U++ framework: U++ SQL » Huge error of Postgresql! Double fields by transmission to the program lose a fractional part!

Home » U++ Library support » U++ SQL » Huge error of Postgresql! Double fields by transmission to the program lose a fractional part!

Show: Today's Messages :: Polls :: Message Navigator
E-mail to friend

Huge error of Postgresql! Double fields by transmission to the program lose a fractional part! [message #47491]

Mon, 23 January 2017 02:28

sergeynikitin
Messages: 748
Registered: January 2008
Location: Moscow, Russia

Contributor

sergeynikitin@jabber.ru

Huge error of Postgresql!
Double fields by transmission to the program from sql table lose a fractional part!

The decision is extremely simple. To scan a string not as int (atoi), but as Double (ScanDouble).
The request to correct quickly as soon as it is possible!

Place to correct:
file PostgreSQL.cpp
Fucntion: void PostgreSQLConnection::GetColumn(int i, Ref f) const
Line: 644
Now:

	switch(info[i].type)
	{
		case INT64_V:
			f.SetValue(ScanInt64(s));
			break;
		case INT_V:
			f.SetValue(atoi(s));
			break;
		case DOUBLE_V:
			f.SetValue(atoi(s));         //<========THIS LINE NEED TO CORRECT
			break;
		case BOOL_V:
			f.SetValue(*s == 't' ? "1" : "0");
			break;
		case DATE_V:
			f.SetValue(sDate(s));
			break;
		case TIME_V: {
				Time t = ToTime(sDate(s));
				t.hour = atoi(s + 11);
				t.minute = atoi(s + 14);
				t.second = atoi(s + 17);
				f.SetValue(t);
			}
			break;
		default: {

   . . . . .

Must be:

	switch(info[i].type)
	{
		case INT64_V:
			f.SetValue(ScanInt64(s));
			break;
		case INT_V:
			f.SetValue(atoi(s));
			break;
		case DOUBLE_V:
			f.SetValue(ScanDouble(s));
			break;
		case BOOL_V:
			f.SetValue(*s == 't' ? "1" : "0");
			break;
		case DATE_V:
			f.SetValue(sDate(s));
			break;
		case TIME_V: {
				Time t = ToTime(sDate(s));
				t.hour = atoi(s + 11);
				t.minute = atoi(s + 14);
				t.second = atoi(s + 17);
				f.SetValue(t);
			}
			break;
		default: {

   . . . . .

Thanks!

SergeyNikitin<U++>( linux, wine )
{
under( Ubuntu || Debian || Raspbian );
}

Report message to a moderator

Re: Huge error of Postgresql! Double fields by transmission to the program lose a fractional part! [message #47492 is a reply to message #47491]

Mon, 23 January 2017 07:34

unodgs
Messages: 1367
Registered: November 2005
Location: Poland

Ultimate Contributor

That proves one thing - noone uses double fields Smile

Smile

Report message to a moderator

Re: Huge error of Postgresql! Double fields by transmission to the program lose a fractional part! [message #47494 is a reply to message #47492]

Mon, 23 January 2017 08:38

sergeynikitin
Messages: 748
Registered: January 2008
Location: Moscow, Russia

Contributor

sergeynikitin@jabber.ru

There is a way that I had to first use. It was necessary to specify in the query string conversion.

SQL.Execute("DBL_FIELD::text FROM SOMETABLE");

But yesterday I thought: Yes, how am I going to suffer. Opened in debugger with incremental execution. And I horrified, how all of this is simple.

SergeyNikitin<U++>( linux, wine )
{
under( Ubuntu || Debian || Raspbian );
}

Report message to a moderator

Re: Huge error of Postgresql! Double fields by transmission to the program lose a fractional part! [message #47496 is a reply to message #47494]

Mon, 23 January 2017 09:38

Klugier
Messages: 1117
Registered: September 2012
Location: Poland, Kraków

Senior Contributor

Hello,

It seems like release blocker Wink

Wink

Added to redmine - I believe we should fix this before release - http://www.ultimatepp.org/redmine/issues/1617.

Sergey can you monitor redmine ticket?

Sincerely,
Klugier

U++ - one framework to rule them all.

