threads / discuss / 5743

Clean up approxidate() in preparation for fixes

Subject: Clean up approxidate() in preparation for fixes

## tl;dr

9 messages between Sep 28, 2006 and Sep 29, 2006.

replies: 8people: 4as markdown or json

Linus Torvalds· Sep 28, 2006, 19:12 UTC · lore

Our approxidate cannot handle simple times like "5 PM yesterday", and to fix that, we will need to add some logic for number handling. This just splits that out into a function of its own (the same way the _real_ date parsing works).

Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---

This should change no code what-so-ever, just split it up in preparation for the next patch..

diff --git a/date.c b/date.c
index e387dcd..4ff6604 100644
--- a/date.c
+++ b/date.c
@@ -712,6 +712,15 @@ static const char *approxidate_alpha(con
 	return end;
 }
 
+static const char *approxidate_digit(const char *date, struct tm *tm, int *num)
+{
+	char *end;
+	unsigned long number = strtoul(date, &end, 10);
+
+	*num = number;
+	return end;
+}
+
 unsigned long approxidate(const char *date)
 {
 	int number = 0;
@@ -731,9 +740,7 @@ unsigned long approxidate(const char *da
 			break;
 		date++;
 		if (isdigit(c)) {
-			char *end;
-			number = strtoul(date-1, &end, 10);
-			date = end;
+			date = approxidate_digit(date-1, &tm, &number);
 			continue;
 		}
 		if (isalpha(c))
Linus Torvalds· Sep 28, 2006, 19:14 UTC · re: Linus Torvalds · lore

Fix approxidate() to understand more extended numbers

You can now say "5:35 PM yesterday", and approxidate() gets the right answer.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
Useful? Of course. In a sick and twisted sort of way.

Try to say "yesterday at 5 PM" before and after this. It _really_ was confused before.

diff --git a/date.c b/date.c
index 4ff6604..db4c185 100644
--- a/date.c
+++ b/date.c
@@ -598,6 +598,32 @@ static void date_tea(struct tm *tm, int 
 	date_time(tm, 17);
 }
 
+static void date_pm(struct tm *tm, int *num)
+{
+	int hour = *num;
+	*num = 0;
+
+	if (hour > 0 && hour < 12) {
+		tm->tm_hour = hour;
+		tm->tm_min = 0;
+		tm->tm_sec = 0;
+	}
+	if (tm->tm_hour > 0 && tm->tm_hour < 12)
+		tm->tm_hour += 12;
+}
+
+static void date_am(struct tm *tm, int *num)
+{
+	int hour = *num;
+	*num = 0;
+
+	if (hour > 0 && hour < 12) {
+		tm->tm_hour = hour;
+		tm->tm_min = 0;
+		tm->tm_sec = 0;
+	}
+}
+
 static const struct special {
 	const char *name;
 	void (*fn)(struct tm *, int *);
@@ -606,6 +632,8 @@ static const struct special {
 	{ "noon", date_noon },
 	{ "midnight", date_midnight },
 	{ "tea", date_tea },
+	{ "PM", date_pm },
+	{ "AM", date_am },
 	{ NULL }
 };
 
@@ -717,6 +745,18 @@ static const char *approxidate_digit(con
 	char *end;
 	unsigned long number = strtoul(date, &end, 10);
 
+	switch (*end) {
+	case ':':
+	case '.':
+	case '/':
+	case '-':
+		if (isdigit(end[1])) {
+			int match = match_multi_number(number, *end, date, end, tm);
+			if (match)
+				return date + match;
+		}
+	}
+
 	*num = number;
 	return end;
 }
Linus Torvalds· Sep 29, 2006, 06:03 UTC · re: Morten Welinder · lore

Re: Fix approxidate() to understand more extended numbers

On Thu, 28 Sep 2006, Morten Welinder wrote:
>
> Just don't hack at 12am or 12pm.

