- html - 出于某种原因,IE8 对我的 Sass 文件中继承的 html5 CSS 不友好?
- JMeter 在响应断言中使用 span 标签的问题
- html - 在 :hover and :active? 上具有不同效果的 CSS 动画
- html - 相对于居中的 html 内容固定的 CSS 重复背景?
我有下面的 DBImporter 类,它工作正常,并且还可以在数据库表中正确插入数据。我正在尝试从 .CSV 文件获取数据并插入到 Oracle 表中。
到目前为止,我只处理目录中的一个文件,并且工作正常。现在我想处理多个文件。因此,在运行第一个文件时正确处理并插入数据,在第二个文件中它开始读取数据并抛出错误,如下所示:
java.lang.IllegalArgumentException: SQL array must not be empty
下面是我的 DBimporter 类。我认为错误是在最终提交批处理期间的某处,但不确定
jdbcTemplate.batchUpdate(sqlBatch.toArray(new String[sqlBatch.size()]));
@Service
public class DBImporter {
private final static Logger log = LoggerFactory.getLogger(DBImporter.class);
private static final List<String> NULL_VALUES = Arrays.asList("", "N.A", "N.A", "UNKNOWN");
private static final List<String> COL_HEADERS = Arrays.asList("ID", "NM", "TYE", "SA");
private static final int BATCH_SIZE = 50;
private boolean eof = false;
private String tableName;
@Autowired
private JdbcTemplate jdbcTemplate;
public void setTableName(String tableName) {
this.tableName = tableName;
}
@Transactional(rollbackFor = IOException.class)
public void processFile(BufferedReader reader, String tableName) {
this.tableName = tableName;
List<String> sqlBatch = new ArrayList<String>(BATCH_SIZE);
log.info("Starte auslesen der Daten");
long t1 = System.currentTimeMillis();
log.info("Start time: " + t1);
jdbcTemplate.execute("DELETE FROM " + tableName);
while (!eof) {
try {
Map<String, ColumnData> dbColumns = getDBColumns();
// Get a list of db column data related to the column headers.
List<ColumnData> columnData = COL_HEADERS.stream().map(dbColumns::get).collect(toList());
// Get the next valid data row if its starts from "FRO" or "BO".
List<String> dataRow = findNextLineStartingWith(reader, "R", "T");
String query = createSql(columnData, dataRow);
sqlBatch.add(query);
// Process batch.
if (sqlBatch.size() >= BATCH_SIZE) {
jdbcTemplate.batchUpdate(sqlBatch.toArray(new String[sqlBatch.size()]));
sqlBatch.clear();
}
} catch (IllegalStateException e) {
break;
} catch (IOException e) {
log.error(e.getLocalizedMessage());
}
}
// Commit the final batch.
jdbcTemplate.batchUpdate(sqlBatch.toArray(new String[sqlBatch.size()]));
sqlBatch.clear();
long delta = System.currentTimeMillis() - t1;
log.info("Total runtime : " + delta / 1000 + " seconds");
}
/**
* Create a SQL insert query using the data row.
*
* @param tableName Name of the table.
* @param columnData Column data list.
* @param dataRow Data row to be inserted.
* @return Generated SQL query string.
*/
private String createSql(List<ColumnData> columnData, List<String> dataRow) {
List<String> values = new ArrayList<>(columnData.size());
for (int i = 0; i < columnData.size(); i++) {
if (NULL_VALUES.contains(dataRow.get(i))) {
values.add("NULL");
} else if (columnData.get(i).getType() >= Types.NUMERIC && columnData.get(i).getType() <= Types.DOUBLE) {
values.add(dataRow.get(i));
} else {
values.add("'" + dataRow.get(i).replace("'", "''") + "'");
}
}
return "INSERT INTO " + tableName + " (" +
columnData.stream().filter(Objects::nonNull).map(ColumnData::getName).collect(joining(", ")) +
", SYSTEM_INSERTED_AT) VALUES (" +
values.stream().collect(joining(", ")) +
", CURRENT_TIMESTAMP)";
}
/**
* Find the next line starting with the given string and split it into columns.
*
* @param reader BufferedReader object to be used.
* @param prefixes A list of prefixes to look for in the string.
* @return List of data objects.
* @throws IOException
*/
private List<String> findNextLineStartingWith(BufferedReader reader, String... prefixes) throws IOException {
while (true) {
String line = readLineOrThrow(reader);
for (String prefix : prefixes)
if (line.startsWith(prefix)) {
ArrayList<String> data = new ArrayList<>();
// Split the line using the delimiter.
data.addAll(Arrays.asList(line.split(";")));
// Build the row to be inserted.
List<String> row = Arrays.asList(data.get(1), data.get(2).trim(), "", "");
return row;
}
}
}
/**
* Read a single line in the file.
*
* @param reader BufferedReader object to be used.
* @return
* @throws IOException
*/
private String readLineOrThrow(BufferedReader reader) throws IOException {
String line = reader.readLine();
if (line == null) {
this.eof = true;
throw new IllegalStateException("Unexpected EOF");
}
return line.trim();
}
/**
* Read database column metadata.
