Merge pull request #334 from ua-parser/benchmark-minor-changes

Benchmark minor changes

Author: thyandrecardoso

…rc/main/resources/regexes_@7388149c.yaml …src/main/resources/regexes_@354aebe.yamlmodules/benchmark/src/main/resources/regexes_@7388149c.yaml renamed to modules/benchmark/src/main/resources/regexes_@354aebe.yaml modules/benchmark/src/main/resources/regexes_@7388149c.yaml renamed to modules/benchmark/src/main/resources/regexes_@354aebe.yaml

@@ -148,7 +148,7 @@ user_agent_parsers:
     family_replacement: 'Pinterestbot'
 
   # Bots
-  - regex: '(CSimpleSpider|Cityreview Robot|CrawlDaddy|CrawlFire|Finderbots|Index crawler|Job Roboter|KiwiStatus Spider|Lijit Crawler|QuerySeekerSpider|ScollSpider|Trends Crawler|USyd-NLP-Spider|SiteCat Webbot|BotName\/\$BotVersion|123metaspider-Bot|1470\.net crawler|50\.nu|8bo Crawler Bot|Aboundex|Accoona-[A-z]{1,30}-Agent|AdsBot-Google(?:-[a-z]{1,30}|)|altavista|AppEngine-Google|archive.{0,30}\.org_bot|archiver|Ask Jeeves|[Bb]ai[Dd]u[Ss]pider(?:-[A-Za-z]{1,30})(?:-[A-Za-z]{1,30}|)|bingbot|BingPreview|blitzbot|BlogBridge|Bloglovin|BoardReader Blog Indexer|BoardReader Favicon Fetcher|boitho.com-dc|BotSeer|BUbiNG|\b\w{0,30}favicon\w{0,30}\b|\bYeti(?:-[a-z]{1,30}|)|Catchpoint(?: bot|)|[Cc]harlotte|Checklinks|clumboot|Comodo HTTP\(S\) Crawler|Comodo-Webinspector-Crawler|ConveraCrawler|CRAWL-E|CrawlConvera|Daumoa(?:-feedfetcher|)|Feed Seeker Bot|Feedbin|findlinks|Flamingo_SearchEngine|FollowSite Bot|furlbot|Genieo|gigabot|GomezAgent|gonzo1|(?:[a-zA-Z]{1,30}-|)Googlebot(?:-[a-zA-Z]{1,30}|)|GoogleOther|Google SketchUp|grub-client|gsa-crawler|heritrix|HiddenMarket|holmes|HooWWWer|htdig|ia_archiver|ICC-Crawler|Icarus6j|ichiro(?:/mobile|)|IconSurf|IlTrovatore(?:-Setaccio|)|InfuzApp|Innovazion Crawler|InternetArchive|IP2[a-z]{1,30}Bot|jbot\b|KaloogaBot|Kraken|Kurzor|larbin|LEIA|LesnikBot|Linguee Bot|LinkAider|LinkedInBot|Lite Bot|Llaut|lycos|Mail\.RU_Bot|masscan|masidani_bot|Mediapartners-Google|Microsoft .{0,30} Bot|mogimogi|mozDex|MJ12bot|msnbot(?:-media {0,2}|)|msrbot|Mtps Feed Aggregation System|netresearch|Netvibes|NewsGator[^/]{0,30}|^NING|Nutch[^/]{0,30}|Nymesis|ObjectsSearch|OgScrper|Orbiter|OOZBOT|PagePeeker|PagesInventory|PaxleFramework|Peeplo Screenshot Bot|PHPCrawl|PlantyNet_WebRobot|Pompos|Qwantify|Read%20Later|Reaper|RedCarpet|Retreiver|Riddler|Rival IQ|scooter|Scrapy|Scrubby|searchsight|seekbot|semanticdiscovery|SemrushBot|Simpy|SimplePie|SEOstats|SimpleRSS|SiteCon|Slackbot-LinkExpanding|Slack-ImgProxy|Slurp|snappy|Speedy Spider|Squrl Java|Stringer|TheUsefulbot|ThumbShotsBot|Thumbshots\.ru|Tiny Tiny RSS|Twitterbot|WhatsApp|URL2PNG|Vagabondo|VoilaBot|^vortex|Votay bot|^voyager|WASALive.Bot|Web-sniffer|WebThumb|WeSEE:[A-z]{1,30}|WhatWeb|WIRE|WordPress|Wotbox|www\.almaden\.ibm\.com|Xenu(?:.s|) Link Sleuth|Xerka [A-z]{1,30}Bot|yacy(?:bot|)|YahooSeeker|Yahoo! Slurp|Yandex\w{1,30}|YodaoBot(?:-[A-z]{1,30}|)|YottaaMonitor|Yowedo|^Zao|^Zao-Crawler|ZeBot_www\.ze\.bz|ZooShot|ZyBorg|ArcGIS Hub Indexer|GPTBot)(?:[ /]v?(\d+)(?:\.(\d+)(?:\.(\d+)|)|)|)'