I think 12pm is correct, but 12am probably isn't (12am should _subtract_ 12, while 12pm does _not_ add 12).

That said, I have a rice cooker that avoids the problem by saying "0:10 PM" for ten minutes past midday ;)

Of course, all sane and civilized countries just use 24-hour format anyway. "Military time" my *ss.

Some day the US will turn metric and 24-hour-format. If the sun doesn't turn into a red giant first, that is.

		Linus
Junio C Hamano· Sep 29, 2006, 06:14 UTC · re: Linus Torvalds · lore

Re: Fix approxidate() to understand more extended numbers

Linus Torvalds <torvalds@osdl.org> writes:
Show 6 quoted lines
> On Thu, 28 Sep 2006, Morten Welinder wrote:
>>
>> Just don't hack at 12am or 12pm.
>
> I think 12pm is correct, but 12am probably isn't (12am should _subtract_ 
> 12, while 12pm does _not_ add 12).

But you have "if (hour > 0 && hour < 12)" in both am and pm so assignment to tm would not trigger...

> That said, I have a rice cooker that avoids the problem by saying "0:10 PM" 
> for ten minutes past midday ;)
You eat rice?
> Of course, all sane and civilized countries just use 24-hour format 
> anyway.

You are referring to the US, but neither is Japan sane nor civilized ;-).

Linus Torvalds· Sep 29, 2006, 06:42 UTC · re: Junio C Hamano · lore

Re: Fix approxidate() to understand more extended numbers

On Thu, 28 Sep 2006, Junio C Hamano wrote:
Show 11 quoted lines
> Linus Torvalds <torvalds@osdl.org> writes:
> 
> > On Thu, 28 Sep 2006, Morten Welinder wrote:
> >>
> >> Just don't hack at 12am or 12pm.
> >
> > I think 12pm is correct, but 12am probably isn't (12am should _subtract_ 
> > 12, while 12pm does _not_ add 12).
> 
> But you have "if (hour > 0 && hour < 12)" in both am and pm so
> assignment to tm would not trigger...
That's not the point.
If you write
	12:30 am

you really _should_ subtract 12, leaving you with 0:30. We don't. So we end up with a 24-hour time of 12:30, which is obviously _pm_, and wrong.

And "12 am" or "12 pm" doesn't work at all.
> > That said, I have a rice cooker that avoids the problem by saying "0:10 PM" 
> > for ten minutes past midday ;)
> 
> You eat rice?
Ok, is it just me, or is that just a very odd question?

I can see the question "You eat uni?". That really _does_ take a bit of getting used to. And Natto I really _really_ don't see the point of.

But rice? Afaik, it's the most common food-staple in the world. It's not exactly odd and exotic..

Show 5 quoted lines
> > Of course, all sane and civilized countries just use 24-hour format 
> > anyway.
> 
> You are referring to the US, but neither is Japan sane nor
> civilized ;-).

Yeah, well, they've been learning bad habits. But at least they are metric.

		Linus
Junio C Hamano· Sep 29, 2006, 07:09 UTC · re: Linus Torvalds · lore

Re: Fix approxidate() to understand more extended numbers

Linus Torvalds <torvalds@osdl.org> writes:
Show 16 quoted lines
>> > I think 12pm is correct, but 12am probably isn't (12am should _subtract_ 
>> > 12, while 12pm does _not_ add 12).
>> 
>> But you have "if (hour > 0 && hour < 12)" in both am and pm so
>> assignment to tm would not trigger...
>
> That's not the point.
>
> If you write
>
> 	12:30 am
>
> you really _should_ subtract 12, leaving you with 0:30. We don't. So we 
> end up with a 24-hour time of 12:30, which is obviously _pm_, and wrong.
>
> And "12 am" or "12 pm" doesn't work at all.

Ah, that's what you meant. My brain a bit too tired from the day job tonight X-<.

