Added max_line_length arg to data.json.write()

Author: adamlui

find-project-root/utils/lib/data/json.py

@@ -35,21 +35,74 @@ def read(input: Union[Path, str], encoding: str = 'utf-8') -> Any:
     else : return json5.loads(input_str)
 
 def write(file_path: Union[Path, str], data: Any, encoding: str = 'utf-8', ensure_ascii: bool = False,
-          style: str = 'pretty', atomic: bool =True):
+          style: str = 'pretty', atomic: bool = True, max_line_length: int = 120) -> None:
     from . import file
+    from typing import Optional, List
+
     Path(file_path).parent.mkdir(parents=True, exist_ok=True)
+
+    def format_compact(obj: Any, indent: int = 0, padded_key: Optional[str] = None) -> List[str]:
+        indent_spaces = '  ' * indent
+        line_prefix = padded_key if padded_key else indent_spaces
+
+        if isinstance(obj, dict):
+
+            # Try fit whole dict in 1 line
+            kv_pairs = [f'"{key}": {json.dumps(val, separators=(",",":"), ensure_ascii=ensure_ascii)}'
+                for key,val in obj.items()]
+            single_line_dict = f'{line_prefix}{{ {", ".join(kv_pairs)} }}'
+            if len(single_line_dict) <= max_line_length:
+                return [single_line_dict]
+
+            # Else split long line up
+            lines = [line_prefix + '{']
+            for idx, (key,val) in enumerate(obj.items()):
+                inner_lines = format_compact(val, indent +1, f'  {indent_spaces}"{key}": ')
+                for line in inner_lines : lines.append(line)
+                if not idx == len(obj) -1 : lines[-1] += ',' # append comma except last line
+            lines.append(indent_spaces + '}')
+            return lines
+
+        elif isinstance(obj, list):
+
+            # Try fit whole list in 1 line
+            single_line_list = line_prefix + json.dumps(obj, separators=(',', ':'), ensure_ascii=ensure_ascii)
+            if len(single_line_list) <= max_line_length:
+                return [single_line_list]
+
+            # Else split long list up
+            lines = [line_prefix + '[']
+            if all(not isinstance(item, (dict, list)) for item in obj): # all items primitives, pack into lines
+                list_items = [json.dumps(item, ensure_ascii=ensure_ascii) for item in obj]
+                inner_indent = '  ' * (indent + 1)
+                current_line_items = []
+                for item in list_items:
+                    candidate_line = ', '.join(current_line_items + [item]) if current_line_items else item
+                    if len(inner_indent + candidate_line) + 1 <= max_line_length : current_line_items.append(item)
+                    else: # current line full, flush/start new line
+                        if current_line_items : lines.append(inner_indent + ', '.join(current_line_items) + ',')
+                        current_line_items = [item]
+                if current_line_items: # flush last line
+                    lines.append(inner_indent + ', '.join(current_line_items))
+            else: # mixed/complex items, format each recursively
+                for idx, item in enumerate(obj):
+                    inner_lines = format_compact(item, indent +1)
+                    for line in inner_lines : lines.append(line)
+                    if not idx == len(obj) -1 : lines[-1] += ','
+            lines.append(indent_spaces + ']')
+            return lines
+
+        else: # primitive
+            return [line_prefix + json.dumps(obj, ensure_ascii=ensure_ascii)]
+
+    # Format JSON
     if style == 'pretty': # single key/val spans multi-lines
         json_str = json.dumps(data, indent=2, ensure_ascii=ensure_ascii)
-    elif style == 'compact': # single key/val per line
-        lines = ['{']
-        items = list(data.items())
-        for idx, (key, val) in enumerate(items):
-            line_end = ',' if idx < len(items) -1 else ''
-            inner = f'{{ {json.dumps(val, ensure_ascii=ensure_ascii)[1:-1]} }}'
-            lines.append(f'  "{key}": {inner}{line_end}')
-        lines.append('}')
-        json_str = '\n'.join(lines)
+    elif style == 'compact': # single key/val per line but honors max_line_length
+        json_str = '\n'.join(format_compact(data))
     else: # minified to single line
         json_str = json.dumps(data, separators=(',', ':'), ensure_ascii=ensure_ascii)
     json_str += '\n'
+
+    # Write to file
     getattr(file, 'atomic_write' if atomic else 'write')(file_path, json_str, encoding=encoding)

