Changelog
Version 1.11 - Worldwide Job Search Support (December 17, 2024)
Major Enhancements
Worldwide Location Support
Added intelligent fallback system to search jobs globally when specific locations return no results
Location input now supports:
Specific locations: "bangalore", "mumbai", "delhi"
All locations: "all", "worldwide", or leave empty
Automatic fallback: If a specific location returns 0 jobs, automatically tries searching all locations
Technical Changes
Search Strategy
Implemented searchAttempts array to track multiple search URL variations
Added fallback logic in Phase 1 crawler:
First attempts location-specific search
If no results on first page, automatically tries broader search without location filter
Enhanced logging to show which search variation is being executed
Input Schema Updates
Updated location field description to clarify worldwide support
Changed default location from empty to "all" for better user experience
Added note about AmbitionBox being primarily India-focused
URL Building
Support for both location-specific URLs: jobs/{role}-jobs-in-{location}-prfloc
Support for worldwide URLs: jobs/{role}-jobs-prf
Automatic construction of role slug for URLs
User Benefits
No Configuration Needed : Users can simply set location to "all" to get worldwide resultsIntelligent Fallback : If searching for a location with no jobs, automatically expands searchBetter Visibility : Enhanced logging shows which search variation is activePreserved Functionality : All existing features (Naukri integration, company info, benefits) work unchanged
Usage Examples
{
"role" : "software engineer" ,
"location" : "all" ,
"maxJobs" : 50
}
{
"role" : "data scientist" ,
"location" : "bangalore" ,
"maxJobs" : 30
}
{
"role" : "python developer" ,
"location" : "worldwide" ,
"maxJobs" : 100
}
Previous Versions
Version 1.10 - Debugging Enhancements
Added comprehensive debugging for NUXT data structure
Improved error messages when no jobs found
Version 1.9 - Request Queue Fix
Changed from addRequests() + run() to run(detailRequests) pattern
Ensures all jobs are processed through Phase 2
Fixed about_this_role extraction using page.evaluate()
Fixed company_info extraction using page.evaluate()
Replaced cheerio with DOM queries for dynamic content
Version 1.7 - Request Limit Fix
Added maxRequestsPerCrawl: totalJobs to Phase 2 crawler
Ensures all extracted jobs are processed (not just 10)
Version 1.6 - Logic Bug Fix
Fixed inverted conditional logic in Naukri URL extraction
Moved Naukri extraction from else to if (naukriUrl) block
Changed from query param extraction to path-based extraction
Pattern: -(naukri|monster|indeed)_(\d+)-jdp
Version 1.3 - Company Slug Regex
Fixed company name extraction from job URLs
New pattern: /\/jobs\/[^/]+-in-([^/]+)-(?:naukri_|monster_|indeed_)[\d]+-jdp/
