From 1aeb8099d053af79d50f4ffee740c29cc10d56fc Mon Sep 17 00:00:00 2001
From: Jonathan Corbet <corbet@lwn.net>
Date: Mon, 30 Jun 2025 11:08:32 -0600
Subject: docs: kdoc: rework type prototype parsing

process_proto_type() is using a complex regex and a "while True" loop to
split a declaration into chunks and, in the end, count brackets.  Switch to
using a simpler regex to just do the split directly, and handle each chunk
as it comes.  The result is, IMO, easier to understand and reason about.

The old algorithm would occasionally elide the space between function
parameters; see struct rng_alg->generate(), foe example.  The only output
difference is to not elide that space, which is more correct.

Signed-off-by: Jonathan Corbet <corbet@lwn.net>
---
 scripts/lib/kdoc/kdoc_parser.py | 43 ++++++++++++++++++++++++-----------------
 1 file changed, 25 insertions(+), 18 deletions(-)

(limited to 'scripts/lib/kdoc/kdoc_parser.py')

diff --git a/scripts/lib/kdoc/kdoc_parser.py b/scripts/lib/kdoc/kdoc_parser.py
index 935f2a3c4b47..61da297df623 100644
--- a/scripts/lib/kdoc/kdoc_parser.py
+++ b/scripts/lib/kdoc/kdoc_parser.py
@@ -1594,30 +1594,37 @@ class KernelDoc:
 
         # Strip C99-style comments and surrounding whitespace
         line = KernRe(r"//.*$", re.S).sub('', line).strip()
+        if not line:
+            return # nothing to see here
 
         # To distinguish preprocessor directive from regular declaration later.
         if line.startswith('#'):
             line += ";"
-
-        r = KernRe(r'([^\{\};]*)([\{\};])(.*)')
-        while True:
-            if r.search(line):
-                if self.entry.prototype:
-                    self.entry.prototype += " "
-                self.entry.prototype += r.group(1) + r.group(2)
-
-                self.entry.brcount += r.group(2).count('{')
-                self.entry.brcount -= r.group(2).count('}')
-
-                if r.group(2) == ';' and self.entry.brcount <= 0:
+        #
+        # Split the declaration on any of { } or ;, and accumulate pieces
+        # until we hit a semicolon while not inside {brackets}
+        #
+        r = KernRe(r'(.*?)([{};])')
+        for chunk in r.split(line):
+            if chunk:  # Ignore empty matches
+                self.entry.prototype += chunk
+                #
+                # This cries out for a match statement ... someday after we can
+                # drop Python 3.9 ...
+                #
+                if chunk == '{':
+                    self.entry.brcount += 1
+                elif chunk == '}':
+                    self.entry.brcount -= 1
+                elif chunk == ';' and self.entry.brcount <= 0:
                     self.dump_declaration(ln, self.entry.prototype)
                     self.reset_state(ln)
-                    break
-
-                line = r.group(3)
-            else:
-                self.entry.prototype += line
-                break
+                    return
+        #
+        # We hit the end of the line while still in the declaration; put
+        # in a space to represent the newline.
+        #
+        self.entry.prototype += ' '
 
     def process_proto(self, ln, line):
         """STATE_PROTO: reading a function/whatever prototype."""
-- 
cgit v1.2.3