Report message to a moderator

Re: Huge error of Postgresql! Double fields by transmission to the program lose a fractional part! [message #47499 is a reply to message #47496]

Mon, 23 January 2017 09:46

sergeynikitin
Messages: 748
Registered: January 2008
Location: Moscow, Russia

Contributor

sergeynikitin@jabber.ru

Yes. I can. I've Add myself to this issue.

SergeyNikitin<U++>( linux, wine )
{
under( Ubuntu || Debian || Raspbian );
}

[Updated on: Mon, 23 January 2017 09:51]

Report message to a moderator

Re: Huge error of Postgresql! Double fields by transmission to the program lose a fractional part! [message #47514 is a reply to message #47499]

Wed, 25 January 2017 11:30

mirek
Messages: 14290
Registered: November 2005

Ultimate Member

sergeynikitin wrote on Mon, 23 January 2017 09:46

Yes. I can. I've Add myself to this issue.

I really wonder what U++ revision you are using, because in trunk it is:

void PostgreSQLConnection::GetColumn(int i, Ref f) const
{
	if(PQgetisnull(result, fetched_row, i))
	{
		f = Null;
		return;
	}
	char *s = PQgetvalue(result, fetched_row, i);
	switch(info[i].type)
	{
		case INT64_V:
			f.SetValue(ScanInt64(s));
			break;
		case INT_V:
			f.SetValue(atoi(s));
			break;
		case DOUBLE_V:
			f.SetValue(atof(s));
			break;

and it is really really unlikely that this bug could get there in last 8 years, because in that period my major project is PGSQL and customers would really notice this within hours...

I have checked svn history and oldest revision available from 2009 has "atof" there...

Mirek

[Updated on: Wed, 25 January 2017 11:31]

Report message to a moderator

Re: Huge error of Postgresql! Double fields by transmission to the program lose a fractional part! [message #47519 is a reply to message #47514]

Wed, 25 January 2017 18:18

sergeynikitin
Messages: 748
Registered: January 2008
Location: Moscow, Russia

Contributor

sergeynikitin@jabber.ru

mirek wrote on Wed, 25 January 2017 13:30

sergeynikitin wrote on Mon, 23 January 2017 09:46
Yes. I can. I've Add myself to this issue.

I really wonder what U++ revision you are using, because in trunk it is:
void PostgreSQLConnection::GetColumn(int i, Ref f) const
{
	if(PQgetisnull(result, fetched_row, i))
	{
		f = Null;
		return;
	}
	char *s = PQgetvalue(result, fetched_row, i);
	switch(info[i].type)
	{
		case INT64_V:
			f.SetValue(ScanInt64(s));
			break;
		case INT_V:
			f.SetValue(atoi(s));
			break;
		case DOUBLE_V:
			f.SetValue(atof(s));
			break;
and it is really really unlikely that this bug could get there in last 8 years, because in that period my major project is PGSQL and customers would really notice this within hours...

I have checked svn history and oldest revision available from 2009 has "atof" there...

Mirek

I use summer svn code. (with c++11 support).

If all correct, - it's OK! I've locally corrected version of PostgreSQL.cpp.

SergeyNikitin<U++>( linux, wine )
{
under( Ubuntu || Debian || Raspbian );
}

Report message to a moderator

Re: Huge error of Postgresql! Double fields by transmission to the program lose a fractional part! [message #47520 is a reply to message #47519]

Wed, 25 January 2017 18:35

sergeynikitin
Messages: 748
Registered: January 2008
Location: Moscow, Russia

Contributor

sergeynikitin@jabber.ru

I've just tested variant of atof(s):
Result on picture:

index.php?t=getfile&id=5183&private=0

index.php?t=getfile&id=5183&private=0

May be ScanDouble with language settings can be accurater?...

Attachment: PostgreSQL_ScanDouble_problem.png
(Size: 38.46KB, Downloaded 875 times)

SergeyNikitin<U++>( linux, wine )
{
under( Ubuntu || Debian || Raspbian );
}

Report message to a moderator

Re: Huge error of Postgresql! Double fields by transmission to the program lose a fractional part! [message #47521 is a reply to message #47520]

Wed, 25 January 2017 18:37

sergeynikitin
Messages: 748
Registered: January 2008
Location: Moscow, Russia

Contributor

sergeynikitin@jabber.ru

Sorry... I do it best regards, and with hope.

