learn_bash_from_session_data 1.0.5 → 1.0.6

package/scripts/quiz_generator.py

@@ -18,6 +18,45 @@ import random
 import re
 import hashlib
 
+try:
+    from scripts.knowledge_base import COMMAND_DB, get_command_info, get_flags_for_command
+except ImportError:
+    try:
+        from knowledge_base import COMMAND_DB, get_command_info, get_flags_for_command
+    except ImportError:
+        COMMAND_DB = {}
+        def get_command_info(name): return None
+        def get_flags_for_command(command): return {}
+
+
+def _get_flags_for_cmd(cmd: str) -> dict[str, str]:
+    """Get merged flags for a command from knowledge_base (primary) and local FLAG_DATABASE (fallback).
+
+    Knowledge_base.py COMMAND_DB is the authoritative source. FLAG_DATABASE provides
+    additional coverage for commands not yet in knowledge_base.
+    """
+    flags = {}
+    # Primary source: knowledge_base COMMAND_DB
+    kb_flags = get_flags_for_command(cmd)
+    if kb_flags:
+        flags.update(kb_flags)
+    # Fallback/supplement: local FLAG_DATABASE
+    if cmd in FLAG_DATABASE:
+        for flag, desc in FLAG_DATABASE[cmd].items():
+            if flag not in flags:
+                flags[flag] = desc
+    return flags
+
+
+def _get_all_flagged_commands() -> set[str]:
+    """Get the set of all commands that have flag data from any source."""
+    cmds = set()
+    for cmd, info in COMMAND_DB.items():
+        if info.get("flags"):
+            cmds.add(cmd)
+    cmds.update(FLAG_DATABASE.keys())
+    return cmds
+
 
 class QuizType(Enum):
     """Types of quiz questions."""
@@ -499,6 +538,19 @@ def _describe_single_command(cmd: str) -> str:
     tokens = cmd.split()
     base_cmd = tokens[0] if tokens else ''
 
+    # Get args (skip flags) for knowledge_base fallback
+    args = [t for t in tokens[1:] if not t.startswith('-')]
+
+    # Check knowledge_base COMMAND_DB for rich description
+    if base_cmd and base_cmd in COMMAND_DB:
+        cmd_info = COMMAND_DB[base_cmd]
+        kb_desc = cmd_info.get('description', '')
+        if kb_desc:
+            # Use knowledge base description but make it contextual with args
+            if args:
+                return f"{kb_desc.lower()} ({' '.join(args[:2])})"
+            return kb_desc.lower()
+
     # Common command descriptions with bash focus
     descriptions = {
         'cd': lambda args: f"changes directory to {args[0] if args else 'specified path'}",
@@ -637,35 +689,35 @@ def _parse_command(cmd_string: str) -> dict:
 
 
 def _get_flag_description(cmd: str, flag: str) -> Optional[str]:
-    """Get description for a flag of a command."""
-    if cmd in FLAG_DATABASE:
-        # Handle flags like -la (combined short flags)
-        if flag in FLAG_DATABASE[cmd]:
-            return FLAG_DATABASE[cmd][flag]
-        # Try individual characters for combined flags
-        if len(flag) > 2 and flag.startswith("-") and not flag.startswith("--"):
-            for char in flag[1:]:
-                single_flag = f"-{char}"
-                if single_flag in FLAG_DATABASE[cmd]:
-                    return FLAG_DATABASE[cmd][single_flag]
+    """Get description for a flag of a command from merged sources."""
+    merged = _get_flags_for_cmd(cmd)
+    if flag in merged:
+        return merged[flag]
+    # Try individual characters for combined flags (e.g., -la -> -l, -a)
+    if len(flag) > 2 and flag.startswith("-") and not flag.startswith("--"):
+        for char in flag[1:]:
+            single_flag = f"-{char}"
+            if single_flag in merged:
+                return merged[single_flag]
     return None
 
 
 def _generate_distractor_flags(cmd: str, correct_flag: str, count: int = 3) -> list[str]:
-    """Generate plausible distractor flags."""
+    """Generate plausible distractor flags from merged knowledge sources."""
     distractors = []
 
-    # Get other flags from the same command
-    if cmd in FLAG_DATABASE:
-        other_flags = [f for f in FLAG_DATABASE[cmd].keys() if f != correct_flag]
+    # Get other flags from the same command (merged sources)
+    cmd_flags = _get_flags_for_cmd(cmd)
+    if cmd_flags:
+        other_flags = [f for f in cmd_flags.keys() if f != correct_flag]
         random.shuffle(other_flags)
         distractors.extend(other_flags[:count])
 
