Merge pull request #22 from AlessioLuciani/dev

v0.4.0

Author: AlessioLuciani

.github/workflows/flutter.yml

@@ -0,0 +1,30 @@
+name: Flutter CI
+
+on:
+  push:
+    branches: [ dev, master ]
+  pull_request:
+    branches: [ dev, master ]
+
+jobs:
+  build:
+
+    runs-on: macos-latest #ubuntu-latest
+    env:
+      working-directory: ./example
+
+    steps:
+    - uses: actions/checkout@v2
+    - uses: actions/setup-java@v1
+      with:
+        java-version: '12.x'
+    - uses: subosito/flutter-action@v1
+      with:
+        channel: 'beta' # or: 'dev' or 'stable'
+    - run: flutter pub get
+      working-directory: ${{env.working-directory}}
+  #  - run: flutter test
+    - run: flutter build appbundle #apk
+      working-directory: ${{env.working-directory}}
+    - run: flutter build ios --release --no-codesign
+      working-directory: ${{env.working-directory}}

CHANGELOG.md

@@ -1,3 +1,9 @@
+## 0.4.0
+
+* The PDFBoxResourceLoader is now used on Android to load PDF documents much faster than before. The fast initialization (i.e. *fastInit*) option has therefore been removed.
+* PDF documents are no longer kept alive in the platform-specific scope. Instead, they are opened and closed at each read with the respective library functions. This does not affect the caching mechanism utilized directly in Dart. This change prevents errors due to multiple document accesses at the same time.
+* Tests have been implemented.
+
 ## 0.3.1
 
 * The possibility to initialize a document faster (without immediately initializing the text stripper engine) on Android has been added.

README.md

@@ -1,7 +1,7 @@
 # PDF Text Plugin
 
 [![Pub Version](https://img.shields.io/pub/v/pdf_text)](https://pub.dev/packages/pdf_text)
-[![GitHub issues](https://img.shields.io/github/issues/AlessioLuciani/flutter-pdf-text)](https://github.com/AlessioLuciani/flutter-pdf-text/issues)
+![Flutter CI](https://github.com/AlessioLuciani/flutter-pdf-text/workflows/Flutter%20CI/badge.svg?branch=master)
 [![GitHub forks](https://img.shields.io/github/forks/AlessioLuciani/flutter-pdf-text)](https://github.com/AlessioLuciani/flutter-pdf-text/network)
 [![GitHub stars](https://img.shields.io/github/stars/AlessioLuciani/flutter-pdf-text)](https://github.com/AlessioLuciani/flutter-pdf-text/stargazers)
 [![GitHub license](https://img.shields.io/github/license/AlessioLuciani/flutter-pdf-text)](https://github.com/AlessioLuciani/flutter-pdf-text/blob/master/LICENSE)
@@ -18,7 +18,7 @@ Add this to your package's `pubspec.yaml` file:
 
 ```yaml
 dependencies:
-  pdf_text: ^0.3.1
+  pdf_text: ^0.4.0
 ```
 
 ## Usage
@@ -29,7 +29,7 @@ Import the package with:
 import 'package:pdf_text/pdf_text.dart';
 ```
 
-*Create a PDF document instance using a File object:*
+**Create a PDF document instance using a File object:**
 
 ```dart
 PDFDoc doc = await PDFDoc.fromFile(file);
@@ -53,13 +53,7 @@ Pass a password for encrypted PDF documents:
 PDFDoc doc = await PDFDoc.fromFile(file, password: password);
 ```
 
-Use faster initialization on Android:
-
-```dart
-PDFDoc doc = await PDFDoc.fromFile(file, fastInit: true);
-```
-
-*Read the text of the entire document:*
+**Read the text of the entire document:**
 
 ```dart
 String docText = await doc.text;
@@ -71,19 +65,19 @@ Retrieve the number of pages of the document:
 int numPages = doc.length;
 ```
 
-*Access a page of the document:*
+**Access a page of the document:**
 
 ```dart
 PDFPage page = doc.pageAt(pageNumber);
 ```
 
-*Read the text of a page of the document:*
+**Read the text of a page of the document:**
 
 ```dart
 String pageText = await page.text;
 ```
 
-*Read the information of the document:*
+**Read the information of the document:**
 