SergeyNikitin<U++>( linux, wine )
{
under( Ubuntu || Debian || Raspbian );
}

Report message to a moderator

Re: Huge error of Postgresql! Double fields by transmission to the program lose a fractional part! [message #47522 is a reply to message #47521]

Wed, 25 January 2017 18:42

Klugier
Messages: 1117
Registered: September 2012
Location: Poland, Kraków

Senior Contributor

Hello Sergey,

The issue was closed on redmine as "Rejected".

Sincerely and thanks for feedback,
Klugier

U++ - one framework to rule them all.

[Updated on: Wed, 25 January 2017 18:43]

Report message to a moderator

Re: Huge error of Postgresql! Double fields by transmission to the program lose a fractional part! [message #47523 is a reply to message #47522]

Wed, 25 January 2017 18:54

sergeynikitin
Messages: 748
Registered: January 2008
Location: Moscow, Russia

Contributor

sergeynikitin@jabber.ru

Why closed? I place there picture, where error is still exists!

SergeyNikitin<U++>( linux, wine )
{
under( Ubuntu || Debian || Raspbian );
}

[Updated on: Wed, 25 January 2017 18:54]

Report message to a moderator

Re: Huge error of Postgresql! Double fields by transmission to the program lose a fractional part! [message #47524 is a reply to message #47523]

Wed, 25 January 2017 19:02

mirek
Messages: 14290
Registered: November 2005

Ultimate Member

sergeynikitin wrote on Wed, 25 January 2017 18:54

Why closed? I place there picture, where error is still exists!

Actually, I have suspiction here...

Can you check how that 's' looks like? (E.g. put DUMP(s) before atof).

I suspect that maybe, your PGSQL is returning decimal comma and atof is unable to handle that, which ScanDouble is.

It would be strange borderline issue, but really worth fixing.

(OTOH, 'atoi' was NEVER there, sorry. You can check svn log....)

Mirek

Report message to a moderator

Re: Huge error of Postgresql! Double fields by transmission to the program lose a fractional part! [message #47526 is a reply to message #47523]

Wed, 25 January 2017 19:03

Klugier
Messages: 1117
Registered: September 2012
Location: Poland, Kraków

Senior Contributor

Hello,

Upps sorry, please close it by yourself when everything will be ok. I will not touch this issue any more.

Sincerely,
Klugier

U++ - one framework to rule them all.

Report message to a moderator

Re: Huge error of Postgresql! Double fields by transmission to the program lose a fractional part! [message #47529 is a reply to message #47524]

Wed, 25 January 2017 19:31

sergeynikitin
Messages: 748
Registered: January 2008
Location: Moscow, Russia

Contributor

sergeynikitin@jabber.ru

When I place there code of, I mistaken, because already correct in my home code. But atof return same result as you can see at picture.

May be will different result if use other Set Language
( I use

	SetLanguage(LNGFromText("RU-RU"));
	
	SetDateFormat("%3:02d.%2:02d.%1:4d");

It's not tested yet. But ScanDouble work fine!

SergeyNikitin<U++>( linux, wine )
{
under( Ubuntu || Debian || Raspbian );
}

[Updated on: Wed, 25 January 2017 19:33]

Report message to a moderator

Re: Huge error of Postgresql! Double fields by transmission to the program lose a fractional part! [message #47530 is a reply to message #47529]

Wed, 25 January 2017 19:49

mirek
Messages: 14290
Registered: November 2005

Ultimate Member

Sure, but I need to know the root of the problem...

Report message to a moderator

Re: Huge error of Postgresql! Double fields by transmission to the program lose a fractional part! [message #47531 is a reply to message #47530]

Wed, 25 January 2017 20:31

sergeynikitin
Messages: 748
Registered: January 2008
Location: Moscow, Russia

Contributor

sergeynikitin@jabber.ru

I'll try to prepare test case tonight .

SergeyNikitin<U++>( linux, wine )
{
under( Ubuntu || Debian || Raspbian );
}

Report message to a moderator

Re: Huge error of Postgresql! Double fields by transmission to the program lose a fractional part! [message #47532 is a reply to message #47531]

Wed, 25 January 2017 20:40

mirek
Messages: 14290
Registered: November 2005

Ultimate Member

sergeynikitin wrote on Wed, 25 January 2017 20:31

I'll try to prepare test case tonight .

