fakedata-python 2.0.2__tar.gz → 2.0.4__tar.gz

fakedata_python-2.0.4/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 Abhay Mourya
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

fakedata_python-2.0.4/MANIFEST.in

@@ -0,0 +1,24 @@
+# Exclude development and Node.js files
+prune .github
+exclude CONTRIBUTING.md
+exclude CODE_OF_CONDUCT.md
+exclude .npmignore
+exclude test.js
+exclude test_py.py
+exclude test_python.py
+
+# Exclude JS source code
+prune src
+exclude package.json
+exclude package-lock.json
+
+# Exclude web and raw data
+prune website
+prune newdata
+prune dist
+prune tests
+
+# Exclude byte-compiled files
+global-exclude *.py[cod]
+global-exclude __pycache__
+global-exclude *.so

{fakedata_python-2.0.2/fakedata_python.egg-info → fakedata_python-2.0.4}/PKG-INFO

@@ -1,7 +1,7 @@
 Metadata-Version: 2.4
 Name: fakedata-python
-Version: 2.0.2
-Summary: The fakedata package generates realistic synthetic user profiles for machine learning, deep learning, data analysis, and data science workflows.
+Version: 2.0.4
+Summary: The fakedata package generates realistic user profiles for machine learning, deep learning, data analysis, and data science workflows.
 Author-email: abhay557 <contact@abhaymourya.in>
 License-Expression: MIT
 Project-URL: Homepage, https://github.com/abhay557/fakedata
@@ -9,6 +9,8 @@ Classifier: Programming Language :: Python :: 3
 Classifier: Operating System :: OS Independent
 Requires-Python: >=3.7
 Description-Content-Type: text/markdown
+License-File: LICENSE
+Dynamic: license-file
 
 # fakedata 
 
@@ -34,52 +36,56 @@ A high-performance, **zero-dependency** synthetic data generation engine, availa
 - **Time-Series Data**: Generate chronological activity logs (logins, page views, purchases) per user for behavioral modeling.
 - **Pipeline Ready**: Export directly to CSV, JSON, or Flat objects (perfect for `pandas.DataFrame`).
 - **CLI Tool**: Generate and export datasets directly from your terminal — no scripting required.
+- **Streaming Generation**: Files are written one record at a time — constant RAM usage regardless of dataset size. Generate 10M+ rows without running out of memory.
 
 ---
 
-##  Python Implementation
+##  Node.js / TypeScript Implementation
 
 ### Installation
 ```bash
-pip install fakedata-python
+npm install @abhay557/fakedata
 ```
 
 ### Quick Start
-```python
-import fakedata.data as data
-import pandas as pd
+```javascript
+const fakedata = require('@abhay557/fakedata');
 
-# Generate 10,000 highly correlated users deterministically
-users = data.users(10000, {"seed": 42})
+// Generate deterministic users with a 5% missing data rate (null injection)
+const users = fakedata.data.users(1000, { seed: 42, missing_rate: 0.05 });
 
-# Or export directly to a Pandas DataFrame
-df = pd.DataFrame(data.users_flat(10000, {"seed": 42}))
-print(df.head())
+// Export directly to CSV format
+const csvString = fakedata.data.usersToCSV(1000, { seed: 42 });
 
-# Create time-series activity data
-ts = data.user_time_series({"days": 30, "events_per_day": 8})
-print(f"Generated {len(ts['activity'])} events for {ts['user']['fullName']}")
+// Time-series activity data
+const ts = fakedata.userTimeSeries({ days: 30, eventsPerDay: 8 });
+console.log(`Generated ${ts.activity.length} events for ${ts.user.fullName}`);
 ```
-##  Node.js / TypeScript Implementation
+
+---
+
+##  Python Implementation
 
 ### Installation
 ```bash
-npm install @abhay557/fakedata
+pip install fakedata-python
 ```
 
 ### Quick Start
-```javascript
-const { data } = require('@abhay557/fakedata');
+```python
+import fakedata
+import pandas as pd
 
-// Generate deterministic users with a 5% missing data rate (null injection)
-const users = data.users(1000, { seed: 42, missing_rate: 0.05 });
+# Generate 10,000 highly correlated users deterministically
+users = fakedata.data.users(10000, {"seed": 42})
 
-// Export directly to CSV format
-const csvString = data.usersToCSV(1000, { seed: 42 });
+# Or export directly to a Pandas DataFrame
+df = pd.DataFrame(fakedata.data.users_flat(10000, {"seed": 42}))
+print(df.head())
 
-// Time-series activity data
-const ts = data.userTimeSeries({ days: 30, eventsPerDay: 8 });
-console.log(`Generated ${ts.activity.length} events for ${ts.user.fullName}`);
+# Create time-series activity data
+ts = fakedata.data.user_time_series({"days": 30, "events_per_day": 8})
+print(f"Generated {len(ts['activity'])} events for {ts['user']['fullName']}")
 ```
 