 ```dart
 PDFDocInfo info = doc.info;
@@ -119,9 +113,9 @@ allows you not to waste time loading text that you will probably not use. When y
 | Return  | Description  |
 |---|---|
 | PDFPage | **pageAt(int pageNumber)** <br> Gets the page of the document at the given page number. |
-| static Future\<PDFDoc> | **fromFile(File file, {String password = "", bool fastInit = false})** <br> Creates a PDFDoc object with a File instance. Optionally, takes a password for encrypted PDF documents. If fastInit is true, the initialization of the document will be faster on Android. In that case, the text stripper engine will not be initialized with this call, but later when some text is read. This means that the first text read will take some time but the document data can be accessed immediately.|
-| static Future\<PDFDoc> | **fromPath(String path, {String password = "", bool fastInit = false})** <br> Creates a PDFDoc object with a file path. Optionally, takes a password for encrypted PDF documents. If fastInit is true, the initialization of the document will be faster on Android. In that case, the text stripper engine will not be initialized with this call, but later when some text is read. This means that the first text read will take some time but the document data can be accessed immediately.|
-| static Future\<PDFDoc> | **fromURL(String url, {String password = "", bool fastInit = false})** <br> Creates a PDFDoc object with a url. Optionally, takes a password for encrypted PDF documents. If fastInit is true, the initialization of the document will be faster on Android. In that case, the text stripper engine will not be initialized with this call, but later when some text is read. This means that the first text read will take some time but the document data can be accessed immediately. It downloads the PDF file located in the given URL and saves it in the app's temporary directory. |
+| static Future\<PDFDoc> | **fromFile(File file, {String password = ""})** <br> Creates a PDFDoc object with a File instance. Optionally, takes a password for encrypted PDF documents.|
+| static Future\<PDFDoc> | **fromPath(String path, {String password = ""})** <br> Creates a PDFDoc object with a file path. Optionally, takes a password for encrypted PDF documents.|
+| static Future\<PDFDoc> | **fromURL(String url, {String password = ""})** <br> Creates a PDFDoc object with a url. Optionally, takes a password for encrypted PDF documents.|
 | void | **deleteFile()** <br> Deletes the file related to this PDFDoc.<br>Throws an exception if the FileSystemEntity cannot be deleted. |
 | static Future | **deleteAllExternalFiles()** <br> Deletes all the files of the documents that have been imported from outside the local file system (e.g. using fromURL). |
 
@@ -155,4 +149,4 @@ class PDFDocInfo {
 ## Contribute
 
 If you have any suggestions, improvements or issues, feel free to contribute to this project.
-You can either submit a new issue or propose a pull request.
+You can either submit a new issue or propose a pull request. Direct your pull requests into the *dev* branch.

android/src/main/kotlin/dev/aluc/pdf_text/PdfTextPlugin.kt

@@ -1,44 +1,30 @@
 package dev.aluc.pdf_text
 
+
 import android.os.Handler
 import android.os.Looper
 import androidx.annotation.NonNull
+import com.tom_roush.pdfbox.pdmodel.PDDocument
+import com.tom_roush.pdfbox.text.PDFTextStripper
+import com.tom_roush.pdfbox.util.PDFBoxResourceLoader
 import io.flutter.embedding.engine.plugins.FlutterPlugin
 import io.flutter.plugin.common.MethodCall
 import io.flutter.plugin.common.MethodChannel
 import io.flutter.plugin.common.MethodChannel.MethodCallHandler
 import io.flutter.plugin.common.MethodChannel.Result
 import io.flutter.plugin.common.PluginRegistry.Registrar
-
 import java.io.File
-
-
-import com.tom_roush.pdfbox.pdmodel.PDDocument
-import com.tom_roush.pdfbox.pdmodel.PDPage
-import com.tom_roush.pdfbox.text.PDFTextStripper
 import kotlin.concurrent.thread
 
 /** PdfTextPlugin */
-public class PdfTextPlugin: FlutterPlugin, MethodCallHandler {
-
-  /**
-   * PDF document cached from the previous use.
-   */
-  private var cachedDoc: PDDocument? = null
-  private var cachedDocPath: String? = null
-
-  /**
-   * PDF text stripper.
-   */
-  private var pdfTextStripper = PDFTextStripper()
+class PdfTextPlugin: FlutterPlugin, MethodCallHandler {
 
   override fun onAttachedToEngine(@NonNull flutterPluginBinding: FlutterPlugin.FlutterPluginBinding) {
-    val channel = MethodChannel(flutterPluginBinding.getFlutterEngine().getDartExecutor(), "pdf_text")
-    channel.setMethodCallHandler(PdfTextPlugin());
+    val channel = MethodChannel(flutterPluginBinding.binaryMessenger, "pdf_text")
+    channel.setMethodCallHandler(PdfTextPlugin())
+    PDFBoxResourceLoader.init(flutterPluginBinding.applicationContext)
   }
 
