Re: [HACKERS] pg_dump and large files - is this a problem?
Tom Lane writes: I think the problem is more accurately described thus: Flex generated files include stdio.h before postgres.h due to the way it lays out the code in the output. stdio.h does something which prevents switching to the large file model later on in postgres.h. (This manifests itself in unistd.h, but unistd.h itself is not the problem per se.) The proposed fix was to include the flex output in some other file (such as the corresponding grammar file) rather than to compile it separately. I have made this change. CVS tip should compile cleanly now on machines where this is an issue. Hmm, sorry for the late response, but I was away on the (long) weekend :-( I think your patch might be the source for Christopher's build problem (Compile problem on FreeBSD/Alpha). Peter already had a patch, that I tested, modified a little, and sent him back for inclusion into CVS. I will attach his patch with my small fixes for cross reference. The issue is, that you need to remove the #include bootstrap_tokens.h line from the lex file. Andreas flex-patch2.gz Description: flex-patch2.gz ---(end of broadcast)--- TIP 3: if posting/reading through Usenet, please send an appropriate subscribe-nomail command to [EMAIL PROTECTED] so that your message can get through to the mailing list cleanly
[HACKERS] Finishing Domains...
I've got a couple of free days coming up, so I want to finish domain support (Check constraints). Is it safe to teach Var how to deal with values that do not originate from a tuple? Or should I create a new primnode to deal with these types of variables. CREATE DOMAIN dom AS integer CHECK(VALUE BETWEEN 6 AND 10); SELECT CAST(8 AS dom); -- Rod Taylor ---(end of broadcast)--- TIP 5: Have you checked our extensive FAQ? http://www.postgresql.org/users-lounge/docs/faq.html
[HACKERS] Float output formatting options
Hi All,
This relates to the recent discussion about floating point output format.
The discussion was at a point where one parameter would be added to
specify the number of extra digits used in fp output formatting.
The parameter would have a 0 default value, a maximum of 2 and the
minimum remained open for discussion.
In a previous message I proposed that for double precision numbers a
minimum value of -13 would be usefful. For single precision numbers this
corresponds to a value of -4.
I downloaded the PG sources and added two parameters (as PGC_USERSET):
int extra_float4_digits, default 0, min -4, max 2
int extra_float8_digits, defualt 0, min -13, max 2
Compiled and tested for these functionalities. It is ok.
The afected files are:
src/backend/utils/adt/float.c
src/backend/utils/misc/guc.c
src/bin/psql/tab-complete.c
src/backend/utils/misc/postgresql.conf.sample
I used sources from Debian source package, postgresql_7.2.1-2woody2.
Diff's produced with diff -u are enclosed as attachments.
Can you comment on this (particularly the min values) ?
Also, if we concluded that there is a need of 2 more digits, should'nt
this be the default ?
Best regards,
Pedro M. Ferreira
--
--
Pedro Miguel Frazao Fernandes Ferreira
Universidade do Algarve
Faculdade de Ciencias e Tecnologia
Campus de Gambelas
8000-117 Faro
Portugal
Tel./Fax: (+351) 289 800950 / 289 819403
http://w3.ualg.pt/~pfrazao
--- old/postgresql-7.2.1/src/backend/utils/adt/float.c Tue Dec 11 02:02:12 2001
+++ postgresql-7.2.1/src/backend/utils/adt/float.c Mon Nov 4 10:32:33 2002
@@ -65,6 +65,12 @@
#include utils/array.h
#include utils/builtins.h
+/*
+ * Configuration options for float4 and float8 extra digits in output format
+ */
+
+int extra_float4_digits;
+int extra_float8_digits;
#if !(NeXT NX_CURRENT_COMPILER_RELEASE NX_COMPILER_RELEASE_3_2)
/* NS3.3 has conflicting declarations of these in math.h */
@@ -237,7 +243,7 @@
if (infflag 0)
PG_RETURN_CSTRING(strcpy(ascii, -Infinity));
- sprintf(ascii, %.*g, FLT_DIG, num);
+ sprintf(ascii, %.*g, FLT_DIG+extra_float4_digits, num);
PG_RETURN_CSTRING(ascii);
}
@@ -299,7 +305,7 @@
if (infflag 0)
PG_RETURN_CSTRING(strcpy(ascii, -Infinity));
- sprintf(ascii, %.*g, DBL_DIG, num);
+ sprintf(ascii, %.*g, DBL_DIG+extra_float8_digits, num);
PG_RETURN_CSTRING(ascii);
}
--- old/postgresql-7.2.1/src/backend/utils/misc/guc.c Tue Oct 30 05:38:56 2001
+++ postgresql-7.2.1/src/backend/utils/misc/guc.c Mon Nov 4 10:53:15 2002
@@ -49,6 +49,11 @@
extern int CommitSiblings;
extern bool FixBTree;
+/* For float4 and float8 extra digits in output format */
+
+extern int extra_float4_digits;
+extern int extra_float8_digits;
+
#ifdef ENABLE_SYSLOG
extern char *Syslog_facility;
extern char *Syslog_ident;
@@ -502,6 +507,19 @@
commit_siblings, PGC_USERSET, CommitSiblings,
5, 1, 1000, NULL, NULL
},
+
+ /*
+* For float4 and float8 extra digits in output format
+*/
+ {
+ float4_extra_digits, PGC_USERSET, extra_float4_digits,
+ 0, -4, 2, NULL, NULL
+ },
+
+ {
+ float8_extra_digits, PGC_USERSET, extra_float8_digits,
+ 0, -13, 2, NULL, NULL
+ },
{
NULL, 0, NULL, 0, 0, 0, NULL, NULL
--- old/postgresql-7.2.1/src/bin/psql/tab-complete.cMon Nov 5 17:46:31 2001
+++ postgresql-7.2.1/src/bin/psql/tab-complete.cMon Nov 4 11:16:01 2002
@@ -266,6 +266,9 @@
default_transaction_isolation,
+ float4_extra_digits,
+ float8_extra_digits,
+
NULL
};
--- old/postgresql-7.2.1/src/backend/utils/misc/postgresql.conf.sample Fri Jan 4
05:50:25 2002
+++ postgresql-7.2.1/src/backend/utils/misc/postgresql.conf.sample Mon Nov 4
+11:02:04 2002
@@ -182,3 +182,9 @@
#password_encryption = false
#sql_inheritance = true
#transform_null_equals = false
+
+#
+# floating point extra digits in output formatting
+#
+#extra_float4_digits = 0 #min -4, max 2
+#extra_float8_digits = 0 #min -13, max 2
---(end of broadcast)---
TIP 2: you can get off all lists at once with the unregister command
(send unregister YourEmailAddressHere to [EMAIL PROTECTED])
Re: [HACKERS] pg_dump and large files - is this a problem?
Zeugswetter Andreas SB SD [EMAIL PROTECTED] writes: The issue is, that you need to remove the #include bootstrap_tokens.h line from the lex file. Good point; I'm surprised gcc doesn't spit up on that. I've made that mod and also added the inclusion-order-correction in pqsignal.c. regards, tom lane ---(end of broadcast)--- TIP 2: you can get off all lists at once with the unregister command (send unregister YourEmailAddressHere to [EMAIL PROTECTED])
Re: [HACKERS] Finishing Domains...
Rod Taylor [EMAIL PROTECTED] writes: Is it safe to teach Var how to deal with values that do not originate from a tuple? Or should I create a new primnode to deal with these types of variables. I have no idea what you're talking about, but it sure sounds like something that should not be in Var ... perhaps you ought to post a more complete design document before you start coding. regards, tom lane ---(end of broadcast)--- TIP 3: if posting/reading through Usenet, please send an appropriate subscribe-nomail command to [EMAIL PROTECTED] so that your message can get through to the mailing list cleanly
Re: [HACKERS] Float output formatting options
Pedro M. Ferreira [EMAIL PROTECTED] writes: int extra_float4_digits, default 0, min -4, max 2 int extra_float8_digits, defualt 0, min -13, max 2 I think a single setting extra_float_digits would be sufficient. Also, if we concluded that there is a need of 2 more digits, should'nt this be the default ? No. pg_dump would want to bump it up on-the-fly. regards, tom lane ---(end of broadcast)--- TIP 4: Don't 'kill -9' the postmaster
Re: [HACKERS] Float output formatting options
Tom Lane wrote: Pedro M. Ferreira [EMAIL PROTECTED] writes: int extra_float4_digits, default 0, min -4, max 2 int extra_float8_digits, defualt 0, min -13, max 2 I think a single setting extra_float_digits would be sufficient. Ok. Assuming, int extra_float_digits, default 0, min -13, max 2 If extra_float_digits==-13 and we are outputing a float4 this results in a negative value for FLT_DIG+extra_float_digits. I dont know if sprintf's behaviour is the same across different libraries for this situation. Should I include the following to handle this case ? if(extra_float_digits-4) sprintf(ascii, %.*g, FLT_DIG-4, num); else sprintf(ascii, %.*g, FLT_DIG+extra_float_digits, num); Also, if we concluded that there is a need of 2 more digits, should'nt this be the default ? No. pg_dump would want to bump it up on-the-fly. regards, tom lane ---(end of broadcast)--- TIP 4: Don't 'kill -9' the postmaster -- -- Pedro Miguel Frazao Fernandes Ferreira Universidade do Algarve Faculdade de Ciencias e Tecnologia Campus de Gambelas 8000-117 Faro Portugal Tel./Fax: (+351) 289 800950 / 289 819403 http://w3.ualg.pt/~pfrazao ---(end of broadcast)--- TIP 1: subscribe and unsubscribe commands go to [EMAIL PROTECTED]
Re: [HACKERS] Float output formatting options
Pedro M. Ferreira [EMAIL PROTECTED] writes: Tom Lane wrote: I think a single setting extra_float_digits would be sufficient. Ok. Assuming, int extra_float_digits, default 0, min -13, max 2 If extra_float_digits==-13 and we are outputing a float4 this results in a negative value for FLT_DIG+extra_float_digits. You would want to clamp the values passed to %g to not less than 1. I'd favor code like int ndig = FLT_DIG + extra_float_digits; if (ndig 1) ndig = 1; sprintf(ascii, %.*g, ndig, num); Probably best to do it this way with float8 too; otherwise we're essentially wiring in the assumption that we know what DBL_DIG is. Which is exactly what we're trying to avoid doing. regards, tom lane ---(end of broadcast)--- TIP 4: Don't 'kill -9' the postmaster
Re: [HACKERS] Float output formatting options
Tom Lane wrote:
Pedro M. Ferreira [EMAIL PROTECTED] writes:
If extra_float_digits==-13 and we are outputing a float4 this results in
a negative value for FLT_DIG+extra_float_digits.
