如何在日志文件中查找根目录下2级深度的文件路径字符串

Got it, let's solve this problem—since find is for locating actual files on your filesystem, it won't help with parsing log strings. Instead, we'll use text-processing tools like grep or awk to target exactly the path format you need: /xxx/xxx/xxx.xxx (root + 2 directories + filename, no deeper levels).

Method 1: Use grep with Regular Expressions #

This is the quickest way if you just need to extract matching paths. We'll use a regex that matches the exact depth and excludes deeper paths.

Command: #

grep -oE '/[^/]+/[^/]+/[^/]+(?!/)' your_log_file.log

Breakdown: #

• -o: Only output the matched part of the line (instead of the whole line)
• -E: Enable extended regular expressions
• /[^/]+: Matches a directory segment (starts with /, followed by one or more non-/ characters)
• We repeat this twice to cover the two directory levels, then add /[^/]+ for the filename
• (?!/): A negative lookahead that ensures there's no trailing / after the filename—this prevents partial matches of deeper paths (like /a/b/c from /a/b/c/d.txt)

If your logs have paths embedded in longer lines and you want to avoid false positives (e.g., paths with extra characters attached), you can tweak the regex to match word boundaries:

grep -oE '\b/[^/]+/[^/]+/[^/]+(?!/)\b' your_log_file.log

Method 2: Use awk for More Flexibility #

If you need to do additional processing (like counting paths, filtering by other log fields), awk is a better choice. We'll split paths by / to check their depth explicitly.

One-Liner Command: #

awk '{
    while (match($0, /\/[^\/]+\/[^\/]+\/[^\/]+(?!\/)/, path_arr)) {
        print path_arr[0]
        $0 = substr($0, RSTART + RLENGTH)
    }
}' your_log_file.log

Breakdown: #

• match($0, regex, path_arr): Finds the first matching path in the current line and stores it in path_arr[0]
• print path_arr[0]: Outputs the matched path
• substr($0, RSTART + RLENGTH): Truncates the line to the part after the matched path, so we can find multiple paths in one line
• The regex works the same way as the grep version—ensuring we only get 2-level deep root paths

Alternatively, if you want to split paths and check depth numerically (useful for debugging edge cases):

awk '{
    for (i=1; i<=NF; i++) {
        split($i, parts, "/")
        # A valid 2-level path will split into 4 elements (e.g., "/a/b/c.txt" → ["", "a", "b", "c.txt"])
        if ($i ~ /^\// && length(parts) == 4 && parts[4] != "") {
            print $i
        }
    }
}' your_log_file.log

Note: This assumes paths are separated by whitespace. If your paths contain spaces, use the match method above instead.

Key Note #

Remember: find's maxdepth/mindepth operate on the filesystem's directory structure. When parsing log strings, we're dealing with text patterns, so text-processing tools are the right approach here.

内容的提问来源于stack exchange，提问作者Eggiderm