似乎
android.database.Cursor.getColumnIndex(String)应该在行循环中不变.
然而,我看到大量代码执行以下操作:
try (Cursor cursor = database.query(...)) {
while (cursor.moveToNext()) {
String aString = cursor.getString(cursor.getColumnIndex(aColumnName));
// ... other loop code and more repeated use of getColumnIndex()
}
}
这看起来很愚蠢和浪费,除非getColumnIndex()的结果真的随着对cursor.moveToNext()的调用而变化 – 而且这个调用应该从循环中提升出来,如:
try (Cursor cursor = database.query(...)) {
int aStringIdx = cursor.getColumnIndex(aColumnName);
while (cursor.moveToNext()) {
String aString = cursor.getString(aStringIdx);
// ... other loop code
}
}
或者在最坏的情况下(假设getColumnIndex()要求您在有效的数据行上):
try (Cursor cursor = database.query(...)) {
if (cursor.moveToFirst()) {
int aStringIdx = cursor.getColumnIndex(aColumnName);
do {
String aString = cursor.getString(aStringIdx);
// ... other loop code
} while (cursor.moveToNext());
}
}
那么这里的真实故事是什么?
最佳答案
I was seeing all did getColumnIndex each time through the row loop, which made no sense at all
也许是因为当我们执行查询时,我们收到一个SQLiteCursor实例,而不是AbstractCursor.如果我们查看源代码,我们可以在内部找到chache:
@Override
public int getColumnIndex(String columnName) {
// Create mColumnNameMap on demand
if (mColumnNameMap == null) {
String[] columns = mColumns;
int columnCount = columns.length;
HashMap<String, Integer> map = new HashMap<String, Integer>(columnCount, 1);
for (int i = 0; i < columnCount; i++) {
map.put(columns[i], i);
}
mColumnNameMap = map;
}
//...
}
另一方面,访问变量(如示例2)比计算哈希键以获取索引更快.可能是相关的,取决于表的大小.
编辑:在具有复杂查询的大型表上进行测试,似乎改进了5%,在我的情况下不值得进行更改.通常,HashMap缓存足够好.