Powershell命令在循环中的两点之间读取并写入文件末尾

Question

有这个 mySQL 文件 (file.sql)，我想复制起点和终点两点之间的文本并写入一个新的 .sql 文件。我在一个循环中需要这个，它将运行一个数组中的 N 个表并写入一个新的 .sql 文件。在每次迭代结束时，我需要添加新行然后将新表的文本附加到文件末尾的过程。

--DO NOT COPY
--DO NOT COPY
-- Table structure for table `address`
--

DROP TABLE IF EXISTS `address`;
/*!40101 SET @saved_cs_client     = @@character_set_client */;
/*!40101 SET character_set_client = utf8 */;
CREATE TABLE `address` (
`ENCODEDKEY` varchar(32) CHARACTER SET utf8 COLLATE utf8_bin NOT NULL,
`ADDRESSTYPE` varchar(256) CHARACTER SET utf8 COLLATE utf8_bin DEFAULT 
NULL,
`CITY` varchar(256) CHARACTER SET utf8 COLLATE utf8_bin DEFAULT NULL
) ENGINE=InnoDB DEFAULT CHARSET=utf8;
/*!40101 SET character_set_client = @saved_cs_client */;

--
-- Dumping data for table `address`
--

 LOCK TABLES `address` WRITE;
/*!40000 ALTER TABLE `address` DISABLE KEYS */;
INSERT INTO `address` VALUES ( 'ENCODEDKEY','ADDRESSTYPE','CITY');
/*!40000 ALTER TABLE `address` ENABLE KEYS */;
UNLOCK TABLES;
--DO NOT COPY
--DO NOT COPY
 
-- Table structure for table `client`
--
DROP TABLE IF EXISTS `client`;
/*!40101 SET @saved_cs_client     = @@character_set_client */;
/*!40101 SET character_set_client = utf8 */;
CREATE TABLE `client` (
`KEY` varchar(32) CHARACTER SET utf8 COLLATE utf8_bin NOT NULL,
`TYPE` varchar(256) CHARACTER SET utf8 COLLATE utf8_bin DEFAULT 
 ) ENGINE=InnoDB DEFAULT CHARSET=utf8;
/*!40101 SET character_set_client = @saved_cs_client */;

--
-- Dumping data for table `client`
--

 LOCK TABLES `client` WRITE;
/*!40000 ALTER TABLE `client` DISABLE KEYS */;
INSERT INTO `client` VALUES ( 'KEY','TYPE');
/*!40000 ALTER TABLE `address` ENABLE KEYS */;
UNLOCK TABLES;
--DO NOT COPY
--DO NOT COPY

 

对于每个迭代，起点应该是 = 的确切字符串

表'find_tablename'的表结构

结束点应该是 = 的确切字符串

解锁表格；

我想复制/读取这 2 点之间的所有文本并写入一个新的 .sql 文件。这个过程应该在一个循环中，循环将运行 N 个表，并在添加新行后附加到同一个 .sql 文件。在这种情况下，它将针对表地址和客户端运行 2 次迭代。

有这个脚本在 PowerShell 中为地址表执行进程。我还考虑在 cmets 中使用如下所示的 Array，它允许我轻松添加/删除我需要的表，然后使用 FOR LOOP 进行迭代。我在 cmets 中的代码是我试图添加以执行此循环的代码。因此位置 tableList[num] 的值将替换 address 。

 # $tableList = New-Object -TypeName 'System.Collections.ArrayList';
 # $tableList.Add("address")
 # $tableList.Add("client")

 # for ($num = 1 ; $num -le $tableList.Count ; $num++)
 # {
$write = $false
$(switch -CaseSensitive -File ("file.sql") {
'-- Table structure for table `address`' { $write = $true; $_ }
'UNLOCK TABLES;' { if ($write) { $_; break; } }
default { if ($write) { $_ }}
 }) | Set-Content "new_file.sql"
 #}

Answer 1

基于现有的解决方案，我创建了这个函数：

function Select-TableSql {
    param (
        [Parameter(Mandatory, Position = 0)]
        [string]$Path,

        [Parameter(Mandatory, Position = 1)]
        [string[]]$TableNames
    )
    $write = $false
    switch -Regex -File ($Path) {
        '-- Table structure for table `(?<TableName>\w+)`' {
            if ($Matches.TableName -in $TableNames) {
                $write = $true
                $_
            }
        }
        'UNLOCK TABLES;' {
            if ($write) {
                $write = $false
                $_
            }
        }
        default {
            if ($write) {
                $_
            }
        }
    }
}

