Cleanup and improve the vokabeln reader.

Original patch by Ilja.
> This patch should make possible to open .voc with longer header. (e.g.
> http://www.vokabeln.de/v1/download/GB_Technik.zip).
> KEduVocDocument::setDocumentComment changed to allow to insert
> multiline file comments from .voc files.
CCMAIL: iljuss@gmail.com

svn path=/trunk/KDE/kdeedu/libkdeedu/; revision=723538

diff --git a/keduvocdocument/keduvocdocument.cpp b/keduvocdocument/keduvocdocument.cpp
index 5cedd1e2a3a2add18755280f28024c252126b2c9..79133bc02b405ecedf8ef5705270391952fb5114 100644 (file)
--- a/keduvocdocument/keduvocdocument.cpp
+++ b/keduvocdocument/keduvocdocument.cpp
@@ -189,7 +189,42 @@ KEduVocDocument::FileType KEduVocDocument::detectFileType( const QString &fileNa
         line2 = ts.readLine();
     }
 
+    /*
+     * Vokabeln.de files:
+    The header seems to be something like this:
+
+    "Name
+    Lang1 - Lang2",123,234,456
+    0
+
+    or something longer:
+
+    "Name
+    Lang1 - Lang2
+    [..]
+    Blah, blah, blah...",123,234,456
+    0
+    */
+
+    QString tmp;
+
+    if ( line1.startsWith(QChar::fromLatin1('"'))) {
+        ts.seek(0);
+        tmp = ts.readLine();
+        // There shouldn't be headers longer than 10 lines.
+        for ( int x=0; x < 10; x++) {
+            if (tmp.contains( "\"," )) {
+                tmp = ts.readLine();
+                if (tmp.endsWith('0')) {
+                    f->close();
+                    return Vokabeln;
+                }
+            }
+            tmp = ts.readLine();
+        }
+    }
     f->close();
+
     if ( line1.startsWith(QString::fromLatin1("<?xml")) ) {
         if ( line2.indexOf( "pauker", 0 ) >  0 ) {
             return Pauker;
@@ -205,18 +240,6 @@ KEduVocDocument::FileType KEduVocDocument::detectFileType( const QString &fileNa
         return Wql;
     }
 
-    /*
-    Vokabeln.de files:
-    The header seems to be something like this:
-    "Name
-    Lang1 - Lang2",123,234,456
-    */
-    if ( line1.startsWith(QChar::fromLatin1('"')) &&
-        ((line1.contains(QRegExp( "\",[0-9]" ))) ||
-        (line2.contains(QRegExp( "\",[0-9]" ))))) {
-        return Vokabeln;
-    }
-
     return Csv;
 }
 
@@ -973,7 +996,7 @@ void KEduVocDocument::setLicense( const QString & s )
 
 void KEduVocDocument::setDocumentComment( const QString & s )
 {
-    d->m_comment = s.simplified();
+    d->m_comment = s.trimmed();
 }
 
 

diff --git a/keduvocdocument/keduvocvokabelnreader.cpp b/keduvocdocument/keduvocvokabelnreader.cpp
index a12310aa469901988757f57f504ad1287103169b..83dde8846047ba6b330f2f655c2eaa2ac278be33 100644 (file)
--- a/keduvocdocument/keduvocvokabelnreader.cpp
+++ b/keduvocdocument/keduvocvokabelnreader.cpp
@@ -52,40 +52,53 @@ bool KEduVocVokabelnReader::readDoc( KEduVocDocument *doc )
     inputStream.setAutoDetectUnicode( false );
     inputStream.seek( 0 );
 
-    QString title,
-    lang1,
-    lang2,
-    expression,
-    original,
-    translation,
-    lessonDescr,
-    temp;
-
-    int     i,
-    wordCount,
-    lesson;
+    QString title;
+    QString lang1;
+    QString lang2;
+    QString expression;
+    QString original;
+    QString translation;
+    QString lessonDescr;
+    QString temp;
+    QString comment;
+
+    int i;
+    int wordCount;
+    int lesson;
 
     int lines = 10000;
 
-    QStringList titles,
-    languages,
-    words;
+    QStringList titles;
+    QStringList languages;
+    QStringList words;
 
-    bool keepGoing = true;
+    temp = inputStream.readLine();
 
-    while ( keepGoing ) {
-        temp = inputStream.readLine();
-        keepGoing = !( temp.indexOf( "\"," ) > 0 );
-        title.append( temp );
-        if ( keepGoing )
-            title.append( " " );
+    if ( temp.startsWith ( QChar::fromLatin1 ( '"' ) ) ) {
+        while ( temp != QChar::fromLatin1('0') ) {
+            if ( title.isEmpty() ) {
+                title = temp;
+            } else {
+                comment.append( temp.append('\n') );
+            }
+            temp = inputStream.readLine();
+        }
+    }
+
+    // 1 line header
+    if (comment.isEmpty()) {
+        titles = title.split( "\"," );
+        m_doc->setTitle(titles[0].mid(1));
+    }
+    // longer header
+    else {
+        titles = comment.split( "\"," );
+        m_doc->setDocumentComment(titles[0]);
+        m_doc->setTitle(title.mid(1));
     }
 
-    titles = title.split( "\"," );
-    m_doc->setTitle( titles[0].mid( 1 ) );
     wordCount = titles[1].section( ',', 0, 0 ).toInt();
 
-    inputStream.readLine();
     inputStream.readLine();
 
     lang1 = inputStream.readLine();
@@ -98,23 +111,25 @@ bool KEduVocVokabelnReader::readDoc( KEduVocDocument *doc )
     m_doc->identifier(1).setLocale( languages[1].mid( 1 ) );
     m_doc->identifier(1).setName( languages[1].mid( 1 ) );
 
-    keepGoing = true;
-    while ( keepGoing )
-        keepGoing = !( inputStream.readLine().indexOf( "8. Lernhilfe" ) > 0 ); //DO NOT translate
+    while ( !temp.contains("8. Lernhilfe") ) {
+        temp = inputStream.readLine(); //DO NOT translate
+    }
 
-    for ( i = 0; i <= 14; i++ )
+    for ( i = 0; i <= 14; i++ ) {
         inputStream.readLine();
+    }
 
     for ( i = 0; i < wordCount - 1; i++ ) {
         int c = 0;
-        expression.resize( 0 );
+        expression.clear();
 
         while ( c < 2 ) {
             temp = inputStream.readLine();
             c+= temp.count( "\",", Qt::CaseSensitive );
             expression.append( temp );
-            if ( c < 2 )
+            if ( c < 2 ) {
                 expression.append( " " );
+            }
         }
 
         words = expression.split( "\"," );
@@ -127,7 +142,6 @@ bool KEduVocVokabelnReader::readDoc( KEduVocDocument *doc )
         kve.setTranslation( 1, translation );
         kve.translation( 1 ).gradeFrom( 0 ).setGrade( 0 );
         kve.translation( 0 ).gradeFrom( 1 ).setGrade( 0 );
-        /// @todo lesson might need a -1. I have no specs or documents to verify.
         kve.setLesson( lesson );
 
         m_doc->appendEntry( &kve );