Skip to contentSkip to conversation

CodingStyle: start flamewar about use of braces

DISCUSSION Randy Dunlap 7 participants 8 messages May 8, 2007 – May 9, 2007 1 message not in the archive

lore
Share & tools
lore

A message that is not in the archive (sent off-list or not yet received): [email protected]

#1 lore raw
Quoted 9 lines · On Tue, 8 May 2007 19:03:10 GMT Linux Kernel Mailing List wrote:
Gitweb:     http://git.kernel.org/git/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=e659ba4a0d2d471c0d73590f78e1a1b5a1eede48
Commit:     e659ba4a0d2d471c0d73590f78e1a1b5a1eede48
Parent:     28be5abb400e5e082f5225105fdc69337ec0c0b4
Author:     Oliver Neukum <[email protected]>
AuthorDate: Tue May 8 00:30:34 2007 -0700
Committer:  Linus Torvalds <[email protected]>
CommitDate: Tue May 8 11:15:12 2007 -0700

CodingStyle: start flamewar about use of braces

It worked somewhat. We never did reach any kind of concensus about this....

Quoted 24 lines
---
 Documentation/CodingStyle |   15 +++++++++++++++
 1 files changed, 15 insertions(+), 0 deletions(-)
Documentation/CodingStyle +8 -0
index 9069189..e7f5fc6 100644
@@ -160,6 +160,21 @@ supply of new-lines on your screen is not a renewable resource (think
 25-line terminal screens here), you have more empty lines to put
 comments on.
 
+Do not unnecessarily use braces where a single statement will do.
+
+if (condition)
+	action();
+
+This does not apply if one branch of a conditional statement is a single
+statement. Use braces in both branches.
+

Documentation/CodingStyle supply of new-lines on your screen is not a renewable resource (think · line 180

Quoted 10 lines
Documentation/CodingStyle +7 -0
+if (condition) {
+	do_this();
+	do_that();
+} else {
+	otherwise();
+}
+
 		3.1:  Spaces
 
 Linux kernel style for use of spaces depends (mostly) on

---
~Randy
*** Remember to use Documentation/SubmitChecklist when testing your code ***

#2 lore raw
Quoted 32 lines · Randy Dunlap
Quoted 9 lines · On Tue, 8 May 2007 19:03:10 GMT Linux Kernel Mailing List wrote:
Gitweb:     http://git.kernel.org/git/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=e659ba4a0d2d471c0d73590f78e1a1b5a1eede48
Commit:     e659ba4a0d2d471c0d73590f78e1a1b5a1eede48
Parent:     28be5abb400e5e082f5225105fdc69337ec0c0b4
Author:     Oliver Neukum <[email protected]>
AuthorDate: Tue May 8 00:30:34 2007 -0700
Committer:  Linus Torvalds <[email protected]>
CommitDate: Tue May 8 11:15:12 2007 -0700

CodingStyle: start flamewar about use of braces

It worked somewhat. We never did reach any kind of concensus about this....

Quoted 16 lines
---
 Documentation/CodingStyle |   15 +++++++++++++++
 1 files changed, 15 insertions(+), 0 deletions(-)
Documentation/CodingStyle +0 -0
index 9069189..e7f5fc6 100644
@@ -160,6 +160,21 @@ supply of new-lines on your screen is not a renewable resource (think
 25-line terminal screens here), you have more empty lines to put
 comments on.
 
Quoted 14 lines
Quoted 4 lines
Documentation/CodingStyle +4 -0
+Do not unnecessarily use braces where a single statement will do.
+
+if (condition)
+	action();

Documentation/CodingStyle supply of new-lines on your screen is not a renewable resource (think · line 176

Quoted 10 lines
Documentation/CodingStyle +10 -0
+
+This does not apply if one branch of a conditional statement is a single
+statement. Use braces in both branches.
+
+if (condition) {
+	do_this();
+	do_that();
+} else {
+	otherwise();
+}

If anyone tries to add braces to my code's 'else' statements where they are not required, that patch will get NAK'd in a heartbeat.

Jeff

#3 lore raw
Quoted 9 lines · Jeff Garzik
Quoted 9 lines
Quoted 9 lines
(quoted diff) +9 -0
+Do not unnecessarily use braces where a single statement will do.
+
+if (condition)
+	action();
+
+This does not apply if one branch of a conditional statement is a single
+statement. Use braces in both branches.
+
+if (condition) {
Quoted 8 lines
Quoted 5 lines

Review

Quoted 5 lines
(quoted diff) +5 -0
+	do_this();
+	do_that();
+} else {
+	otherwise();
+}

If anyone tries to add braces to my code's 'else' statements where they are not required, that patch will get NAK'd in a heartbeat.

Oh isn't coding style fun. I personally hate code that doesn't ALWAYS have the braces everywhere since it makes adding a print statement or other debuging to the condition such a pain since you then have to add braces to the condition to avoid breaking the code just to insert a print statement. It is one of the few things I disagree with in the linux kernel coding style.

--
Len Sorensen

#4 lore raw
Quoted 30 lines · Lennart Sorensen
Quoted 9 lines · Jeff Garzik
Quoted 9 lines
Quoted 9 lines
(quoted diff) +9 -0
+Do not unnecessarily use braces where a single statement will do.
+
+if (condition)
+	action();
+
+This does not apply if one branch of a conditional statement is a single
+statement. Use braces in both branches.
+
+if (condition) {
Quoted 10 lines
Quoted 7 lines

Review

Quoted 7 lines
(quoted diff) +5 -0
+	do_this();
+	do_that();
+} else {
+	otherwise();
+}

If anyone tries to add braces to my code's 'else' statements where they are not required, that patch will get NAK'd in a heartbeat.

