文本处理：如何逆序提取Pattern A至首个Pattern B匹配区间行

Let's break down practical solutions for both of your text extraction tasks using standard Unix tools (awk, tac) — they're widely available and ideal for these jobs.

Problem 1: Extract from Pattern A to the First Pattern B #

If you need to grab all lines starting from the first match of Pattern A up to (and including) the first subsequent Pattern B, here's a straightforward awk one-liner:

awk '/Pattern A/{flag=1} flag; /Pattern B/{flag=0}' example_file.txt

How this works: #

• When awk hits a line matching Pattern A, it sets a flag to 1 (active).
• The flag; line tells awk to print the current line only if the flag is on.
• Once we encounter a line with Pattern B, the flag is turned off, stopping further printing.

Edge Case Adjustments: #

• No Pattern B after A: This will print everything from Pattern A to the end of the file.
• Only want the first occurrence: Add exit after turning off the flag to stop processing once the first Pattern B is found:

  awk '/Pattern A/{flag=1} flag; /Pattern B/{flag=0; exit}' example_file.txt
  

Problem 2: Bottom-Up Interval Extraction (AK5*R to AK2) #

For this task, we need to scan from the end of the file upwards, find every line matching AK5*R, and extract the lines from that line up to the first preceding AK2. We'll name these intervals E1 (last occurrence), E2 (second-last), etc.

Full Step-by-Step Solution: #

1. First, reverse the file so we can process from bottom to top using tac:

   tac example_file.txt > reversed_file.txt
   

2. Use awk to split the reversed file into target blocks and save them to numbered files:

   awk '
   /AK5\*R/ {
       if (block) { print block > "E" cnt; cnt++ }
       block = $0 "\n"
       next
   }
   /AK2/ {
       block = block $0 "\n"
       print block > "E" cnt
       cnt++
       block = ""
       next
   }
   { if (block) block = block $0 "\n" }
   END { if (block) print block > "E" cnt }
   ' reversed_file.txt
   

3. Reverse each E file to restore the original line order:

   for file in E*; do tac "$file" > "$file.tmp" && mv "$file.tmp" "$file"; done
   

How this works: #

• tac reverses the input file, so the last line of the original becomes the first line we process.
• The awk script collects lines into a block when it hits AK5*R. It keeps adding lines until it finds AK2, then saves the block to a numbered E file. If there's an unfinished block at the end (no AK2 after the last AK5*R), it saves that too.
• Finally, we reverse each E file to get the lines back in their original order (since we processed reversed input).

Result: #

• E1 contains the last occurrence of AK5*R in the original file, along with all lines up to the first AK2 above it.
• E2 contains the second-last AK5*R interval, and so on.

内容的提问来源于stack exchange，提问作者WashichawbachaW