CSV（在字段值中有额外的引号）到ColdFusion中的数组

Question

我正在使用这篇文章将 CSV 文件转换为数组。一切都很好。但是我得到了一个文件，其中包含字段值中的额外引号，例如：

“bash:”快捷方式“是”和

< code>"bash: \ "快捷键\ "是" 。

所以我尝试像这样替换这些引号：

<cffile action="read" file="#filePath#" variable="csvContent">
<cfset csvContent = reReplace(csvContent, '(?:[^,\r\n])"(?:[^,\r\n])', '&quot;', 'ALL')>

<--- Then do the conversion --->
<cfset array = csvToArray(csv = csvContent)>

但是非捕获组不工作。我做错了什么？

还有别的办法吗？

编辑1：

我还尝试使用cfhttp并得到以下错误：

<cfhttp name="csvToQuery" method="get" url="#url#" />

详细信息：验证列属性和目标文件中指定的列数

消息:行中的列数不正确。

stack trace:cold fusion . tagext . net . http tag $ InvalidColumnsException:行中的列数不正确。在coldfusion.tagext . net . httptag . conn helper(http tag . Java:1149)在cold fusion . tagext . net . http tag . doendtag(http tag . Java:1219)在cfmfhttp2 ecfm 308364137 . run page(C:\ inetpub \ wwwroot \ mfhttp . cfm:1)在cold fusion . runtime . cfjsppage . invoke(cfjsppage . Java:244)在cold fusion . tagextcfm servlet . service(cfm servlet . Java:219)at cold fusion . bootstrap . bootstrap servlet . service(bootstrap servlet . Java:89)at org . Apache . catalina . core . applicationfilterchain . internal dofilter(applicationfilter chain . Java:303)at org . Apache . catalina . core . applicationfilter chain . do filter(applicationfilter chain . Java:208)at cold fusion . monitor . event . monitoring servlet filter . do filter(monitoring servlet filter

Answer 1

哦，你不能这么容易地自己修复这种输入。正则表达式会进一步破坏你的数据。

你能在Java中创建一个小脚本来处理这个问题吗？如果你这样做了，然后使用uniVardy-parser来读取你的CSV输入，并用正确的引号转义写回它：

这是唯一一个可以处理断引号转义的CSV解析器。请尝试以下示例：

import com.univocity.parsers.csv;

import java.io.*;
import java.util.*;

public class Test {

    public static void main(String ... args){
        CsvParserSettings settings = new CsvParserSettings();
        settings.getFormat().setLineSeparator("\r\n");
        settings.setParseUnescapedQuotes(true); // THIS IS IMPORTANT FOR YOU
        CsvParser parser = new CsvParser(settings);

        String line1 = "something,\"a quoted value \"with unescaped quotes\" can be parsed\", something\r\n";
        System.out.println("Input line: " + line1);

        String line2 = "\"after the newline \r\n you will find \" more stuff\r\n";
        System.out.println("Input line: " + line2);

        List<String[]> allInputLines = parser.parseAll(new StringReader(line1 + line2));

        System.out.println("===============\nParsed input values\n===============");
        int count = 0;
        for(String[] line : allInputLines){
            System.out.println("From line " + ++count + ":");
            for(String element : line){
                System.out.println("\t" + element);

            }
            System.out.println();
        }

        //Let's write your output CSV
        StringWriter output = new StringWriter();
        CsvWriterSettings writerSettings = new CsvWriterSettings();
        writerSettings.getFormat().setLineSeparator("\r\n");
        writerSettings.getFormat().setQuoteEscape('\\'); //it seems you are using backslash as quote escape
        writerSettings.getFormat().setCharToEscapeQuoteEscaping('\\'); //when your quote escape character is not the same as the quote character, you might need to escape the escape character as well
        writerSettings.setQuoteAllFields(true); //let's force quotes on all fields so whatever is parsing your input file has more  chance of doing it properly
        CsvWriter writer = new CsvWriter(output, writerSettings);

        for(String[] row : allInputLines){
            writer.writeRow(row);
        }
        writer.close();

        System.out.println("===============\nNicely formatted output\n===============");
        System.out.println(output.toString());

    }

}

此代码将产生以下输出(可能会被您的数据导入工具读取):

Input line: something,"a quoted value "with unescaped quotes" can be parsed", something

Input line: "after the newline 
you will find " more stuff

===============
Parsed input values
===============
From line 1:
    something
    a quoted value "with unescaped quotes" can be parsed
    something

From line 2:
    after the newline 
you will find " more stuff


===============
Nicely formatted output
===============
"something","a quoted value \"with unescaped quotes\" can be parsed","something"

"after the newline 
 you will find \" more stuff"

披露：我是这个库的作者。它是开源和免费的（Apache V2.0许可证）。

ColdFusion 10示例：

>

将罐子装入Application.cfc

this.javaSettings = { loadPaths: ["C:\path\to\univocity-parsers-1.5.6.jar" ]};

使用 createObject 创建解析器类的实例：

filePath = "c:\path\to\yourFile.csv";
settings = createObject("java", "com.univocity.parsers.csv.CsvParserSettings").init();
settings.getFormat().setLineSeparator(chr(13)& chr(10));
settings.getFormat().setQuoteEscape("\");
settings.setParseUnescapedQuotes(true); // THIS IS IMPORTANT FOR YOU
parser = createObject("java", "com.univocity.parsers.csv.CsvParser").init(settings);
reader = createObject("java", "java.io.StringReader").init(fileRead(filePath));
arrayOfLines = parser.parseAll(reader);

// display results
counter = 1;
for (line in arrayOfLines) {
    writeOutput("<br>From line "& (counter++) & ":");
    for (element in line) {
       writeOutput("<br>"& element);
    }
}