You would want to clamp the values passed to %g to not less than 1.
I'd favor code like
int ndig = FLT_DIG + extra_float_digits;
if (ndig 1)
ndig = 1;
sprintf(ascii, %.*g, ndig, num);
Probably best to do it this way with float8 too; otherwise we're
essentially wiring in the assumption that we know what DBL_DIG is.
Which is exactly what we're trying to avoid doing.
Good.
Corrected this, compiled and tested it. Works fine.
I am attaching the diff's made with diff -u. Sources were from Debian
source package Version 7.2.1-2woody2.
Best regards,
Pedro
regards, tom lane
--
--
Pedro Miguel Frazao Fernandes Ferreira
Universidade do Algarve
Faculdade de Ciencias e Tecnologia
Campus de Gambelas
8000-117 Faro
Portugal
Tel./Fax: (+351) 289 800950 / 289 819403
http://w3.ualg.pt/~pfrazao
--- old/postgresql-7.2.1/src/backend/utils/adt/float.c Tue Dec 11 02:02:12 2001
+++ postgresql-7.2.1/src/backend/utils/adt/float.c Mon Nov 4 15:50:19 2002
@@ -65,6 +65,11 @@
#include utils/array.h
#include utils/builtins.h
+/*
+ * Configuration options for float4 and float8 extra digits in output format
+ */
+
+int extra_float_digits;
#if !(NeXT NX_CURRENT_COMPILER_RELEASE NX_COMPILER_RELEASE_3_2)
/* NS3.3 has conflicting declarations of these in math.h */
@@ -228,6 +233,7 @@
float4 num = PG_GETARG_FLOAT4(0);
char *ascii = (char *) palloc(MAXFLOATWIDTH + 1);
int infflag;
+ int ndig = FLT_DIG + extra_float_digits;
if (isnan(num))
PG_RETURN_CSTRING(strcpy(ascii, NaN));
@@ -237,7 +243,10 @@
if (infflag 0)
PG_RETURN_CSTRING(strcpy(ascii, -Infinity));
- sprintf(ascii, %.*g, FLT_DIG, num);
+ if (ndig 1)
+ ndig = 1;
+
+ sprintf(ascii, %.*g, ndig, num);
PG_RETURN_CSTRING(ascii);
}
@@ -290,6 +299,7 @@
float8 num = PG_GETARG_FLOAT8(0);
char *ascii = (char *) palloc(MAXDOUBLEWIDTH + 1);
int infflag;
+ int ndig = DBL_DIG + extra_float_digits;
if (isnan(num))
PG_RETURN_CSTRING(strcpy(ascii, NaN));
@@ -299,7 +309,10 @@
if (infflag 0)
PG_RETURN_CSTRING(strcpy(ascii, -Infinity));
- sprintf(ascii, %.*g, DBL_DIG, num);
+if (ndig 1)
+ ndig = 1;
+
+ sprintf(ascii, %.*g, ndig, num);
PG_RETURN_CSTRING(ascii);
}
--- old/postgresql-7.2.1/src/backend/utils/misc/guc.c Tue Oct 30 05:38:56 2001
+++ postgresql-7.2.1/src/backend/utils/misc/guc.c Mon Nov 4 15:41:24 2002
@@ -49,6 +49,10 @@
extern int CommitSiblings;
extern bool FixBTree;
+/* For float4 and float8 extra digits in output format */
+
+extern int extra_float_digits;
+
#ifdef ENABLE_SYSLOG
extern char *Syslog_facility;
extern char *Syslog_ident;
@@ -502,6 +506,14 @@
commit_siblings, PGC_USERSET, CommitSiblings,
5, 1, 1000, NULL, NULL
},
+
+ /*
+* For float4 and float8 extra digits in output format
+*/
+ {
+ float_extra_digits, PGC_USERSET, extra_float_digits,
+ 0, -13, 2, NULL, NULL
+ },
{
NULL, 0, NULL, 0, 0, 0, NULL, NULL
--- old/postgresql-7.2.1/src/bin/psql/tab-complete.cMon Nov 5 17:46:31 2001
+++ postgresql-7.2.1/src/bin/psql/tab-complete.cMon Nov 4 15:38:52 2002
@@ -266,6 +266,8 @@
default_transaction_isolation,
+ float_extra_digits,
+
NULL
};
--- old/postgresql-7.2.1/src/backend/utils/misc/postgresql.conf.sample Fri Jan 4
05:50:25 2002
+++ postgresql-7.2.1/src/backend/utils/misc/postgresql.conf.sample Mon Nov 4
+15:39:55 2002
@@ -182,3 +182,8 @@
#password_encryption = false
#sql_inheritance = true
#transform_null_equals = false
+
+#
+# floating point extra digits in output formatting
+#
+#extra_float_digits = 0#min -13, max 2
---(end of broadcast)---
TIP 5: Have you checked our extensive FAQ?
http://www.postgresql.org/users-lounge/docs/faq.html
Re: [HACKERS] Float output formatting options
Pedro M. Ferreira [EMAIL PROTECTED] writes: I am attaching the diff's made with diff -u. Sources were from Debian source package Version 7.2.1-2woody2. Looks good, will keep to apply after we branch for 7.4 development. BTW, did you check to see if this affects the geometric types or not? I am not sure that they go through float8out; they may need similar adjustments in their output routines. regards, tom lane ---(end of broadcast)--- TIP 6: Have you searched our list archives? http://archives.postgresql.org
Re: [HACKERS] Float output formatting options
Tom Lane wrote: Looks good, will keep to apply after we branch for 7.4 development. BTW, did you check to see if this affects the geometric types or not? I am not sure that they go through float8out; they may need similar adjustments in their output routines. Only checked arrays. I will check this and get you posted about it ASAP. Best regards, Pedro regards, tom lane -- -- Pedro Miguel Frazao Fernandes Ferreira Universidade do Algarve Faculdade de Ciencias e Tecnologia Campus de Gambelas 8000-117 Faro Portugal Tel./Fax: (+351) 289 800950 / 289 819403 http://w3.ualg.pt/~pfrazao ---(end of broadcast)--- TIP 3: if posting/reading through Usenet, please send an appropriate subscribe-nomail command to [EMAIL PROTECTED] so that your message can get through to the mailing list cleanly
[HACKERS] v7.3Beta4 Tag'd and Packaged ...