 ---
@@ -133,6 +139,9 @@ fakedata generate -n 500 -l in --seed 42 -o india.json
 # Fraud detection dataset with 5% anomalies
 fakedata generate -n 10000 -a 0.05 -f csv -o fraud_data.csv
 
+# Generate 1 million rows without running out of memory (streaming)
+fakedata generate -n 1000000 -f csv -o big_dataset.csv
+
 # Preview a single user in the console
 fakedata preview
 
@@ -140,6 +149,23 @@ fakedata preview
 fakedata generate -n 100 --timeseries --days 60 -o activity.json
 ```
 
+### Streaming Architecture
+
+When writing to a file (`-o`), the CLI uses a **streaming write** strategy:
+
+- The output file is **created first**, before any data is generated.
+- Each user is generated **one at a time** and written immediately to disk.
+- The generated object is then **discarded** — it is never held in a large array.
+- **RAM usage stays constant** (O(1)) regardless of how many records you generate.
+- A live progress counter is printed every 10,000 records for large jobs.
+
+This means you can generate **tens of millions of rows** without hitting Node.js heap limits or Python memory errors.
+
+```
+Before (old):  generate ALL → hold in RAM → write to file   ❌ OOM at ~500k rows
+After  (new):  open file → generate 1 → write → discard → repeat  ✅ unlimited
+```
+
 ---
 ### sample output - one user
 ```fakedata.data.user()``` 
@@ -432,3 +458,21 @@ Distributed under the **MIT License**. See `LICENSE` for more information.
 **Maintainer**: [abhay557](https://github.com/abhay557)
 
 - Project Commit History - `https://github.com/abhay557/random-api.xyz`
+
+---
+
+## Contributing
+
+Contributions are welcome! Whether it's a bug fix, a new feature, or improved docs — every bit helps.
+
+- Read the [Contributing Guide](./CONTRIBUTING.md) before submitting a PR.
+- Use the [Bug Report](https://github.com/abhay557/fakedata/issues/new?template=bug_report.md) template to report issues.
+- Use the [Feature Request](https://github.com/abhay557/fakedata/issues/new?template=feature_request.md) template to suggest ideas.
+- Please follow our [Code of Conduct](./CODE_OF_CONDUCT.md) in all interactions.
+
+```bash
+# Fork the repo, then:
+git clone https://github.com/YOUR_USERNAME/fakedata.git
+git checkout -b feature/my-improvement
+# Make your changes, then open a Pull Request!
+```

fakedata_python-2.0.2/PKG-INFO → fakedata_python-2.0.4/README.md

@@ -1,15 +1,3 @@
-Metadata-Version: 2.4
-Name: fakedata-python
-Version: 2.0.2
-Summary: The fakedata package generates realistic synthetic user profiles for machine learning, deep learning, data analysis, and data science workflows.
-Author-email: abhay557 <contact@abhaymourya.in>
-License-Expression: MIT
-Project-URL: Homepage, https://github.com/abhay557/fakedata
-Classifier: Programming Language :: Python :: 3
-Classifier: Operating System :: OS Independent
-Requires-Python: >=3.7
-Description-Content-Type: text/markdown
-
 # fakedata 
 
 [![NPM Version](https://img.shields.io/npm/v/@abhay557/fakedata?color=red&label=npm)](https://www.npmjs.com/package/@abhay557/fakedata)
@@ -34,52 +22,56 @@ A high-performance, **zero-dependency** synthetic data generation engine, availa
 - **Time-Series Data**: Generate chronological activity logs (logins, page views, purchases) per user for behavioral modeling.
 - **Pipeline Ready**: Export directly to CSV, JSON, or Flat objects (perfect for `pandas.DataFrame`).
 - **CLI Tool**: Generate and export datasets directly from your terminal — no scripting required.
+- **Streaming Generation**: Files are written one record at a time — constant RAM usage regardless of dataset size. Generate 10M+ rows without running out of memory.
 
