ocrgrep 0.1.0__py3-none-any.whl

ocrgrep-0.1.0.dist-info/METADATA

@@ -0,0 +1,89 @@
+Metadata-Version: 2.4
+Name: ocrgrep
+Version: 0.1.0
+Summary: grep-like OCR tool for images and videos.
+Author-email: tmctmt <tmctmt@proton.me>
+License: MIT License
+        
+        Copyright (c) 2026 tmctmt
+        
+        Permission is hereby granted, free of charge, to any person obtaining a copy
+        of this software and associated documentation files (the "Software"), to deal
+        in the Software without restriction, including without limitation the rights
+        to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+        copies of the Software, and to permit persons to whom the Software is
+        furnished to do so, subject to the following conditions:
+        
+        The above copyright notice and this permission notice shall be included in all
+        copies or substantial portions of the Software.
+        
+        THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+        IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+        FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+        AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+        LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+        OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+        SOFTWARE.
+        
+Keywords: ocr,grep,screen ai,chrome,locro
+Classifier: Development Status :: 4 - Beta
+Classifier: Environment :: Console
+Classifier: Intended Audience :: End Users/Desktop
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Natural Language :: English
+Classifier: Operating System :: OS Independent
+Classifier: Programming Language :: Python
+Classifier: Topic :: Utilities
+Requires-Python: >=3.8
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: tqdm~=4.0
+Requires-Dist: opencv-python~=4.0
+Requires-Dist: filetype~=1.0
+Requires-Dist: Pillow~=12.0
+Dynamic: license-file
+
+## ocrgrep
+A grep-like OCR tool for image and video files, utilizing the fast and accurate [Google Screen AI](https://chromium.googlesource.com/chromium/src/+/refs/tags/126.0.6452.4/services/screen_ai/README.md) engine via [locro](https://github.com/sergiocorreia/clv-locro/tree/master).
+
+```
+$ ocrgrep -ih 'grep-like' screenshot.png    
+rep README.md in main Preview ocrgrep A grep-like OCR tool for image and video files, uti
+```
+
+```
+usage: ocrgrep.py [-i] [-m NUM] [-p] [-r] [-w NUM] [-C NUM] [-F] [-h] [-t] [--include GLOB]
+                  [--exclude GLOB] [--no-image] [--no-video] [--video-max-msec NUM]
+                  [--video-step-msec NUM] [--help]
+                  pattern files [files ...]
+
+positional arguments:
+  pattern
+  files
+
+options:
+  -i, --ignore-case     ignore case distinctions in patterns and data
+  -m, --max-count NUM   stop after NUM selected lines
+  -p, --progress        show progress bar
+  -r, --recursive       scan subfiles in directories
+  -w, --workers NUM     concurrency (default: 16)
+  -C, --context NUM     print NUM characters of output context (default: 40)
+  -F, --fixed-strings   PATTERN is a string
+  -h, --no-filename     suppress the file name prefix on output
+  -t, --no-timestamp    suppress the timestamp prefix on output for videos
+  --include GLOB        search only files that match GLOB (a file pattern)
+  --exclude GLOB        skip files that match GLOB
+  --no-image            ignore image files
+  --no-video            ignore video files
+  --video-max-msec NUM  stop after NUM milliseconds of video
+  --video-step-msec NUM
+                        scan a frame for every NUM milliseconds of video (default: 1000)
+  --help                show this help message and exit
+```
+  
+# Install
+```
+pip install git+https://github.com/sergiocorreia/clv-locro.git
+locro download
+pip install ocrgrep
+```

ocrgrep-0.1.0.dist-info/RECORD

@@ -0,0 +1,7 @@
+ocrgrep.py,sha256=4NglUohr1KhJNKFg49jqNBvIvnY3OPbLbbIM4JyAj9I,6294
+ocrgrep-0.1.0.dist-info/licenses/LICENSE,sha256=66VAXGOhqfQi17P07j1_OpVRj31u2nfq6DM1vG1dVJ8,1063
+ocrgrep-0.1.0.dist-info/METADATA,sha256=dfpvqo9XBfMIu9AdqUnGUOpntCIGkuLe3zpD9qT6Vl8,3843
+ocrgrep-0.1.0.dist-info/WHEEL,sha256=aeYiig01lYGDzBgS8HxWXOg3uV61G9ijOsup-k9o1sk,91
+ocrgrep-0.1.0.dist-info/entry_points.txt,sha256=msBseBUajJu-jI2TuEPBFxDQs97HJ-aS83wUkVWTvC4,40
+ocrgrep-0.1.0.dist-info/top_level.txt,sha256=962MaTHe0Gfz3PRDYfIokmrhKbBLwy3WQAYt3SZyyBA,8
+ocrgrep-0.1.0.dist-info/RECORD,,

ocrgrep-0.1.0.dist-info/WHEEL

@@ -0,0 +1,5 @@
+Wheel-Version: 1.0
+Generator: setuptools (82.0.1)
+Root-Is-Purelib: true
+Tag: py3-none-any
+

ocrgrep-0.1.0.dist-info/entry_points.txt

@@ -0,0 +1,2 @@
+[console_scripts]
+ocrgrep = ocrgrep:cli

ocrgrep-0.1.0.dist-info/licenses/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 tmctmt
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