Morning all ... As I was unable to figure out a way of merging HEAD into a BRANCH (BRANCH into HEAD is easy), I moved the REL7_3_STABLE tag up to today, so that the BRANCH is in a more appropriate location. I've packaged up beta4 based on the branch, and its currently up and available at the ftp site ... please download and test. For those with commit access, please note that v7.3 related stuff should, from now on, be committed *on* the appropriate branch, and v7.4 should be acceptable to add in ... Let me know if there are any problems with the package, especially in the docs area, so that I can put out an announce later on this evening ... ---(end of broadcast)--- TIP 6: Have you searched our list archives? http://archives.postgresql.org
Re: [HACKERS] Float output formatting options
Tom Lane wrote: BTW, did you check to see if this affects the geometric types or not? I am not sure that they go through float8out; they may need similar adjustments in their output routines. In fact they need adjustments. The *_out routines (in src/backend/utils/adt/geo_ops.c) for the geometric types rely on two functions to output data: static int pair_encode(float8 x, float8 y, char *str); static int single_encode(float8 x, char *str); These functions produce output with (for pair_encode): sprintf(str, %.*g,%.*g, digits8, x, digits8, y); digits8 is defined as , #define P_MAXDIG DBL_DIG static int digits8 = P_MAXDIG; I think it would be done the same way as for float4_out and float8_out: extern int extra_float_digits; int ndig = digits8 + extra_float_digits; if (ndig 1) ndig = 1; sprintf(str, %.*g,%.*g, ndig, x, ndig, y); There a bunch of other places where output is produced. They are all within #ifdef GEODEBUG / #enfif blocks. Should these be corrected the same way ? Regards, Pedro ---(end of broadcast)--- TIP 4: Don't 'kill -9' the postmaster
Re: [HACKERS] Float output formatting options
Pedro M. Ferreira [EMAIL PROTECTED] writes: These functions produce output with (for pair_encode): sprintf(str, %.*g,%.*g, digits8, x, digits8, y); digits8 is defined as , #define P_MAXDIG DBL_DIG static int digits8 = P_MAXDIG; I think it would be done the same way as for float4_out and float8_out: Yeah. In fact I'd be inclined to remove the static variable and make the code match float8out exactly (do DBL_DIG + extra_float_digits). There a bunch of other places where output is produced. They are all within #ifdef GEODEBUG / #enfif blocks. Should these be corrected the same way ? Up to you. Personally I'd just leave them alone... regards, tom lane ---(end of broadcast)--- TIP 6: Have you searched our list archives? http://archives.postgresql.org
Re: [HACKERS] Float output formatting options
Tom Lane wrote: Yeah. In fact I'd be inclined to remove the static variable and make the code match float8out exactly (do DBL_DIG + extra_float_digits). P_MAXDIG is only used in the lines below: #define P_MAXDIG DBL_DIG #define P_MAXLEN (2*(P_MAXDIG+7)+1) static int digits8 = P_MAXDIG; Is it ok to remove #define P_MAXDIG DBL_DIG, change P_MAXLEN to 2*(DBL_DIG+7)+1) and remove the line 'static int digits8 = P_MAXDIG;' ? Would then change the two geo output functions and replace digits8 by DBL_DIG in the #ifdef GEODEBUG / #enfif output stuff. There a bunch of other places where output is produced. They are all within #ifdef GEODEBUG / #enfif blocks. Should these be corrected the same way ? ---(end of broadcast)--- TIP 3: if posting/reading through Usenet, please send an appropriate subscribe-nomail command to [EMAIL PROTECTED] so that your message can get through to the mailing list cleanly
Re: [HACKERS] v7.3Beta4 Tag'd and Packaged ...
Marc G. Fournier [EMAIL PROTECTED] writes: As I was unable to figure out a way of merging HEAD into a BRANCH (BRANCH into HEAD is easy), I moved the REL7_3_STABLE tag up to today, so that the BRANCH is in a more appropriate location. It looks like this was only 99.44% right. There are some files and whole subdirectories that were deleted since REL7_3_STABLE was originally branched, but your update did not mark them as deleted in the branch: Only in REL7_3/pgsql/contrib: retep Only in REL7_3/pgsql/contrib/xml: README Only in REL7_3/pgsql/contrib/xml: pgxml.source Only in REL7_3/pgsql/contrib/xml: pgxml_dom.source Only in REL7_3/pgsql/doc/src/sgml: libpgeasy.sgml Only in REL7_3/pgsql/doc/src/sgml: odbc.sgml Only in REL7_3/pgsql/doc/src/sgml: recovery.sgml Only in REL7_3/pgsql/doc/src/sgml: version.sgml Only in REL7_3/pgsql/doc/src/sgml: y2k.sgml Only in REL7_3/pgsql/src/interfaces/jdbc: CHANGELOG Only in REL7_3/pgsql/src/interfaces/jdbc: Implementation Only in REL7_3/pgsql/src/interfaces/jdbc: jdbc.jpx Only in REL7_3/pgsql/src/interfaces/jdbc: utils Only in REL7_3/pgsql/src/test/regress/expected: geometry-bsdi-precision.out Perhaps easiest to just re-delete them in the branch? Or is there a cleaner way? For those with commit access, please note that v7.3 related stuff should, from now on, be committed *on* the appropriate branch, and v7.4 should be acceptable to add in ... Right-o: HEAD is 7.4devel now ... regards, tom lane ---(end of broadcast)--- TIP 1: subscribe and unsubscribe commands go to [EMAIL PROTECTED]
Re: [HACKERS] Float output formatting options
Pedro M. Ferreira [EMAIL PROTECTED] writes: Is it ok to remove #define P_MAXDIG DBL_DIG, change P_MAXLEN to 2*(DBL_DIG+7)+1) and remove the line 'static int digits8 = P_MAXDIG;' ? Perhaps P_MAXLEN now needs to be (2*(DBL_DIG+2+7)+1), considering that we'll allow extra_float_digits to be up to 2. What's it used for? regards, tom lane ---(end of broadcast)--- TIP 6: Have you searched our list archives? http://archives.postgresql.org
Re: [HACKERS] Float output formatting options
Tom Lane wrote: Pedro M. Ferreira [EMAIL PROTECTED] writes: Is it ok to remove #define P_MAXDIG DBL_DIG, change P_MAXLEN to 2*(DBL_DIG+7)+1) and remove the line 'static int digits8 = P_MAXDIG;' ? Perhaps P_MAXLEN now needs to be (2*(DBL_DIG+2+7)+1), considering that we'll allow extra_float_digits to be up to 2. What's it used for? Yes. I guess so, because it is used in what I think is a memory allocation function. P_MAXLEN is only used twice: * 1st use in path_encode() (allmost all the geo *_out functions use path_encode): intsize = npts * (P_MAXLEN + 3) + 2; /* Check for integer overflow */ if ((size - 2) / npts != (P_MAXLEN + 3)) elog(ERROR, Too many points requested); * 2nd use in circle_out(PG_FUNCTION_ARGS): result = palloc(3 * (P_MAXLEN + 1) + 3); I will do the changes tomorrow and send in the appropriate diff's. Regards, Pedro ---(end of broadcast)--- TIP 2: you can get off all lists at once with the unregister command (send unregister YourEmailAddressHere to [EMAIL PROTECTED])
Re: [HACKERS] Is my Internet connection slow
Peter Eisentraut [EMAIL PROTECTED] writes: Why do we have two doc builds anyway? Mainly because Bruce is willing to expend the cycles to rebuild his nearly on-demand. Last I checked, the build on hub is only updated once or twice a day, so it's not as useful for verifying doc checkins. regards, tom lane ---(end of broadcast)--- TIP 6: Have you searched our list archives? http://archives.postgresql.org
Re: [HACKERS] [CYGWIN] ipc-daemon
Peter Eisentraut [EMAIL PROTECTED] writes: Jason Tishler writes: 1. If ipc-daemon is not running, then cygipc's shmget() will return EACCES. 2. This causes PostgreSQL's InternalIpcMemoryCreate() to return NULL. 3. This causes PostgreSQL's PGSharedMemoryCreate() to spin looking for accessible shared memory which it will never find. To me, this is a bug in PostgreSQL. I disagree: just because cygipc returns error codes chosen at random doesn't mean that we should neglect the clear meaning of an error code. If a normal Unix system were to return EACCES here, the clear implication would be that there is an existing segment that we do not have permission to access. I don't see how cygipc isn't running can reasonably be translated into permission denied. After reading the shmget() man page, I choose to return ENOSPC. Is there a better choice? Such as ENOMEM? My first thought was ENOSYS (system call not implemented -- what BSD kernels tend to return if you didn't compile them with SysV IPC support), but that isn't a clearly superior choice either. If you can detect that cygipc is not running, then ENOSYS seems the best choice for reporting that. (ENOSPC would be misleading too.) If it's impractical to fix cygipc then I'd grudgingly go along with if (errno == EEXIST #ifndef __CYWGIN__ /* cygipc is broken */ || errno == EACCES #endif #ifdef EIDRM || errno == EIDRM #endif ) return NULL; regards, tom lane ---(end of broadcast)--- TIP 4: Don't 'kill -9' the postmaster
Re: [HACKERS] Request for supported platforms
On Fri, Nov 01, 2002 at 10:22:42AM -0500, Tom Lane wrote: I've committed fixes for the other issues too, so CVS tip should now pass cleanly on your platform. I've built the snapshot from CVS tip and now the regression tests pass with both GCC 2.95.1 and Compaq C V6.4-216 (dtk) on Digital UNIX V4.0G (Rev. 1530). Thank you very much for your help! -- Alessio F. Bragadini[EMAIL PROTECTED] APL Financial Services http://village.albourne.com Nicosia, Cyprus phone: +357-22-755750 It is more complicated than you think -- The Eighth Networking Truth from RFC 1925 ---(end of broadcast)--- TIP 6: Have you searched our list archives? http://archives.postgresql.org
[HACKERS] missing const in PQescapeBytea/PQunescapeBytea in 7.3b3
Hi,
I just discovered, that there is missing a const when passing a buffer to
PQescapeBytea and PQunescapeBytea. I fixed it and tried to create a usable
diff (I'm not so familar to diff).