You do not understand...

I need to know what your PostgreSQL instance returns (before atof), so that I am sure we are doing the right fix.

Because very likely the problem is that your PostgreSQL is configured somehow differently than my PostgreSQL.

So pretty please, place DUMP(s) before atof (or ScanDouble), like

switch(info[i].type)
	{
		case INT64_V:
			f.SetValue(ScanInt64(s));
			break;
		case INT_V:
			f.SetValue(atoi(s));
			break;
		case DOUBLE_V:
DUMP(s);
			f.SetValue(ScanDouble(s));
			break;

then post here a log (a couple of lines is enough).

Thanks,

Mirek

[Updated on: Wed, 25 January 2017 20:48]

Report message to a moderator

Re: Huge error of Postgresql! Double fields by transmission to the program lose a fractional part! [message #47534 is a reply to message #47532]

Thu, 26 January 2017 01:00

sergeynikitin
Messages: 748
Registered: January 2008
Location: Moscow, Russia

Contributor

sergeynikitin@jabber.ru

I see, I will.

SergeyNikitin<U++>( linux, wine )
{
under( Ubuntu || Debian || Raspbian );
}

Report message to a moderator

Re: Huge error of Postgresql! Double fields by transmission to the program lose a fractional part! [message #47535 is a reply to message #47534]

Thu, 26 January 2017 01:25

sergeynikitin
Messages: 748
Registered: January 2008
Location: Moscow, Russia

Contributor

sergeynikitin@jabber.ru

Modified code (while lang settings SetLanguage(LNGFromText("RU-RU"));:

	char *s = PQgetvalue(result, fetched_row, i);
	switch(info[i].type)
	{
		case INT64_V:
			f.SetValue(ScanInt64(s));
			break;
		case INT_V:
			f.SetValue(atoi(s));
			break;
		case DOUBLE_V:
			RLOG("===start analysis===");
			RDUMP(s);
			double da,dS;
			da = atof(s);
			dS = ScanDouble(s);
			RDUMP(da);
			RDUMP(dS);
			static String test_format_and_lang_settings;
			test_format_and_lang_settings = Format("%f",dS);
			RDUMP(test_format_and_lang_settings);
			RLOG("====end analysis====");
			RLOG(" ");
			f.SetValue(ScanDouble(s));
			break;
		case BOOL_V:
			f.SetValue(*s == 't' ? "1" : "0");
			break;
		case DATE_V:

Log:

26.01.2017 03:22:15 ===start analysis===
26.01.2017 03:22:15 s = 90
26.01.2017 03:22:15 da = 90
26.01.2017 03:22:15 dS = 90
26.01.2017 03:22:15 test_format_and_lang_settings = 90,000000
26.01.2017 03:22:15 ====end analysis====
26.01.2017 03:22:15  
26.01.2017 03:22:15 ===start analysis===
26.01.2017 03:22:15 s = 40
26.01.2017 03:22:15 da = 40
26.01.2017 03:22:15 dS = 40
26.01.2017 03:22:15 test_format_and_lang_settings = 40,000000
26.01.2017 03:22:15 ====end analysis====
26.01.2017 03:22:15  
26.01.2017 03:22:15 ===start analysis===
26.01.2017 03:22:15 s = 269.96336996337
26.01.2017 03:22:15 da = 269
26.01.2017 03:22:15 dS = 269.96336996337
26.01.2017 03:22:15 test_format_and_lang_settings = 269,963370
26.01.2017 03:22:15 ====end analysis====
26.01.2017 03:22:15  
26.01.2017 03:22:15 ===start analysis===
26.01.2017 03:22:15 s = 381.556455240666
26.01.2017 03:22:15 da = 381
26.01.2017 03:22:15 dS = 381.556455240666
26.01.2017 03:22:15 test_format_and_lang_settings = 381,556455
26.01.2017 03:22:15 ====end analysis====
26.01.2017 03:22:15  
26.01.2017 03:22:15 ===start analysis===
26.01.2017 03:22:15 s = 282.083413662361
26.01.2017 03:22:15 da = 282
26.01.2017 03:22:15 dS = 282.083413662361
26.01.2017 03:22:15 test_format_and_lang_settings = 282,083414
26.01.2017 03:22:15 ====end analysis====