+  - regex: '(CSimpleSpider|Cityreview Robot|CrawlDaddy|CrawlFire|Finderbots|Index crawler|Job Roboter|KiwiStatus Spider|Lijit Crawler|QuerySeekerSpider|ScollSpider|Trends Crawler|USyd-NLP-Spider|SiteCat Webbot|BotName\/\$BotVersion|123metaspider-Bot|1470\.net crawler|50\.nu|8bo Crawler Bot|Aboundex|Accoona-[A-z]{1,30}-Agent|AdsBot-Google(?:-[a-z]{1,30}|)|altavista|AppEngine-Google|archive.{0,30}\.org_bot|archiver|Ask Jeeves|[Bb]ai[Dd]u[Ss]pider(?:-[A-Za-z]{1,30})(?:-[A-Za-z]{1,30}|)|bingbot|BingPreview|blitzbot|BlogBridge|Bloglovin|BoardReader Blog Indexer|BoardReader Favicon Fetcher|boitho.com-dc|BotSeer|BUbiNG|\b\w{0,30}favicon\w{0,30}\b|\bYeti(?:-[a-z]{1,30}|)|Catchpoint(?: bot|)|[Cc]harlotte|Checklinks|clumboot|Comodo HTTP\(S\) Crawler|Comodo-Webinspector-Crawler|ConveraCrawler|CRAWL-E|CrawlConvera|Daumoa(?:-feedfetcher|)|Feed Seeker Bot|Feedbin|findlinks|Flamingo_SearchEngine|FollowSite Bot|furlbot|Genieo|gigabot|GomezAgent|gonzo1|(?:[a-zA-Z]{1,30}-|)Googlebot(?:-[a-zA-Z]{1,30}|)|GoogleOther|Google SketchUp|grub-client|gsa-crawler|heritrix|HiddenMarket|holmes|HooWWWer|htdig|ia_archiver|ICC-Crawler|Icarus6j|ichiro(?:/mobile|)|IconSurf|IlTrovatore(?:-Setaccio|)|InfuzApp|Innovazion Crawler|InternetArchive|IP2[a-z]{1,30}Bot|jbot\b|KaloogaBot|Kraken|Kurzor|larbin|LEIA|LesnikBot|Linguee Bot|LinkAider|LinkedInBot|Lite Bot|Llaut|lycos|Mail\.RU_Bot|masscan|masidani_bot|Mediapartners-Google|Microsoft .{0,30} Bot|mogimogi|mozDex|MJ12bot|msnbot(?:-media {0,2}|)|msrbot|Mtps Feed Aggregation System|netresearch|Netvibes|NewsGator[^/]{0,30}|^NING|Nutch[^/]{0,30}|Nymesis|ObjectsSearch|OgScrper|Orbiter|OOZBOT|PagePeeker|PagesInventory|PaxleFramework|Peeplo Screenshot Bot|PHPCrawl|PlantyNet_WebRobot|Pompos|Qwantify|Read%20Later|Reaper|RedCarpet|Retreiver|Riddler|Rival IQ|scooter|Scrapy|Scrubby|searchsight|seekbot|semanticdiscovery|SemrushBot|Simpy|SimplePie|SEOstats|SimpleRSS|SiteCon|Slackbot-LinkExpanding|Slack-ImgProxy|Slurp|snappy|Speedy Spider|Squrl Java|Stringer|TheUsefulbot|ThumbShotsBot|Thumbshots\.ru|Tiny Tiny RSS|Twitterbot|WhatsApp|URL2PNG|Vagabondo|VoilaBot|^vortex|Votay bot|^voyager|WASALive.Bot|Web-sniffer|WebThumb|WeSEE:[A-z]{1,30}|WhatWeb|WIRE|WordPress|Wotbox|www\.almaden\.ibm\.com|Xenu(?:.s|) Link Sleuth|Xerka [A-z]{1,30}Bot|yacy(?:bot|)|YahooSeeker|Yahoo! Slurp|Yandex\w{1,30}|YodaoBot(?:-[A-z]{1,30}|)|YottaaMonitor|Yowedo|^Zao|^Zao-Crawler|ZeBot_www\.ze\.bz|ZooShot|ZyBorg|ArcGIS Hub Indexer|GPTBot|Google-InspectionTool)(?:[ /]v?(\d+)(?:\.(\d+)(?:\.(\d+)|)|)|)'
 
