⚡ Optimize repeated Regex compilation in string replacement utilities

app/src/main/java/io/aatricks/novelscraper/data/repository/SummaryService.kt

@@ -21,7 +21,11 @@ class SummaryService @Inject constructor(
     @ApplicationContext private val context: Context
 ) {
 
-    private val TAG = "SummaryService"
+    companion object {
+        private const val TAG = "SummaryService"
+        private val WHITESPACE_REGEX = Regex("\\s+")
+    }
+
     private var modelFile: File? = null
     private var isInitialized = false
     private var isInitializing = false
@@ -79,7 +83,7 @@ class SummaryService @Inject constructor(
             val selectedContent = selectKeyContent(content, maxWords = 300)
             val prompt = buildPrompt(chapterTitle, selectedContent)
 
-            Log.d(TAG, "Generating summary (${selectedContent.split(Regex("\\s+")).size} words, ~${(selectedContent.length + prompt.length) / 4 + 200} tokens)")
+            Log.d(TAG, "Generating summary (${selectedContent.split(WHITESPACE_REGEX).size} words, ~${(selectedContent.length + prompt.length) / 4 + 200} tokens)")
 
             generateWithRetry(prompt, selectedContent, content, onProgress)
         }.onFailure { e ->
@@ -151,7 +155,7 @@ class SummaryService @Inject constructor(
     private fun selectKeyContent(content: List<String>, maxWords: Int): String {
         if (content.isEmpty()) return ""
 
-        val wordsPerParagraph = content.map { it.split(Regex("\\s+")) }
+        val wordsPerParagraph = content.map { it.split(WHITESPACE_REGEX) }
         val totalWords = wordsPerParagraph.sumOf { it.size }
         if (totalWords <= maxWords) return content.joinToString("\n\n")
 

app/src/main/java/io/aatricks/novelscraper/data/repository/source/BaseJsoupSource.kt

@@ -13,6 +13,11 @@ abstract class BaseJsoupSource(
     protected open val preferencesManager: PreferencesManager? = null,
     protected open val okHttpClient: okhttp3.OkHttpClient? = null
 ) : NovelSource {
+
+    companion object {
+        private val MULTIPLE_SLASHES_REGEX = Regex("/+")
+    }
+
     protected open val userAgent = "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36"
     protected open val timeout = 15000L
 
@@ -73,6 +78,6 @@ abstract class BaseJsoupSource(
             path.startsWith("//") -> "https:$path"
             path.startsWith("/") -> "$baseUrl$path"
             else -> if (path.startsWith(baseUrl)) path else "$baseUrl/$path"
-        }.replace(Regex("/+"), "/").replace("https:/", "https://").replace("http:/", "http://")
+        }.replace(MULTIPLE_SLASHES_REGEX, "/").replace("https:/", "https://").replace("http:/", "http://")
     }
 }

app/src/main/java/io/aatricks/novelscraper/data/repository/source/MangaBatSource.kt

@@ -14,6 +14,10 @@ class MangaBatSource @Inject constructor(
     override val name = "MangaBat"
     override val baseUrl = "https://www.mangabats.com"
 
+    companion object {
+        private val SUMMARY_REGEX = Regex(".*summary: ", RegexOption.IGNORE_CASE)
+    }
+
     override suspend fun getPopularNovels(page: Int, tags: List<String>): List<ExploreItem> = io {
         val url = if (tags.isNotEmpty()) {
             val tagSlug = tags.first().lowercase().replace(" ", "-")
@@ -89,7 +93,7 @@ class MangaBatSource @Inject constructor(
         val author = extractAuthor(document)
         val summary = document.select("#contentBox, .panel-story-info-description, .story-info-description")
             .first()?.text()?.replace("Description :", "")
-            ?.replace(Regex(".*summary: ", RegexOption.IGNORE_CASE), "")?.trim()
+            ?.replace(SUMMARY_REGEX, "")?.trim()
 
         val coverUrl = extractCoverUrl(document)
 

app/src/main/java/io/aatricks/novelscraper/data/repository/source/NovelFireSource.kt

@@ -21,14 +21,25 @@ class NovelFireSource @Inject constructor(
     override val name = "NovelFire"
     override val baseUrl = "https://novelfire.net"
 