Tommi
*** postgresql-7.3b3/src/interfaces/libpq/fe-exec.c 2002-09-04 22:31:47.0 +0200
--- postgresql-7.3b3tm1/src/interfaces/libpq/fe-exec.c 2002-11-02 22:28:34.0 +0100
***
*** 118,126
* anything = 0x80 --- \\ooo (where ooo is an octal expression)
*/
unsigned char *
! PQescapeBytea(unsigned char *bintext, size_t binlen, size_t *bytealen)
{
! unsigned char *vp;
unsigned char *rp;
unsigned char *result;
size_t i;
--- 118,126
* anything = 0x80 --- \\ooo (where ooo is an octal expression)
*/
unsigned char *
! PQescapeBytea(const unsigned char *bintext, size_t binlen, size_t *bytealen)
{
! const unsigned char *vp;
unsigned char *rp;
unsigned char *result;
size_t i;
***
*** 202,213
* 6 \\
*/
unsigned char *
! PQunescapeBytea(unsigned char *strtext, size_t *retbuflen)
{
size_t buflen;
unsigned char *buffer,
- *sp,
*bp;
unsigned int state = 0;
if (strtext == NULL)
--- 202,213
* 6 \\
*/
unsigned char *
! PQunescapeBytea(const unsigned char *strtext, size_t *retbuflen)
{
size_t buflen;
unsigned char *buffer,
*bp;
+ const unsigned char *sp;
unsigned int state = 0;
if (strtext == NULL)
*** postgresql-7.3b3/src/interfaces/libpq/libpq-fe.h 2002-09-04 22:31:47.0 +0200
--- postgresql-7.3b3tm1/src/interfaces/libpq/libpq-fe.h 2002-11-02 22:27:24.0 +0100
***
*** 249,257
/* Quoting strings before inclusion in queries. */
extern size_t PQescapeString(char *to, const char *from, size_t length);
! extern unsigned char *PQescapeBytea(unsigned char *bintext, size_t binlen,
size_t *bytealen);
! extern unsigned char *PQunescapeBytea(unsigned char *strtext,
size_t *retbuflen);
--- 249,257
/* Quoting strings before inclusion in queries. */
extern size_t PQescapeString(char *to, const char *from, size_t length);
! extern unsigned char *PQescapeBytea(const unsigned char *bintext, size_t binlen,
size_t *bytealen);
! extern unsigned char *PQunescapeBytea(const unsigned char *strtext,
size_t *retbuflen);
---(end of broadcast)---
TIP 4: Don't 'kill -9' the postmaster
[HACKERS] TPCC test in postgreSQL 7.3 beta3
Hello, I used TPCC to test postgreSQL 7.3 beta3 version. My test steps are showed as follow, 1st, the number of TPCC simulation processes is 100, run time is 7200 seconds. The consistency test was got through successfully. 2nd, the number of TPCC simulation processes is 100, run time is 3600 seconds. But, the consistency test couldn't be got through. The phenomenon are same in TPCC test of postgreSQL 7.3 beta2 version. Why the consistency test can't not be got through in long time TPCC test? Thanks. __ Do you Yahoo!? HotJobs - Search new jobs daily now http://hotjobs.yahoo.com/ ---(end of broadcast)--- TIP 5: Have you checked our extensive FAQ? http://www.postgresql.org/users-lounge/docs/faq.html
Re: [HACKERS] v7.3Beta4 Tag'd and Packaged ...
On Mon, 4 Nov 2002, Tom Lane wrote: Marc G. Fournier [EMAIL PROTECTED] writes: As I was unable to figure out a way of merging HEAD into a BRANCH (BRANCH into HEAD is easy), I moved the REL7_3_STABLE tag up to today, so that the BRANCH is in a more appropriate location. It looks like this was only 99.44% right. There are some files and whole subdirectories that were deleted since REL7_3_STABLE was originally branched, but your update did not mark them as deleted in the branch: Only in REL7_3/pgsql/contrib: retep Only in REL7_3/pgsql/contrib/xml: README Only in REL7_3/pgsql/contrib/xml: pgxml.source Only in REL7_3/pgsql/contrib/xml: pgxml_dom.source Only in REL7_3/pgsql/doc/src/sgml: libpgeasy.sgml Only in REL7_3/pgsql/doc/src/sgml: odbc.sgml Only in REL7_3/pgsql/doc/src/sgml: recovery.sgml Only in REL7_3/pgsql/doc/src/sgml: version.sgml Only in REL7_3/pgsql/doc/src/sgml: y2k.sgml Only in REL7_3/pgsql/src/interfaces/jdbc: CHANGELOG Only in REL7_3/pgsql/src/interfaces/jdbc: Implementation Only in REL7_3/pgsql/src/interfaces/jdbc: jdbc.jpx Only in REL7_3/pgsql/src/interfaces/jdbc: utils Only in REL7_3/pgsql/src/test/regress/expected: geometry-bsdi-precision.out Perhaps easiest to just re-delete them in the branch? Or is there a cleaner way? Easiest is to re-delete them ... from what i can tell from teh CVS docs, we did this backwards ... we should have worked on the branch itself, and then merges into the working tree ... something to remember for next time ---(end of broadcast)--- TIP 1: subscribe and unsubscribe commands go to [EMAIL PROTECTED]
Re: [HACKERS] v7.3Beta4 Tag'd and Packaged ...
Marc G. Fournier [EMAIL PROTECTED] writes: Perhaps easiest to just re-delete them in the branch? Or is there a cleaner way? Easiest is to re-delete them ... Okay, done that. REL7_3_STABLE branch now matches CVS tip --- or did until I branded HEAD's configure.in as 7.4devel, a few moments ago. Do you want to regenerate the 7.3b4 tarball so it omits the deleted files? Probably not really essential to do so. Meanwhile, the floodgates are open for 7.4 development on HEAD. regards, tom lane ---(end of broadcast)--- TIP 6: Have you searched our list archives? http://archives.postgresql.org
[HACKERS] protocol change in 7.4
There has been some previous discussion of changing the FE/BE protocol
in 7.4, in order to fix several problems. I think this is worth doing:
if we can resolve all these issues in a single release, it will lessen
the upgrade difficulties for users.
I'm aware of the following problems that need a protocol change to fix
them:
(1) Add an optional textual message to NOTIFY
(2) Remove the hard-coded limits on database and user names
(SM_USER, SM_DATABASE), replace them with variable-length
fields.
(3) Remove some legacy elements in the startup packet
('unused' can go -- perhaps 'tty' as well). I think the
'length' field of the password packet is also not used,
but I'll need to double-check that.
(4) Fix the COPY protocol (Tom?)
(5) Fix the Fastpath protocol (Tom?)
(6) Protocol-level support for prepared queries, in order to
bypass the parser (and maybe be more compatible with the
implementation of prepared queries in other databases).
(7) Include the current transaction status, since it's
difficult for the client app to determine it for certain
(Tom/Bruce?)
If I've missed anything or if there is something you think we should
add, please let me know.
I can implement (1), (2), (3), and possibly (7), if someone can tell
me exactly what is required (my memory of the discussion relating to
this is fuzzy). The rest is up for grabs.
Finally, how should we manage the transition? I wasn't around for the
earlier protocol changes, so I'd appreciate any input on steps we can
take to improve backward-compatibility.
Cheers,
Neil
--
Neil Conway [EMAIL PROTECTED] || PGP Key ID: DB3C29FC
---(end of broadcast)---
TIP 3: if posting/reading through Usenet, please send an appropriate
subscribe-nomail command to [EMAIL PROTECTED] so that your
message can get through to the mailing list cleanly
Re: [HACKERS] protocol change in 7.4
On Mon, Nov 04, 2002 at 08:10:29PM -0500, Mike Mascari wrote: Actually, I was thinking along the lines of a true CREATE DATABASE LINK implementation, where multiple databases could participate in a distributed transaction. That would require the backend in which the main query is executing to act as the coordinator and each of the other participating databases to act as cohorts. And would require a protocol change to support the PREPARE, COMMIT-VOTE/ABORT-VOTE reply, and an ACK message following the completion of the distributed COMMIT or ABORT. Right, you need TPC in order for pgsql to participate in transactions that span anything outside the DB proper. A DB link is one example, or an external transaction manager that coordinates DB and filesystem updates, for example. Zope could use this, to coordinate the DB with it's internal object store. Ross ---(end of broadcast)--- TIP 6: Have you searched our list archives? http://archives.postgresql.org
Re: [HACKERS] Finishing Domains...