   # AWS S3 Clients
   # must come before "Bots General matcher" to catch "boto"/"boto3" before "bot"
@@ -443,12 +443,6 @@ user_agent_parsers:
   - regex: '(coc_coc_browser)/(\d+)\.(\d+)(?:\.(\d+)|)'
     family_replacement: 'Coc Coc'
 
-  # Baidu Browsers (desktop spoofs chrome & IE, explorer is mobile)
-  - regex: '(baidubrowser)[/\s](\d+)(?:\.(\d+)|)(?:\.(\d+)|)'
-    family_replacement: 'Baidu Browser'
-  - regex: '(FlyFlow)/(\d+)\.(\d+)'
-    family_replacement: 'Baidu Explorer'
-
   # MxBrowser is Maxthon. Must go before Mobile Chrome for Android
   - regex: '(MxBrowser)/(\d+)\.(\d+)(?:\.(\d+)|)'
     family_replacement: 'Maxthon'
@@ -477,6 +471,12 @@ user_agent_parsers:
   - regex: 'Mozilla.{1,200}Android.{1,200}(GSA)/(\d+)\.(\d+)\.(\d+)'
     family_replacement: 'Google'
 
+  # Baidu Browsers (desktop spoofs chrome & IE, explorer is mobile)
+  - regex: '(baidubrowser)[/\s](\d+)(?:\.(\d+)|)(?:\.(\d+)|)'
+    family_replacement: 'Baidu Browser'
+  - regex: '(FlyFlow|flyflow|baiduboxapp)/(\d+)\.(\d+)(?:\.(\d+)|)(?:\.(\d+)|)'
+    family_replacement: 'Baidu Explorer'
+
   # QQ Browsers
   - regex: '(MQQBrowser/Mini)(?:(\d+)(?:\.(\d+)|)(?:\.(\d+)|)|)'
     family_replacement: 'QQ Browser Mini'
@@ -506,10 +506,22 @@ user_agent_parsers:
     family_replacement: 'Ecosia Android'
 
   # VivoBrowser
-  - regex: '(VivoBrowser)\/(\d+)\.(\d+)\.(\d+)\.(\d+)'
+  - regex: '(VivoBrowser)\/(\d+)\.(\d+)\.(\d+)(?:\.(\d+)|)'
 
   # HiBrowser
-  - regex: '(HiBrowser)\/v(\d+)\.(\d+)\.(\d+)\.(\d+)'
+  - regex: '(H[Ii]Browser)\/v(\d+)\.(\d+)\.(\d+)\.(\d+)'
+    family_replacement: 'HiBrowser'
+
+  # Honor Browser
+  - regex: '(bdhonorbrowser)/(\d+)\.(\d+)\.(\d+)\.(\d+)'
+    family_replacement: 'Honor Browser'
+
+  # Weibo
+  # Must before Chrome Mobile WebView
+  - regex: '(weibo)__(\d+)\.(\d+)\.(\d+)'
+    family_replacement: 'Weibo'
+  - regex: '(WeiboliteiOS|WeiboIntliOS)'
+    family_replacement: 'Weibo'
 