Show 6 quoted lines
>> > That said, I have a rice cooker that avoids the problem by saying "0:10 PM" 
>> > for ten minutes past midday ;)
>> 
>> You eat rice?
>
> Ok, is it just me, or is that just a very odd question?

Having rice cooker implied eating the stuff regularly and I just did not expect that from a north european. Just showing my ignorance -- I've never been to Europe.

Johannes Schindelin· Sep 29, 2006, 14:04 UTC · re: Junio C Hamano · lore

Re: Fix approxidate() to understand more extended numbers

Hi,
On Fri, 29 Sep 2006, Junio C Hamano wrote:
> Having rice cooker implied eating the stuff regularly and I just
> did not expect that from a north european.  Just showing my
> ignorance -- I've never been to Europe.

Actually, yesterday I saw a digital wrist watch for the first time in my life! Amazing! ;-)

There is even a running joke about Great Britain: "Oh yes, I _love_ English dishes; especially the Chinese ones!"

Ciao, Dscho

Linus Torvalds· Sep 29, 2006, 19:36 UTC · re: Junio C Hamano · lore

Re: Fix approxidate() to understand more extended numbers

On Fri, 29 Sep 2006, Junio C Hamano wrote:
Show 13 quoted lines
> Linus Torvalds <torvalds@osdl.org> writes:
> >
> > If you write
> >
> > 	12:30 am
> >
> > you really _should_ subtract 12, leaving you with 0:30. We don't. So we 
> > end up with a 24-hour time of 12:30, which is obviously _pm_, and wrong.
> >
> > And "12 am" or "12 pm" doesn't work at all.
> 
> Ah, that's what you meant.  My brain a bit too tired from the
> day job tonight X-<.
Here's a patch that should work. It just simplifies the whole thing to say 
	"hour = (hour % 12) + X"
where X is 12 for PM and 0 for AM.

It also fixes the "exact date" parsing, which didn't parse AM at all, and as such would do the same "12:30 AM" means "12:30 24-hour-format" bug. Of course, I hope that no exact dates use AM/PM anyway, but since we support the PM format, let's just get it right.

Not hugely tested, but I did test some of it, and it all _looks_ sane.
		Linus
---
diff --git a/date.c b/date.c
index db4c185..1825922 100644
--- a/date.c
+++ b/date.c
@@ -256,8 +256,12 @@ static int match_alpha(const char *date,
 	}
 
 	if (match_string(date, "PM") == 2) {
-		if (tm->tm_hour > 0 && tm->tm_hour < 12)
-			tm->tm_hour += 12;
+		tm->tm_hour = (tm->tm_hour % 12) + 12;
+		return 2;
+	}
+
+	if (match_string(date, "AM") == 2) {
+		tm->tm_hour = (tm->tm_hour % 12) + 0;
 		return 2;
 	}
 
@@ -600,28 +604,30 @@ static void date_tea(struct tm *tm, int 
 
 static void date_pm(struct tm *tm, int *num)
 {
-	int hour = *num;
+	int hour, n = *num;
 	*num = 0;
 
-	if (hour > 0 && hour < 12) {
-		tm->tm_hour = hour;
+	hour = tm->tm_hour;
+	if (n) {
+		hour = n;
 		tm->tm_min = 0;
 		tm->tm_sec = 0;
 	}
-	if (tm->tm_hour > 0 && tm->tm_hour < 12)
-		tm->tm_hour += 12;
+	tm->tm_hour = (hour % 12) + 12;
 }
 
 static void date_am(struct tm *tm, int *num)
 {
-	int hour = *num;
+	int hour, n = *num;
 	*num = 0;
 
-	if (hour > 0 && hour < 12) {
-		tm->tm_hour = hour;
+	hour = tm->tm_hour;
+	if (n) {
+		hour = n;
 		tm->tm_min = 0;
 		tm->tm_sec = 0;
 	}
+	tm->tm_hour = (hour % 12);
 }
 
 static const struct special {

← back to recent threads