get-min-py/src/get_min_py/cli/lib/data/json.py

@@ -35,21 +35,74 @@ def read(input: Union[Path, str], encoding: str = 'utf-8') -> Any:
     else : return json5.loads(input_str)
 
 def write(file_path: Union[Path, str], data: Any, encoding: str = 'utf-8', ensure_ascii: bool = False,
-          style: str = 'pretty', atomic: bool =True):
+          style: str = 'pretty', atomic: bool = True, max_line_length: int = 120) -> None:
     from . import file
+    from typing import Optional, List
+
     Path(file_path).parent.mkdir(parents=True, exist_ok=True)
+
+    def format_compact(obj: Any, indent: int = 0, padded_key: Optional[str] = None) -> List[str]:
+        indent_spaces = '  ' * indent
+        line_prefix = padded_key if padded_key else indent_spaces
+
+        if isinstance(obj, dict):
+
+            # Try fit whole dict in 1 line
+            kv_pairs = [f'"{key}": {json.dumps(val, separators=(",",":"), ensure_ascii=ensure_ascii)}'
+                for key,val in obj.items()]
+            single_line_dict = f'{line_prefix}{{ {", ".join(kv_pairs)} }}'
+            if len(single_line_dict) <= max_line_length:
+                return [single_line_dict]
+
+            # Else split long line up
+            lines = [line_prefix + '{']
+            for idx, (key,val) in enumerate(obj.items()):
+                inner_lines = format_compact(val, indent +1, f'  {indent_spaces}"{key}": ')
+                for line in inner_lines : lines.append(line)
+                if not idx == len(obj) -1 : lines[-1] += ',' # append comma except last line
+            lines.append(indent_spaces + '}')
+            return lines
+
+        elif isinstance(obj, list):
+
+            # Try fit whole list in 1 line
+            single_line_list = line_prefix + json.dumps(obj, separators=(',', ':'), ensure_ascii=ensure_ascii)
+            if len(single_line_list) <= max_line_length:
+                return [single_line_list]
+
+            # Else split long list up
+            lines = [line_prefix + '[']
+            if all(not isinstance(item, (dict, list)) for item in obj): # all items primitives, pack into lines
+                list_items = [json.dumps(item, ensure_ascii=ensure_ascii) for item in obj]
+                inner_indent = '  ' * (indent + 1)
+                current_line_items = []
+                for item in list_items:
+                    candidate_line = ', '.join(current_line_items + [item]) if current_line_items else item
+                    if len(inner_indent + candidate_line) + 1 <= max_line_length : current_line_items.append(item)
+                    else: # current line full, flush/start new line
+                        if current_line_items : lines.append(inner_indent + ', '.join(current_line_items) + ',')
+                        current_line_items = [item]
+                if current_line_items: # flush last line
+                    lines.append(inner_indent + ', '.join(current_line_items))
+            else: # mixed/complex items, format each recursively
+                for idx, item in enumerate(obj):
+                    inner_lines = format_compact(item, indent +1)
+                    for line in inner_lines : lines.append(line)
+                    if not idx == len(obj) -1 : lines[-1] += ','
+            lines.append(indent_spaces + ']')
+            return lines
+
+        else: # primitive
+            return [line_prefix + json.dumps(obj, ensure_ascii=ensure_ascii)]
+
+    # Format JSON
     if style == 'pretty': # single key/val spans multi-lines
         json_str = json.dumps(data, indent=2, ensure_ascii=ensure_ascii)
-    elif style == 'compact': # single key/val per line
-        lines = ['{']
-        items = list(data.items())
-        for idx, (key, val) in enumerate(items):
-            line_end = ',' if idx < len(items) -1 else ''
-            inner = f'{{ {json.dumps(val, ensure_ascii=ensure_ascii)[1:-1]} }}'
-            lines.append(f'  "{key}": {inner}{line_end}')
-        lines.append('}')
-        json_str = '\n'.join(lines)
+    elif style == 'compact': # single key/val per line but honors max_line_length
+        json_str = '\n'.join(format_compact(data))
     else: # minified to single line
         json_str = json.dumps(data, separators=(',', ':'), ensure_ascii=ensure_ascii)
     json_str += '\n'
+
+    # Write to file
     getattr(file, 'atomic_write' if atomic else 'write')(file_path, json_str, encoding=encoding)

