c ".\env\scripts\activate & python -m PyInstaller .\utsushis-charm.spec --o
 
 copy %skill_corrections% "dist\%skill_corrections%"
 
-if not exist %tesseract_name% (
-    echo "Tesseract installer missing, downloading..."
-    curl https://digi.bib.uni-mannheim.de/tesseract/tesseract-ocr-w64-setup-v4.1.0.20190314.exe -o %tesseract_name%
-)
-
-copy %tesseract_name% "dist\%tesseract_name%"
-
 7z a -tzip %archive_name% ".\dist\*"

skill_corrections.csv

@@ -901,3 +901,14 @@ slugs",slugger
 SDSIS,spare
 Free:,Free
 Free.,Free
+Aftack,Attack
+Affnity,Affinity
+Siding,Sliding
+Wide—Range,Wide-Range
+Artilery,Artillery
+Ballstics,Ballistics
+Fortity,Fortify
+Eyo,Eye
+Sholls,Shells
+Recoll,Recoil
+Consttution,Constitution

src/charm_extraction.py

@@ -20,6 +20,7 @@
 import json
 import cv2
 import os
+from pathlib import Path
 DEBUG = False
 
 
@@ -32,19 +33,30 @@
 spell.load_dictionary(get_resource_path("skill_dict"), 0, 1)
 
 
-known_corrections = {}
-with open(get_resource_path('skill_corrections'), encoding='utf-8') as scf:
-    for line in scf.readlines():
-        line = line.strip()
-        w, r = line.split(',')
-        known_corrections[w] = r
+def load_corrections(known_corrections=None):
+    known_corrections = known_corrections or {}
+    corrections_path = get_resource_path('skill_corrections')
+    Path(corrections_path).touch()  # if not exists
+    with open(corrections_path, encoding='utf-8') as scf:
+        for line in scf.readlines():
+            line = line.strip()
+            w, r = line.split(',')
+            known_corrections[w] = r
 
+    return known_corrections
 
-all_skills = {}
-with open(get_resource_path('skill_list')) as slf:
-    for line in slf.readlines():
-        skill_name = line.strip()
-        all_skills[skill_name.lower()] = skill_name
+
+def load_all_skills(all_skills=None):
+    all_skills = all_skills or {}
+    with open(get_resource_path('skill_list')) as slf:
+        for line in slf.readlines():
+            skill_name = line.strip()
+            all_skills[skill_name.lower()] = skill_name
+    return all_skills
+
+
+known_corrections = load_corrections()
+all_skills = load_all_skills()
 
 
 def is_skill(skill_dict, skill_name):

src/manual_tesseract_bindings.py → src/tesseract/Tesseract.py

@@ -9,69 +9,11 @@
 import shutil
 import platform
 import logging
+from pathlib import Path
+from .tesseract_utils import *
 logger = logging.getLogger(__name__)
 
-
-class TesseractError(Exception):
-    pass
-
-
-def find_tesseract():
-    # TODO: Make this resilient to "change" (tesseract version), probably not necessary
-    locations = [
-        ctypes.util.find_library("libtesseract-4"),  # win32
-        ctypes.util.find_library("libtesseract302"),  # win32 version 3.2
-        ctypes.util.find_library("tesseract"),  # others
-    ]
-
-    if platform.system() == "Windows":
-        locations += [
-            os.path.join(os.getenv("ProgramW6432"),
-                         "Tesseract-OCR", "libtesseract-4.dll"),
-            os.path.join(os.getenv('LOCALAPPDATA'),
-                         "Tesseract-OCR", "libtesseract-4.dll"),
-            os.path.join(os.getenv("ProgramFiles"),
-                         "Tesseract-OCR", "libtesseract-4.dll"),
-            os.path.join(os.getenv("programfiles(x86)"),
-                         "Tesseract-OCR", "libtesseract-4.dll"),
-        ]
-    elif platform.system() == "Darwin":  # MacOS
-        locations += [
-            # add potential environment paths here:
-            # Example:
-            # os.path.join(os.getenv("MACOS_ENV_NAME"), "Tesseract-OCR", "libtesseract-4.dll"),
-        ]
-    elif platform.system() == "Linux":
-        locations += [
-            # add potential environment paths here:
-            # Example:
-            # os.path.join(os.getenv("LINUX_ENV_NAME"), "Tesseract-OCR", "libtesseract-4.dll"),
-        ]
-
-    for potential in filter(lambda x: x, locations):
-        if os.path.isfile(potential):
-            logger.debug(f"Using tesseract at {potential}")
-            return potential
-
-    raise TesseractError(
-        'Tesseract library was not found on your system. Please install it')
-
-
-def set_tessdata():
-    if 'TESSDATA_PREFIX' in os.environ:
-        return
-    path = find_tesseract()
-    path = os.path.dirname(path)
-    TESSDATA_PREFIX = os.path.join(path, 'tessdata')
-    os.environ['TESSDATA_PREFIX'] = TESSDATA_PREFIX
-    logger.debug(f"Set 'TESSDATA_PREFIX' to {TESSDATA_PREFIX}")
-
-
-def get_datapath():
-    if 'TESSDATA_PREFIX' not in os.environ:
-        set_tessdata()
-    return os.environ['TESSDATA_PREFIX']
-
+from .TesseractError import TesseractError
 
 class Tesseract(object):
     _lib = None
@@ -130,6 +72,8 @@ def __init__(self, language='eng', datapath=None, lib_path=None):
             self.setup_lib(lib_path)
         self._api = self._lib.TessBaseAPICreate()
 
+        download_language_data(language)
+
         # required windows nonsense
         encoded_lang = language.encode("utf-8")
         datapath = get_datapath() if datapath is None else datapath
@@ -182,36 +126,13 @@ def set_variable(self, key, val):
         self._lib.TessBaseAPISetVariable(self._api, key, val)
 
 
-def convert_to_grayscale(image_data):
-    return cv2.cvtColor(image_data, cv2.COLOR_BGR2GRAY)
-
-
-def process_image_with_tesseract(tesseract, image):
-    whitelist = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz1234567890\'/-"
-
-    height, width = image.shape[:2]
-    if len(image.shape) == 2:
-        depth = 1
-    else:
-        depth = image.shape[2]
-
-    # Forcing obnoxious type conversion, probably some windows BS
-    image = image.astype(np.uint8)
-
-    tesseract.set_image(image.ctypes, width, height, depth)
-    tesseract.set_variable("whitelist", whitelist)
-    tesseract.set_resolution()
-    text = tesseract.get_text()
-    return text.strip()
-
-
 if __name__ == '__main__':
     set_tessdata()
     PACKAGE_PARENT = '..'
     SCRIPT_DIR = os.path.dirname(os.path.realpath(
         os.path.join(os.getcwd(), os.path.expanduser(__file__))))
     sys.path.append(os.path.normpath(os.path.join(SCRIPT_DIR, PACKAGE_PARENT)))
-    from src.utils import remove_non_skill_info, apply_trunc_threshold, get_skills, _trim_image_past_skill_name
+    from ..utils import remove_non_skill_info, apply_trunc_threshold, get_skills, _trim_image_past_skill_name
 
     test_img = [
         "frames/frame0.png",

src/tesseract/TesseractError.py

@@ -0,0 +1,3 @@
+
+class TesseractError(Exception):
+    pass