https://github.com/semanticdata/regex-address-extractor
Simple webapp to extract addresses using regex. Removes duplicates and invalid addresses.
https://github.com/semanticdata/regex-address-extractor
address-parser addresses html-css-javascript regex
Last synced: about 1 year ago
JSON representation
Simple webapp to extract addresses using regex. Removes duplicates and invalid addresses.
- Host: GitHub
- URL: https://github.com/semanticdata/regex-address-extractor
- Owner: semanticdata
- License: mit
- Created: 2024-10-18T15:32:33.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2025-01-22T19:37:12.000Z (about 1 year ago)
- Last Synced: 2025-02-01T12:46:15.807Z (about 1 year ago)
- Topics: address-parser, addresses, html-css-javascript, regex
- Language: HTML
- Homepage: https://semanticdata.github.io/regex-address-extractor/
- Size: 79.1 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Regex Address Extractor
A web-based tool that extracts street addresses from any text input. This tool uses regular expressions to identify and extract US street addresses, making it useful for parsing addresses from emails, documents, or any text content.
## Features
- Extract addresses from any pasted text
- Supports various street types (Avenue, Street, Road, etc.)
- Handles directional suffixes (North, South, East, West)
- Supports hyphenated building numbers
- Alphabetically sorts extracted addresses
- One-click copy of all extracted addresses
- Visual feedback for all actions
- Mobile-responsive design
## How It Works
The tool preprocesses the input text by normalizing common address separators (converting them to newlines):
- "and" between addresses
- Commas
- Semicolons
Then it uses an enhanced regular expression pattern to identify common US street address formats:
```regex
\d+(?:-\d+)?\s+(?:[A-Za-z0-9\.']+\s?)+(?:Avenue|Ave|Street|St|Drive|Dr|Road|Rd|Boulevard|Blvd|Lane|Ln|Place|Pl|Court|Ct|Circle|Cir|Way|Parkway|Pkwy|Highway|Hwy)(?:\s+[A-Za-z]+)?(?:\s+(?:North|South|East|West|N|S|E|W))?
```
The pattern matches:
- Street numbers (including hyphenated ranges like "123-125")
- Street names (including those with periods and apostrophes)
- Common street type abbreviations and full names
- Optional directional suffixes
- Various street types:
- Avenue/Ave
- Street/St
- Drive/Dr
- Road/Rd
- Boulevard/Blvd
- Lane/Ln
- Place/Pl
- Court/Ct
- Circle/Cir
- Way
- Parkway/Pkwy
- Highway/Hwy
## Usage
1. Open the HTML file in any modern web browser
2. Paste your text containing addresses into the input area
3. Click "Extract Addresses" to process the text
4. View the extracted addresses in alphabetical order
5. Use the "Copy Addresses" button to copy all extracted addresses to your clipboard
## Example Input
```plaintext
Here's our new office at 4741 Welcome Ave N, right next to the park.
The delivery should go to 6124 Lakeland Ave.
Other locations include 4856 Maryland Ave N and 3002 Jersey Ave N.
Meeting locations: 3517 Perry Ave N, 5909 29th Pl N
Warehouse: 7100 Medicine Lake Rd
Branch offices:
- 3344 Adair Ave N
- 6118 34th Ave N
- 3617 Vera Cruz Ave N
```
```plaintext
4741 Welcome Ave N
6124 Lakeland Ave
4856 Maryland Ave N
3002 Jersey Ave N
3517 Perry Ave N
5909 29th Pl N
7100 Medicine Lake Rd
3344 Adair Ave N
6118 34th Ave N
3617 Vera Cruz Ave N
```
## Technical Details
- Pure JavaScript implementation (no dependencies)
- Uses the Clipboard API for modern copy functionality
- Responsive CSS design
- Error handling for invalid inputs
- User feedback system for all actions
## Development
To modify the address matching patterns:
1. Edit the `addressRegex` constant in the JavaScript code
2. Test new patterns on [regex101](https://regex101.com/)
3. Consider adding new street types or address formats as needed
## Testing
The project includes a comprehensive test suite using Jest. The tests cover:
- Basic address extraction (single and multiple addresses)
- Address variations (different street types, hyphenated numbers)
- Directional suffixes (North, South, East, West, N, S, E, W)
- Edge cases (addresses with periods, apostrophes)
- Input validation (handling invalid addresses)
- Separator handling (commas, "and", semicolons)
- Performance testing with large inputs
To run the tests:
```bash
# Install dependencies
npm install
# Run tests once
npm test
# Run tests in watch mode
npm run test:watch
```
### Test Categories
1. **Basic Functionality Tests**
- Single address extraction
- Multiple comma-separated addresses
- Different street type recognition
2. **Address Variations**
- Support for 11 different street types
- Hyphenated building numbers (e.g., "123-125")
- Directional suffixes in both full (North) and abbreviated (N) forms
3. **Edge Cases**
- Addresses with periods (e.g., "St. John's Street")
- Mixed valid and invalid addresses
- Invalid address formats
- Empty or malformed inputs
4. **Performance**
- Processing of 1000 addresses under 1 second
- Memory efficient address extraction
## License
Source code in this repository is available under the [MIT License](./LICENSE).