public inbox for [email protected]  
help / color / mirror / Atom feed
[PATCH v4] popcount
2+ messages / 2 participants
[nested] [flat]

* [PATCH v4] popcount
@ 2020-12-30 10:51  David Fetter <[email protected]>
  0 siblings, 0 replies; 2+ messages in thread

From: David Fetter @ 2020-12-30 10:51 UTC (permalink / raw)


Now it's accessible to SQL for the BIT VARYING and BYTEA types.

diff --git doc/src/sgml/func.sgml doc/src/sgml/func.sgml
index aa99665e2e..7626edeee7 100644
--- doc/src/sgml/func.sgml
+++ doc/src/sgml/func.sgml
@@ -4030,6 +4030,23 @@ SELECT format('Testing %3$s, %2$s, %s', 'one', 'two', 'three');
       </para></entry>
      </row>
 
+      <row>
+       <entry role="func_table_entry"><para role="func_signature">
+        <indexterm>
+         <primary>count_set_bits</primary>
+        </indexterm>
+        <function>count_set_bits</function> ( <parameter>bytes</parameter> <type>bytea</type> )
+        <returnvalue>bigint</returnvalue>
+       </para>
+       <para>
+        Counts the number of bits set in a binary string.
+       </para>
+       <para>
+        <literal>count_set_bits('\xdeadbeef'::bytea)</literal>
+        <returnvalue>24</returnvalue>
+       </para></entry>
+      </row>
+
       <row>
        <entry role="func_table_entry"><para role="func_signature">
         <indexterm>
@@ -4830,6 +4847,23 @@ SELECT format('Testing %3$s, %2$s, %s', 'one', 'two', 'three');
        </para></entry>
       </row>
 
+      <row>
+       <entry role="func_table_entry"><para role="func_signature">
+        <indexterm>
+         <primary>count_set_bits</primary>
+        </indexterm>
+        <function>count_set_bits</function> ( <parameter>bits</parameter> <type>bit</type> )
+        <returnvalue>bigint</returnvalue>
+       </para>
+       <para>
+        Counts the bits set in a bit string.
+       </para>
+       <para>
+        <literal>count_set_bits(B'101010101010101010')</literal>
+        <returnvalue>9</returnvalue>
+       </para></entry>
+      </row>
+
       <row>
        <entry role="func_table_entry"><para role="func_signature">
         <indexterm>
@@ -4869,6 +4903,7 @@ SELECT format('Testing %3$s, %2$s, %s', 'one', 'two', 'three');
         <returnvalue>101010001010101010</returnvalue>
        </para></entry>
       </row>
+
      </tbody>
     </tgroup>
    </table>
diff --git src/include/catalog/pg_proc.dat src/include/catalog/pg_proc.dat
index b5f52d4e4a..416ee0c2fb 100644
--- src/include/catalog/pg_proc.dat
+++ src/include/catalog/pg_proc.dat
@@ -1446,6 +1446,9 @@
 { oid => '752', descr => 'substitute portion of string',
   proname => 'overlay', prorettype => 'bytea',
   proargtypes => 'bytea bytea int4', prosrc => 'byteaoverlay_no_len' },
+{ oid => '8436', descr => 'count set bits',
+  proname => 'count_set_bits', prorettype => 'int8', proargtypes => 'bytea',
+  prosrc => 'byteapopcount'},
 
 { oid => '725',
   proname => 'dist_pl', prorettype => 'float8', proargtypes => 'point line',
@@ -3865,6 +3868,9 @@
 { oid => '3033', descr => 'set bit',
   proname => 'set_bit', prorettype => 'bit', proargtypes => 'bit int4 int4',
   prosrc => 'bitsetbit' },
+{ oid => '8435', descr => 'count set bits',
+  proname => 'count_set_bits', prorettype => 'int8', proargtypes => 'bit',
+  prosrc => 'bitpopcount'},
 