On Mon, 2002-11-04 at 09:44, Tom Lane wrote: Rod Taylor [EMAIL PROTECTED] writes: Is it safe to teach Var how to deal with values that do not originate from a tuple? Or should I create a new primnode to deal with these types of variables. I have no idea what you're talking about, but it sure sounds like something that should not be in Var ... perhaps you ought to post a more complete design document before you start coding. Sorry, had planned on it. Anyway, below is a summary of what I think I need to do. Goal: CREATE DOMAIN dom AS integer CHECK(VALUE = 4 OR VALUE = 10); ^^ The constraint will be stored in pg_constraint with the conrelid set to 0, and contypid set to the domain pg_type oid. Domain check constraints will be pulled out of the table and inserted into the executor tree in coerce_type_constraints(), for execution by ExecEvalConstraintTest(). In the parser, I intend to represent the 'VALUE' term by a DomainConstraintValue node which inherits the properties (TypId, TypMod) of the domain base type. Requires some muckary in transformExpr() to carry the current datatype through it, much like carrying a context through the executor. It has been mentioned a few times that parse nodes should not be used in the executor as they tend to be bug prone, so I don't want to use DomainConstraintValue in the executor. So, I need a way to represent VALUE in both the stored pg_constraint.conbin column and the executor, which is replaced with the Datum being operated on during ExecEvalConstraintTest(). The Datum is created in ExecEvalConstraintTest, so it's easy to add it to the local context, and call ExecEvalExpr on the (VALUE = 4 OR VALUE = 10) expression. ExecEvalExpr runs through the Node passed and fills in subexpressions (Var nodes for example). Do I teach ExecEvalVar how to pull out an arbitrary value from ExprContext (since this is technically a variable), or do I create a different 'Var' for managing variables which do not come in the Tuple format. All I need is the datatype, and a little knowledge of the variable thats in ExprContext. -- Rod Taylor ---(end of broadcast)--- TIP 6: Have you searched our list archives? http://archives.postgresql.org
Re: [HACKERS] Float output formatting options
Perhaps P_MAXLEN now needs to be (2*(DBL_DIG+2+7)+1), considering that we'll allow extra_float_digits to be up to 2. What's it used for? Yes. I guess so, because it is used in what I think is a memory allocation function. P_MAXLEN is only used twice: Curse my slowness, but what's the actual problem being fixed here? Chris ---(end of broadcast)--- TIP 5: Have you checked our extensive FAQ? http://www.postgresql.org/users-lounge/docs/faq.html
Re: [HACKERS] protocol change in 7.4
I'm now implementing 2PC replication and distributed transaction. My 2PC needs some supports in startup packet to establish a replication session or a recovery session. BTW, my 2PC replication is working, and I'm implementing 2PC recovery now. I would like to here more about your implementation. Do you have some documentation that I could read? If not, perhaps (if you have the time) you could put together a post describing your work. Like Is it an internal or external solution. Are you sending SQL or tuples in your update messages. How are you handling failure detection? Is this partial or full replication? Please forgive me for asking so many questions, but I'm rather intrigued by database replication. Darren ---(end of broadcast)--- TIP 2: you can get off all lists at once with the unregister command (send unregister YourEmailAddressHere to [EMAIL PROTECTED])
Re: [HACKERS] protocol change in 7.4
Darren Johnson [EMAIL PROTECTED] wrote: I would like to here more about your implementation. Do you have some documentation that I could read? Documentation is not available, but I have some slides for my presentation. http://snaga.org/pgsql/20021018_2pc.pdf Some answers for your questions may be in these slides. And a current source code is available from: http://snaga.org/pgsql/pgsql-20021025.tgz If not, perhaps (if you have the time) you could put together a post describing your work. Like Is it an internal or external solution. Are you sending SQL or tuples in your update messages. How are you handling failure detection? Is this partial or full replication? It is an internal solution. In 2PC, pre-commit and commit are required. So my implementation has some internal modifications on transaction handling, log recording and else. -- NAGAYASU Satoshi [EMAIL PROTECTED] ---(end of broadcast)--- TIP 6: Have you searched our list archives? http://archives.postgresql.org
[HACKERS] Another bug in tsearch?
Hi Oleg Teodor, This behaviour is causing problems in my search engine: australia=# select 'banana/pineapple'::mquery_txt; mquery_txt 'banana/pineapple' (1 row) In our case the forward slash symbol should really be treated as a word break. Are there any cases where it shouldn't be? I can maybe only think of 3/4, etc. Chris ---(end of broadcast)--- TIP 4: Don't 'kill -9' the postmaster
[HACKERS] rebuilding beta4 ... fails in docs build ...
one too many things removed? this build worked earlier today, so its only since removing those directories that something has broken ... openjade -D . -D ./ref -c /usr/local/share/sgml/docbook/dsssl/modular/catalog - d stylesheet.dsl -i output-html -t sgml postgres.sgml openjade:installation.sgml:1535:2:E: character data is not allowed here openjade:installation.sgml:1535:30:E: document type does not allow element EMAI L here; assuming missing ENTRY start-tag openjade:/usr/local/share/sgml/docbook/dsssl/modular/html/../common/dbtable.dsl: 224:13:E: 2nd argument for primitive ancestor of wrong type: #unknown object 151425216 not a singleton node list gmake[1]: *** [postgres.html] Error 1 gmake[1]: Leaving directory `/usr/local/pgsql/beta/pgsql/doc/src/sgml' gmake: *** [postgres.tar] Error 2 ---(end of broadcast)--- TIP 6: Have you searched our list archives? http://archives.postgresql.org
Re: [HACKERS] Float output formatting options
Christopher Kings-Lynne [EMAIL PROTECTED] writes: Curse my slowness, but what's the actual problem being fixed here? Two things: * allow pg_dump to accurately dump and restore float quantities (setting float_extra_digits to 2 during the dump will accomplish this, at least on systems with reasonable float I/O routines). * allow us to get out from under the geometry regression test's platform dependency problems (setting float_extra_digits to -2 or so during the test should make most or all of the variations go away). This proposal is the first one I've seen that solves both these problems without introducing any compatibility issues of its own. regards, tom lane ---(end of broadcast)--- TIP 2: you can get off all lists at once with the unregister command (send unregister YourEmailAddressHere to [EMAIL PROTECTED])
Re: [HACKERS] rebuilding beta4 ... fails in docs build ...
On Mon, 2002-11-04 at 23:10, Marc G. Fournier wrote: one too many things removed? this build worked earlier today, so its only since removing those directories that something has broken ... Bruce left an extra /entry tag at on line 1534 of installation.sgml :) Can't make a patch because anoncvs hasn't updated yet :) openjade -D . -D ./ref -c /usr/local/share/sgml/docbook/dsssl/modular/catalog - d stylesheet.dsl -i output-html -t sgml postgres.sgml openjade:installation.sgml:1535:2:E: character data is not allowed here openjade:installation.sgml:1535:30:E: document type does not allow element EMAI L here; assuming missing ENTRY start-tag openjade:/usr/local/share/sgml/docbook/dsssl/modular/html/../common/dbtable.dsl: 224:13:E: 2nd argument for primitive ancestor of wrong type: #unknown object 151425216 not a singleton node list gmake[1]: *** [postgres.html] Error 1 gmake[1]: Leaving directory `/usr/local/pgsql/beta/pgsql/doc/src/sgml' gmake: *** [postgres.tar] Error 2 ---(end of broadcast)--- TIP 6: Have you searched our list archives? http://archives.postgresql.org -- Rod Taylor ---(end of broadcast)--- TIP 4: Don't 'kill -9' the postmaster
Re: [HACKERS] rebuilding beta4 ... fails in docs build ...
Marc G. Fournier [EMAIL PROTECTED] writes: one too many things removed? Just broken SGML markup ... fixed I think ... regards, tom lane ---(end of broadcast)--- TIP 5: Have you checked our extensive FAQ? http://www.postgresql.org/users-lounge/docs/faq.html
Re: [HACKERS] protocol change in 7.4
I don't see why 2PC would require any protocol-level change. I would think that the API would be something like BEGIN; issue some commands ... PRECOMMIT; -- if the above does not return an error, then COMMIT; In other words, 2PC would require some new commands, but a new command doesn't affect the protocol layer. regards, tom lane ---(end of broadcast)--- TIP 5: Have you checked our extensive FAQ? http://www.postgresql.org/users-lounge/docs/faq.html
Re: [HACKERS] protocol change in 7.4
Questions have arisen during discussions about errors relating
to how to support error codes without changing the FE/BE
protocols. (see TODO.detail/error)
Now that the protocol is up for revision, how about supporting
sql state strings, error codes, and other information directly in
the protocol.
Regards,
Grant
Neil Conway wrote:
There has been some previous discussion of changing the FE/BE protocol
in 7.4, in order to fix several problems. I think this is worth doing:
if we can resolve all these issues in a single release, it will lessen
the upgrade difficulties for users.
I'm aware of the following problems that need a protocol change to fix
them:
(1) Add an optional textual message to NOTIFY
(2) Remove the hard-coded limits on database and user names
(SM_USER, SM_DATABASE), replace them with variable-length
fields.
(3) Remove some legacy elements in the startup packet
('unused' can go -- perhaps 'tty' as well). I think the
'length' field of the password packet is also not used,
but I'll need to double-check that.
(4) Fix the COPY protocol (Tom?)
(5) Fix the Fastpath protocol (Tom?)
(6) Protocol-level support for prepared queries, in order to
bypass the parser (and maybe be more compatible with the
implementation of prepared queries in other databases).
(7) Include the current transaction status, since it's
difficult for the client app to determine it for certain
(Tom/Bruce?)
If I've missed anything or if there is something you think we should
add, please let me know.
I can implement (1), (2), (3), and possibly (7), if someone can tell
me exactly what is required (my memory of the discussion relating to
this is fuzzy). The rest is up for grabs.
Finally, how should we manage the transition? I wasn't around for the
earlier protocol changes, so I'd appreciate any input on steps we can
take to improve backward-compatibility.