 ---
 
-##  Python Implementation
+##  Node.js / TypeScript Implementation
 
 ### Installation
 ```bash
-pip install fakedata-python
+npm install @abhay557/fakedata
 ```
 
 ### Quick Start
-```python
-import fakedata.data as data
-import pandas as pd
+```javascript
+const fakedata = require('@abhay557/fakedata');
 
-# Generate 10,000 highly correlated users deterministically
-users = data.users(10000, {"seed": 42})
+// Generate deterministic users with a 5% missing data rate (null injection)
+const users = fakedata.data.users(1000, { seed: 42, missing_rate: 0.05 });
 
-# Or export directly to a Pandas DataFrame
-df = pd.DataFrame(data.users_flat(10000, {"seed": 42}))
-print(df.head())
+// Export directly to CSV format
+const csvString = fakedata.data.usersToCSV(1000, { seed: 42 });
 
-# Create time-series activity data
-ts = data.user_time_series({"days": 30, "events_per_day": 8})
-print(f"Generated {len(ts['activity'])} events for {ts['user']['fullName']}")
+// Time-series activity data
+const ts = fakedata.userTimeSeries({ days: 30, eventsPerDay: 8 });
+console.log(`Generated ${ts.activity.length} events for ${ts.user.fullName}`);
 ```
-##  Node.js / TypeScript Implementation
+
+---
+
+##  Python Implementation
 
 ### Installation
 ```bash
-npm install @abhay557/fakedata
+pip install fakedata-python
 ```
 
 ### Quick Start
-```javascript
-const { data } = require('@abhay557/fakedata');
+```python
+import fakedata
+import pandas as pd
 
-// Generate deterministic users with a 5% missing data rate (null injection)
-const users = data.users(1000, { seed: 42, missing_rate: 0.05 });
+# Generate 10,000 highly correlated users deterministically
+users = fakedata.data.users(10000, {"seed": 42})
 
-// Export directly to CSV format
-const csvString = data.usersToCSV(1000, { seed: 42 });
+# Or export directly to a Pandas DataFrame
+df = pd.DataFrame(fakedata.data.users_flat(10000, {"seed": 42}))
+print(df.head())
 
-// Time-series activity data
-const ts = data.userTimeSeries({ days: 30, eventsPerDay: 8 });
-console.log(`Generated ${ts.activity.length} events for ${ts.user.fullName}`);
+# Create time-series activity data
+ts = fakedata.data.user_time_series({"days": 30, "events_per_day": 8})
+print(f"Generated {len(ts['activity'])} events for {ts['user']['fullName']}")
 ```
 
 ---
@@ -133,6 +125,9 @@ fakedata generate -n 500 -l in --seed 42 -o india.json
 # Fraud detection dataset with 5% anomalies
 fakedata generate -n 10000 -a 0.05 -f csv -o fraud_data.csv
 
+# Generate 1 million rows without running out of memory (streaming)
+fakedata generate -n 1000000 -f csv -o big_dataset.csv
+
 # Preview a single user in the console
 fakedata preview
 
@@ -140,6 +135,23 @@ fakedata preview
 fakedata generate -n 100 --timeseries --days 60 -o activity.json
 ```
 
+### Streaming Architecture
+
+When writing to a file (`-o`), the CLI uses a **streaming write** strategy:
+
+- The output file is **created first**, before any data is generated.
+- Each user is generated **one at a time** and written immediately to disk.
+- The generated object is then **discarded** — it is never held in a large array.
+- **RAM usage stays constant** (O(1)) regardless of how many records you generate.
+- A live progress counter is printed every 10,000 records for large jobs.
+
+This means you can generate **tens of millions of rows** without hitting Node.js heap limits or Python memory errors.
+
+```
+Before (old):  generate ALL → hold in RAM → write to file   ❌ OOM at ~500k rows
+After  (new):  open file → generate 1 → write → discard → repeat  ✅ unlimited
+```
+
 ---
 ### sample output - one user
 ```fakedata.data.user()``` 
@@ -432,3 +444,21 @@ Distributed under the **MIT License**. See `LICENSE` for more information.
 **Maintainer**: [abhay557](https://github.com/abhay557)
 
