small fixes

Readme.org

@@ -1,7 +1,7 @@
 #+title:  Script
 #+author: Sébastien Miquel
 #+date:   14-03-2026
-# Time-stamp: <08-08-26 13:18>
+# Time-stamp: <08-08-26 16:44>
 #+OPTIONS:
 
 * Méta
@@ -149,6 +149,8 @@ Mettre les copies scannées au format pdf dans =Interro=.
     Découpe la partie gauche des copies, là où il devrait y avoir les
     labels des exercices/questions.
 
+    =python cutleft.py Interro --fullpage= to use the full page always.
+
     Rerun on a single file with =python cutleft.py Interro/Copies/Copie01.pdf=
 
 ** Labelisation et regroupement
@@ -174,11 +176,11 @@ Set proxy with ~export HTTPS_PROXY="http://10.0.0.1:3128"~
     =python plotting.py Interro/Copies/Copie01.pdf=
 
     It also generates les =Copie01.json=, à partir des =Copie01_01.json=
-    1. En cas de soucis, (par exemple les pages ne sont pas dans le bon ordre)
-       - Réordonner les pages du fichier pdf
-       - Rerun =python cutleft.py Interro/Copie{id}=
-       - Rerun =python gemini_dir_batching.py Interro/Copie{id}= ?? À
-         vérifier, pas sûr que ça marche.
+    En cas de soucis, (par exemple les pages ne sont pas dans le bon ordre)
+     - Réordonner les pages du fichier pdf
+     - Rerun =python cutleft.py Interro/Copie{id}=
+     - Rerun =python gemini_dir_batching.py Interro/Copie{id}=
+       ?? À vérifier, pas sûr que ça marche.
  3. =python splitting_int.py Interro=
 
     Découpe les copies suivant les exercices

annotating_by_label.py

@@ -1,6 +1,7 @@
 import sys
 import os
 import json
+import utils
 import shutil
 import argparse
 import concurrent.futures
@@ -124,6 +125,7 @@ def main():
         with open(label_groups, 'w') as f:
             for items in groups.values():
                 f.write(",".join(items) + "\n")
+        utils.edit_file_and_enter(label_groups)
 
     with open(label_groups, "r") as f:
         lines = [line.strip() for line in f if line.strip()]

cutleft.py

@@ -3,6 +3,7 @@ from functools import lru_cache
 import os
 import time
 import json  # Added for schema output
+import argparse
 import tkinter as tk
 from threading import Thread
 from queue import Queue, Empty
@@ -14,10 +15,13 @@ DELIMITER_WIDTH = 5
 DELIMITER_COLOR = (0, 0, 0)
 OUTPUT_SIZE = (1800, 1000)
 
-if len(sys.argv) < 2:
-    sys.exit("Usage: python script.py <directory_path_or_file_path>")
+parser = argparse.ArgumentParser(description="PDF Cropper")
+parser.add_argument("path", help="Directory path or PDF file path")
+parser.add_argument("--fullpage", action="store_true", help="Process all files in full page mode (1 page per output file)")
+args = parser.parse_args()
 
-path_arg = sys.argv[1]
+path_arg = args.path
+fullpage_mode = args.fullpage
 files = []
 INPUT_DIR = ""
 COPIES_DIR = ""
@@ -121,7 +125,6 @@ def process_single_pdf(filename, shift_offset=0, max_per_file=5):
     Converts PDF to stitched images.
     Returns a tuple: (preview_image_resized, list_of_split_images, schema_dict)
     """
-    # pdf_path = os.path.join(INPUT_DIR, filename)
     try:
         pages = get_pdf_pages(filename)
         cropped_images = []
@@ -164,7 +167,6 @@ def process_single_pdf(filename, shift_offset=0, max_per_file=5):
 