+    companion object {
+        private val BRACKET_NUMBER_REGEX = Regex("^\\[\\d+\\]\\s*")
+        private val R_NUMBER_REGEX = Regex("^R\\s*\\d+(\\.\\d+)?\\s*")
+        private val RANK_PREFIX_REGEX = Regex("^Rank\\s*\\d+\\s*", RegexOption.IGNORE_CASE)
+        private val RANK_REGEX = Regex("RANK\\s+(\\d+)", RegexOption.IGNORE_CASE)
+        private val RATING_REGEX = Regex("Average score is\\s+([0-9.]+)", RegexOption.IGNORE_CASE)
+        private val CHAPTERS_COUNT_REGEX = Regex("(\\d+)\\s*Chapters", RegexOption.IGNORE_CASE)
+        private val TIME_AGO_REGEX = Regex("\\d+\\s+(year|month|day|hour|minute|second)s?\\s+ago.*$")
+        private val LEADING_NUM_REGEX = Regex("^(\\d+)\\s+(Chapter\\s+\\1.*)")
+    }
+
     private fun cleanNovelTitle(title: String): String {
         var clean = title
         // Remove [123] at start
-        clean = clean.replace(Regex("^\\[\\d+\\]\\s*"), "")
+        clean = clean.replace(BRACKET_NUMBER_REGEX, "")
         // Remove R 14.8 or R 123 at start
-        clean = clean.replace(Regex("^R\\s*\\d+(\\.\\d+)?\\s*"), "")
+        clean = clean.replace(R_NUMBER_REGEX, "")
         // Remove Rank 123 at start
-        clean = clean.replace(Regex("^Rank\\s*\\d+\\s*", RegexOption.IGNORE_CASE), "")
+        clean = clean.replace(RANK_PREFIX_REGEX, "")
         return clean.trim()
     }
 
@@ -155,8 +166,8 @@ class NovelFireSource @Inject constructor(
 
         val infoText = document.text()
         val chapterCount = extractChapterCount(infoText)
-        val rank = Regex("RANK\\s+(\\d+)", RegexOption.IGNORE_CASE).find(infoText)?.groupValues?.get(1)
-        val rating = Regex("Average score is\\s+([0-9.]+)", RegexOption.IGNORE_CASE).find(infoText)?.groupValues?.get(1)
+        val rank = RANK_REGEX.find(infoText)?.groupValues?.get(1)
+        val rating = RATING_REGEX.find(infoText)?.groupValues?.get(1)
 
         val chaptersUrl = getChaptersUrl(url, document)
         val firstPageDoc = runCatching { getDocument(chaptersUrl) }.getOrDefault(document)
@@ -195,8 +206,7 @@ class NovelFireSource @Inject constructor(
     }
 
     private fun extractChapterCount(infoText: String): Int {
-        return Regex("(\\d+)\\s*Chapters", RegexOption.IGNORE_CASE)
-            .find(infoText)?.groupValues?.get(1)?.toIntOrNull() ?: 0
+        return CHAPTERS_COUNT_REGEX.find(infoText)?.groupValues?.get(1)?.toIntOrNull() ?: 0
     }
 
     private fun getChaptersUrl(url: String, document: org.jsoup.nodes.Document): String {
@@ -217,9 +227,8 @@ class NovelFireSource @Inject constructor(
                 element.select(".chapter-title").text().ifBlank { element.text() }
             }
 
-            var cleanTitle = rawTitle.replace(Regex("\\d+\\s+(year|month|day|hour|minute|second)s?\\s+ago.*$"), "").trim()
-            val leadingNumRegex = Regex("^(\\d+)\\s+(Chapter\\s+\\1.*)")
-            leadingNumRegex.find(cleanTitle)?.let { match ->
+            var cleanTitle = rawTitle.replace(TIME_AGO_REGEX, "").trim()
+            LEADING_NUM_REGEX.find(cleanTitle)?.let { match ->
                 cleanTitle = match.groupValues[2]
             }
 

app/src/main/java/io/aatricks/novelscraper/ui/viewmodel/ReaderViewModel.kt

@@ -33,6 +33,10 @@ class ReaderViewModel @Inject constructor(
     private val preferencesManager: PreferencesManager
 ) : BaseViewModel<ReaderViewModel.ReaderUiState>(ReaderUiState()) {
 
+    companion object {
+        private val DOUBLE_NEWLINE_REGEX = Regex("""\n\s*\n""")
+    }
+
     // Current library item ID being read
     private var currentLibraryItemId: String? = null
 
@@ -613,7 +617,7 @@ class ReaderViewModel @Inject constructor(
             if (textBuffer.isEmpty()) return
             val joined = textBuffer.joinToString("\n\n")
             val formatted = TextUtils.formatChapterText(joined)
-            val parts = formatted.split(Regex("""\n\s*\n""")).map { it.trim() }.filter { it.isNotBlank() }
+            val parts = formatted.split(DOUBLE_NEWLINE_REGEX).map { it.trim() }.filter { it.isNotBlank() }
             parts.forEach { p -> formattedElements.add(ContentElement.Text(p)) }
             textBuffer.clear()
         }

app/src/test/java/io/aatricks/novelscraper/data/repository/source/NovelFireSourceBenchmarkTest.kt