SergeyNikitin<U++>( linux, wine )
{
under( Ubuntu || Debian || Raspbian );
}

[Updated on: Thu, 26 January 2017 01:28]

Report message to a moderator

Re: Huge error of Postgresql! Double fields by transmission to the program lose a fractional part! [message #47536 is a reply to message #47535]

Thu, 26 January 2017 01:33

sergeynikitin
Messages: 748
Registered: January 2008
Location: Moscow, Russia

Contributor

sergeynikitin@jabber.ru

While lang settings SetLanguage(LNGFromText("EN-US");:

26.01.2017 03:26:34 ===start analysis===
26.01.2017 03:26:34 s = 40
26.01.2017 03:26:34 da = 40
26.01.2017 03:26:34 dS = 40
26.01.2017 03:26:34 test_format_and_lang_settings = 40,000000
26.01.2017 03:26:34 ====end analysis====
26.01.2017 03:26:34  
26.01.2017 03:26:34 ===start analysis===
26.01.2017 03:26:34 s = 269.96336996337
26.01.2017 03:26:34 da = 269
26.01.2017 03:26:34 dS = 269.96336996337
26.01.2017 03:26:34 test_format_and_lang_settings = 269,963370
26.01.2017 03:26:34 ====end analysis====
26.01.2017 03:26:34  
26.01.2017 03:26:34 ===start analysis===
26.01.2017 03:26:34 s = 59.95115995116
26.01.2017 03:26:34 da = 59
26.01.2017 03:26:34 dS = 59.95115995116
26.01.2017 03:26:34 test_format_and_lang_settings = 59,951160
26.01.2017 03:26:34 ====end analysis====
26.01.2017 03:26:34  
26.01.2017 03:26:34 ===start analysis===
26.01.2017 03:26:34 s = 222.435897435897
26.01.2017 03:26:34 da = 222
26.01.2017 03:26:34 dS = 222.435897435897
26.01.2017 03:26:34 test_format_and_lang_settings = 222,435897
26.01.2017 03:26:34 ====end analysis====
26.01.2017 03:26:34

Both test under UBUNTU 16.04 (with Russian main language)

SergeyNikitin<U++>( linux, wine )
{
under( Ubuntu || Debian || Raspbian );
}

[Updated on: Thu, 26 January 2017 01:52]

Report message to a moderator

Re: Huge error of Postgresql! Double fields by transmission to the program lose a fractional part! [message #47537 is a reply to message #47536]

Thu, 26 January 2017 01:43

sergeynikitin
Messages: 748
Registered: January 2008
Location: Moscow, Russia

Contributor

sergeynikitin@jabber.ru

But if I start

LC_ALL=en_US.UTF-8 ./ProgramName

or

LANG=en_US.UTF-8 ./ProgramName

than all correct.

26.01.2017 03:41:15 ===start analysis===
26.01.2017 03:41:15 s = 40
26.01.2017 03:41:15 da = 40
26.01.2017 03:41:15 dS = 40
26.01.2017 03:41:15 test_format_and_lang_settings = 40.000000
26.01.2017 03:41:15 ====end analysis====
26.01.2017 03:41:15  
26.01.2017 03:41:15 ===start analysis===
26.01.2017 03:41:15 s = 269.96336996337
26.01.2017 03:41:15 da = 269.96336996337
26.01.2017 03:41:15 dS = 269.96336996337
26.01.2017 03:41:15 test_format_and_lang_settings = 269.963370
26.01.2017 03:41:15 ====end analysis====
26.01.2017 03:41:15  
26.01.2017 03:41:15 ===start analysis===
26.01.2017 03:41:15 s = 59.95115995116
26.01.2017 03:41:15 da = 59.95115995116
26.01.2017 03:41:15 dS = 59.95115995116
26.01.2017 03:41:15 test_format_and_lang_settings = 59.951160
26.01.2017 03:41:15 ====end analysis====
26.01.2017 03:41:15

SergeyNikitin<U++>( linux, wine )
{
under( Ubuntu || Debian || Raspbian );
}

[Updated on: Thu, 26 January 2017 01:50]