Cheers,
Neil
smime.p7s
Description: S/MIME Cryptographic Signature
Re: [HACKERS] protocol change in 7.4
Grant Finnemore [EMAIL PROTECTED] writes: Now that the protocol is up for revision, how about supporting sql state strings, error codes, and other information directly in the protocol. Ah, thanks for pointing that out. Error codes would be another thing we can ideally support in 7.4, and we'd need a protocol change to do it properly, AFAICS. IIRC, Peter E. expressed some interest in doing this... Cheers, Neil -- Neil Conway [EMAIL PROTECTED] || PGP Key ID: DB3C29FC ---(end of broadcast)--- TIP 1: subscribe and unsubscribe commands go to [EMAIL PROTECTED]
Re: [HACKERS] protocol change in 7.4
Tom Lane wrote: I don't see why 2PC would require any protocol-level change. I would think that the API would be something like BEGIN; issue some commands ... PRECOMMIT; -- if the above does not return an error, then COMMIT; In other words, 2PC would require some new commands, but a new command doesn't affect the protocol layer. I think a precommit-vote-commit phase of 2PC can be implemented in command-lavel or protocol-level. In command-level 2PC, an user application (or application programmer) must know the DBMS is clustered or not (to use PRECOMMIT command). In protocol-layer 2PC, no new SQL command is required. A precommit-vote-commit phase will be called implicitly. It means an user application can be used without any modification. An application can use a traditional way (BEGIN...COMMIT). So I made my decision to use protocol-layer implementation. It doesn't affect the SQL command layer. -- NAGAYASU Satoshi [EMAIL PROTECTED] ---(end of broadcast)--- TIP 3: if posting/reading through Usenet, please send an appropriate subscribe-nomail command to [EMAIL PROTECTED] so that your message can get through to the mailing list cleanly
Re: [HACKERS] Is my Internet connection slow
On Mon, 4 Nov 2002, Bruce Momjian wrote: Sure, script attached. It uses a few of my other scripts so it will need customization. I can easily supply those missing scripts too. I run it every 7 minutes from cron with: */7 * * * * root /letc/pgsgml /dev/null 21 I can even install the whole thing on postgresql.org if you wish. It requires a personal crontab, of course, and a directory to put the result. I already have a ~momjian web directory for such stuff. Actually, from my read through the docs on CVS itself, there is a way of setting it up so that when something is committed, it runs an external program ... I'll have to do some further research on it, but I should be able to set it up so that when someone commits something to docs, it does the build ... Can you setup your scripts on postgresql.org, and let me know where/what to run, and I'll see if I can get that setup? ---(end of broadcast)--- TIP 3: if posting/reading through Usenet, please send an appropriate subscribe-nomail command to [EMAIL PROTECTED] so that your message can get through to the mailing list cleanly
Re: [HACKERS] Request for supported platforms
Ports list updated: http://candle.pha.pa.us/main/writings/pgsql/sgml/supported-platforms.html --- Sean Chittenden wrote: Or at least it's *trying* to apply it for 4.7 --- as near as I can tell without testing, the above scrap of resultmap code is wrong because both of the i.86 lines will match on FreeBSD 4.7, and I think the pg_regress coding will take the last match. Larry, did you actually test the CVS-tip resultmap to make sure it picks the right comparison file on your box? Yes, just did and it *FAILS*. you need the order you have below. Sorry... Okay, I've updated the CVS tip to look like this: geometry/i.86-.*-freebsd=geometry-positive-zeros-bsd geometry/i.86-.*-freebsd4.7=geometry-bsd-precision geometry/i.86-.*-freebsd5=geometry-bsd-precision geometry/alpha.*-freebsd=geometry-positive-zeros Sean, would you verify this works for you? It does, thank you. I've just updated the -devel port to 7.3b4, hopefully the mirrors will pick up the bits soon. -sc -- Sean Chittenden -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road + Christ can be your backup.| Newtown Square, Pennsylvania 19073 ---(end of broadcast)--- TIP 3: if posting/reading through Usenet, please send an appropriate subscribe-nomail command to [EMAIL PROTECTED] so that your message can get through to the mailing list cleanly
Re: [HACKERS] Is my Internet connection slow
On Mon, 2002-11-04 at 21:40, Marc G. Fournier wrote: On Mon, 4 Nov 2002, Bruce Momjian wrote: Sure, script attached. It uses a few of my other scripts so it will need customization. I can easily supply those missing scripts too. I run it every 7 minutes from cron with: */7 * * * * root /letc/pgsgml /dev/null 21 I can even install the whole thing on postgresql.org if you wish. It requires a personal crontab, of course, and a directory to put the result. I already have a ~momjian web directory for such stuff. Actually, from my read through the docs on CVS itself, there is a way of setting it up so that when something is committed, it runs an external One way is to change the DEFAULT in CVSROOT/loginfo. Normally its a log accumulator or email system of somekind. But there is no reason it couldn't be more complex and kick off additional jobs. Just make sure whatever it is execs a background jobs (commits are held during execution of the script) and protect against simultaneous runs of the background job. Rod Taylor ---(end of broadcast)--- TIP 6: Have you searched our list archives? http://archives.postgresql.org
[HACKERS] PostgreSQL IRC Channel... who's the Admin?
Hi all, Who's an Admin for the PostgreSQL IRC Channel (#postgresql) on irc.openprojects.net? The topic there is advertising some items for some guy on eBay, claiming it supports the PostgreSQL channel, and no-one knows anything about it. There also doesn't appear to be any IRC Admin's about. So, does anyone know who the Admin's are, so we can get things fixed up? :-) Regards and best wishes, Justin Clift -- My grandfather once told me that there are two kinds of people: those who work and those who take the credit. He told me to try to be in the first group; there was less competition there. - Indira Gandhi ---(end of broadcast)--- TIP 2: you can get off all lists at once with the unregister command (send unregister YourEmailAddressHere to [EMAIL PROTECTED])
Re: [HACKERS] [CYGWIN] ipc-daemon
Tom, Peter, On Mon, Nov 04, 2002 at 02:43:01PM -0500, Tom Lane wrote: Peter Eisentraut [EMAIL PROTECTED] writes: To me, this is a bug in PostgreSQL. I disagree: just because cygipc returns error codes chosen at random doesn't mean that we should neglect the clear meaning of an error code. If a normal Unix system were to return EACCES here, the clear implication would be that there is an existing segment that we do not have permission to access. I don't see how cygipc isn't running can reasonably be translated into permission denied. [snip] My first thought was ENOSYS (system call not implemented -- what BSD kernels tend to return if you didn't compile them with SysV IPC support), but that isn't a clearly superior choice either. If you can detect that cygipc is not running, then ENOSYS seems the best choice for reporting that. (ENOSPC would be misleading too.) If it's impractical to fix cygipc then I'd grudgingly go along with if (errno == EEXIST #ifndef __CYWGIN__ /* cygipc is broken */ || errno == EACCES #endif #ifdef EIDRM || errno == EIDRM #endif ) return NULL; Thanks for your feedback. I will take this to the Cygwin list and see what happens. Unfortunately, the cygipc maintainer is AWOL right now because of Ph.D. thesis commitments. Hence, even if I can get the Cygwin community to agree to this change, there may not be an official cygipc release for a while. Thanks, Jason -- GPG key available on key servers or http://www.tishler.net/jason/gpg.txt msg24667/pgp0.pgp Description: PGP signature
Re: [CYGWIN] [HACKERS] Request for supported platforms
Peter, On Wed, Oct 30, 2002 at 07:36:40PM +0100, Peter Eisentraut wrote: Dave Page writes: Hackers: As the Cygwin release that is actively supported is the binary distribution that Jason builds, I would think this is OK to be listed as supported if no-one disagrees... I disagree. We document as supported those platforms that build out of the box, not those that build somehow, somewhere, by someone. Rather than advocating methods to manually edit your system headers we should try to find out what the problem really is, such as by analyzing config.log. Did you miss the following? http://archives.postgresql.org/pgsql-hackers/2002-10/msg01303.php As you can see, I have already performed root cause analysis of theses problems *and* have taken the proper steps to ensure that PostgreSQL builds OOTB under Cygwin (after the next Cygwin release). Jason ---(end of broadcast)--- TIP 3: if posting/reading through Usenet, please send an appropriate subscribe-nomail command to [EMAIL PROTECTED] so that your message can get through to the mailing list cleanly