*
* @param tableName Name of the table to process.
* @return A map containing column information.
*/
private Map<String, ColumnData> getDBColumns() {
Map<String, ColumnData> result = new HashMap<>();
try (Connection connection = jdbcTemplate.getDataSource().getConnection()) {
ResultSet rs = connection.getMetaData().getColumns(null, null, tableName, null);
while (rs.next()) {
String columnName = rs.getString(4).toUpperCase();
int type = rs.getInt(5);
result.put(columnName, new ColumnData(columnName, type));
}
return result;
} catch (SQLException e) {
throw new RuntimeException(e);
}
}
}
最佳答案
请尝试以下更改:
// Commit the final batch.
if (sqlBatch.size() > 0){
jdbcTemplate.batchUpdate(sqlBatch.toArray(new String[sqlBatch.size()]));
sqlBatch.clear();
}
还有
@Transactional(rollbackFor = IOException.class)
public void processFile(BufferedReader reader, String tableName) {
eof = false;
...
但是如果您想要一个更清晰、更安全的解决方案,请对代码进行如下更改:
public class DBImporter {
private final static Logger log = LoggerFactory.getLogger(DBImporter.class);
private static final List<String> NULL_VALUES = Arrays.asList("", "N.A", "N.A", "UNKNOWN");
private static final List<String> COL_HEADERS = Arrays.asList("USER_ID", "NAME", "TYPE", "SRC_DATA");
private static final int BATCH_SIZE = 50;
@Autowired
private JdbcTemplate jdbcTemplate;
@Transactional(rollbackFor = IOException.class)
public void processFile(BufferedReader reader, String tableName) {
AtomicBoolean eof = new AtomicBoolean(false);
List<String> sqlBatch = new ArrayList<String>(BATCH_SIZE);
log.info("Starte auslesen der Daten");
long t1 = System.currentTimeMillis();
log.info("Start time: " + t1);
jdbcTemplate.execute("DELETE FROM " + tableName);
while (!eof.get()) {
try {
Map<String, ColumnData> dbColumns = getDBColumns(tableName);
// Get a list of db column data related to the column headers.
List<ColumnData> columnData = COL_HEADERS.stream().map(dbColumns::get).collect(toList());
// Get the next valid data row if its starts from "R" or "T".
List<String> dataRow = findNextLineStartingWith(reader, eof, "R", "T");
String query = createSql(tableName, columnData, dataRow);
sqlBatch.add(query);
// Process batch.
if (sqlBatch.size() >= BATCH_SIZE) {
jdbcTemplate.batchUpdate(sqlBatch.toArray(new String[sqlBatch.size()]));
sqlBatch.clear();
}
} catch (IllegalStateException e) {
break;
} catch (IOException e) {
log.error(e.getLocalizedMessage());
}
}
// Commit the final batch.
jdbcTemplate.batchUpdate(sqlBatch.toArray(new String[sqlBatch.size()]));
sqlBatch.clear();
long delta = System.currentTimeMillis() - t1;
log.info("Total runtime : " + delta / 1000 + " seconds");
}
/**
* Create a SQL insert query using the data row.
*
* @param tableName Name of the table.
* @param columnData Column data list.
* @param dataRow Data row to be inserted.
* @return Generated SQL query string.
*/
private String createSql(String tableName, List<ColumnData> columnData, List<String> dataRow) {
List<String> values = new ArrayList<>(columnData.size());
for (int i = 0; i < columnData.size(); i++) {
if (NULL_VALUES.contains(dataRow.get(i))) {
values.add("NULL");
} else if (columnData.get(i).getType() >= Types.NUMERIC && columnData.get(i).getType() <= Types.DOUBLE) {
values.add(dataRow.get(i));
} else {
values.add("'" + dataRow.get(i).replace("'", "''") + "'");
}
}
return "INSERT INTO " + tableName + " (" +
columnData.stream().filter(Objects::nonNull).map(ColumnData::getName).collect(joining(", ")) +
", SYSTEM_INSERTED_AT) VALUES (" +
values.stream().collect(joining(", ")) +
", CURRENT_TIMESTAMP)";
}
/**
* Find the next line starting with the given string and split it into columns.
*
* @param reader BufferedReader object to be used.
* @param prefixes A list of prefixes to look for in the string.
* @return List of data objects.
* @throws IOException
*/
private List<String> findNextLineStartingWith(BufferedReader reader, AtomicBoolean eof, String... prefixes) throws IOException {
while (true) {
String line = readLineOrThrow(reader, eof);
for (String prefix : prefixes)
if (line.startsWith(prefix)) {
ArrayList<String> data = new ArrayList<>();
// Split the line using the delimiter.
data.addAll(Arrays.asList(line.split(";")));
// Build the row to be inserted.
List<String> row = Arrays.asList(data.get(1), data.get(2).trim(), "", "");
// Insert type depending on the prefix.
if (prefix.equals("R"))
row.set(2, "USER");
else if (prefix.equals("T"))
row.set(2, "PERM");
row.set(3, String.join(";", row.subList(0, 3)));
return row;
}
}
}
/**
* Read a single line in the file.