     # If we need more, get common flags from other commands
     if len(distractors) < count:
-        for other_cmd, flags in FLAG_DATABASE.items():
+        for other_cmd in _get_all_flagged_commands():
             if other_cmd != cmd:
-                for flag in flags:
+                for flag in _get_flags_for_cmd(other_cmd):
                     if flag not in distractors and flag != correct_flag:
                         distractors.append(flag)
                         if len(distractors) >= count:
@@ -680,10 +732,10 @@ def _generate_distractor_descriptions(correct_desc: str, count: int = 3) -> list
     """Generate plausible wrong descriptions."""
     distractors = []
 
-    # Collect all descriptions from FLAG_DATABASE
+    # Collect all descriptions from merged sources
     all_descriptions = []
-    for cmd_flags in FLAG_DATABASE.values():
-        all_descriptions.extend(cmd_flags.values())
+    for cmd in _get_all_flagged_commands():
+        all_descriptions.extend(_get_flags_for_cmd(cmd).values())
 
     # Remove duplicates and the correct answer
     all_descriptions = list(set(all_descriptions))
@@ -797,16 +849,17 @@ def generate_which_flag_quiz(
     parsed = _parse_command(cmd_string)
     base_cmd = parsed["base"]
 
-    if base_cmd not in FLAG_DATABASE or not parsed["flags"]:
+    cmd_flags = _get_flags_for_cmd(base_cmd)
+    if not cmd_flags or not parsed["flags"]:
         return None
 
     # Pick a flag to quiz on
-    available_flags = [f for f in parsed["flags"] if f in FLAG_DATABASE.get(base_cmd, {})]
+    available_flags = [f for f in parsed["flags"] if f in cmd_flags]
     if not available_flags:
         return None
 
     target_flag = random.choice(available_flags)
-    flag_desc = FLAG_DATABASE[base_cmd][target_flag]
+    flag_desc = cmd_flags[target_flag]
 
     # Generate distractor flags
     distractor_flags = _generate_distractor_flags(base_cmd, target_flag, 3)
@@ -831,13 +884,13 @@ def generate_which_flag_quiz(
             correct_id = opt_id
 
         # Get description for option explanation
-        flag_explanation = FLAG_DATABASE.get(base_cmd, {}).get(flag, "Unknown flag")
+        flag_explanation = cmd_flags.get(flag, "Unknown flag")
 
         options.append(QuizOption(
             id=opt_id,
             text=flag,
             is_correct=is_correct,
-            explanation=f"{flag}: {flag_explanation}" if flag in FLAG_DATABASE.get(base_cmd, {}) else f"{flag}: Not a standard flag for {base_cmd}"
+            explanation=f"{flag}: {flag_explanation}" if flag in cmd_flags else f"{flag}: Not a standard flag for {base_cmd}"
         ))
 
     question_id = _generate_id(f"which_flag_{base_cmd}_{target_flag}")
@@ -896,7 +949,7 @@ def generate_build_command_quiz(
         distractors.append(" ".join(missing_flag))
 
     # Distractor 3: Wrong flag
-    if parsed["flags"] and base_cmd in FLAG_DATABASE:
+    if parsed["flags"] and _get_flags_for_cmd(base_cmd):
         wrong_flags = _generate_distractor_flags(base_cmd, parsed["flags"][0], 1)
         if wrong_flags:
             wrong_flag_cmd = [base_cmd] + [wrong_flags[0]] + parsed["flags"][1:] + parsed["args"]
@@ -1065,14 +1118,15 @@ def _create_similar_command_variant(command: dict) -> Optional[dict]:
     parsed = _parse_command(cmd_string)
     base_cmd = parsed["base"]
 
-    if base_cmd not in FLAG_DATABASE:
+    variant_flags = _get_flags_for_cmd(base_cmd)
+    if not variant_flags:
         return None
 
     # Strategy: add, remove, or change a flag
     strategies = []
 
     # Can add a flag
-    available_flags = [f for f in FLAG_DATABASE[base_cmd].keys() if f not in parsed["flags"]]
+    available_flags = [f for f in variant_flags.keys() if f not in parsed["flags"]]
     if available_flags:
         strategies.append("add")