Update app.py

app.py

@@ -91,8 +91,28 @@ FILE_EMOJIS = {
     "mp3": "🎵",
 }
 
-# 5. High-Information Content Extraction
-def get_high_info_terms(text: str) -> list:
+def get_high_info_terms(text: str, prioritize_start=True) -> list:
+    """🧠 #1 - The Neural Network for Filenames (but way simpler and probably underpaid)
+
+    Scans text like a caffeinated librarian on a mission, hunting for words that actually
+    mean something. Filters out boring words like 'the' and 'and' (sorry old friends),
+    while preserving the good stuff like 'quantum' and 'neural' (party time! 🎉).
+
+    Think of it as a bouncer for your filenames - if a word isn't cool enough,
+    it's not getting in. But key phrases? VIP access, baby! 🎭
+
+    Args:
+        text (str): The text to strip mine for linguistic gold
+        prioritize_start (bool): If True, treats the start like the cool kids' table
+                               (default behavior because we're not monsters)
+
+    Returns:
+        list: The VIP list of words that made the cut. Maximum of 8 terms if we're
+              prioritizing the start (because YOLO), 5 otherwise (because sanity).
+              
+    Warning: May occasionally let through a word that sounds smart but is actually
+            just showing off. We're working on its ego. 🎭
+    """
     stop_words = set([
         'the', 'a', 'an', 'and', 'or', 'but', 'in', 'on', 'at', 'to', 'for', 'of', 'with',
         'by', 'from', 'up', 'about', 'into', 'over', 'after', 'is', 'are', 'was', 'were',
@@ -100,7 +120,9 @@ def get_high_info_terms(text: str) -> list:
         'should', 'could', 'might', 'must', 'shall', 'can', 'may', 'this', 'that', 'these',
         'those', 'i', 'you', 'he', 'she', 'it', 'we', 'they', 'what', 'which', 'who',
         'when', 'where', 'why', 'how', 'all', 'any', 'both', 'each', 'few', 'more', 'most',
-        'other', 'some', 'such', 'than', 'too', 'very', 'just', 'there'
+        'other', 'some', 'such', 'than', 'too', 'very', 'just', 'there', 'please', 'tell',
+        'explain', 'show', 'give', 'write', 'provide', 'need', 'want', 'would', 'could',
+        'lets', 'let', 'try', 'use', 'make', 'help'
     ])
 
     key_phrases = [
@@ -113,23 +135,48 @@ def get_high_info_terms(text: str) -> list:
         'research paper', 'scientific study', 'empirical analysis'
     ]
 
-    preserved_phrases = []
+    # First check for key phrases at the start of the text
+    start_phrases = []
     lower_text = text.lower()
+    text_start = lower_text[:100]  # Look at first 100 chars for starting phrases
+    for phrase in key_phrases:
+        if text_start.startswith(phrase) or text_start.find(f" {phrase}") >= 0:
+            start_phrases.append(phrase)
+            text = text.replace(phrase, '')
+
+    # Then check for key phrases in the rest of the text
+    preserved_phrases = []
     for phrase in key_phrases:
-        if phrase in lower_text:
+        if phrase in lower_text and phrase not in start_phrases:
             preserved_phrases.append(phrase)
             text = text.replace(phrase, '')
 
-    words = re.findall(r'\b\w+(?:-\w+)*\b', text)
+    # Get the first ~50 words to analyze the start more carefully
+    start_words = text.split()[:50]
+    words_with_pos = []
+    for pos, word in enumerate(start_words):
+        word = re.sub(r'[^\w\s-]', '', word.lower())
+        if (len(word) > 3 and 
+            word not in stop_words and 
+            not word.isdigit() and 
+            any(c.isalpha() for c in word)):
+            words_with_pos.append((pos, word))
+
+    # Get remaining high-info words from the rest of the text
+    remaining_words = re.findall(r'\b\w+(?:-\w+)*\b', text[100:])
     high_info_words = [
-        word.lower() for word in words 
+        word.lower() for word in remaining_words 
         if len(word) > 3
         and word.lower() not in stop_words
         and not word.isdigit()
         and any(c.isalpha() for c in word)
     ]
 
-    all_terms = preserved_phrases + high_info_words
+    # Combine terms prioritizing start content
+    start_terms = [word for _, word in sorted(words_with_pos, key=lambda x: x[0])][:3]
+    all_terms = (start_phrases + start_terms + preserved_phrases + high_info_words)
+
+    # Remove duplicates while preserving order
     seen = set()
     unique_terms = []
     for term in all_terms:
@@ -137,22 +184,50 @@ def get_high_info_terms(text: str) -> list:
             seen.add(term)
             unique_terms.append(term)
 
-    max_terms = 5
+    max_terms = 8 if prioritize_start else 5
     return unique_terms[:max_terms]
 
-# 6. Filename Generation
 def generate_filename(content, file_type="md"):
+    """🎯 #2 - The File Naming Sommelier (pairs well with frustrated developers)
+
+    Takes your content and turns it into a filename that's actually readable by humans!
+    A revolutionary concept, we know. Combines timestamps with meaningful words,
+    because '20231218_quantum_research' beats 'asdfg123.md' any day of the week.
+
+    Think of it as your personal file naming barista - takes your raw content beans
+    and turns them into a smooth, well-crafted filename. No foam art though, sorry! ☕
+
+    Args:
+        content (str): Your beautiful text that needs a home(name)
+        file_type (str): The file extension (defaults to "md" because we're markdown 
+                        hipsters at heart)
+
+    Returns:
+        str: A filename that won't make you question your life choices when you see it 
+            in 6 months. Limited to 120 chars because we're not writing a novel here.
+            
+    Pro Tip: If your filename ends up being just 'file.md', either your content was 
+            empty or we've failed spectacularly. Please file a bug report or just 
+            laugh it off. 🎪
+    """
     prefix = datetime.now().strftime("%y%m_%H%M") + "_"
-    info_terms = get_high_info_terms(content)
+    
+    # Get high-info terms with start prioritization
+    info_terms = get_high_info_terms(content, prioritize_start=True)
+    
+    # Create filename with terms
     name_text = '_'.join(term.replace(' ', '-') for term in info_terms) if info_terms else 'file'
     
-    max_length = 100
+    # Ensure reasonable length
+    max_length = 120  # Increased to allow more meaningful content
     if len(name_text) > max_length:
         name_text = name_text[:max_length]
     
     filename = f"{prefix}{name_text}.{file_type}"
     return filename
 
+    
+
 # 7. Audio Processing
 def clean_for_speech(text: str) -> str:
     text = text.replace("\n", " ")