Mo parser

Author: cah4a

.gitignore

@@ -0,0 +1,3 @@
+.dart_tool
+.idea
+.packages

CHANGELOG.md

@@ -0,0 +1,3 @@
+## 0.0.1
+
+- Initial version

README.md

@@ -0,0 +1,8 @@
+# MO parser for UTF-8 encoded *.mo files
+
+## Example
+
+```dart
+Parser parser = new Parser(buffer);
+Map translateTable = parser.parse();
+```

analysis_options.yaml

@@ -0,0 +1,14 @@
+analyzer:
+#   exclude:
+#     - path/to/excluded/files/**
+
+# Lint rules and documentation, see http://dart-lang.github.io/linter/lints
+linter:
+  rules:
+    - cancel_subscriptions
+    - hash_and_equals
+    - iterable_contains_unrelated_type
+    - list_remove_unrelated_type
+    - test_types_in_equals
+    - unrelated_type_equality_checks
+    - valid_regexps

example/README.md

@@ -0,0 +1,8 @@
+# Parse
+
+## MO files
+
+```dart
+Parser parser = new Parser(buffer);
+Map translateTable = parser.parse();
+```

lib/mo_parser.dart

@@ -0,0 +1,10 @@
+library mo_parser;
+
+import './src/parser.dart';
+export './src/parser.dart';
+
+/// Parse *.mo file to Map object
+Map parser (List<int> buffer) {
+  Parser parser = new Parser(buffer);
+  return parser.parse();
+}

lib/src/parser.dart

@@ -0,0 +1,151 @@
+import 'dart:convert';
+import 'dart:typed_data';
+
+import './utils/parse_header.dart';
+
+/// Parser class
+/// Contains decoding and parse logic for passed List<int>
+class Parser {
+  // Magic constant to check the endianness of the input file
+  static const _MAGIC = 0x950412de;
+
+  // Default endian for read/write
+  Endian _endian = Endian.little;
+  ByteData _fileContents;
+  Map _table;
+  // Offset position for original strings table
+  int _offsetOriginals;
+  // Offset position for translation strings table
+  int _offsetTranslations;
+  // GetText revision nr, usually 0
+  int _revision;
+  // Total count of translated strings
+  int _total;
+
+  Parser(List<int> fileContent) {
+    this._fileContents = ByteData.view(Uint8List.fromList(fileContent).buffer);
+
+    this._table = {
+      'charset': 'utf-8',
+      'headers': null,
+      'translations': {},
+    };
+  }
+
+  // Checks if number values in the input file are in big- or littleendian format.
+  bool _checkMagick () {
+    if (this._fileContents.getUint32(0, Endian.little) == Parser._MAGIC) {
+      this._endian = Endian.little;
+      return true;
+    } else if (this._fileContents.getUint32(0, Endian.big) == Parser._MAGIC) {
+      this._endian = Endian.big;
+      return true;
+    } else {
+      return false;
+    }
+  }
+
+  // Read the original strings and translations from the input MO file. Use the
+  // first translation string in the file as the header.
+  void _loadTranslationTable () {
+    int offsetOriginals = this._offsetOriginals;
+    int offsetTranslations = this._offsetTranslations;
+    int position, length;
+    String msgid, msgstr;
+    Iterable msgidRange, msgstrRange;
+
+    for (int i = 0; i < this._total; i++) {
+      // msgid string
+      length = this._fileContents.getUint32(offsetOriginals, this._endian);
+      offsetOriginals += 4;
+      position = this._fileContents.getUint32(offsetOriginals, this._endian);
+      offsetOriginals += 4;
+      msgidRange = this._fileContents.buffer.asUint8List().getRange(position, position + length);
+
+      // matching msgstr
+      length = this._fileContents.getUint32(offsetTranslations, this._endian);
+      offsetTranslations += 4;
+      position = this._fileContents.getUint32(offsetTranslations, this._endian);
+      offsetTranslations += 4;
+      msgstrRange = this._fileContents.buffer.asUint8List().getRange(position, position + length);
+
+      if (i == 0 && msgidRange.toList().isEmpty) {
+        this._handleCharset(msgstrRange);
+      }
+
+      /**
+       * dart:convert support limited quantity of charsets
+       * https://api.dartlang.org/dev/2.1.1-dev.0.1/dart-convert/dart-convert-library.html
+       *
+       * More about issue
+       * https://stackoverflow.com/questions/21142985/convert-a-string-from-iso-8859-2-to-utf-8-in-the-dart-language
+       * https://stackoverflow.com/questions/51148729/how-to-manually-convert-between-latin-5-and-unicode-code-points
+       */
+      msgid = utf8.decode(msgidRange.toList());
+      msgstr = utf8.decode(msgstrRange.toList());
+
+      this._addString(msgid, msgstr);
+    }
+
+    // dump the file contents object
+    this._fileContents = null;
+  }
+
+  void _handleCharset (Iterable headers) {
+    String headersParsed = utf8.decode(headers.toList());
+
+    this._table['headers'] = parseHeader(headersStr: headersParsed);
+  }
+
+  void _addString (dynamic msgid, dynamic msgstr) {
+    final Map translation = {};
+    List<String> parts;
+    String msgctxt, msgidPlural;
+
+    msgid = msgid.split('\u0004');
+    if (msgid.length > 1) {
+      msgctxt = msgid.first;
+      msgid.removeAt(0);
+      translation['msgctxt'] = msgctxt;
+    } else {
+      msgctxt = '';
+    }
+    msgid = msgid.join('\u0004');
+
+    parts = msgid.split('\u0000');
+    msgid = parts.first;
+    parts.removeAt(0);
+
+    translation['msgid'] = msgid;
+    msgidPlural = parts.join('\u0000');
+
+    if (!msgidPlural.isEmpty) {
+      translation['msgid_plural'] = msgidPlural;
+    }
+
+    msgstr = msgstr.split('\u0000');
+    translation['msgstr'] = msgstr;
+
+    if (!this._table['translations'].containsKey(msgctxt)) {
+      this._table['translations'][msgctxt] = {};
+    }
+
+    this._table['translations'][msgctxt][msgid] = translation;
+  }
+
+  /// Parses the MO object and returns translation table
+  Map parse () {
+    if (!this._checkMagick()) {
+      return null;
+    }
+
+    this._revision = this._fileContents.getUint32(4, this._endian);
+    this._total = this._fileContents.getUint32(8, this._endian);
+    this._offsetOriginals = this._fileContents.getUint32(12, this._endian);
+    this._offsetTranslations = this._fileContents.getUint32(16, this._endian);
+
+    this._loadTranslationTable();
+
+    return this._table;
+  }
+}

lib/src/utils/parse_header.dart

@@ -0,0 +1,20 @@
+/// Parses a header string into an object of key-value pairs
+Map parseHeader({String headersStr = ''}) {
+  final List<String> lines = headersStr.split('\n');
+  final Map headers = {};
+
+  lines.forEach((line) {
+    final List<String> parts = line.trim().split(':');
+    final String key = parts.first.trim().toLowerCase();
+    parts.removeAt(0);
+    final String value = parts.join(':').trim();
+
+    if (key.isEmpty) {
+      return;
+    }
+
+    headers[key] = value;
+  });
+
+  return headers;
+}