-
-
   // This static function is optional and equivalent to onAttachedToEngine. It supports the old
   // pre-Flutter-1.12 Android projects. You are encouraged to continue supporting
   // plugin registration via this function while apps migrate to use the new Android APIs
@@ -53,36 +39,38 @@ public class PdfTextPlugin: FlutterPlugin, MethodCallHandler {
     fun registerWith(registrar: Registrar) {
       val channel = MethodChannel(registrar.messenger(), "pdf_text")
       channel.setMethodCallHandler(PdfTextPlugin())
+      PDFBoxResourceLoader.init(registrar.context())
     }
   }
 
   override fun onMethodCall(@NonNull call: MethodCall, @NonNull result: Result) {
     thread (start = true) {
       when (call.method) {
           "initDoc" -> {
-            val args = call.arguments as Map<String, Any>
+            val args = call.arguments as Map<*, *>
             val path = args["path"] as String
             val password = args["password"] as String
-            val fastInit = args["fastInit"] as Boolean
-            initDoc(result, path, password, fastInit)
+            initDoc(result, path, password)
           }
           "getDocPageText" -> {
-            val args = call.arguments as Map<String, Any>
+            val args = call.arguments as Map<*, *>
             val path = args["path"] as String
             val pageNumber = args["number"] as Int
-            getDocPageText(result, path, pageNumber)
+            val password = args["password"] as String
+            getDocPageText(result, path, pageNumber, password)
           }
           "getDocText" -> {
-            val args = call.arguments as Map<String, Any>
+            val args = call.arguments as Map<*, *>
             val path = args["path"] as String
+            @Suppress("UNCHECKED_CAST")
             val missingPagesNumbers = args["missingPagesNumbers"] as List<Int>
-            getDocText(result, path, missingPagesNumbers)
+            val password = args["password"] as String
+            getDocText(result, path, missingPagesNumbers, password)
           }
           else -> {
             Handler(Looper.getMainLooper()).post {
               result.notImplemented()
             }
-
           }
       }
     }
@@ -94,34 +82,35 @@ public class PdfTextPlugin: FlutterPlugin, MethodCallHandler {
   /**
     Initializes the PDF document and returns some information into the channel.
    */
-  private fun initDoc(result: Result, path: String, password: String, fastInit: Boolean) {
-    val doc = getDoc(result, path, password, !fastInit) ?: return
-    // Getting the length of the PDF document in pages.
-    val length = doc.numberOfPages
+  private fun initDoc(result: Result, path: String, password: String) {
+    getDoc(result, path, password)?.use { doc ->
+      // Getting the length of the PDF document in pages.
+      val length = doc.numberOfPages
 
-    val info = doc.documentInformation
+      val info = doc.documentInformation
 
-    var creationDate: String? = null
-    if (info.creationDate != null) {
-      creationDate = info.creationDate.time.toString()
-    }
-    var modificationDate: String? = null
-    if (info.modificationDate != null) {
-      modificationDate = info.modificationDate.time.toString()
-    }
-    val data = hashMapOf<String, Any>(
-            "length" to length,
-            "info" to hashMapOf("author" to info.author,
-                    "creationDate" to creationDate,
-                    "modificationDate" to modificationDate,
-                    "creator" to info.creator, "producer" to info.producer,
-                    "keywords" to splitKeywords(info.keywords),
-                    "title" to info.title, "subject" to info.subject
-            )
-    )
-
-    Handler(Looper.getMainLooper()).post {
-      result.success(data)
+      var creationDate: String? = null
+      if (info.creationDate != null) {
+        creationDate = info.creationDate.time.toString()
+      }
+      var modificationDate: String? = null
+      if (info.modificationDate != null) {
+        modificationDate = info.modificationDate.time.toString()
+      }
+      val data = hashMapOf<String, Any>(
+              "length" to length,
+              "info" to hashMapOf("author" to info.author,
+                      "creationDate" to creationDate,
+                      "modificationDate" to modificationDate,
+                      "creator" to info.creator, "producer" to info.producer,
+                      "keywords" to splitKeywords(info.keywords),
+                      "title" to info.title, "subject" to info.subject
+              )
+      )
+      doc.close()
+      Handler(Looper.getMainLooper()).post {
+        result.success(data)
+      }
     }
   }
 
