Wait for a canceled downloader job to finish before starting another

Page writes block, so a canceled job could still be writing pages when the next one started on the same pages, leaving duplicate "(1)" files and failed chapters.

app/src/main/java/eu/kanade/tachiyomi/data/download/Downloader.kt

@@ -22,6 +22,7 @@ import kotlinx.coroutines.CancellationException
 import kotlinx.coroutines.CoroutineScope
 import kotlinx.coroutines.Dispatchers
 import kotlinx.coroutines.Job
+import kotlinx.coroutines.NonCancellable
 import kotlinx.coroutines.SupervisorJob
 import kotlinx.coroutines.async
 import kotlinx.coroutines.delay
@@ -41,6 +42,7 @@ import kotlinx.coroutines.flow.transformLatest
 import kotlinx.coroutines.flow.update
 import kotlinx.coroutines.launch
 import kotlinx.coroutines.supervisorScope
+import kotlinx.coroutines.withContext
 import logcat.LogPriority
 import mihon.core.archive.ZipWriter
 import nl.adaptivity.xmlutil.serialization.XML
@@ -91,6 +93,8 @@ class Downloader(
     val queueState = _queueState.asStateFlow()
 
     private val scope = CoroutineScope(SupervisorJob() + Dispatchers.IO)
+
+    // Kept when canceled, so the next job can wait for it to finish
     private var downloaderJob: Job? = null
 
     /**
@@ -183,7 +187,12 @@ class Downloader(
     private fun launchDownloaderJob() {
         if (isRunning) return
 
+        val previousJob = downloaderJob
         downloaderJob = scope.launch {
+            // Page writes block, so a canceled job can still be writing the pages this one would start on. Not
+            // cancelable, so a job canceled while waiting still finishes after the one before it.
+            withContext(NonCancellable) { previousJob?.join() }
+
             val activeDownloadsFlow = combine(
                 queueState,
                 downloadPreferences.parallelSourceLimit.changes(),
@@ -253,7 +262,6 @@ class Downloader(
      */
     private fun cancelDownloaderJob() {
         downloaderJob?.cancel()
-        downloaderJob = null
     }
 
     /**