@@ -0,0 +1,76 @@
+package io.aatricks.novelscraper.data.repository.source
+
+import org.junit.Test
+import kotlin.system.measureTimeMillis
+
+class NovelFireSourceBenchmarkTest {
+
+    private fun cleanNovelTitleOriginal(title: String): String {
+        var clean = title
+        // Remove [123] at start
+        clean = clean.replace(Regex("^\\[\\d+\\]\\s*"), "")
+        // Remove R 14.8 or R 123 at start
+        clean = clean.replace(Regex("^R\\s*\\d+(\\.\\d+)?\\s*"), "")
+        // Remove Rank 123 at start
+        clean = clean.replace(Regex("^Rank\\s*\\d+\\s*", RegexOption.IGNORE_CASE), "")
+        return clean.trim()
+    }
+
+    companion object {
+        private val BRACKET_NUMBER_REGEX = Regex("^\\[\\d+\\]\\s*")
+        private val R_NUMBER_REGEX = Regex("^R\\s*\\d+(\\.\\d+)?\\s*")
+        private val RANK_PREFIX_REGEX = Regex("^Rank\\s*\\d+\\s*", RegexOption.IGNORE_CASE)
+    }
+
+    private fun cleanNovelTitleOptimized(title: String): String {
+        var clean = title
+        // Remove [123] at start
+        clean = clean.replace(BRACKET_NUMBER_REGEX, "")
+        // Remove R 14.8 or R 123 at start
+        clean = clean.replace(R_NUMBER_REGEX, "")
+        // Remove Rank 123 at start
+        clean = clean.replace(RANK_PREFIX_REGEX, "")
+        return clean.trim()
+    }
+
+    @Test
+    fun benchmarkCleanNovelTitle() {
+        val titles = listOf(
+            "[123] Some Novel Name",
+            "R 14.8 Another Novel Name",
+            "Rank 42 A Third Novel Name",
+            "Just a regular name",
+            "Rank 123 [123] Double trouble name",
+            "Nothing to replace here 123"
+        )
+
+        // Warmup
+        for (i in 0 until 1000) {
+            for (title in titles) {
+                cleanNovelTitleOriginal(title)
+                cleanNovelTitleOptimized(title)
+            }
+        }
+
+        val iterations = 100_000
+        val timeOriginal = measureTimeMillis {
+            for (i in 0 until iterations) {
+                for (title in titles) {
+                    cleanNovelTitleOriginal(title)
+                }
+            }
+        }
+
+        val timeOptimized = measureTimeMillis {
+            for (i in 0 until iterations) {
+                for (title in titles) {
+                    cleanNovelTitleOptimized(title)
+                }
+            }
+        }
+
+        println("Baseline - Time to run $iterations iterations: $timeOriginal ms")
+        println("Optimized - Time to run $iterations iterations: $timeOptimized ms")
+        println("Improvement: ${timeOriginal - timeOptimized} ms")
+    }
+}

