Sed处理多行文件内容变量时报错，求解决方案

I see the issue here—when your variable $a contains newlines, embedding it directly into the sed substitution command breaks things. Sed parses commands line-by-line, so the newlines in $a split the s command into incomplete chunks, hence the "unterminated `s' command" error.

Let’s break down practical solutions to fix this, depending on whether you want to flatten the multi-line content into a single line or preserve its original line breaks:

Solution 1: Flatten Multi-Line Content Into a Single Line #

If you want to merge all lines from the file into one continuous string (matching your expected output format), first strip newlines and escape characters that would break sed’s substitution logic:

# Read file, remove newlines, and escape sed special characters (&, \, /)
a=$(tr -d '\n' < file | sed -e 's/\\/\\\\/g' -e 's/\//\\\//g' -e 's/&/\\&/g')

# Use | as the sed delimiter to avoid conflicts with / in content
sed "/^[[:blank:]]*\"data_$2\": \[$/,/^[[:blank:]]*],*$/ {s|^[[:blank:]]*}$|&$a|}" "$1"

Why this works: #

• tr -d '\n' removes all newlines from the file content, ensuring $a is a single line.
• The sed escape steps handle characters with special meaning in substitutions (backslashes, slashes, and ampersands) to prevent unexpected behavior.
• Using | instead of / as the substitution delimiter avoids conflicts if your file content contains slashes.

Solution 2: Preserve Original Line Breaks #

If you want to keep the file’s line breaks (even if it makes the JSON technically invalid), escape newlines as \n so sed interprets them correctly:

# Read entire file into one line, escape special chars, replace newlines with \n
a=$(sed -e ':a' -e 'N' -e '$!ba' -e 's/\\/\\\\/g' -e 's/\//\\\//g' -e 's/&/\\&/g' -e 's/\n/\\n/g' file)

sed "/^[[:blank:]]*\"data_$2\": \[$/,/^[[:blank:]]*],*$/ {s|^[[:blank:]]*}$|&$a|}" "$1"

How this works: #

• :a;N;$!ba is a sed trick to read the entire file into a single buffer (instead of processing line-by-line).
• We escape special characters first, then replace each newline with \n—sed will render this as a literal newline in the output.

Solution 3: Avoid Variables Entirely (Simplest Approach) #

Skip storing the file content in a variable altogether, and have sed directly read the file at the target position using the r command:

sed "/^[[:blank:]]*\"data_$2\": \[$/,/^[[:blank:]]*],*$/ {
  /^[[:blank:]]*}$/ {
    # Append file content right after the closing }
    s/^[[:blank:]]*}$/&/
    r file
  }
}" "$1"

Note: #

This inserts the file content on a new line after the }. If you need it on the same line as the }, stick with Solutions 1 or 2.

Testing With Your Example #

For a file with content:

yeah
is
test
my
script

Solution 1 will produce exactly your expected output:

{ "mytable_test": [ { "test1": "abc", "test2": "def", "test2": "ghi" }yeah is test my script ], "mytable": [ "test1", "test2", "test3" ] }

内容的提问来源于stack exchange，提问作者Mathie