is-unicode-supported/src/is_unicode_supported/cli/lib/data/json.py

@@ -35,21 +35,74 @@ def read(input: Union[Path, str], encoding: str = 'utf-8') -> Any:
     else : return json5.loads(input_str)
 
 def write(file_path: Union[Path, str], data: Any, encoding: str = 'utf-8', ensure_ascii: bool = False,
-          style: str = 'pretty', atomic: bool =True):
+          style: str = 'pretty', atomic: bool = True, max_line_length: int = 120) -> None:
     from . import file
+    from typing import Optional, List
+
     Path(file_path).parent.mkdir(parents=True, exist_ok=True)
+
+    def format_compact(obj: Any, indent: int = 0, padded_key: Optional[str] = None) -> List[str]:
+        indent_spaces = '  ' * indent
+        line_prefix = padded_key if padded_key else indent_spaces
+
+        if isinstance(obj, dict):
+
+            # Try fit whole dict in 1 line
+            kv_pairs = [f'"{key}": {json.dumps(val, separators=(",",":"), ensure_ascii=ensure_ascii)}'
+                for key,val in obj.items()]
+            single_line_dict = f'{line_prefix}{{ {", ".join(kv_pairs)} }}'
+            if len(single_line_dict) <= max_line_length:
+                return [single_line_dict]
+
+            # Else split long line up
+            lines = [line_prefix + '{']
+            for idx, (key,val) in enumerate(obj.items()):
+                inner_lines = format_compact(val, indent +1, f'  {indent_spaces}"{key}": ')
+                for line in inner_lines : lines.append(line)
+                if not idx == len(obj) -1 : lines[-1] += ',' # append comma except last line
+            lines.append(indent_spaces + '}')
+            return lines
+
+        elif isinstance(obj, list):
+
+            # Try fit whole list in 1 line
+            single_line_list = line_prefix + json.dumps(obj, separators=(',', ':'), ensure_ascii=ensure_ascii)
+            if len(single_line_list) <= max_line_length:
+                return [single_line_list]
+
+            # Else split long list up
+            lines = [line_prefix + '[']
+            if all(not isinstance(item, (dict, list)) for item in obj): # all items primitives, pack into lines
+                list_items = [json.dumps(item, ensure_ascii=ensure_ascii) for item in obj]
+                inner_indent = '  ' * (indent + 1)
+                current_line_items = []
+                for item in list_items:
+                    candidate_line = ', '.join(current_line_items + [item]) if current_line_items else item
+                    if len(inner_indent + candidate_line) + 1 <= max_line_length : current_line_items.append(item)
+                    else: # current line full, flush/start new line
+                        if current_line_items : lines.append(inner_indent + ', '.join(current_line_items) + ',')
+                        current_line_items = [item]
+                if current_line_items: # flush last line
+                    lines.append(inner_indent + ', '.join(current_line_items))
+            else: # mixed/complex items, format each recursively
+                for idx, item in enumerate(obj):
+                    inner_lines = format_compact(item, indent +1)
+                    for line in inner_lines : lines.append(line)
+                    if not idx == len(obj) -1 : lines[-1] += ','
+            lines.append(indent_spaces + ']')
+            return lines
+
+        else: # primitive
+            return [line_prefix + json.dumps(obj, ensure_ascii=ensure_ascii)]
+
+    # Format JSON
     if style == 'pretty': # single key/val spans multi-lines
         json_str = json.dumps(data, indent=2, ensure_ascii=ensure_ascii)
-    elif style == 'compact': # single key/val per line
-        lines = ['{']
-        items = list(data.items())
-        for idx, (key, val) in enumerate(items):
-            line_end = ',' if idx < len(items) -1 else ''
-            inner = f'{{ {json.dumps(val, ensure_ascii=ensure_ascii)[1:-1]} }}'
-            lines.append(f'  "{key}": {inner}{line_end}')
-        lines.append('}')
-        json_str = '\n'.join(lines)
+    elif style == 'compact': # single key/val per line but honors max_line_length
+        json_str = '\n'.join(format_compact(data))
     else: # minified to single line
         json_str = json.dumps(data, separators=(',', ':'), ensure_ascii=ensure_ascii)
     json_str += '\n'
+
+    # Write to file
     getattr(file, 'atomic_write' if atomic else 'write')(file_path, json_str, encoding=encoding)