@@ -132,7 +121,7 @@ public class PdfTextPlugin: FlutterPlugin, MethodCallHandler {
     if (keywordsString == null) {
       return null
     }
-    var keywords = keywordsString.split(",").toMutableList()
+    val keywords = keywordsString.split(",").toMutableList()
     for (i in keywords.indices) {
       var keyword = keywords[i]
       keyword = keyword.dropWhile { it == ' ' }
@@ -145,51 +134,45 @@ public class PdfTextPlugin: FlutterPlugin, MethodCallHandler {
   /**
     Gets the text  of a document page, given its number.
    */
-  private fun getDocPageText(result: Result, path: String, pageNumber: Int) {
-    val doc = getDoc(result, path) ?: return
-    pdfTextStripper.startPage = pageNumber
-    pdfTextStripper.endPage = pageNumber
-    val text = pdfTextStripper.getText(doc)
-    Handler(Looper.getMainLooper()).post {
-      result.success(text)
+  private fun getDocPageText(result: Result, path: String, pageNumber: Int, password: String) {
+    getDoc(result, path, password)?.use { doc ->
+      val stripper = PDFTextStripper();
+      stripper.startPage = pageNumber
+      stripper.endPage = pageNumber
+      val text = stripper.getText(doc)
+      doc.close()
+      Handler(Looper.getMainLooper()).post {
+        result.success(text)
+      }
     }
   }
 
   /**
   Gets the text of the entire document.
   In order to improve the performance, it only retrieves the pages that are currently missing.
    */
-  private fun getDocText(result: Result, path: String, missingPagesNumbers: List<Int>) {
-    val doc = getDoc(result, path) ?: return
-    var missingPagesTexts = arrayListOf<String>()
-    missingPagesNumbers.forEach {
-      pdfTextStripper.startPage = it
-      pdfTextStripper.endPage = it
-      missingPagesTexts.add(pdfTextStripper.getText(doc))
-    }
-    Handler(Looper.getMainLooper()).post {
-      result.success(missingPagesTexts)
+  private fun getDocText(result: Result, path: String, missingPagesNumbers: List<Int>, password: String) {
+    getDoc(result, path, password)?.use { doc ->
+      val missingPagesTexts = arrayListOf<String>()
+      val stripper = PDFTextStripper();
+      missingPagesNumbers.forEach {
+        stripper.startPage = it
+        stripper.endPage = it
+        missingPagesTexts.add(stripper.getText(doc))
+      }
+      doc.close()
+      Handler(Looper.getMainLooper()).post {
+        result.success(missingPagesTexts)
+      }
     }
   }
 
   /**
   Gets a PDF document, given its path.
-  Initializes the text stripper engine if initTextStripper is true.
    */
-  private fun getDoc(result: Result, path: String, password: String = "",
-    initTextStripper: Boolean = true): PDDocument? {
-    // Checking for cached document
-    if (cachedDoc != null && cachedDocPath == path) {
-      return cachedDoc
-    }
+  private fun getDoc(result: Result, path: String, password: String = ""): PDDocument? {
     return try {
-      val doc = PDDocument.load(File(path), password)
-      cachedDoc = doc
-      cachedDocPath = path
-      if (initTextStripper) {
-        initTextStripperEngine(doc)
-      }
-      doc
+      PDDocument.load(File(path), password)
     } catch (e: Exception) {
       Handler(Looper.getMainLooper()).post {
         result.error("INVALID_PATH",
@@ -199,13 +182,4 @@ public class PdfTextPlugin: FlutterPlugin, MethodCallHandler {
       null
     }
   }
-
-  /**
-   * Initializes the text stripper engine. This can take some time.
-   */
-  private fun initTextStripperEngine(doc: PDDocument) {
-    pdfTextStripper.startPage = 1
-    pdfTextStripper.endPage = 1
-    pdfTextStripper.getText(doc)
-  }
 }