 # for macaddr type support
 { oid => '436', descr => 'I/O',
diff --git src/backend/utils/adt/varbit.c src/backend/utils/adt/varbit.c
index 2235866244..c9c6c73422 100644
--- src/backend/utils/adt/varbit.c
+++ src/backend/utils/adt/varbit.c
@@ -36,6 +36,7 @@
 #include "libpq/pqformat.h"
 #include "nodes/nodeFuncs.h"
 #include "nodes/supportnodes.h"
+#include "port/pg_bitutils.h"
 #include "utils/array.h"
 #include "utils/builtins.h"
 #include "utils/varbit.h"
@@ -1878,3 +1879,30 @@ bitgetbit(PG_FUNCTION_ARGS)
 	else
 		PG_RETURN_INT32(0);
 }
+
+/*
+ * bitpopcount
+ *
+ * Returns the number of bits set in a bit string.
+ *
+ */
+Datum
+bitpopcount(PG_FUNCTION_ARGS)
+{
+	/* There's really no chance of an overflow here because
+	 * to get to INT64_MAX set bits, an object would have to be
+	 * an exbibyte long, exceeding what PostgreSQL can currently
+	 * store by a factor of 2^28
+	 */
+	int64		popcount;
+	VarBit		*arg1 = PG_GETARG_VARBIT_P(0);
+	bits8		*p;
+	int			len;
+
+	p = VARBITS(arg1);
+	len = (VARBITLEN(arg1) + BITS_PER_BYTE - 1) / BITS_PER_BYTE;
+
+	popcount = pg_popcount((char *)p, len);
+
+	PG_RETURN_INT64(popcount);
+}
diff --git src/backend/utils/adt/varlena.c src/backend/utils/adt/varlena.c
index 479ed9ae54..3f1179a0e8 100644
--- src/backend/utils/adt/varlena.c
+++ src/backend/utils/adt/varlena.c
@@ -3439,6 +3439,22 @@ bytea_overlay(bytea *t1, bytea *t2, int sp, int sl)
 	return result;
 }
 
+/*
+ * popcount
+ */
+Datum
+byteapopcount(PG_FUNCTION_ARGS)
+{
+	bytea	*t1 = PG_GETARG_BYTEA_PP(0);
+	int		len;
+	int64	result;
+
+	len = VARSIZE_ANY_EXHDR(t1);
+	result = pg_popcount(VARDATA_ANY(t1), len);
+
+	PG_RETURN_INT64(result);
+}
+
 /*
  * byteapos -
  *	  Return the position of the specified substring.
diff --git src/test/regress/expected/bit.out src/test/regress/expected/bit.out
index a7f95b846d..1187076a8d 100644
--- src/test/regress/expected/bit.out
+++ src/test/regress/expected/bit.out
@@ -710,6 +710,19 @@ SELECT overlay(B'0101011100' placing '001' from 20);
  0101011100001
 (1 row)
 
+-- Popcount
+SELECT count_set_bits(B'0101011100'::bit(10));
+ count_set_bits 
+----------------
+              5
+(1 row)
+
+SELECT count_set_bits(B'1111111111'::bit(10));
+ count_set_bits 
+----------------
+             10
+(1 row)
+
 -- This table is intentionally left around to exercise pg_dump/pg_upgrade
 CREATE TABLE bit_defaults(
   b1 bit(4) DEFAULT '1001',
diff --git src/test/regress/expected/strings.out src/test/regress/expected/strings.out
index 7c91afa6e4..e94e76000b 100644
--- src/test/regress/expected/strings.out
+++ src/test/regress/expected/strings.out
@@ -2179,3 +2179,10 @@ SELECT encode(overlay(E'Th\\000omas'::bytea placing E'\\002\\003'::bytea from 5
  Th\000o\x02\x03
 (1 row)
 
+SET bytea_output TO hex;
+SELECT count_set_bits(E'\\xdeadbeef'::bytea);
+ count_set_bits 
+----------------
+             24
+(1 row)
+
diff --git src/test/regress/sql/bit.sql src/test/regress/sql/bit.sql
index ea01742c4a..8893d38b0d 100644
--- src/test/regress/sql/bit.sql
+++ src/test/regress/sql/bit.sql
@@ -215,6 +215,10 @@ SELECT overlay(B'0101011100' placing '101' from 6);
 SELECT overlay(B'0101011100' placing '001' from 11);
 SELECT overlay(B'0101011100' placing '001' from 20);
 
+-- Popcount
+SELECT count_set_bits(B'0101011100'::bit(10));
+SELECT count_set_bits(B'1111111111'::bit(10));
+
 -- This table is intentionally left around to exercise pg_dump/pg_upgrade
 CREATE TABLE bit_defaults(
   b1 bit(4) DEFAULT '1001',
diff --git src/test/regress/sql/strings.sql src/test/regress/sql/strings.sql
index ef4bfb008a..a4f7879b16 100644
--- src/test/regress/sql/strings.sql
+++ src/test/regress/sql/strings.sql
@@ -730,3 +730,6 @@ SELECT btrim(E'\\000trim\\000'::bytea, ''::bytea);
 SELECT encode(overlay(E'Th\\000omas'::bytea placing E'Th\\001omas'::bytea from 2),'escape');
 SELECT encode(overlay(E'Th\\000omas'::bytea placing E'\\002\\003'::bytea from 8),'escape');
 SELECT encode(overlay(E'Th\\000omas'::bytea placing E'\\002\\003'::bytea from 5 for 3),'escape');
+
+SET bytea_output TO hex;
+SELECT count_set_bits(E'\\xdeadbeef'::bytea);


