Mac端TextWrangler正则处理：两类字符串截取与删除需求

Got it, let's tackle these two questions one by one with clear, step-by-step solutions tailored for TextWrangler:

TextWrangler uses grep-style regex for find-and-replace, so here's how to handle both scenarios:

Scenario 1: Keep only the first two comma-separated fields (remove everything after the second comma) #

1. Open your target file in TextWrangler, press Cmd + F to launch the Find/Replace window.
2. Check the Grep box in the top-right corner to enable regex mode.
3. In the "Find" field, paste this regex:

   ^([^,]*,[^,]*),.*
   

4. In the "Replace" field, enter:

   \1
   

5. Click "Replace All" (or use "Replace" first to test individual lines and make sure it works as expected).

How it works:

• ^ anchors the match to the start of the line.
• ([^,]*,[^,]*) captures the first two fields (everything up to and including the content right before the second comma).
• ,.* matches the second comma and all text after it. Replacing with \1 keeps only the captured first two fields.

Scenario 2: Remove text after the last comma in each line #

1. Follow steps 1-2 above to enable Grep mode.
2. "Find" regex:

   (.*),[^,]*$
   

3. "Replace" with:

   \1
   

4. Run replace all or test individual lines as needed.

How it works:

• (.*), captures all text up to (and including) the last comma in the line.
• [^,]*$ matches any non-comma characters from the last comma to the end of the line. Replacing with \1 strips that trailing part entirely.

Since you mentioned attaching original and expected results, let's use a common example to demonstrate. Suppose your original string looks like this:

"apple","banana","cherry","date","elderberry","fig","grape"

And you want the result to be:

"apple","banana","date","elderberry","grape"

Here's how to do this in TextWrangler (again using Grep mode):

1. Open Find/Replace, enable Grep.
2. "Find" regex:

   ((?:[^"]*"[^"]*",){2})[^"]*"[^"]*",((?:[^"]*"[^"]*",){2})[^"]*"[^"]*",(.*)
   

3. "Replace" with:

   \1\2\3
   

4. Run the replace operation.

How it works:

• (?:[^"]*"[^"]*",) is a non-capturing group that matches one full quoted string plus its trailing comma.
• {2} repeats that group twice to capture the first two quoted strings ("apple","banana",).
• [^"]*"[^"]*", skips the 3rd quoted string (we don't capture this part, so it gets dropped).
• The next ((?:...) {2}) captures the 4th and 5th strings ("date","elderberry",).
• Another [^"]*"[^"]*", skips the 6th string.
• (.*) captures everything after the 6th string (the 7th string here).
• Replacing with \1\2\3 stitches together the captured parts, leaving out the 3rd and 6th strings entirely.

If your actual string has a different structure (like no trailing comma, or more/less total strings), just adjust the regex group counts accordingly—this pattern scales as long as you need to skip the 3rd and 6th quoted entries.

内容的提问来源于stack exchange，提问作者joe4java