   # Chrome Mobile
   - regex: 'Version/.{1,300}(Chrome)/(\d+)\.(\d+)\.(\d+)\.(\d+)'
@@ -592,7 +604,7 @@ user_agent_parsers:
   - regex: '(115Browser)/(\d+)\.(\d+)\.(\d+)\.(\d+)'
     family_replacement: '115 Browser'
 
- # Avira
+  # Avira
   - regex: '(Avira)/(\d+)\.(\d+)\.(\d+)\.(\d+)'
     family_replacement: 'Avira'
 
@@ -614,7 +626,7 @@ user_agent_parsers:
   # Smart Lenovo Browser
   - regex: '(SLBrowser)/(\d+)\.(\d+)\.(\d+)\.(\d+) SLBChan/(\d+)'
     family_replacement: 'Smart Lenovo Browser'
-  
+
   # Atom Browser
   - regex: '(Atom)/(\d+)\.(\d+)\.(\d+)\.(\d+)'
     family_replacement: 'Atom Browser'
@@ -667,6 +679,23 @@ user_agent_parsers:
   - regex: '(JiSu)/(\d+)\.(\d+)\.(\d+)'
     family_replacement: 'JiSu Browser'
 
+  # Wolvic Browser
+  - regex: '(Wolvic)/(\d+)\.(\d+)\.(\d+)'
+    family_replacement: 'Wolvic Browser'
+
+  # HeyTap Browser
+  - regex: '(HeyTapBrowser)/(\d+)\.(\d+)\.(\d+)\.(\d+)'
+    family_replacement: 'HeyTap Browser'
+
+  # SmartTV WebBrowser
+  - regex: '(Thano)/(\d+)\.(\d+)'
+    family_replacement: 'SmartTV WebBrowser'
+
+  # WeChat Browser
+  - regex: '(MicroMessenger)/(\d+)\.(\d+)\.(\d+)\.(\d+)'
+    family_replacement: 'WeChat Browser'
+
+
   #### END SPECIAL CASES TOP ####
 
   #### MAIN CASES - this catches > 50% of all browsers ####
@@ -1485,7 +1514,7 @@ os_parsers:
   # Box Drive and Box Sync on Mac OS X use OSX version numbers, not Darwin
   - regex: '^Box.{0,200};(Darwin)/(10)\.(1\d)(?:\.(\d+)|)'
     os_replacement: 'Mac OS X'
-  
+
   ##########
   # Hashicorp API
   # APN/1.0 HashiCorp/1.0 Terraform/1.8.0 (+https://www.terraform.io) terraform-provider-aws/4.67.0 (+https://registry.terraform.io/providers/hashicorp/aws) aws-sdk-go/1.44.261 (go1.19.8; darwin; arm64)
@@ -2013,7 +2042,7 @@ device_parsers:
   # Mobile Spiders
   # Catch the mobile crawler before checking for iPhones / Androids.
   #########
-  - regex: '^.{0,100}?(?:(?:iPhone|Windows CE|Windows Phone|Android).{0,300}(?:(?:Bot|Yeti)-Mobile|YRSpider|BingPreview|bots?/\d|(?:bot|spider)\.html)|AdsBot-Google-Mobile.{0,200}iPhone)'
+  - regex: '^.{0,100}?(?:(?:iPhone|Windows CE|Windows Phone|Android).{0,300}(?:(?:Bot|Yeti)-Mobile|YRSpider|BingPreview|bots?/\d|(?:bot|spider)\.html|Google-InspectionTool)|AdsBot-Google-Mobile.{0,200}iPhone)'
     regex_flag: 'i'
     device_replacement: 'Spider'
     brand_replacement: 'Spider'
@@ -3215,7 +3244,7 @@ device_parsers:
     device_replacement: 'HTC $1'
     brand_replacement: 'HTC'
     model_replacement: '$1'
-  - regex: '; {0,2}(ADR6200|ADR6400L|ADR6425LVW|Amaze|DesireS?|EndeavorU|Eris|EVO|Evo\d[A-Z]+|HD2|IncredibleS?|Inspire[A-Z0-9]*|Inspire[A-Z0-9]*|Sensation[A-Z0-9]*|Wildfire)[ _-](.{1,200}?)(?:[/;\)]|Build|MIUI|1\.0)'
+  - regex: '; {0,2}(ADR6200|ADR6400L|ADR6425LVW|Amaze|DesireS?|EndeavorU|Eris|EVO|Evo\d[A-Z]+|HD2|IncredibleS?|Inspire[A-Z0-9]*|Sensation[A-Z0-9]*|Wildfire)[ _-](.{1,200}?)(?:[/;\)]|Build|MIUI|1\.0)'
     regex_flag: 'i'
     device_replacement: 'HTC $1 $2'
     brand_replacement: 'HTC'
@@ -5619,7 +5648,6 @@ device_parsers:
     brand_replacement: 'Asus'
     model_replacement: '$1'
 
