# Changelog of Decision Maker Name & Email Extractor (`dominic-quaiser/decision-maker-name-email-extractor`) Actor

- **URL**: https://apify.com/dominic-quaiser/decision-maker-name-email-extractor/changelog.md
- **Full Actor documentation**: https://apify.com/dominic-quaiser/decision-maker-name-email-extractor.md

## Changelog

### \[0.9.0—beta] — 2026-07-03

#### Changed

- **External NER API**: Name recognition now uses the external NER API (configured via the `NER_API_KEY` and `NER_API_BASE_URL` environment variables) instead of a locally loaded Hugging Face model. This removes the `spacy`, `transformers`, and `torch` dependencies, making the image much smaller and startup much faster.

#### Fixed

- Pinned `pydantic < 2.12` and `browserforge < 1.2.4`: the crawlee 0.6 line pulled in by `apify < 3.0` crashes at import time with the latest releases of both.
- Modernized the boilerplate spider/downloader middlewares for Scrapy 2.13+: `process_spider_output` is now an async generator (required by the current engine), the deprecated `spider` arguments and the no-longer-called `process_start_requests` were removed.

### \[0.8.0—beta] — 2025-07-23

#### Added

- Added name validation through CSS classes.

### \[0.7.0—beta] — 2025-07-23

#### Added

- **Custom Request Headers**: Added `defaultRequestHeaders` input to allow custom HTTP headers for all requests
- **Language Filtering**: Added `languageCountryCode` input to specify language/country codes for URL filtering
- **Improved Error Handling**: Added more robust error handling throughout the pipeline

#### Changed

- **Code Refactoring**: Restructured codebase for better maintainability and clarity
- **Name Extraction Improvements**: Enhanced name validation and extraction logic to reduce false positives
- **Email Matching Optimization**: Improved accuracy of email-person matching algorithm
- **URL Processing**: Optimized URL processing and domain extraction logic
- **HTML Parsing**: Enhanced BeautifulSoup parsing with better context detection

### \[0.6.0—alpha] — 2025-07-19

#### Added

- New user inputs for `explorationMode`, `depth`, `pagecount`, `priorityLinkKeywords`, `useCstomJobTitles` and `customJobTitles`.

#### Chnaged

- Improvements to the name extraction and validation.

### \[0.5.1—alpha] — 2025-07-15

#### Fixed

- **Minor Bug Fixes**: Addressed issues in name extraction logic to ensure more consistent and accurate results.

### \[0.5.0—alpha] — 2025-07-13

#### Added

- **New Input Settings**: Introduced `jobTitles` and `obeyRobotstxt` settings to enhance customization and compliance.
  - `jobTitles`: Allows users to specify target job titles for more precise extraction.
  - `obeyRobotstxt`: Ensures adherence to website robots.txt files for ethical crawling.

#### Changed

- **Updated Job Title Lists**: Refined and expanded the internal job title database to improve accuracy and relevance of extracted decision-maker information.

#### Fixed

- **Extraction and Processing Improvements**: Resolved multiple bugs affecting data extraction and processing, resulting in more reliable and accurate outputs.

### \[0.4.0—alpha] — 2025-07-11

#### Changed

- **Enhanced Name Processing**: Optimized the handling of extracted names to improve accuracy, reduce false positives, and better support complex name formats.

#### Fixed

- **Minor Bug Fixes**: Addressed issues in name extraction logic to ensure more consistent and accurate results.

### \[0.3.0—alpha] — 2025-07-11

#### Changed:

- Expanded FALSE\_POSITIVE\_SELECTORS to include broader matching for testimonial-like content, including elements with class or ID containing "customer", "client", "review", "quote", "press", and related keywords.

### \[0.2.0—alpha] — 2025-07-09

#### Changed:

- Replaced the NER model with a slightly smaller one from Hugging Face. This new model offers better performance and improved support for multilingual names, paving the way for future internationalization features.

#### Added:

- Support for bulk URL input. Each URL is processed independently.

### \[0.1.0—alpha] — 2025-07-08

#### Added:

- Initial alpha release of the **Decision Maker Name & Email Extractor**.