benchmark_output.txt

@@ -0,0 +1,50 @@
+> Task :app:preBuild UP-TO-DATE
+> Task :app:preDebugBuild UP-TO-DATE
+> Task :app:checkKotlinGradlePluginConfigurationErrors SKIPPED
+> Task :app:checkDebugAarMetadata UP-TO-DATE
+> Task :app:processDebugNavigationResources UP-TO-DATE
+> Task :app:compileDebugNavigationResources UP-TO-DATE
+> Task :app:generateDebugResValues UP-TO-DATE
+> Task :app:mapDebugSourceSetPaths UP-TO-DATE
+> Task :app:generateDebugResources UP-TO-DATE
+> Task :app:mergeDebugResources UP-TO-DATE
+> Task :app:packageDebugResources UP-TO-DATE
+> Task :app:parseDebugLocalResources UP-TO-DATE
+> Task :app:createDebugCompatibleScreenManifests UP-TO-DATE
+> Task :app:extractDeepLinksDebug UP-TO-DATE
+> Task :app:processDebugMainManifest UP-TO-DATE
+> Task :app:processDebugManifest UP-TO-DATE
+> Task :app:processDebugManifestForPackage UP-TO-DATE
+> Task :app:processDebugResources UP-TO-DATE
+> Task :app:kspDebugKotlin UP-TO-DATE
+> Task :app:compileDebugKotlin UP-TO-DATE
+> Task :app:javaPreCompileDebug UP-TO-DATE
+> Task :app:compileDebugJavaWithJavac UP-TO-DATE
+> Task :app:hiltAggregateDepsDebug UP-TO-DATE
+> Task :app:hiltJavaCompileDebug
+> Task :app:preDebugUnitTestBuild UP-TO-DATE
+> Task :app:processDebugJavaRes
+> Task :app:javaPreCompileDebugUnitTest UP-TO-DATE
+> Task :app:bundleDebugClassesToCompileJar
+> Task :app:transformDebugClassesWithAsm
+> Task :app:bundleDebugClassesToRuntimeJar
+> Task :app:kspDebugUnitTestKotlin
+
+> Task :app:compileDebugUnitTestKotlin
+w: file:///app/app/src/test/java/io/aatricks/novelscraper/data/repository/ContentRepositoryEpubTest.kt:43:19 'fun createTempDir(prefix: String = ..., suffix: String? = ..., directory: File? = ...): File' is deprecated. Avoid creating temporary directories in the default temp location with this function due to too wide permissions on the newly created directory. Use kotlin.io.path.createTempDirectory instead.
+
+> Task :app:compileDebugUnitTestJavaWithJavac NO-SOURCE
+> Task :app:hiltAggregateDepsDebugUnitTest
+> Task :app:hiltJavaCompileDebugUnitTest NO-SOURCE
+> Task :app:processDebugUnitTestJavaRes
+> Task :app:transformDebugUnitTestClassesWithAsm
+
+> Task :app:testDebugUnitTest
+
+NovelFireSourceBenchmarkTest > benchmarkCleanNovelTitle STANDARD_OUT
+    Baseline - Time to run 100000 iterations: 1806 ms
+
+NovelFireSourceBenchmarkTest > benchmarkCleanNovelTitle PASSED
+
+BUILD SUCCESSFUL in 37s
+32 actionable tasks: 11 executed, 21 up-to-date

benchmark_output_2.txt

@@ -0,0 +1,49 @@
+> Task :app:preBuild UP-TO-DATE
+> Task :app:preDebugBuild UP-TO-DATE
+> Task :app:checkKotlinGradlePluginConfigurationErrors SKIPPED
+> Task :app:checkDebugAarMetadata UP-TO-DATE
+> Task :app:processDebugNavigationResources UP-TO-DATE
+> Task :app:compileDebugNavigationResources UP-TO-DATE
+> Task :app:generateDebugResValues UP-TO-DATE
+> Task :app:mapDebugSourceSetPaths UP-TO-DATE
+> Task :app:generateDebugResources UP-TO-DATE
+> Task :app:mergeDebugResources UP-TO-DATE
+> Task :app:packageDebugResources UP-TO-DATE
+> Task :app:parseDebugLocalResources UP-TO-DATE
+> Task :app:createDebugCompatibleScreenManifests UP-TO-DATE
+> Task :app:extractDeepLinksDebug UP-TO-DATE
+> Task :app:processDebugMainManifest UP-TO-DATE
+> Task :app:processDebugManifest UP-TO-DATE
+> Task :app:processDebugManifestForPackage UP-TO-DATE
+> Task :app:processDebugResources UP-TO-DATE
+> Task :app:javaPreCompileDebug UP-TO-DATE
+> Task :app:preDebugUnitTestBuild UP-TO-DATE
+> Task :app:javaPreCompileDebugUnitTest UP-TO-DATE
+> Task :app:kspDebugKotlin
+> Task :app:compileDebugKotlin
+> Task :app:compileDebugJavaWithJavac
+> Task :app:hiltAggregateDepsDebug UP-TO-DATE
+> Task :app:hiltJavaCompileDebug
+> Task :app:processDebugJavaRes
+> Task :app:bundleDebugClassesToCompileJar
+> Task :app:transformDebugClassesWithAsm
+> Task :app:bundleDebugClassesToRuntimeJar
+> Task :app:kspDebugUnitTestKotlin
+> Task :app:compileDebugUnitTestKotlin
+> Task :app:compileDebugUnitTestJavaWithJavac NO-SOURCE
+> Task :app:hiltAggregateDepsDebugUnitTest UP-TO-DATE
+> Task :app:hiltJavaCompileDebugUnitTest NO-SOURCE
+> Task :app:processDebugUnitTestJavaRes UP-TO-DATE
+> Task :app:transformDebugUnitTestClassesWithAsm
+
+> Task :app:testDebugUnitTest
+
+NovelFireSourceBenchmarkTest > benchmarkCleanNovelTitle STANDARD_OUT
+    Baseline - Time to run 100000 iterations: 1788 ms
+    Optimized - Time to run 100000 iterations: 315 ms
+    Improvement: 1473 ms
+
+NovelFireSourceBenchmarkTest > benchmarkCleanNovelTitle PASSED
+
+BUILD SUCCESSFUL in 26s
+32 actionable tasks: 12 executed, 20 up-to-date