         # 3. Generate Preview (All stitched together, Resized)
         full_stitch = stitch_images(cropped_images)
-        # preview_resized = full_stitch.resize(OUTPUT_SIZE, Image.LANCZOS)
         preview_resized = full_stitch.resize(OUTPUT_SIZE, Image.BILINEAR)
 
         schema = {
@@ -190,21 +192,15 @@ def save_results(result_tuple, filename):
     # --- Cleanup: Delete existing files for this PDF ---
     for f in os.listdir(OUTPUT_DIR):
         file_path = os.path.join(OUTPUT_DIR, f)
-        # 1. Delete schema file
         if f == f"{base_name}_schema.json":
             os.remove(file_path)
-        # 2. Delete image files (pattern: basename_01.jpg, etc.)
         elif f.startswith(f"{base_name}_") and f.endswith(".jpg"):
-            # Check if the suffix is strictly numeric (e.g. "01") to avoid
-            # deleting unrelated files like "file_v2_01.jpg" when processing "file.pdf"
             suffix = f[len(base_name)+1:-4]
             if suffix.isdigit():
                 os.remove(file_path)
-    # ---------------------------------------------------
 
     # Save Images
     for i, img in enumerate(splits):
-        # Suffix _01, _02, etc.
         suffix = f"_{i+1:02d}"
         output_filename = f"{base_name}{suffix}.jpg"
         output_path = os.path.join(OUTPUT_DIR, output_filename)
@@ -217,14 +213,17 @@ def save_results(result_tuple, filename):
     with open(json_path, 'w') as f:
         json.dump(schema, f, indent=4)
     print(f"Saved schema: {json_filename}")
+
+
 # --- GUI Application ---
 
 class ImageReviewer:
-    def __init__(self, file_list):
+    def __init__(self, file_list, default_max_per_file=5):
         self.files = file_list
         self.index = 0
         self.current_shift = 0
-        self.current_max_per_file = 5
+        self.default_max_per_file = default_max_per_file
+        self.current_max_per_file = default_max_per_file
         self.current_preview = None # Only stores the resized preview for GUI
         self.is_processing = False
 
@@ -247,8 +246,7 @@ class ImageReviewer:
         self.root.bind('n', lambda e: self.on_shift(50))
         self.root.bind('N', lambda e: self.on_shift(100))
         self.root.bind('t', lambda e: self.on_shift(-50))
-        self.root.bind('1', lambda e: self.on_set_max_pages(1)) # New Binding
-
+        self.root.bind('1', lambda e: self.on_set_max_pages(1))
 
         # Start background pre-fetcher
         self.bg_thread = Thread(target=self.prefetch_worker, daemon=True)
@@ -266,7 +264,6 @@ class ImageReviewer:
             return
         self.current_max_per_file = count
         print(f"Setting max pages per file: {count}")
-        # Trigger reprocessing with current settings
         self.trigger_processing(self.files[self.index], self.current_shift)
 
     def prefetch_worker(self):
@@ -276,7 +273,7 @@ class ImageReviewer:
             target = self.index + 1
             if target < len(self.files) and target != idx_to_process:
                 fname = self.files[target]
-                get_pdf_pages(fname)  # Just calling it warms the lru_cache
+                get_pdf_pages(fname)
                 idx_to_process = target
             time.sleep(0.05)
 
@@ -290,7 +287,6 @@ class ImageReviewer:
         self.is_processing = False
         self.current_shift = 0
 
-        # Always trigger processing. If prefetched, get_pdf_pages returns instantly.
         self.trigger_processing(filename, self.current_shift)
 
     def trigger_processing(self, filename, shift):
@@ -317,7 +313,6 @@ class ImageReviewer:
                 self.load_current_image(use_prefetch=True)
             self.is_processing = False
         except Empty:
-            # Check again in 100ms
             self.root.after(100, lambda: self.check_manual_queue(filename))
 
     def handle_processing_result(self, result, filename):
@@ -325,7 +320,6 @@ class ImageReviewer:
         preview, splits, schema = result
         self.current_preview = preview
 
-        # Save in a background thread so the GUI updates instantly
         Thread(target=save_results, args=(result, filename), daemon=True).start()
 
         self.update_display(filename, schema)
@@ -350,7 +344,7 @@ class ImageReviewer:
 
     def on_shift(self, amount):
         if self.is_processing:
-            return # Ignore keys while processing
+            return
         self.current_shift += amount
         print(f"Applying shift: {self.current_shift}")
         self.trigger_processing(self.files[self.index], self.current_shift)
@@ -360,12 +354,13 @@ class ImageReviewer:
             return
         self.index += 1
         self.current_shift = 0
-        self.current_max_per_file = 5 # Reset to default
+        self.current_max_per_file = self.default_max_per_file
         self.load_current_image(use_prefetch=True)
 
+
 # --- Entry Point ---
 if __name__ == "__main__":
     if not files:
         print("No PDF files found.")
     else:
-        app = ImageReviewer(files)
+        app = ImageReviewer(files, default_max_per_file=1 if fullpage_mode else 5)

gemini_for_enonce.py

@@ -1,6 +1,6 @@
-import shlex
 import re
 import os
+import utils
 import subprocess
 import sys
 import argparse
@@ -119,7 +119,7 @@ PROMPT_3 = """I am providing:
 2. The source code of the exam questions (`enonce` file).
 