-
   ##########
   # Bird
   ##########
@@ -5819,7 +5847,6 @@ device_parsers:
     brand_replacement: 'Motorola'
     model_replacement: '$2'
 
-
   ##########
   # nintendo
   ##########
@@ -6023,7 +6050,7 @@ device_parsers:
   ##########
   # Spiders (this is a hack...)
   ##########
-  - regex: '^.{0,100}(bot|BUbiNG|zao|borg|DBot|oegp|silk|Xenu|zeal|^NING|CCBot|crawl|htdig|lycos|slurp|teoma|voila|yahoo|Sogou|CiBra|Nutch|^Java/|^JNLP/|Daumoa|Daum|Genieo|ichiro|larbin|pompos|Scrapy|snappy|speedy|spider|msnbot|msrbot|vortex|^vortex|crawler|favicon|indexer|Riddler|scooter|scraper|scrubby|WhatWeb|WinHTTP|bingbot|BingPreview|openbot|gigabot|furlbot|polybot|seekbot|^voyager|archiver|Icarus6j|mogimogi|Netvibes|blitzbot|altavista|charlotte|findlinks|Retreiver|TLSProber|WordPress|SeznamBot|ProoXiBot|wsr\-agent|Squrl Java|EtaoSpider|PaperLiBot|SputnikBot|A6\-Indexer|netresearch|searchsight|baiduspider|YisouSpider|ICC\-Crawler|http%20client|Python-urllib|dataparksearch|converacrawler|Screaming Frog|AppEngine-Google|YahooCacheSystem|fast\-webcrawler|Sogou Pic Spider|semanticdiscovery|Innovazion Crawler|facebookexternalhit|Google.{0,200}/\+/web/snippet|Google-HTTP-Java-Client|BlogBridge|IlTrovatore-Setaccio|InternetArchive|GomezAgent|WebThumbnail|heritrix|NewsGator|PagePeeker|Reaper|ZooShot|holmes|NL-Crawler|Pingdom|StatusCake|WhatsApp|masscan|Google Web Preview|Qwantify|Yeti|OgScrper|RecipeRadar|GPTBot)'
+  - regex: '^.{0,100}(bot|BUbiNG|zao|borg|DBot|oegp|silk|Xenu|zeal|^NING|CCBot|crawl|htdig|lycos|slurp|teoma|voila|yahoo|Sogou|CiBra|Nutch|^Java/|^JNLP/|Daumoa|Daum|Genieo|ichiro|larbin|pompos|Scrapy|snappy|speedy|spider|msnbot|msrbot|vortex|^vortex|crawler|favicon|indexer|Riddler|scooter|scraper|scrubby|WhatWeb|WinHTTP|bingbot|BingPreview|openbot|gigabot|furlbot|polybot|seekbot|^voyager|archiver|Icarus6j|mogimogi|Netvibes|blitzbot|altavista|charlotte|findlinks|Retreiver|TLSProber|WordPress|SeznamBot|ProoXiBot|wsr\-agent|Squrl Java|EtaoSpider|PaperLiBot|SputnikBot|A6\-Indexer|netresearch|searchsight|baiduspider|YisouSpider|ICC\-Crawler|http%20client|Python-urllib|dataparksearch|converacrawler|Screaming Frog|AppEngine-Google|YahooCacheSystem|fast\-webcrawler|Sogou Pic Spider|semanticdiscovery|Innovazion Crawler|facebookexternalhit|Google.{0,200}/\+/web/snippet|Google-HTTP-Java-Client|BlogBridge|IlTrovatore-Setaccio|InternetArchive|GomezAgent|WebThumbnail|heritrix|NewsGator|PagePeeker|Reaper|ZooShot|holmes|NL-Crawler|Pingdom|StatusCake|WhatsApp|masscan|Google Web Preview|Qwantify|Yeti|OgScrper|RecipeRadar|GPTBot|Google-InspectionTool)'
     regex_flag: 'i'
     device_replacement: 'Spider'
     brand_replacement: 'Spider'