Oh isn't coding style fun. I personally hate code that doesn't ALWAYS have the braces everywhere since it makes adding a print statement or other debuging to the condition such a pain since you then have to add braces to the condition to avoid breaking the code just to insert a print statement. It is one of the few things I disagree with in the linux kernel coding style.

Quoted 10 lines

--
Len Sorensen
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

I agree - it is merely defensive programming to always have the braces, it prevents someone from sticking in a diag and forgetting to add braces so things go haywire.

My $.02
Steve

Signature · 3 lines
"They that give up essential liberty to obtain temporary safety, deserve neither liberty nor safety."  (Ben Franklin)

"The course of history shows that as a government grows, liberty decreases."  (Thomas Jefferson)
#5 lore raw
Quoted 11 lines · Stephen Clark
Quoted 9 lines · Lennart Sorensen
Quoted 1 line · Jeff Garzik

If anyone tries to add braces to my code's 'else' statements where they are not required, that patch will get NAK'd in a heartbeat.

Oh isn't coding style fun. I personally hate code that doesn't ALWAYS have the braces everywhere since it makes adding a print statement or other debuging to the condition such a pain since you then have to add braces to the condition to avoid breaking the code just to insert a print statement. It is one of the few things I disagree with in the linux kernel coding style.

Quoted 2 lines

I agree - it is merely defensive programming to always have the braces, it prevents someone from sticking in a diag and forgetting to add braces so things go haywire.

If you compulsively add braces where not needed, the number of lines in the source file explodes. Net result is code that is more difficult to read, because far fewer lines of code fit on a single page/screen.

But hey, it's personal preference. I'm not going to enforce the make-code-more-readable rule on others' code... :)

Jeff

#6 lore raw
In reply to Jeff Garzik
Quoted 42 lines · Jeff Garzik
Quoted 41 lines · Randy Dunlap
Quoted 9 lines · On Tue, 8 May 2007 19:03:10 GMT Linux Kernel Mailing List wrote:
Gitweb:     http://git.kernel.org/git/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=e659ba4a0d2d471c0d73590f78e1a1b5a1eede48
Commit:     e659ba4a0d2d471c0d73590f78e1a1b5a1eede48
Parent:     28be5abb400e5e082f5225105fdc69337ec0c0b4
Author:     Oliver Neukum <[email protected]>
AuthorDate: Tue May 8 00:30:34 2007 -0700
Committer:  Linus Torvalds <[email protected]>
CommitDate: Tue May 8 11:15:12 2007 -0700

CodingStyle: start flamewar about use of braces

It worked somewhat. We never did reach any kind of concensus about this....

Quoted 25 lines
---
 Documentation/CodingStyle |   15 +++++++++++++++
 1 files changed, 15 insertions(+), 0 deletions(-)
Documentation/CodingStyle +9 -0
index 9069189..e7f5fc6 100644
@@ -160,6 +160,21 @@ supply of new-lines on your screen is not a renewable resource (think
 25-line terminal screens here), you have more empty lines to put
 comments on.
 
+Do not unnecessarily use braces where a single statement will do.
+
+if (condition)
+    action();
+
+This does not apply if one branch of a conditional statement is a single
+statement. Use braces in both branches.
+
+if (condition) {
Quoted 8 lines
Quoted 5 lines

Documentation/CodingStyle supply of new-lines on your screen is not a renewable resource (think · line 176

Quoted 5 lines
Documentation/CodingStyle +5 -0
+    do_this();
+    do_that();
+} else {
+    otherwise();
+}

If anyone tries to add braces to my code's 'else' statements where they are not required, that patch will get NAK'd in a heartbeat.

Yes. This patch shouldn't go in. There's no ambiguity in Documentation/CodingStyle that it fixes anyway, only *imposes* a "style", which:

(1) there is no consensus about *at all*, and

(2) blatantly goes against _all_ similar examples in K&R.

It's quite funny to worship them as "prophets" and adopt their style for some stuff (with a simple "Rationale: K&R.") one instant and then suggest a style completely opposite to theirs one screenful later in the same file.

#7 lore raw
Quoted 4 lines · Satyam Sharma

It's quite funny to worship them as "prophets" and adopt their style for some stuff (with a simple "Rationale: K&R.") one instant and then suggest a style completely opposite to theirs one screenful later in the same file.

That's what happens to a text which is incrementally written by people with different writing styles (and different opinions).

As a slightly off-topic side note: CodingStyle was once a guideline which sort of worked because it was a compact and entertaining read; so some people took the time to read it and didn't mind too much to adopt the preferred style. It appears that CodingStyle is now being transformed into a norm which the authors want to make work by being comprehensive, and by being able to point offenders to a chapter/ section/ clause which they breached. Alas only a few people will be able and willing to memorize CodingStyle's rules then.

I don't know which form is better suited to the goal of well maintainable code --- the compact, easy to read, winning guideline, or the comprehensive norm.

Signature · 3 lines
Stefan Richter
-=====-=-=== -=-= -=--=
http://arcgraph.de/sr/
#8 lore raw
Quoted 3 lines · Stefan Richter

I don't know which form is better suited to the goal of well maintainable code --- the compact, easy to read, winning guideline, or the comprehensive norm.

Bah, it's a slippery slope. Too little guidelines and we end up people submitting code that is "obviously" violating Linux coding standards on the basis of that it's not forbidden by Documentation/CodingStyle. And with too much detail, we end up with something that nobody wants to read let alone follow. So what CodingStyle should say, really, is that maintainers have different preference and you're usually better off following them than fighting against them ;-).