Re: [HACKERS] [GENERAL] Database replication... - Mission Critical DBMS's --
On Tue, 2002-10-29 at 17:56, Greg Patnude wrote: Does anyone know of a reasonable means of synchronizing two (or more) postgreSQL database servers in order to maintain two COMPLETE concurrent copies of the same database ? I'm not sure how complete it is (just starting to look at this myself) but 'dbbalancer' at least makes some claims to this. It acts as (basically) a postgresql proxy server, and can send queries to multiple servers. My presumption would be that if you initialize 2 databases to a known identical start, have all the same triggers and rules on both, then send all queries to both databases, you will have 2 identical databases at the end. Don't know how well that will work in practice tho. I should know more in the next couple of weeks. b.g. ---(end of broadcast)--- TIP 2: you can get off all lists at once with the unregister command (send unregister YourEmailAddressHere to [EMAIL PROTECTED])
Re: [HACKERS] Turning the PLANNER off
On Wed, Oct 30, 2002 at 10:59:39PM -0500, Tom Lane wrote: What's the basis for your assertion that it's planning things that don't need it? Given a JOIN-constrained query I do not believe the planner will look at any cases other than the intended join order. Well, that was a loose choice of words - let's say the planner seems to be taking awfully long to build an execution tree with only one choice available. He can do whatever he wants, as long as he has no illusions about getting it accepted back into the sources ;-). Understood - this would be a hacking tool only. What would probably be more useful is to do some profiling to understand why the planner is taking longer than he wants even with a JOIN-constrained query. I should think this would be pretty quick. Yup, that targets the same question as above - anything 'uneeded' actually happening in the planner? I'll send him off with this suggestion. Ross ---(end of broadcast)--- TIP 5: Have you checked our extensive FAQ? http://www.postgresql.org/users-lounge/docs/faq.html
Re: [HACKERS] Is my Internet connection slow
On Mon, 4 Nov 2002, Tom Lane wrote: Peter Eisentraut [EMAIL PROTECTED] writes: Why do we have two doc builds anyway? Mainly because Bruce is willing to expend the cycles to rebuild his nearly on-demand. Last I checked, the build on hub is only updated once or twice a day, so it's not as useful for verifying doc checkins. 'k, is there a reason why it can't be run more often? What command on hub has to be run? ---(end of broadcast)--- TIP 5: Have you checked our extensive FAQ? http://www.postgresql.org/users-lounge/docs/faq.html
[HACKERS] disable a single trigger
What is the status of the tgenabled field in pg_trigger? It would be so nice to be able to disable just a single trigger on a table rather than having to disable all triggers. The docs say not presently checked everywhere it should be, so disabling a trigger by setting this false does not work reliably. Will it ever be reliable? Thanks, -- Laurette Cisneros The Database Group (510) 420-3137 NextBus Information Systems, Inc. www.nextbus.com -- My other vehicle is my imagination. - bumper sticker ---(end of broadcast)--- TIP 2: you can get off all lists at once with the unregister command (send unregister YourEmailAddressHere to [EMAIL PROTECTED])
Re: [HACKERS] Is my Internet connection slow
Marc G. Fournier [EMAIL PROTECTED] writes: Mainly because Bruce is willing to expend the cycles to rebuild his nearly on-demand. Last I checked, the build on hub is only updated once or twice a day, so it's not as useful for verifying doc checkins. 'k, is there a reason why it can't be run more often? What command on hub has to be run? AFAICT, Bruce does a cvs update every ten or fifteen minutes and then rebuilds the docs if anything changed in the doc subtree. Bruce, maybe you could send Marc the script you use? regards, tom lane ---(end of broadcast)--- TIP 1: subscribe and unsubscribe commands go to [EMAIL PROTECTED]
Re: [HACKERS] disable a single trigger
Laurette Cisneros [EMAIL PROTECTED] writes: What is the status of the tgenabled field in pg_trigger? It would be so nice to be able to disable just a single trigger on a table rather than having to disable all triggers. The docs say not presently checked everywhere it should be, so disabling a trigger by setting this false does not work reliably. I believe it does work for BEFORE triggers, but not AFTER triggers. There was some question about when it should be checked for an AFTER trigger (when the event is queued? Or fired?). regards, tom lane ---(end of broadcast)--- TIP 3: if posting/reading through Usenet, please send an appropriate subscribe-nomail command to [EMAIL PROTECTED] so that your message can get through to the mailing list cleanly
Re: [HACKERS] Is my Internet connection slow
Sure, script attached. It uses a few of my other scripts so it will need customization. I can easily supply those missing scripts too. I run it every 7 minutes from cron with: */7 * * * * root /letc/pgsgml /dev/null 21 I can even install the whole thing on postgresql.org if you wish. It requires a personal crontab, of course, and a directory to put the result. I already have a ~momjian web directory for such stuff. --- Tom Lane wrote: Marc G. Fournier [EMAIL PROTECTED] writes: Mainly because Bruce is willing to expend the cycles to rebuild his nearly on-demand. Last I checked, the build on hub is only updated once or twice a day, so it's not as useful for verifying doc checkins. 'k, is there a reason why it can't be run more often? What command on hub has to be run? AFAICT, Bruce does a cvs update every ten or fifteen minutes and then rebuilds the docs if anything changed in the doc subtree. Bruce, maybe you could send Marc the script you use? regards, tom lane ---(end of broadcast)--- TIP 1: subscribe and unsubscribe commands go to [EMAIL PROTECTED] -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road + Christ can be your backup.| Newtown Square, Pennsylvania 19073 : cd /pgsgmlbuild [ -f /tmp/pgsql_sgml.lock ] exit TRAPEXTRAFILE=/tmp/pgsql_sgml.lock . traprm touch /tmp/pgsql_sgml.lock $TMP/1 while : do CONT=N if [ X$1 = X-f ] thenCONT=Y shift elsepgcvs update sgml $TMP/2 21 BG=$!; export BG (sleep 300; kill $BG /dev/null 21) BG2=$!; export BG2 wait $BG if ! kill0 $BG2 thenCONT=N elsekill $BG2 /dev/null 21 CONT=Y cat $TMP/2 | grep '^[A-Z] sgml/' $TMP/1 [ $? -ne 0 ] CONT=N fi fi if [ $CONT = N ] thenbreak fi # continue echo \nBuild: `date` build.dates cat $TMP/1 build.dates echo PostgreSQL CVS Documentation Build $TMP/0 echo --\n $TMP/0 echo Build started: `date` $TMP/0 gmake postgres.tar.gz 21 | grep -v DTDDECL $TMP/2 echo Build completed: `date`\n $TMP/0 echo Changes in this build: $TMP/0 cat $TMP/1 $TMP/0 echo \nA full copy of this documenation is at ftp://candle.pha.pa.us/pub/postgresql/postgresql-docs.tar.gz\n; $TMP/0 if grep -qi 'error' $TMP/2 || grep -qi ':E:' $TMP/2 thenecho \nThere were errors in this build. They appear in red.\n $TMP/0 cat $TMP/2 $TMP/0 elseecho \nBuild completed successfully. $TMP/0 fi pipe sed 's;HTML.manifest:;HTML.manifest :;g' $TMP/0 txt2html -m -s 100 -p 100 --title PostgreSQL CVS Docs built `date` \ --link /u/txt2html/txt2html.dict \ --append_head /u/txt2html/BODY $TMP/0 build.html pipe sed 's;^.*[Ee]rror.*$;FONT COLOR=RED/FONT;' build.html pipe sed 's;^.*:E:.*$;FONT COLOR=RED/FONT;' build.html rm -f /var/www/docs/main/writings/pgsql/sgml/* mv sgml/*.html build.html /var/www/docs/main/writings/pgsql/sgml cp sgml/*.css sgml/*.gif /var/www/docs/main/writings/pgsql/sgml cp postgres.tar.gz /pgftp/postgresql-docs.tar.gz gmake clean /dev/null 21 [ X$1 = X-f ] break done ---(end of broadcast)--- TIP 4: Don't 'kill -9' the postmaster
Re: [HACKERS] v7.3Beta4 Tag'd and Packaged ...
On Mon, 2002-11-04 at 15:52, Tom Lane wrote: Marc G. Fournier [EMAIL PROTECTED] writes: Perhaps easiest to just re-delete them in the branch? Or is there a cleaner way? Easiest is to re-delete them ... Okay, done that. REL7_3_STABLE branch now matches CVS tip --- or did until I branded HEAD's configure.in as 7.4devel, a few moments ago. Do you want to regenerate the 7.3b4 tarball so it omits the deleted files? Probably not really essential to do so. Out of curiousity, is there an easy (heh) way to upgrade(change?) a checked out tree to be off the branch? -- Larry Rosenman http://www.lerctr.org/~ler Phone: +1 972-414-9812 E-Mail: [EMAIL PROTECTED] US Mail: 1905 Steamboat Springs Drive, Garland, TX 75044-6749 ---(end of broadcast)--- TIP 5: Have you checked our extensive FAQ? http://www.postgresql.org/users-lounge/docs/faq.html
Re: [HACKERS] pg_dump and large files - is this a problem?