modules/benchmark/src/main/scala/org/uaparser/scala/benchmark/BenchmarkSupport.scala

@@ -0,0 +1,29 @@
+package org.uaparser.scala.benchmark
+
+import scala.io.Source
+
+import org.uaparser.scala.Parser
+
+object BenchmarkSupport {
+
+  def loadLinesFromResource(resourceName: String): Array[String] = {
+    val source = Source.fromResource(resourceName)
+    try
+      source.getLines().map(_.trim).filter(_.nonEmpty).toArray
+    finally
+      source.close()
+  }
+
+  def loadParserForPinnedRegexes(): Parser = {
+    val inputStream = getClass.getClassLoader.getResourceAsStream("regexes_@354aebe.yaml")
+    try
+      Parser.fromInputStream(inputStream).get
+    finally
+      inputStream.close()
+  }
+
+  @inline def incrementIndex(i: Int, n: Int): Int = {
+    val j = i + 1
+    if (j == n) 0 else j
+  }
+}

modules/benchmark/src/main/scala/org/uaparser/scala/benchmark/UapScalaAllBenchmarks.scala

@@ -1,34 +1,47 @@
 package org.uaparser.scala.benchmark
 
-import scala.io.Source
+import java.util.concurrent.TimeUnit
 
-import org.openjdk.jmh.annotations.{Benchmark, Scope, State}
+import org.openjdk.jmh.annotations.*
 import org.openjdk.jmh.infra.Blackhole
 
 import org.uaparser.scala.Parser
 
-@State(Scope.Benchmark)
+@BenchmarkMode(Array(Mode.Throughput))
+@OutputTimeUnit(TimeUnit.SECONDS)
+@State(Scope.Thread)
 class UapScalaAllBenchmarks {
 
-  // an entire bundle of strings taken from the current suite of tests
-  val allUserAgentStrings: List[String] = Source.fromResource("all-user-agents.txt").getLines().toList
+  private var parser: Parser = _
+  private var uas: Array[String] = _
+  private var idx: Int = 0
 
-  var parser: Parser =
-    Parser.fromInputStream(Thread.currentThread.getContextClassLoader.getResourceAsStream("regexes_@7388149c.yaml")).get
+  @Setup(Level.Trial)
+  def setup(): Unit = {
+    parser = BenchmarkSupport.loadParserForPinnedRegexes()
+    uas = BenchmarkSupport.loadLinesFromResource("all-user-agents.txt")
+    idx = 0
+  }
 
-  @Benchmark
-  def measureAllStrDeviceParser(bh: Blackhole): Unit =
-    allUserAgentStrings.foreach(s => bh.consume(parser.deviceParser.parse(s)))
+  @Setup(Level.Iteration)
+  def resetCursor(): Unit =
+    idx = 0
 
-  @Benchmark
-  def measureAllStrOsParser(bh: Blackhole): Unit =
-    allUserAgentStrings.foreach(s => bh.consume(parser.osParser.parse(s)))
+  @inline private def nextUA(): String = {
+    val s = uas(idx)
+    idx = BenchmarkSupport.incrementIndex(idx, uas.length)
+    s
+  }
 
