[PATCH -perfbook v2 0/3] punctcheck.pl updates

[PATCH -perfbook v2 0/3] punctcheck.pl updates

[Akira Yokosawa]

Hi Paul,

Trying again.

So, this series adds patterns in punctcheck.pl to detect misuses
of punctuation markers "\@.", ".\@", and its friends.

Patch 1/3 adds the new pattern.
Patch 2/3 adds hint messages to help you and other contributors
see what is wrong.  Hopefully the added messages are clear enough.
Patch 3/3 fixes existing misuses detected by the added pattern.

        Thanks, Akira
--
Akira Yokosawa (3):
  punctcheck.pl: Add pattern to catch reverse use of period markers
  punctcheck.pl: Add hints in complaint messages
  Fix period markers

 glossary.tex            |  2 +-
 memorder/memorder.tex   |  2 +-
 together/applyrcu.tex   |  2 +-
 utilities/punctcheck.pl | 15 +++++++++++++++
 4 files changed, 18 insertions(+), 3 deletions(-)


base-commit: 5f28b7cb0d4c447c9d6397770906c7cb4c96832c
-- 
2.25.1

[PATCH -perfbook v2 1/3] punctcheck.pl: Add pattern to catch reverse use of period markers

[Akira Yokosawa]

End-of-sentence markers ("\@.") make sense when they follow
uppercase letters.
Mid-sentence markers (".\@") make sense when they follow
lowercase letters.

Add patterns to cache reverse uses.

Signed-off-by: Akira Yokosawa <akiyks@gmail.com>
---
 utilities/punctcheck.pl | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/utilities/punctcheck.pl b/utilities/punctcheck.pl
index 7d23cc57..8820bf9c 100755
--- a/utilities/punctcheck.pl
+++ b/utilities/punctcheck.pl
@@ -91,6 +91,12 @@ sub check_line {
 	if ($line =~ /^(?=[\s]*+[^%])[^%]*\\\@[\.\?\!\:][\)\}\']*\s+[^%]/){
 	    $ng += 1;
 	}
+	if ($line =~ /[a-z][\)\}\']*\\\@[\.\?\!\:][\)\}\']*\s/) {
+	    $ng += 1;
+	}
+	if ($line =~ /[A-Z][\)\}\']*[\.\?\!\:]\\\@[\)\}\']*\s/) {
+	    $ng += 1;
+	}
 	if ($ng) {
 	    print $ARGV[0], ':', $line_num, ':', $line_raw;
 	}
-- 
2.25.1

[PATCH -perfbook v2 2/3] punctcheck.pl: Add hints in complaint messages

[Akira Yokosawa]

To help Paul (and other contributors), print a hint in front of
the violating line.

Signed-off-by: Akira Yokosawa <akiyks@gmail.com>
---
 utilities/punctcheck.pl | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/utilities/punctcheck.pl b/utilities/punctcheck.pl
index 8820bf9c..94cbafc9 100755
--- a/utilities/punctcheck.pl
+++ b/utilities/punctcheck.pl
@@ -71,6 +71,8 @@ sub check_line {
 		$next_line =~ /\\E?QuickQuizEnd[BEM]?/ ||
 		$next_line =~ /\\end\{(quot|enum|item|sequ)/ ) {
 		$ng -= 1;
+	    } else {
+		print "vvv Hint: sentence ending with an uppercase needs \"\\\@.\" vvv\n";
 	    }
 	}
 	if ($line =~ /^(?=[\s]*+[^%])[^%]*[a-z][\)\}\']*[\.\?\!][\)\}\']*\s+[^%]/ ||
@@ -80,21 +82,28 @@ sub check_line {
 	    if ($line =~ /^(?=[\s]*+[^%])[^%]*[a-z][\)\}\']*[\.\?\!][\)\}\']*\s+\\\\/ ||
 		$line =~ /^(?=[\s]*+[^%])[^%]*.*[\.:][\)\}\']*\s+\\\\/ ) {
 		$ng -= 1;
+	    } else {
+		print "vvv Hint: end of sentence needs new line vvv\n";
 	    }
 	}
 	if ($line =~ /^(?=[\s]*+[^%])[^%]*[^~]\\cite/) {
 	    $ng += 1;
 	    if ($line =~ /^(?=[\s]*+[^%])[^%]*~\(\\cite/) {
 		$ng -= 1;
+	    } else {
+		print "vvv Hint: \\cite{} needs a \"~\" in front vvv\n";
 	    }
 	}
 	if ($line =~ /^(?=[\s]*+[^%])[^%]*\\\@[\.\?\!\:][\)\}\']*\s+[^%]/){
 	    $ng += 1;
+	    print "vvv Hint: end of sentence needs new line vvv\n";
 	}
 	if ($line =~ /[a-z][\)\}\']*\\\@[\.\?\!\:][\)\}\']*\s/) {
+	    print "vvv Hint: should be \"xxx.\\\@\" vvv\n";
 	    $ng += 1;
 	}
 	if ($line =~ /[A-Z][\)\}\']*[\.\?\!\:]\\\@[\)\}\']*\s/) {
+	    print "vvv Hint: should be \"XXX\\\@.\" vvv\n";
 	    $ng += 1;
 	}
 	if ($ng) {
-- 
2.25.1

[PATCH -perfbook v2 3/3] Fix period markers

[Akira Yokosawa]

These are caught by patterns added in punctcheck.pl.