Does this resolve our AIX compile problem? --- Tom Lane wrote: Zeugswetter Andreas SB SD [EMAIL PROTECTED] writes: The issue is, that you need to remove the #include bootstrap_tokens.h line from the lex file. Good point; I'm surprised gcc doesn't spit up on that. I've made that mod and also added the inclusion-order-correction in pqsignal.c. regards, tom lane -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road + Christ can be your backup.| Newtown Square, Pennsylvania 19073 ---(end of broadcast)--- TIP 5: Have you checked our extensive FAQ? http://www.postgresql.org/users-lounge/docs/faq.html
Re: [HACKERS] Request for supported platforms
Ports list updated: http://candle.pha.pa.us/main/writings/pgsql/sgml/supported-platforms.html --- Alessio Bragadini wrote: On Fri, Nov 01, 2002 at 10:22:42AM -0500, Tom Lane wrote: I've committed fixes for the other issues too, so CVS tip should now pass cleanly on your platform. I've built the snapshot from CVS tip and now the regression tests pass with both GCC 2.95.1 and Compaq C V6.4-216 (dtk) on Digital UNIX V4.0G (Rev. 1530). Thank you very much for your help! -- Alessio F. Bragadini [EMAIL PROTECTED] APL Financial Serviceshttp://village.albourne.com Nicosia, Cyprus phone: +357-22-755750 It is more complicated than you think -- The Eighth Networking Truth from RFC 1925 ---(end of broadcast)--- TIP 6: Have you searched our list archives? http://archives.postgresql.org -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road + Christ can be your backup.| Newtown Square, Pennsylvania 19073 ---(end of broadcast)--- TIP 5: Have you checked our extensive FAQ? http://www.postgresql.org/users-lounge/docs/faq.html
Re: [HACKERS] missing const in PQescapeBytea/PQunescapeBytea in 7.3b3
This has been saved for the 7.4 release: http:/momjian.postgresql.org/cgi-bin/pgpatches2 --- Tommi Maekitalo wrote: Hi, I just discovered, that there is missing a const when passing a buffer to PQescapeBytea and PQunescapeBytea. I fixed it and tried to create a usable diff (I'm not so familar to diff). Tommi [ Attachment, skipping... ] [ Attachment, skipping... ] ---(end of broadcast)--- TIP 4: Don't 'kill -9' the postmaster -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road + Christ can be your backup.| Newtown Square, Pennsylvania 19073 ---(end of broadcast)--- TIP 3: if posting/reading through Usenet, please send an appropriate subscribe-nomail command to [EMAIL PROTECTED] so that your message can get through to the mailing list cleanly
Re: [HACKERS] [CYGWIN] ipc-daemon
Jason Tishler writes: 1. If ipc-daemon is not running, then cygipc's shmget() will return EACCES. 2. This causes PostgreSQL's InternalIpcMemoryCreate() to return NULL. 3. This causes PostgreSQL's PGSharedMemoryCreate() to spin looking for accessible shared memory which it will never find. To me, this is a bug in PostgreSQL. A comment in InternalIpcMemoryCreate() says * Fail quietly if error indicates a collision with existing * segment. One would expect EEXIST, given that we said IPC_EXCL, * but perhaps we could get a permission violation instead? I tend to think that the answer to that question is No. After reading the shmget() man page, I choose to return ENOSPC. Is there a better choice? Such as ENOMEM? My first thought was ENOSYS (system call not implemented -- what BSD kernels tend to return if you didn't compile them with SysV IPC support), but that isn't a clearly superior choice either. Fixing PostgreSQL is probably better and quicker to yield a return. -- Peter Eisentraut [EMAIL PROTECTED] ---(end of broadcast)--- TIP 5: Have you checked our extensive FAQ? http://www.postgresql.org/users-lounge/docs/faq.html
Re: [HACKERS] Is my Internet connection slow
Bruce Momjian writes: Yeah. I don't use your docs build anymore unless I really have to... the one at hub.org is much more responsive. Oh, that is a problem. Thanks. Why do we have two doc builds anyway? -- Peter Eisentraut [EMAIL PROTECTED] ---(end of broadcast)--- TIP 2: you can get off all lists at once with the unregister command (send unregister YourEmailAddressHere to [EMAIL PROTECTED])
Re: [HACKERS] protocol change in 7.4
Neil Conway wrote: There has been some previous discussion of changing the FE/BE protocol in 7.4, in order to fix several problems. I think this is worth doing: if we can resolve all these issues in a single release, it will lessen the upgrade difficulties for users. snip If I've missed anything or if there is something you think we should add, please let me know. Is there any thought about changing the protocol to support two-phase commit? Not that 2PC and distributed transactions would be implemented in 7.4, but to prevent another protocol change in the future? Mike Mascari [EMAIL PROTECTED] ---(end of broadcast)--- TIP 4: Don't 'kill -9' the postmaster
Re: [HACKERS] protocol change in 7.4
Mike Mascari [EMAIL PROTECTED] writes: Is there any thought about changing the protocol to support two-phase commit? Not that 2PC and distributed transactions would be implemented in 7.4, but to prevent another protocol change in the future? My understanding is that 2PC is one way to implement multi-master replication. If that's what you're referring to, then I'm not sure I see the point: the multi-master replication project (pgreplication) doesn't use 2PC, due to apparent scalability problems (not to mention that it also uses a separate channel for communications between backends on different nodes). Cheers, Neil -- Neil Conway [EMAIL PROTECTED] || PGP Key ID: DB3C29FC ---(end of broadcast)--- TIP 4: Don't 'kill -9' the postmaster
Re: [HACKERS] protocol change in 7.4
Neil Conway wrote: Mike Mascari [EMAIL PROTECTED] writes: Is there any thought about changing the protocol to support two-phase commit? Not that 2PC and distributed transactions would be implemented in 7.4, but to prevent another protocol change in the future? My understanding is that 2PC is one way to implement multi-master replication. If that's what you're referring to, then I'm not sure I see the point: the multi-master replication project (pgreplication) doesn't use 2PC, due to apparent scalability problems (not to mention that it also uses a separate channel for communications between backends on different nodes). Actually, I was thinking along the lines of a true CREATE DATABASE LINK implementation, where multiple databases could participate in a distributed transaction. That would require the backend in which the main query is executing to act as the coordinator and each of the other participating databases to act as cohorts. And would require a protocol change to support the PREPARE, COMMIT-VOTE/ABORT-VOTE reply, and an ACK message following the completion of the distributed COMMIT or ABORT. Mike Mascari [EMAIL PROTECTED] ---(end of broadcast)--- TIP 5: Have you checked our extensive FAQ? http://www.postgresql.org/users-lounge/docs/faq.html
Re: [HACKERS] protocol change in 7.4
Hi, Mike Mascari [EMAIL PROTECTED] wrote: Is there any thought about changing the protocol to support two-phase commit? Not that 2PC and distributed transactions would be implemented in 7.4, but to prevent another protocol change in the future? I'm now implementing 2PC replication and distributed transaction. My 2PC needs some supports in startup packet to establish a replication session or a recovery session. BTW, my 2PC replication is working, and I'm implementing 2PC recovery now. -- NAGAYASU Satoshi [EMAIL PROTECTED] ---(end of broadcast)--- TIP 4: Don't 'kill -9' the postmaster
Re: [HACKERS] Request for supported platforms
Or at least it's *trying* to apply it for 4.7 --- as near as I can tell without testing, the above scrap of resultmap code is wrong because both of the i.86 lines will match on FreeBSD 4.7, and I think the pg_regress coding will take the last match. Larry, did you actually test the CVS-tip resultmap to make sure it picks the right comparison file on your box? Yes, just did and it *FAILS*. you need the order you have below. Sorry... Okay, I've updated the CVS tip to look like this: geometry/i.86-.*-freebsd=geometry-positive-zeros-bsd geometry/i.86-.*-freebsd4.7=geometry-bsd-precision geometry/i.86-.*-freebsd5=geometry-bsd-precision geometry/alpha.*-freebsd=geometry-positive-zeros Sean, would you verify this works for you? It does, thank you. I've just updated the -devel port to 7.3b4, hopefully the mirrors will pick up the bits soon. -sc -- Sean Chittenden ---(end of broadcast)--- TIP 4: Don't 'kill -9' the postmaster
Re: [HACKERS] PostgreSQL IRC Channel... who's the Admin?
Justin Clift [EMAIL PROTECTED] writes: Hi all, Who's an Admin for the PostgreSQL IRC Channel (#postgresql) on irc.openprojects.net? The only person I've seen with ops is someone called 'raja', who is only rarely (i.e. once every couple months) in the channel. I talked to lilo about getting Bruce ops, and he talked to Bruce about it. Not sure what came out of that... Cheers, Neil -- Neil Conway [EMAIL PROTECTED] || PGP Key ID: DB3C29FC ---(end of broadcast)--- TIP 2: you can get off all lists at once with the unregister command (send unregister YourEmailAddressHere to [EMAIL PROTECTED])
Re: [HACKERS] v7.3Beta4 Tag'd and Packaged ...
As I was unable to figure out a way of merging HEAD into a BRANCH (BRANCH into HEAD is easy), I moved the REL7_3_STABLE tag up to today, so that the BRANCH is in a more appropriate location. It is easy to move both directions; I've done it. You need to have the right stuff in your checked-out tree (probably the branch), and you need to do the merge from HEAD. What went wrong in trying to do that? - Thomas ---(end of broadcast)--- TIP 1: subscribe and unsubscribe commands go to [EMAIL PROTECTED]
Re: [HACKERS] v7.3Beta4 Tag'd and Packaged ...
Out of curiousity, is there an easy (heh) way to upgrade(change?) a checked out tree to be off the branch? cvs update -rYOUR_BRANCH_NAME_HERE . - Thomas ---(end of broadcast)--- TIP 1: subscribe and unsubscribe commands go to [EMAIL PROTECTED]
Re: [HACKERS] PostgreSQL IRC Channel... who's the Admin?
Hi Neil, Cool. It's fixed now. We've just recreated the channel and started giving operator access to the right people. :-) Regards and best wishes, Justin Clift Neil Conway wrote: Justin Clift [EMAIL PROTECTED] writes: Hi all, Who's an Admin for the PostgreSQL IRC Channel (#postgresql) on irc.openprojects.net? The only person I've seen with ops is someone called 'raja', who is only rarely (i.e. once every couple months) in the channel. I talked to lilo about getting Bruce ops, and he talked to Bruce about it. Not sure what came out of that... Cheers, Neil -- Neil Conway [EMAIL PROTECTED] || PGP Key ID: DB3C29FC -- My grandfather once told me that there are two kinds of people: those who work and those who take the credit. He told me to try to be in the first group; there was less competition there. - Indira Gandhi ---(end of broadcast)--- TIP 2: you can get off all lists at once with the unregister command (send unregister YourEmailAddressHere to [EMAIL PROTECTED])