 - Project Commit History - `https://github.com/abhay557/random-api.xyz`
+
+---
+
+## Contributing
+
+Contributions are welcome! Whether it's a bug fix, a new feature, or improved docs — every bit helps.
+
+- Read the [Contributing Guide](./CONTRIBUTING.md) before submitting a PR.
+- Use the [Bug Report](https://github.com/abhay557/fakedata/issues/new?template=bug_report.md) template to report issues.
+- Use the [Feature Request](https://github.com/abhay557/fakedata/issues/new?template=feature_request.md) template to suggest ideas.
+- Please follow our [Code of Conduct](./CODE_OF_CONDUCT.md) in all interactions.
+
+```bash
+# Fork the repo, then:
+git clone https://github.com/YOUR_USERNAME/fakedata.git
+git checkout -b feature/my-improvement
+# Make your changes, then open a Pull Request!
+```

{fakedata_python-2.0.2 → fakedata_python-2.0.4}/fakedata/cli.py

@@ -89,44 +89,93 @@ EXAMPLES:
             'anomaly_rate': args.anomaly_rate,
             'missing_rate': args.missing_rate,
         }
-        # Remove None values so defaults are used inside the engine
         options = {k: v for k, v in options.items() if v is not None and v != 0.0}
 
+        count = args.count
         start = time.time()
-
-        if args.timeseries:
-            results = [
-                data.user_time_series({**options, 'days': args.days, 'events_per_day': args.events_per_day})
-                for _ in range(args.count)
-            ]
-            output = json.dumps(results, indent=2 if args.pretty else None)
-
-        elif args.format == 'csv':
-            output = data.users_to_csv(args.count, options if options else None)
-
-        elif args.format == 'flat':
-            rows = data.users_flat(args.count, options if options else None)
-            output = json.dumps(rows, indent=2 if args.pretty else None)
-
-        else:  # json
-            if args.pretty:
-                output = data.users_to_json(args.count, options if options else None)
+        PROGRESS_INTERVAL = 10000
+
+        # ── stdout: buffer is fine for small terminal output ──────────────
+        if not args.output:
+            if args.timeseries:
+                results = [
+                    data.user_time_series({**options, 'days': args.days, 'events_per_day': args.events_per_day})
+                    for _ in range(count)
+                ]
+                print(json.dumps(results, indent=2 if args.pretty else None))
+            elif args.format == 'csv':
+                print(data.users_to_csv(count, options if options else None))
+            elif args.format == 'flat':
+                rows = data.users_flat(count, options if options else None)
+                print(json.dumps(rows, indent=2 if args.pretty else None))
             else:
-                output = json.dumps(data.users(args.count, options if options else None))
+                if args.pretty:
+                    print(data.users_to_json(count, options if options else None))
+                else:
+                    print(json.dumps(data.users(count, options if options else None)))
+            return
+
+        # ── File: STREAMING — open file first, write one record at a time ──
+        out_path = os.path.abspath(args.output)
+
+        with open(out_path, 'w', encoding='utf-8') as f:
+
+            if args.timeseries:
+                f.write('[\n')
+                for i in range(count):
+                    rec = data.user_time_series({**options, 'days': args.days, 'events_per_day': args.events_per_day})
+                    line = json.dumps(rec, indent=2 if args.pretty else None)
+                    f.write(line + (',' if i < count - 1 else '') + '\n')
+                    if (i + 1) % PROGRESS_INTERVAL == 0:
+                        print(f"\r  ⏳ {i+1:,} / {count:,} written...", end='', file=sys.stderr)
+                f.write(']\n')
+
+            elif args.format == 'csv':
+                # Write header from first record
+                first = data.user(options if options else None)
+                header = ','.join(f'"{k}"' for k in first.keys())
+                f.write(header + '\n')
+                f.write(_user_to_csv_row(first) + '\n')
+                for i in range(1, count):
+                    u = data.user(options if options else None)
+                    f.write(_user_to_csv_row(u) + '\n')
+                    if (i + 1) % PROGRESS_INTERVAL == 0:
+                        print(f"\r  ⏳ {i+1:,} / {count:,} written...", end='', file=sys.stderr)
+
+            else:  # json / flat
+                f.write('[\n')
+                for i in range(count):
+                    u = data.user(options if options else None)
+                    line = json.dumps(u, indent=2 if args.pretty else None)
+                    f.write(line + (',' if i < count - 1 else '') + '\n')
+                    if (i + 1) % PROGRESS_INTERVAL == 0:
+                        print(f"\r  ⏳ {i+1:,} / {count:,} written...", end='', file=sys.stderr)
+                f.write(']\n')
 