latin-locales/utils/lib/data/json.py

@@ -35,21 +35,74 @@ def read(input: Union[Path, str], encoding: str = 'utf-8') -> Any:
     else : return json5.loads(input_str)
 
 def write(file_path: Union[Path, str], data: Any, encoding: str = 'utf-8', ensure_ascii: bool = False,
-          style: str = 'pretty', atomic: bool =True):
+          style: str = 'pretty', atomic: bool = True, max_line_length: int = 120) -> None:
     from . import file
+    from typing import Optional, List
+
     Path(file_path).parent.mkdir(parents=True, exist_ok=True)
+
+    def format_compact(obj: Any, indent: int = 0, padded_key: Optional[str] = None) -> List[str]:
+        indent_spaces = '  ' * indent
+        line_prefix = padded_key if padded_key else indent_spaces
+
+        if isinstance(obj, dict):
+
+            # Try fit whole dict in 1 line
+            kv_pairs = [f'"{key}": {json.dumps(val, separators=(",",":"), ensure_ascii=ensure_ascii)}'
+                for key,val in obj.items()]
+            single_line_dict = f'{line_prefix}{{ {", ".join(kv_pairs)} }}'
+            if len(single_line_dict) <= max_line_length:
+                return [single_line_dict]
+
+            # Else split long line up
+            lines = [line_prefix + '{']
+            for idx, (key,val) in enumerate(obj.items()):
+                inner_lines = format_compact(val, indent +1, f'  {indent_spaces}"{key}": ')
+                for line in inner_lines : lines.append(line)
+                if not idx == len(obj) -1 : lines[-1] += ',' # append comma except last line
+            lines.append(indent_spaces + '}')
+            return lines
+
+        elif isinstance(obj, list):
+
+            # Try fit whole list in 1 line
+            single_line_list = line_prefix + json.dumps(obj, separators=(',', ':'), ensure_ascii=ensure_ascii)
+            if len(single_line_list) <= max_line_length:
+                return [single_line_list]
+
+            # Else split long list up
+            lines = [line_prefix + '[']
+            if all(not isinstance(item, (dict, list)) for item in obj): # all items primitives, pack into lines
+                list_items = [json.dumps(item, ensure_ascii=ensure_ascii) for item in obj]
+                inner_indent = '  ' * (indent + 1)
+                current_line_items = []
+                for item in list_items:
+                    candidate_line = ', '.join(current_line_items + [item]) if current_line_items else item
+                    if len(inner_indent + candidate_line) + 1 <= max_line_length : current_line_items.append(item)
+                    else: # current line full, flush/start new line
+                        if current_line_items : lines.append(inner_indent + ', '.join(current_line_items) + ',')
+                        current_line_items = [item]
+                if current_line_items: # flush last line
+                    lines.append(inner_indent + ', '.join(current_line_items))
+            else: # mixed/complex items, format each recursively
+                for idx, item in enumerate(obj):
+                    inner_lines = format_compact(item, indent +1)
+                    for line in inner_lines : lines.append(line)
+                    if not idx == len(obj) -1 : lines[-1] += ','
+            lines.append(indent_spaces + ']')
+            return lines
+
+        else: # primitive
+            return [line_prefix + json.dumps(obj, ensure_ascii=ensure_ascii)]
+
+    # Format JSON
     if style == 'pretty': # single key/val spans multi-lines
         json_str = json.dumps(data, indent=2, ensure_ascii=ensure_ascii)
-    elif style == 'compact': # single key/val per line
-        lines = ['{']
-        items = list(data.items())
-        for idx, (key, val) in enumerate(items):
-            line_end = ',' if idx < len(items) -1 else ''
-            inner = f'{{ {json.dumps(val, ensure_ascii=ensure_ascii)[1:-1]} }}'
-            lines.append(f'  "{key}": {inner}{line_end}')
-        lines.append('}')
-        json_str = '\n'.join(lines)
+    elif style == 'compact': # single key/val per line but honors max_line_length
+        json_str = '\n'.join(format_compact(data))
     else: # minified to single line
         json_str = json.dumps(data, separators=(',', ':'), ensure_ascii=ensure_ascii)
     json_str += '\n'
+
+    # Write to file
     getattr(file, 'atomic_write' if atomic else 'write')(file_path, json_str, encoding=encoding)