ocrgrep-0.1.0.dist-info/top_level.txt

@@ -0,0 +1 @@
+ocrgrep

ocrgrep.py

@@ -0,0 +1,164 @@
+from dataclasses import dataclass
+from datetime import timedelta
+from fnmatch import fnmatch
+from functools import partial
+from multiprocessing import Pool, cpu_count
+from pathlib import Path
+import argparse
+import re
+import sys
+
+from locro import ScreenAI
+from tqdm import tqdm
+import cv2
+import filetype
+import PIL
+
+RED = '\033[31m'
+YELLOW = '\033[33m'
+MAGENTA = '\033[35m'
+RESET = '\033[39m'
+
+@dataclass(frozen=True)
+class Result:
+    path: Path
+    text: str
+
+@dataclass(frozen=True)
+class VideoResult(Result):
+    msec: float
+
+def ocr(path: Path, args: argparse.Namespace):
+    if 'engine' not in globals():
+        global engine
+        engine = ScreenAI()
+
+    results: list[Result] = []
+
+    if filetype.is_image(path) and not args.no_image:
+        image = PIL.Image.open(path)
+        text = engine.ocr_pil_image(image).text
+        results.append(Result(path, text))
+    
+    if filetype.is_video(path) and not args.no_video:
+        cap = cv2.VideoCapture(path)
+        prev_msec = None
+
+        if cap.get(cv2.CAP_PROP_FRAME_COUNT) == -1:
+            return results
+
+        while True:
+            msec = cap.get(cv2.CAP_PROP_POS_MSEC)
+
+            if args.video_max_msec and msec > args.video_max_msec:
+                break
+
+            if prev_msec and (msec - prev_msec) < args.video_step_msec:
+                success = cap.grab()
+                if success:
+                    continue
+                break
+
+            prev_msec = msec
+
+            success, image = cap.read()
+            if not success:
+                break
+
+            image = PIL.Image.fromarray(image)
+            text = engine.ocr_pil_image(image).text
+            results.append(VideoResult(path, text, msec))
+
+        cap.release()
+            
+    return results
+
+def cli():
+    p = argparse.ArgumentParser(
+        description='grep-like OCR tool for images and videos.',
+        epilog='example: %(prog)s -i "hello world" video.mp4 screenshot.png',
+        formatter_class=argparse.RawDescriptionHelpFormatter,
+        add_help=False
+    )
+    p.add_argument('pattern')
+    p.add_argument('files', nargs='+')
+    p.add_argument('-i', '--ignore-case', action='store_true',
+                   help='ignore case distinctions in patterns and data')
+    p.add_argument('-m', '--max-count', type=int, metavar='NUM',
+                   help='stop after NUM selected lines')
+    p.add_argument('-p', '--progress', action='store_true',
+                   help='show progress bar')
+    p.add_argument('-r', '--recursive', action='store_true',
+                   help='scan subfiles in directories')
+    p.add_argument('-w', '--workers', type=int, default=int(cpu_count()//2), metavar='NUM',
+                   help='concurrency (default: %(default)s)')
+    p.add_argument('-C', '--context', type=int, default=40, metavar='NUM',
+                   help='print NUM characters of output context (default: %(default)s)')
+    p.add_argument('-F', '--fixed-strings', action='store_true',
+                   help='PATTERN is a string')
+    p.add_argument('-h', '--no-filename', action='store_true',
+                   help='suppress the file name prefix on output')
+    p.add_argument('-t', '--no-timestamp', action='store_true',
+                   help='suppress the timestamp prefix on output for videos')
+    p.add_argument('--include', action='append', default=[], metavar='GLOB',
+                   help='search only files that match GLOB (a file pattern)')
+    p.add_argument('--exclude', action='append', default=[], metavar='GLOB',
+                   help='skip files that match GLOB')
+    p.add_argument('--no-image', action='store_true',
+                   help='ignore image files')
+    p.add_argument('--no-video', action='store_true',
+                   help='ignore video files')
+    p.add_argument('--video-max-msec', type=int, metavar='NUM',
+                   help='stop after NUM milliseconds of video')
+    p.add_argument('--video-step-msec', type=int, default=1000, metavar='NUM',
+                   help='scan a frame for every NUM milliseconds of video (default: %(default)s)')
+    p.add_argument('--help', action='help',
+                   help='show this help message and exit')
+    
+    args = p.parse_args()
+    flags = re.IGNORECASE if args.ignore_case else 0
+
+    def should_include(path: Path):
+        if args.exclude and any(fnmatch(path, pat) for pat in args.exclude):
+            return False
+        if args.include and not any(fnmatch(path, pat) for pat in args.include):
+            return False
+        return True
+
+    files = []
+    for path in map(Path, args.files):
+        if path.is_file() and should_include(path):
+            files.append(path)
+        elif path.is_dir() and args.recursive:
+            files.extend(s for s in path.rglob('*') if s.is_file() and should_include(s))
+        elif path.is_dir():
+            print(f'ocrgrep: {path}: Is a directory', file=sys.stderr)
+        else:
+            print(f'ocrgrep: {path}: No such file or directory', file=sys.stderr)
+
+    with Pool(args.workers) as pool, tqdm(total=len(files), disable=not args.progress) as pbar:
+        for results in pool.imap_unordered(partial(ocr, args=args), files):
+            count = 0
+            for result in results:
+                text = re.sub(r'\s+', r' ', result.text)
+                match = re.search(args.pattern, text, flags=flags)
+                if match:
+                    line = ''
+                    if not args.no_filename:
+                        line += MAGENTA + str(result.path) + RESET + ':'
+                    if not args.no_timestamp and isinstance(result, VideoResult):
+                        line += YELLOW + str(timedelta(milliseconds=result.msec))[:-3] + RESET + ':'
+                    line += re.sub(
+                        re.escape(args.pattern) if args.fixed_strings else args.pattern,
+                        lambda m: RED + m.group(0) + RESET,
+                        text[max(0, match.start() - args.context) : match.end() + args.context].strip(),
+                        flags=flags
+                    )
+                    pbar.write(line)
+                    count += 1
+                    if args.max_count and count == args.max_count:
+                        break
+            pbar.update()
+
+if __name__ == '__main__':
+    cli()