-  @Benchmark
-  def measureAllStrUserAgentParser(bh: Blackhole): Unit =
-    allUserAgentStrings.foreach(s => bh.consume(parser.userAgentParser.parse(s)))
+  @Benchmark def allDevice(bh: Blackhole): Unit =
+    bh.consume(parser.deviceParser.parse(nextUA()))
 
-  @Benchmark
-  def measureAllStrAllParser(bh: Blackhole): Unit =
-    allUserAgentStrings.foreach(s => bh.consume(parser.parse(s)))
+  @Benchmark def allOs(bh: Blackhole): Unit =
+    bh.consume(parser.osParser.parse(nextUA()))
+
+  @Benchmark def allUserAgent(bh: Blackhole): Unit =
+    bh.consume(parser.userAgentParser.parse(nextUA()))
+
+  @Benchmark def allAll(bh: Blackhole): Unit =
+    bh.consume(parser.parse(nextUA()))
 }

modules/benchmark/src/main/scala/org/uaparser/scala/benchmark/UapScalaSampleBenchmarks.scala

@@ -1,34 +1,46 @@
 package org.uaparser.scala.benchmark
 
-import scala.io.Source
+import java.util.concurrent.TimeUnit
 
-import org.openjdk.jmh.annotations.{Benchmark, Scope, State}
+import org.openjdk.jmh.annotations.*
 import org.openjdk.jmh.infra.Blackhole
 
 import org.uaparser.scala.Parser
 
-@State(Scope.Benchmark)
+@BenchmarkMode(Array(Mode.Throughput))
+@OutputTimeUnit(TimeUnit.SECONDS)
+@State(Scope.Thread)
 class UapScalaSampleBenchmarks {
-
-  // an entire bundle of strings taken from the current suite of tests
-  val sampleUserAgentStrings: List[String] = Source.fromResource("sample-user-agents.txt").getLines().toList
-
-  var parser: Parser =
-    Parser.fromInputStream(Thread.currentThread.getContextClassLoader.getResourceAsStream("regexes_@7388149c.yaml")).get
-
-  @Benchmark
-  def measureSampleStrDeviceParser(bh: Blackhole): Unit =
-    sampleUserAgentStrings.foreach(s => bh.consume(parser.deviceParser.parse(s)))
-
-  @Benchmark
-  def measureSampleStrOsParser(bh: Blackhole): Unit =
-    sampleUserAgentStrings.foreach(s => bh.consume(parser.osParser.parse(s)))
-
-  @Benchmark
-  def measureSampleStrUserAgentParser(bh: Blackhole): Unit =
-    sampleUserAgentStrings.foreach(s => bh.consume(parser.userAgentParser.parse(s)))
-
-  @Benchmark
-  def measureSampleStrAllParser(bh: Blackhole): Unit =
-    sampleUserAgentStrings.foreach(s => bh.consume(parser.parse(s)))
+  private var parser: Parser = _
+  private var uas: Array[String] = _
+  private var idx: Int = 0
+
+  @Setup(Level.Trial)
+  def setup(): Unit = {
+    parser = BenchmarkSupport.loadParserForPinnedRegexes()
+    uas = BenchmarkSupport.loadLinesFromResource("sample-user-agents.txt")
+    idx = 0
+  }
+
+  @Setup(Level.Iteration)
+  def resetCursor(): Unit =
+    idx = 0
+
+  @inline private def nextUA(): String = {
+    val s = uas(idx)
+    idx = BenchmarkSupport.incrementIndex(idx, uas.length)
+    s
+  }
+
+  @Benchmark def sampleDevice(bh: Blackhole): Unit =
+    bh.consume(parser.deviceParser.parse(nextUA()))
+
+  @Benchmark def sampleOs(bh: Blackhole): Unit =
+    bh.consume(parser.osParser.parse(nextUA()))
+
+  @Benchmark def sampleUserAgent(bh: Blackhole): Unit =
+    bh.consume(parser.userAgentParser.parse(nextUA()))
+
+  @Benchmark def sampleAll(bh: Blackhole): Unit =
+    bh.consume(parser.parse(nextUA()))
 }