Report message to a moderator

Re: Huge error of Postgresql! Double fields by transmission to the program lose a fractional part! [message #47538 is a reply to message #47537]

Thu, 26 January 2017 01:45

sergeynikitin
Messages: 748
Registered: January 2008
Location: Moscow, Russia

Contributor

sergeynikitin@jabber.ru

It's repeated same results on different computer (but always under UBUNTU 16.04)

SergeyNikitin<U++>( linux, wine )
{
under( Ubuntu || Debian || Raspbian );
}

Report message to a moderator

Re: Huge error of Postgresql! Double fields by transmission to the program lose a fractional part! [message #47539 is a reply to message #47538]

Thu, 26 January 2017 01:57

sergeynikitin
Messages: 748
Registered: January 2008
Location: Moscow, Russia

Contributor

sergeynikitin@jabber.ru

Postgresql (Server v. 9.5) language settings are default.

fragment of postgres config:

# These settings are initialized by initdb, but they can be changed.
lc_messages = 'ru_RU.UTF-8'			# locale for system error message
					# strings
lc_monetary = 'ru_RU.UTF-8'			# locale for monetary formatting
lc_numeric = 'ru_RU.UTF-8'			# locale for number formatting
lc_time = 'ru_RU.UTF-8'				# locale for time formatting

SergeyNikitin<U++>( linux, wine )
{
under( Ubuntu || Debian || Raspbian );
}

Report message to a moderator

Re: Huge error of Postgresql! Double fields by transmission to the program lose a fractional part! [message #47540 is a reply to message #47539]

Thu, 26 January 2017 02:04

sergeynikitin
Messages: 748
Registered: January 2008
Location: Moscow, Russia

Contributor

sergeynikitin@jabber.ru

I change postgresql setting to ( /etc/postgresql/9.5/main/postgresql.conf )
en_US:

# These settings are initialized by initdb, but they can be changed.
#lc_messages = 'ru_RU.UTF-8'			# locale for system error message
					# strings
#lc_monetary = 'ru_RU.UTF-8'			# locale for monetary formatting
#lc_numeric = 'ru_RU.UTF-8'			# locale for number formatting
#lc_time = 'ru_RU.UTF-8'				# locale for time formatting


lc_messages = 'en_US.UTF-8'			# locale for system error message
					# strings
lc_monetary = 'en_US.UTF-8'			# locale for monetary formatting
lc_numeric = 'en_US.UTF-8'			# locale for number formatting
lc_time = 'en_US.UTF-8'				# locale for time formatting

Result is zero!

26.01.2017 03:59:52 ===start analysis===
26.01.2017 03:59:52 s = 40
26.01.2017 03:59:52 da = 40
26.01.2017 03:59:52 dS = 40
26.01.2017 03:59:52 test_format_and_lang_settings = 40,000000
26.01.2017 03:59:52 ====end analysis====
26.01.2017 03:59:52  
26.01.2017 03:59:52 ===start analysis===
26.01.2017 03:59:52 s = 269.96336996337
26.01.2017 03:59:52 da = 269
26.01.2017 03:59:52 dS = 269.96336996337
26.01.2017 03:59:52 test_format_and_lang_settings = 269,963370
26.01.2017 03:59:52 ====end analysis====
26.01.2017 03:59:52  
26.01.2017 03:59:52 ===start analysis===
26.01.2017 03:59:52 s = 59.95115995116
26.01.2017 03:59:52 da = 59
26.01.2017 03:59:52 dS = 59.95115995116
26.01.2017 03:59:52 test_format_and_lang_settings = 59,951160
26.01.2017 03:59:52 ====end analysis====

SergeyNikitin<U++>( linux, wine )
{
under( Ubuntu || Debian || Raspbian );
}

Report message to a moderator

Re: Huge error of Postgresql! Double fields by transmission to the program lose a fractional part! [message #47541 is a reply to message #47540]

Thu, 26 January 2017 02:23

sergeynikitin
Messages: 748
Registered: January 2008
Location: Moscow, Russia

Contributor

sergeynikitin@jabber.ru

I test timings:

Test code:

#include <Core/Core.h>

using namespace Upp;

