jsonunwrap 0.2.1__tar.gz → 0.2.2__tar.gz

{jsonunwrap-0.2.1 → jsonunwrap-0.2.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: jsonunwrap
-Version: 0.2.1
+Version: 0.2.2
 Summary: A small python package that unpacks data from a JSON url and converts it into a csv file.
 Author-email: njuedominic <njuemugodominic@gmail.com>
 Project-URL: Homepage, https://github.com/njuedominic/json-unwrap

jsonunwrap-0.2.2/jsonunwrap/core.py

@@ -0,0 +1,99 @@
+"""
+This module contains the core functionality for the json_unwrap package.
+The main function is `json_to_csv`, which converts a JSON file to a CSV file.
+"""
+import os
+from typing import Any, Dict, List, Union
+import pandas as pd
+import requests
+
+def unwrap_data(data: Union[Dict[str, Any], List[Any]]) -> pd.DataFrame:
+    """
+    Normalizes and deeply flattens semi-structured JSON data into a pandas DataFrame.
+    """
+    # 1. Ensure we start with a clean record list
+    if isinstance(data, dict):
+        # Handle cases where the data is inside a wrapper key (like {"products": [...]})
+        list_keys = [k for k, v in data.items() if isinstance(v, list)]
+        if list_keys and len(data) <= 4:
+            main_data = data[list_keys[0]]
+        else:
+            main_data = [data]
+    else:
+        main_data = data
+
+    # 2. Base normalization
+    df = pd.json_normalize(main_data)
+
+    # 3. Clean Linear Pass: Avoid infinite loops by tracking column states directly
+    columns_to_process = list(df.columns)
+    
+    while columns_to_process:
+        col = columns_to_process.pop(0)
+        
+        # Guard check if the column was dropped in a previous iteration
+        if col not in df.columns:
+            continue
+            
+        non_null_vals = df[col].dropna()
+        if non_null_vals.empty:
+            continue
+
+        # Check for nested dictionaries
+        if any(isinstance(val, dict) for val in non_null_vals):
+            nested_df = pd.json_normalize(non_null_vals).set_index(non_null_vals.index)
+            # Add new sub-columns back into the processing queue
+            new_cols = [f"{c}_{col}" for c in nested_df.columns]
+            df = df.drop(columns=[col]).join(nested_df, rsuffix=f"_{col}")
+            columns_to_process.extend(new_cols)
+
+        # Check for nested lists (But do not loop back if it's just raw strings/ints)
+        elif any(isinstance(val, list) for val in non_null_vals):
+            # Check if the list contains dictionaries before exploding heavily
+            first_list = next((v for v in non_null_vals if isinstance(v, list) and v), None)
+            
+            df = df.explode(col)
+            
+            # If the inner elements were dictionaries, we need to flatten them on the next pass
+            if first_list and isinstance(first_list[0], dict):
+                columns_to_process.append(col)
+
+    return df
+
+ 
+def fetch_json(url: str, **kwargs: Any) -> Union[Dict[str, Any], List[Any]]:
+    """
+    Fetches raw JSON data from a URL.
+    
+    Args:
+        url: The endpoint web target.
+        **kwargs: Additional arguments passed directly to requests.get (e.g., headers, auth).
+    """
+    response = requests.get(url, **kwargs)
+    response.raise_for_status()
+    return response.json()
+
+
+def json_to_csv(url: str, output_path: str) -> pd.DataFrame:
+    """
+    Fetches JSON from a URL, deeply flattens it, and saves it directly to a CSV file.
+    
+    Args:
+        url: The endpoint URL containing the target JSON data.
+        output_path: Target filesystem path where the CSV will be written.
+        
+    Returns:
+        The generated pandas DataFrame.
+    """
+    # Ensure the parent output directory exists safely
+    directory = os.path.dirname(output_path)
+    if directory:
+        os.makedirs(directory, exist_ok=True)
+        
+    raw_data = fetch_json(url)
+    df = unwrap_data(raw_data)
+    
+    df.to_csv(output_path, index=False)
+    return df
+
+

{jsonunwrap-0.2.1 → jsonunwrap-0.2.2}/jsonunwrap.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: jsonunwrap
-Version: 0.2.1
+Version: 0.2.2
 Summary: A small python package that unpacks data from a JSON url and converts it into a csv file.
 Author-email: njuedominic <njuemugodominic@gmail.com>
 Project-URL: Homepage, https://github.com/njuedominic/json-unwrap

{jsonunwrap-0.2.1 → jsonunwrap-0.2.2}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "jsonunwrap"
-version = "0.2.1"
+version = "0.2.2"
 authors = [
     { name = "njuedominic", email="njuemugodominic@gmail.com"}
 ]

jsonunwrap-0.2.1/jsonunwrap/core.py

@@ -1,80 +0,0 @@
-"""
-This module contains the core functionality for the json_unwrap package.
-The main function is `json_to_csv`, which converts a JSON file to a CSV file.
-"""
-import os
-from typing import Any, Dict, List, Union
-import pandas as pd
-import requests
-
-def unwrap_data(data: Union[Dict[str, Any], List[Any]]) -> pd.DataFrame:
-    """
-    Normalizes and deeply flattens semi-structured JSON data into a pandas DataFrame.
-    """
-    if isinstance(data, dict):
-        main_data = [data]
-    else:
-        main_data = data
-
-    df = pd.json_normalize(main_data)
-
-    # Professional Fix: Track already exploded primitive columns to prevent loops
-    exploded_columns = set()
-
-    changed = True
-    while changed:
-        changed = False
-        for col in list(df.columns):
-            # 1. Explode lists (Only if we haven't exploded this exact column name yet)
-            if col not in exploded_columns and any(isinstance(val, list) for val in df[col].dropna()):
-                df = df.explode(col)
-                exploded_columns.add(col)  # Mark as done so we don't repeat it!
-                changed = True
-                break  # Refresh columns list
-
-            # 2. Normalize and merge nested dictionaries
-            if any(isinstance(val, dict) for val in df[col].dropna()):
-                nested_df = pd.json_normalize(df[col]).set_index(df.index)
-                df = df.drop(columns=[col]).join(nested_df, rsuffix=f"_{col}")
-                changed = True
-                break
-                
-    return df
-
- 
-def fetch_json(url: str, **kwargs: Any) -> Union[Dict[str, Any], List[Any]]:
-    """
-    Fetches raw JSON data from a URL.
-    
-    Args:
-        url: The endpoint web target.
-        **kwargs: Additional arguments passed directly to requests.get (e.g., headers, auth).
-    """
-    response = requests.get(url, **kwargs)
-    response.raise_for_status()
-    return response.json()
-
-
-def json_to_csv(url: str, output_path: str) -> pd.DataFrame:
-    """
-    Fetches JSON from a URL, deeply flattens it, and saves it directly to a CSV file.
-    
-    Args:
-        url: The endpoint URL containing the target JSON data.
-        output_path: Target filesystem path where the CSV will be written.
-        
-    Returns:
-        The generated pandas DataFrame.
-    """
-    # Ensure the parent output directory exists safely
-    directory = os.path.dirname(output_path)
-    if directory:
-        os.makedirs(directory, exist_ok=True)
-        
-    raw_data = fetch_json(url)
-    df = unwrap_data(raw_data)
-    
-    df.to_csv(output_path, index=False)
-    return df
-
-