         elapsed = round(time.time() - start, 2)
-
-        if args.output:
-            out_path = os.path.abspath(args.output)
-            with open(out_path, 'w', encoding='utf-8') as f:
-                f.write(output)
-            size_kb = round(len(output.encode('utf-8')) / 1024, 1)
-            print(
-                f"✔ Done! Generated {args.count:,} users in {elapsed}s → {out_path} ({size_kb} KB)",
-                file=sys.stderr
-            )
+        size_bytes = os.path.getsize(out_path)
+        size_label = f"{size_bytes / 1048576:.1f} MB" if size_bytes >= 1048576 else f"{size_bytes / 1024:.1f} KB"
+        print('\r', end='', file=sys.stderr)  # clear progress line
+        print(
+            f"✔ Done! Generated {count:,} users in {elapsed}s → {out_path} ({size_label})",
+            file=sys.stderr
+        )
+
+
+def _user_to_csv_row(u):
+    """Serialize a single user dict to a CSV row string."""
+    import json as _json
+    parts = []
+    for v in u.values():
+        if v is None:
+            parts.append('')
+        elif isinstance(v, (dict, list)):
+            parts.append('"' + _json.dumps(v).replace('"', '""') + '"')
+        elif isinstance(v, str):
+            parts.append('"' + v.replace('"', '""') + '"')
         else:
-            print(output)
+            parts.append(str(v))
+    return ','.join(parts)
 
 
 if __name__ == '__main__':

fakedata_python-2.0.2/README.md → fakedata_python-2.0.4/fakedata_python.egg-info/PKG-INFO

@@ -1,3 +1,17 @@
+Metadata-Version: 2.4
+Name: fakedata-python
+Version: 2.0.4
+Summary: The fakedata package generates realistic user profiles for machine learning, deep learning, data analysis, and data science workflows.
+Author-email: abhay557 <contact@abhaymourya.in>
+License-Expression: MIT
+Project-URL: Homepage, https://github.com/abhay557/fakedata
+Classifier: Programming Language :: Python :: 3
+Classifier: Operating System :: OS Independent
+Requires-Python: >=3.7
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Dynamic: license-file
+
 # fakedata 
 
 [![NPM Version](https://img.shields.io/npm/v/@abhay557/fakedata?color=red&label=npm)](https://www.npmjs.com/package/@abhay557/fakedata)
@@ -22,52 +36,56 @@ A high-performance, **zero-dependency** synthetic data generation engine, availa
 - **Time-Series Data**: Generate chronological activity logs (logins, page views, purchases) per user for behavioral modeling.
 - **Pipeline Ready**: Export directly to CSV, JSON, or Flat objects (perfect for `pandas.DataFrame`).
 - **CLI Tool**: Generate and export datasets directly from your terminal — no scripting required.
+- **Streaming Generation**: Files are written one record at a time — constant RAM usage regardless of dataset size. Generate 10M+ rows without running out of memory.
 
 ---
 
-##  Python Implementation
+##  Node.js / TypeScript Implementation
 
 ### Installation
 ```bash
-pip install fakedata-python
+npm install @abhay557/fakedata
 ```
 
 ### Quick Start
-```python
-import fakedata.data as data
-import pandas as pd
+```javascript
+const fakedata = require('@abhay557/fakedata');
 
-# Generate 10,000 highly correlated users deterministically
-users = data.users(10000, {"seed": 42})
+// Generate deterministic users with a 5% missing data rate (null injection)
+const users = fakedata.data.users(1000, { seed: 42, missing_rate: 0.05 });
 
-# Or export directly to a Pandas DataFrame
-df = pd.DataFrame(data.users_flat(10000, {"seed": 42}))
-print(df.head())
+// Export directly to CSV format
+const csvString = fakedata.data.usersToCSV(1000, { seed: 42 });
 