*
* @param reader BufferedReader object to be used.
* @return
* @throws IOException
*/
private String readLineOrThrow(BufferedReader reader, AtomicBoolean eof) throws IOException {
String line = reader.readLine();
if (line == null) {
eof.set(true);
throw new IllegalStateException("Unexpected EOF");
}
return line.trim();
}
/**
* Read database column metadata.
*
* @param tableName Name of the table to process.
* @return A map containing column information.
*/
private Map<String, ColumnData> getDBColumns(String tableName) {
Map<String, ColumnData> result = new HashMap<>();
try (Connection connection = jdbcTemplate.getDataSource().getConnection()) {
ResultSet rs = connection.getMetaData().getColumns(null, null, tableName, null);
while (rs.next()) {
String columnName = rs.getString(4).toUpperCase();
int type = rs.getInt(5);
result.put(columnName, new ColumnData(columnName, type));
}
return result;
} catch (SQLException e) {
throw new RuntimeException(e);
}
}
}
关于java.lang.IllegalArgumentException : SQL array must not be empty,我们在Stack Overflow上找到一个类似的问题: https://stackoverflow.com/questions/58822069/
在 C 中: int a[10]; printf("%p\n", a); printf("%p\n", &a[0]); 产量: 0x7fff5606c600 0x7fff5606c600 这是我所期望
我一直在尝试运行此循环来更改基于数组的元素的位置,但出现以下错误。不太确定哪里出了问题。任何想法或想法!谢谢。 var population = [[98, 8, 45, 34, 56], [9, 1
我正在尝试获取一个 Ruby 数组数组并将其分组以计算其值。 数组有一个月份和一个 bool 值: array = [["June", false], ["June", false], ["June"
所以我们的目标是在遇到某个元素时将数组分割成子数组下面的示例 array.split("stop here") ["haii", "keep", "these in the same array bu
在this问题已经回答了两个表达式是相等的,但在这种情况下它们会产生不同的结果。对于给定的 int[] 分数,为什么会这样: Arrays.stream(scores) .forEac
我认为我需要的是哈希数组的数组,但我不知道如何制作它。 Perl 能做到吗? 如果是这样,代码会是什么样子? 最佳答案 perldoc perldsc是了解 Perl 数据结构的好文档。 关于arra
我遇到了这个问题,从 API 中我得到一个扩展 JSON,其中包含一个名为坐标的对象,该对象是一个包含数组 o 数组的数组。 为了更清楚地看这个例子: "coordinates": [
postgres 中有(v 9.5,如果重要的话): create table json_test( id varchar NOT NULL, data jsonb NOT NULL, PRIM
我用 echo "${array[@]}" 和 echo "${array[*]}" 得到了相同的结果。 如果我这样做: mkdir 假音乐; touch fakemusic/{Beatles,Sto
我正在尝试创建 typealias 对象的数组数组 - 但我收到“表达式类型不明确,没有更多上下文”编译错误。这是我的代码: typealias TestClosure = ((message: St
如果您在 Python 中创建一维数组,使用 NumPy 包有什么好处吗? 最佳答案 这完全取决于您打算如何处理数组。如果您所做的只是创建简单数据类型的数组并进行 I/O,array模块就可以了。 另
当我将数组推送到只有一个数组作为其唯一元素的数组数组时,为什么会得到这种数据结构? use v6; my @d = ( [ 1 .. 3 ] ); @d.push( [ 4 .. 6 ] ); @d.
在 Julia 中,我想将定义为二维数组向量的数据转换为二维矩阵数组。 如下例所述,我想把数据s转换成数据t,但是至今没有成功。 我该如何处理这个案子? julia> s = [[1 2 3], [4
C 没有elementsof 关键字来获取数组的元素数。所以这通常由计算 sizeof(Array)/sizeof(Array[0]) 代替但这需要重复数组变量名。1[&Array] 是指向数组后第一
所以,假设我有一个像这样的(愚蠢的)函数: function doSomething(input: number|string): boolean { if (input === 42 || in
我有以下数组: a = [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16] 我将它用于一些像这样的视觉内容: 1 2 3 4 5 6 7 8 9 10
我想知道数组中的 .toList 与 .to[List] 之间有什么区别。我在spark-shell中做了这个测试,结果没有区别,但我不知道用什么更好。任何意见? scala> val l = Arr
我很难获得完全相同对象的多个元素的当前元素索引: $b = "A","D","B","D","C","E","D","F" $b | ? { $_ -contains "D" } 替代版本: $b =
我正在尝试使用来自我的 API 的 v-select 执行 options,我将数据放在数组数组中。 Array which I got from API 它应该是一个带有搜索的 select,因为它
这个问题在这里已经有了答案: String literals: pointer vs. char array (1 个回答) 4 个月前关闭。 当我执行下一个代码时 int main() {
我是一名优秀的程序员,十分优秀!