Signed-off-by: Akira Yokosawa <akiyks@gmail.com>
---
 glossary.tex          | 2 +-
 memorder/memorder.tex | 2 +-
 together/applyrcu.tex | 2 +-
 3 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/glossary.tex b/glossary.tex
index 46c68bbe..6038c9ed 100644
--- a/glossary.tex
+++ b/glossary.tex
@@ -143,7 +143,7 @@
 	takes a pointer, and old value, and a new value.
 	If the pointed-to value is equal to the old value, it is atomically
 	replaced with the new value.
-	There is some variety in CAS API.\@
+	There is some variety in CAS API\@.
 	One variation returns the actual pointed-to value, so that the
 	caller compares the CAS return value to the specified old value,
 	with equality indicating a successful CAS operation.
diff --git a/memorder/memorder.tex b/memorder/memorder.tex
index 9792099f..cb91de93 100644
--- a/memorder/memorder.tex
+++ b/memorder/memorder.tex
@@ -3303,7 +3303,7 @@ As it turns out, quite a bit:
 	same lock, and vice versa?
 \item	Suppose that a given CPU or thread executes one access
 	(call it ``A''), releases a lock, reacquires that same lock,
-	then executes another access (call it ``B'').\@
+	then executes another access (call it ``B'')\@.
 	Is some other CPU or thread not holding that lock guaranteed to
 	see A and B in order?
 \item	As above, but with the lock reacquisition carried out by some
diff --git a/together/applyrcu.tex b/together/applyrcu.tex
index 934fbed5..a3ee559e 100644
--- a/together/applyrcu.tex
+++ b/together/applyrcu.tex
@@ -816,7 +816,7 @@ For example, in Linux kernel v5.16, the \co{khugepaged_scan_file()}
 function checks to see if some other task needs the current CPU
 using \co{need_resched()}, and if so invokes \co{xas_pause()} to
 adjust the traveral's iterator appropriately, and then invokes
-\co{cond_resched_rcu()} to yield the CPU.\@
+\co{cond_resched_rcu()} to yield the CPU\@.
 In turn, the \co{cond_resched_rcu()} function invokes \co{rcu_read_unlock()},
 \co{cond_resched()}, and finally \co{rcu_read_lock()} to drop out of
 the RCU read-side critical section in order to yield the CPU.
-- 
2.25.1

[PATCH -perfbook v2 4/3] together/applyrcu: Fix typo

[Akira Yokosawa]

Signed-off-by: Akira Yokosawa <akiyks@gmail.com>
---
And I noticed this in one of the hunks in Patch 3/3.
Sending as a follow up.

        Thanks, Akira
--
 together/applyrcu.tex | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/together/applyrcu.tex b/together/applyrcu.tex
index a3ee559e..2a5a3c12 100644
--- a/together/applyrcu.tex
+++ b/together/applyrcu.tex
@@ -815,7 +815,7 @@ way of stopping in the middle and resuming later.
 For example, in Linux kernel v5.16, the \co{khugepaged_scan_file()}
 function checks to see if some other task needs the current CPU
 using \co{need_resched()}, and if so invokes \co{xas_pause()} to
-adjust the traveral's iterator appropriately, and then invokes
+adjust the travesal's iterator appropriately, and then invokes
 \co{cond_resched_rcu()} to yield the CPU\@.
 In turn, the \co{cond_resched_rcu()} function invokes \co{rcu_read_unlock()},
 \co{cond_resched()}, and finally \co{rcu_read_lock()} to drop out of
-- 
2.25.1

Re: [PATCH -perfbook v2 0/3] punctcheck.pl updates

[Paul E. McKenney]

On Mon, Apr 04, 2022 at 09:17:34AM +0900, Akira Yokosawa wrote:
> Hi Paul,
> 
> Trying again.
> 
> So, this series adds patterns in punctcheck.pl to detect misuses
> of punctuation markers "\@.", ".\@", and its friends.
> 
> Patch 1/3 adds the new pattern.
> Patch 2/3 adds hint messages to help you and other contributors
> see what is wrong.  Hopefully the added messages are clear enough.
> Patch 3/3 fixes existing misuses detected by the added pattern.

Much nicer!  I queued and pushed all four, thank you!!!

							Thanx, Paul

>         Thanks, Akira
> --
> Akira Yokosawa (3):
>   punctcheck.pl: Add pattern to catch reverse use of period markers
>   punctcheck.pl: Add hints in complaint messages
>   Fix period markers
> 
>  glossary.tex            |  2 +-
>  memorder/memorder.tex   |  2 +-
>  together/applyrcu.tex   |  2 +-
>  utilities/punctcheck.pl | 15 +++++++++++++++
>  4 files changed, 18 insertions(+), 3 deletions(-)
> 
> 
> base-commit: 5f28b7cb0d4c447c9d6397770906c7cb4c96832c
> -- 
> 2.25.1
>