CONSOLE_APP_MAIN
{
	double a;
	String s = "269.96336996337";
	#define LOOPS 1000000
	for(int i = 0; i < LOOPS; i++){
		RTIMING("Using atof");
		a = atof(s);
	}
	for(int i = 0; i < LOOPS; i++){
		RTIMING("Using ScanDouble");
		a = ScanDouble(s);
	}
}

TIMING results:

TIMING Using ScanDouble:  1.12 s  -  1.12 us ( 1.22 s  / 1000000 ), min:  0.00 ns, max:  1.00 ms, nesting: 1 - 1000000
TIMING Using atof     : 337.45 ms - 337.45 ns (439.00 ms / 1000000 ), min:  0.00 ns, max:  1.00 ms, nesting: 1 - 1000000

It's not dramatically slower! But standard settings (on various environments) show right results!

SergeyNikitin<U++>( linux, wine )
{
under( Ubuntu || Debian || Raspbian );
}

Report message to a moderator

Re: Huge error of Postgresql! Double fields by transmission to the program lose a fractional part! [message #47543 is a reply to message #47536]

Thu, 26 January 2017 09:09

mirek
Messages: 14290
Registered: November 2005

Ultimate Member

OK, so it seems like the moral of the story is to never trust or use 'atof'.

Makes me want to do add to Core.h either

#define atof @ // never use atof, as it has unpredictable results

or even

#define atof Atof

Atof being U++ upcoming equivalent of atof, basically Nvl(ScanDouble(x)).

[Updated on: Thu, 26 January 2017 09:09]

Report message to a moderator

Re: Huge error of Postgresql! Double fields by transmission to the program lose a fractional part! [message #47544 is a reply to message #47541]

Thu, 26 January 2017 14:10

mirek
Messages: 14290
Registered: November 2005

Ultimate Member

sergeynikitin wrote on Thu, 26 January 2017 02:23

I test timings:

Test code:

#include <Core/Core.h>

using namespace Upp;

CONSOLE_APP_MAIN
{
	double a;
	String s = "269.96336996337";
	#define LOOPS 1000000
	for(int i = 0; i < LOOPS; i++){
		RTIMING("Using atof");
		a = atof(s);
	}
	for(int i = 0; i < LOOPS; i++){
		RTIMING("Using ScanDouble");
		a = ScanDouble(s);
	}
}

TIMING results:

TIMING Using ScanDouble:  1.12 s  -  1.12 us ( 1.22 s  / 1000000 ), min:  0.00 ns, max:  1.00 ms, nesting: 1 - 1000000
TIMING Using atof     : 337.45 ms - 337.45 ns (439.00 ms / 1000000 ), min:  0.00 ns, max:  1.00 ms, nesting: 1 - 1000000

It's not dramatically slower! But standard settings (on various environments) show right results!

BTW, have you tried in Release? MSC14 32-bit release I am getting:

TIMING Using ScanDouble: 120.72 ms - 120.72 ns (148.00 ms / 1000000 ), min:  0.00 ns, max:  2.00 ms, nesting: 1 - 1000000
TIMING Using atof     : 281.72 ms - 281.72 ns (309.00 ms / 1000000 ), min:  0.00 ns, max:  1.00 ms, nesting: 1 - 1000000

I really would not like U++ being slower... Smile

Smile

Report message to a moderator

Re: Huge error of Postgresql! Double fields by transmission to the program lose a fractional part! [message #47545 is a reply to message #47544]

Thu, 26 January 2017 15:01

mirek
Messages: 14290
Registered: November 2005

Ultimate Member

rc4 should have this fix. Can you check please so that we can close this issue?

Mirek

Report message to a moderator

Re: Huge error of Postgresql! Double fields by transmission to the program lose a fractional part! [message #47546 is a reply to message #47545]

Thu, 26 January 2017 15:25

sergeynikitin
Messages: 748
Registered: January 2008
Location: Moscow, Russia

Contributor

sergeynikitin@jabber.ru

Ok, I'll ckeck it tonight (a few later).

SergeyNikitin<U++>( linux, wine )
{
under( Ubuntu || Debian || Raspbian );
}

Report message to a moderator

Previous Topic:	I propose to Add Postgresql to topic header
Next Topic:	How can .sch express a many-to-many relationship ?

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

PDF

]

Current Time: Mon Apr 27 10:11:36 GMT+2 2026

Total time taken to generate the page: 0.01393 seconds