你可以这样使用它：

Select-TableSql file.sql client, address | Set-Content new_file.sql

（注意：这里直接在文件上使用switch 的目的是因为它在大文件上比与Get-Content 的任何组合都快得多。）

Answer 2

我不会在此使用switch，而是使用-split、-replace 和最后-join 的组合：

# read the original file as single multiline string
$sql = Get-Content -Path 'D:\Test\SqlFile.sql' -Raw

# 1) split the whole content on 'UNLOCK TABLES'
# 2) keep only the text blocks that contain '-- Table structure for table'
# 3) remove all text above '-- Table structure for table' from each text block. 
# 4) Append the final 'UNLOCK TABLES;' sql command we've removed using the first split

$result = $sql -split '(?ms)^UNLOCK TABLES;\s\r?\n' | Where-Object { $_ -match '-- Table structure for table' } | ForEach-Object {
    ($_ -replace '(?ms)^.*(-- Table structure for table .*)', '$1') + "UNLOCK TABLES;"
}

您现在可以将生成的数组保存为单个文件，但如果我正确理解问题，您希望将数组合并到一个新文件中

$result -join "`r`n`r`n" | Set-Content -Path 'D:\Test\newSqlFile.sql'

生成的文件将如下所示：

-- Table structure for table `address`
--

DROP TABLE IF EXISTS `address`;
/*!40101 SET @saved_cs_client     = @@character_set_client */;
/*!40101 SET character_set_client = utf8 */;
CREATE TABLE `address` (
`ENCODEDKEY` varchar(32) CHARACTER SET utf8 COLLATE utf8_bin NOT NULL,
`ADDRESSTYPE` varchar(256) CHARACTER SET utf8 COLLATE utf8_bin DEFAULT 
NULL,
`CITY` varchar(256) CHARACTER SET utf8 COLLATE utf8_bin DEFAULT NULL
) ENGINE=InnoDB DEFAULT CHARSET=utf8;
/*!40101 SET character_set_client = @saved_cs_client */;

--
-- Dumping data for table `address`
--

 LOCK TABLES `address` WRITE;
/*!40000 ALTER TABLE `address` DISABLE KEYS */;
INSERT INTO `address` VALUES ( 'ENCODEDKEY','ADDRESSTYPE','CITY');
/*!40000 ALTER TABLE `address` ENABLE KEYS */;
UNLOCK TABLES;

-- Table structure for table `client`
--
DROP TABLE IF EXISTS `client`;
/*!40101 SET @saved_cs_client     = @@character_set_client */;
/*!40101 SET character_set_client = utf8 */;
CREATE TABLE `client` (
`KEY` varchar(32) CHARACTER SET utf8 COLLATE utf8_bin NOT NULL,
`TYPE` varchar(256) CHARACTER SET utf8 COLLATE utf8_bin DEFAULT 
 ) ENGINE=InnoDB DEFAULT CHARSET=utf8;
/*!40101 SET character_set_client = @saved_cs_client */;

--
-- Dumping data for table `client`
--

 LOCK TABLES `client` WRITE;
/*!40000 ALTER TABLE `client` DISABLE KEYS */;
INSERT INTO `client` VALUES ( 'KEY','TYPE');
/*!40000 ALTER TABLE `address` ENABLE KEYS */;
UNLOCK TABLES;

---

编辑

确实，我没有在您的问题中阅读有关仅过滤出一组固定表的任何内容，但这很容易实现：

# the list of table names you want to retrieve from the file
$tablesToExtract = 'address','client','someothertable' # etc.

# read the original file as single multiline string
$sql = Get-Content -Path 'D:\Test\SqlFile.sql' -Raw

# 1) split the whole content on 'UNLOCK TABLES'
# 2) keep only the text blocks that contain '-- Table structure for table'
# 3) parse out the table name and check if that can be found in the $tablesToExtract array
# 4) remove all text above '-- Table structure for table' from each text block. 
# 5) Append the final 'UNLOCK TABLES;' sql command we've removed using the first split
$result = $sql -split '(?ms)^UNLOCK TABLES;\s\r?\n' | Where-Object { $_ -match '-- Table structure for table' } | ForEach-Object {
    # check if the table is in the list
    $tableName = ([regex]'-- Table structure for table `(\w+)`').Match($_).Groups[1].Value
    if ($tablesToExtract -contains $tableName) {
        ($_ -replace '(?ms)^.*(-- Table structure for table .*)', '$1') + "UNLOCK TABLES;"
    }
}

您现在可以将生成的数组保存为单个文件，但如果我正确理解问题，您希望将数组合并到一个新文件中

$result -join "`r`n`r`n" | Set-Content -Path 'D:\Test\newSqlFile.sql'