 Your task:
- Extract important information necessary to understand the questions (e.g., definitions of objects, global notations, hypotheses, context) that are NOT part of the question texts themselves. Often, this information can be in a previous \\item that is not itself a question, but contains the question items. 
+ Extract important information necessary to understand the questions (e.g., definitions of objects, global notations, hypotheses, context) that are NOT part of the question texts themselves. Often, this information can be in a previous \\item that is not itself a question, but contains the question items.
 
 For example, given LaTeX code like
 
@@ -148,7 +148,9 @@ def process_exam(folder_path: str, restart: bool = False):
     folder = Path(folder_path)
 
     cache_dir = folder / "Cache"
+    tmp_dir = folder / "Tmp"
     cache_dir.mkdir(exist_ok=True)
+    tmp_dir.mkdir(exist_ok=True)
 
     cache_q_file = cache_dir / "gemini_questions.json"
     cache_s_file = cache_dir / "gemini_solutions.json"
@@ -341,8 +343,8 @@ def process_exam(folder_path: str, restart: bool = False):
     # ==========================================
     # INITIAL GROUPING COMPUTATION
     # ==========================================
-    items_file = folder / "exam_items.txt"
-    full_items_file = folder / "exam_items_full.txt"
+    items_file = tmp_dir / "exam_items.txt"
+    full_items_file = tmp_dir / "exam_items_full.txt"
     trunc_map = {}
 
     # --- INITIAL GROUPING COMPUTATION ---
@@ -469,20 +471,7 @@ def process_exam(folder_path: str, restart: bool = False):
     # --- OPEN EDITOR AND PARSE ---
     while True:
         print("Opening items file for editing...")
-        editor = os.environ.get("EDITOR")
-        try:
-            if editor:
-                subprocess.run(shlex.split(editor) + [str(items_file)])
-            else:
-                if sys.platform.startswith("linux"):
-                    subprocess.run(["xdg-open", str(items_file)])
-                elif sys.platform == "darwin":
-                    subprocess.run(["open", str(items_file)])
-                else:
-                    os.startfile(str(items_file))
-            input("Press ENTER here once you have saved and closed the text file...")
-        except Exception as e:
-            print(f"Error running editor: {e}")
+        utils.edit_file_and_enter(items_file)
 
         print(f"Parsing edited items from {items_file.name}...")
         with open(items_file, "r", encoding="utf-8") as f:

gemini_for_labels.py

@@ -180,6 +180,7 @@ def generate_request(file, labels, names, context_labels, wrong_labels):
 parser = argparse.ArgumentParser(description="Process a directory or specific files using Gemini.")
 parser.add_argument("input_paths", nargs='+', help="The input directory or specific files")
 parser.add_argument("--overwrite", action="store_true", help="Regenerate output even if it exists")
+
 args = parser.parse_args()
 
 # input_arg = Path(args.input_path)

utils.py

@@ -23,6 +23,24 @@ def enonce_total(base_dir):
     return "".join(output)
 
 import os
+import shlex
+
+def edit_file_and_enter(file):
+    editor = os.environ.get("EDITOR")
+    try:
+        if editor:
+            subprocess.run(shlex.split(editor) + [str(file)])
+        else:
+            if sys.platform.startswith("linux"):
+                subprocess.run(["xdg-open", str(file)])
+            elif sys.platform == "darwin":
+                subprocess.run(["open", str(file)])
+            else:
+                os.startfile(str(file))
+        input("Press ENTER here once you have saved and closed the text file...")
+    except Exception as e:
+        print(f"Error running editor: {e}")
+
 
 def pdf_image_of_enonce(root_dir, label):
     pdf_path = os.path.join(root_dir, "Text2", f"{label}.pdf")