-# Create time-series activity data
-ts = data.user_time_series({"days": 30, "events_per_day": 8})
-print(f"Generated {len(ts['activity'])} events for {ts['user']['fullName']}")
+// Time-series activity data
+const ts = fakedata.userTimeSeries({ days: 30, eventsPerDay: 8 });
+console.log(`Generated ${ts.activity.length} events for ${ts.user.fullName}`);
 ```
-##  Node.js / TypeScript Implementation
+
+---
+
+##  Python Implementation
 
 ### Installation
 ```bash
-npm install @abhay557/fakedata
+pip install fakedata-python
 ```
 
 ### Quick Start
-```javascript
-const { data } = require('@abhay557/fakedata');
+```python
+import fakedata
+import pandas as pd
 
-// Generate deterministic users with a 5% missing data rate (null injection)
-const users = data.users(1000, { seed: 42, missing_rate: 0.05 });
+# Generate 10,000 highly correlated users deterministically
+users = fakedata.data.users(10000, {"seed": 42})
 
-// Export directly to CSV format
-const csvString = data.usersToCSV(1000, { seed: 42 });
+# Or export directly to a Pandas DataFrame
+df = pd.DataFrame(fakedata.data.users_flat(10000, {"seed": 42}))
+print(df.head())
 
-// Time-series activity data
-const ts = data.userTimeSeries({ days: 30, eventsPerDay: 8 });
-console.log(`Generated ${ts.activity.length} events for ${ts.user.fullName}`);
+# Create time-series activity data
+ts = fakedata.data.user_time_series({"days": 30, "events_per_day": 8})
+print(f"Generated {len(ts['activity'])} events for {ts['user']['fullName']}")
 ```
 
 ---
@@ -121,6 +139,9 @@ fakedata generate -n 500 -l in --seed 42 -o india.json
 # Fraud detection dataset with 5% anomalies
 fakedata generate -n 10000 -a 0.05 -f csv -o fraud_data.csv
 
+# Generate 1 million rows without running out of memory (streaming)
+fakedata generate -n 1000000 -f csv -o big_dataset.csv
+
 # Preview a single user in the console
 fakedata preview
 
@@ -128,6 +149,23 @@ fakedata preview
 fakedata generate -n 100 --timeseries --days 60 -o activity.json
 ```
 
+### Streaming Architecture
+
+When writing to a file (`-o`), the CLI uses a **streaming write** strategy:
+
+- The output file is **created first**, before any data is generated.
+- Each user is generated **one at a time** and written immediately to disk.
+- The generated object is then **discarded** — it is never held in a large array.
+- **RAM usage stays constant** (O(1)) regardless of how many records you generate.
+- A live progress counter is printed every 10,000 records for large jobs.
+
+This means you can generate **tens of millions of rows** without hitting Node.js heap limits or Python memory errors.
+
+```
+Before (old):  generate ALL → hold in RAM → write to file   ❌ OOM at ~500k rows
+After  (new):  open file → generate 1 → write → discard → repeat  ✅ unlimited
+```
+
 ---
 ### sample output - one user
 ```fakedata.data.user()``` 
@@ -420,3 +458,21 @@ Distributed under the **MIT License**. See `LICENSE` for more information.
 **Maintainer**: [abhay557](https://github.com/abhay557)
 
 - Project Commit History - `https://github.com/abhay557/random-api.xyz`
+
+---
+
+## Contributing
+
+Contributions are welcome! Whether it's a bug fix, a new feature, or improved docs — every bit helps.
+
+- Read the [Contributing Guide](./CONTRIBUTING.md) before submitting a PR.
+- Use the [Bug Report](https://github.com/abhay557/fakedata/issues/new?template=bug_report.md) template to report issues.
+- Use the [Feature Request](https://github.com/abhay557/fakedata/issues/new?template=feature_request.md) template to suggest ideas.
+- Please follow our [Code of Conduct](./CODE_OF_CONDUCT.md) in all interactions.
+
+```bash
+# Fork the repo, then:
+git clone https://github.com/YOUR_USERNAME/fakedata.git
+git checkout -b feature/my-improvement
+# Make your changes, then open a Pull Request!
+```

{fakedata_python-2.0.2 → fakedata_python-2.0.4}/fakedata_python.egg-info/SOURCES.txt

@@ -1,3 +1,5 @@
+LICENSE
+MANIFEST.in
 README.md
 pyproject.toml
 fakedata/__init__.py

{fakedata_python-2.0.2 → fakedata_python-2.0.4}/pyproject.toml

@@ -4,11 +4,11 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "fakedata-python"
-version = "2.0.2"
+version = "2.0.4"
 authors = [
   { name="abhay557", email="contact@abhaymourya.in" },
 ]
-description = "The fakedata package generates realistic synthetic user profiles for machine learning, deep learning, data analysis, and data science workflows."
+description = "The fakedata package generates realistic user profiles for machine learning, deep learning, data analysis, and data science workflows."
 readme = "README.md"
 license = "MIT"
 requires-python = ">=3.7"