^ permalink  raw  reply  [nested|flat] 2+ messages in thread

* IPC::Run::time[r|out] vs our TAP tests
@ 2024-04-04 21:24  Tom Lane <[email protected]>
  0 siblings, 0 replies; 2+ messages in thread

From: Tom Lane @ 2024-04-04 21:24 UTC (permalink / raw)
  To: [email protected]

TIL that IPC::Run::timer is not the same as IPC::Run::timeout.
With a timer object you have to check $timer->is_expired to see
if the timeout has elapsed, but with a timeout object you don't
because it will throw a Perl exception upon timing out, probably
killing your test program.

It appears that a good chunk of our TAP codebase has not read this
memo, because I see plenty of places that are checking is_expired
in the naive belief that they'll still have control after a timeout
has fired.

The particular thing that started me down this road was wondering
why we are getting no useful failure details from buildfarm member
tanager's struggles with the tab-completion test case added by commit
927332b95 [1].  Apparently it's not seeing a match to what it expects
so it eventually times out, but all we get in the log is

[03:03:42.595](0.002s) ok 82 - complete an interpolated psql variable name
[03:03:42.597](0.002s) ok 83 - \\r works
IPC::Run: timeout on timer #1 at /usr/share/perl5/IPC/Run.pm line 2944.
# Postmaster PID for node "main" is 17308
### Stopping node "main" using mode immediate

We would have learned something useful if control had returned to
pump_until, or even better 010_tab_completion.pl's check_completion,
but it doesn't.

A minimum fix that seems to make this work better is as attached,
but I feel like somebody ought to examine all the IPC::Run::timer
and IPC::Run::timeout calls and see which ones are mistaken.
It's a little scary to convert a timeout to a timer because of
the hazard that someplace that would need to be checking for
is_expired isn't.

Also, the debug printout code at the bottom of check_completion
is quite useless, because control can never reach it since
BackgroundPsql::query_until will "die" on failure.  I think that
that code worked when written, and I'm suspicious that 664d75753
broke it, but I've not dug deeply into the history.

			regards, tom lane

[1] https://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=tanager&dt=2024-04-04%2016%3A56%3A14



Attachments:

  [text/x-diff] use-timer-not-timeout-in-BackgroundPsql.patch (1.0K, ../../[email protected]/2-use-timer-not-timeout-in-BackgroundPsql.patch)
  download | inline diff:
diff --git a/src/test/perl/PostgreSQL/Test/BackgroundPsql.pm b/src/test/perl/PostgreSQL/Test/BackgroundPsql.pm
index 4091c311b8..72bb21e160 100644
--- a/src/test/perl/PostgreSQL/Test/BackgroundPsql.pm
+++ b/src/test/perl/PostgreSQL/Test/BackgroundPsql.pm
@@ -96,7 +96,7 @@ sub new
 	  "Forbidden caller of constructor: package: $package, file: $file:$line"
 	  unless $package->isa('PostgreSQL::Test::Cluster');
 
-	$psql->{timeout} = IPC::Run::timeout(
+	$psql->{timeout} = IPC::Run::timer(
 		defined($timeout)
 		? $timeout
 		: $PostgreSQL::Test::Utils::timeout_default);
diff --git a/src/test/perl/PostgreSQL/Test/Utils.pm b/src/test/perl/PostgreSQL/Test/Utils.pm
index 42d5a50dc8..8fd0426d30 100644
--- a/src/test/perl/PostgreSQL/Test/Utils.pm
+++ b/src/test/perl/PostgreSQL/Test/Utils.pm
@@ -429,6 +429,7 @@ Pump until string is matched on the specified stream, or timeout occurs.
 sub pump_until
 {
 	my ($proc, $timeout, $stream, $until) = @_;
+	die "timeout should be a timer" if defined($timeout->exception);
 	$proc->pump_nb();
 	while (1)
 	{


^ permalink  raw  reply  [nested|flat] 2+ messages in thread


end of thread, other threads:[~2024-04-04 21:24 UTC | newest]

Thread overview: 2+ messages (download: mbox mbox.gz follow: Atom feed)
-- links below jump to the message on this page --
2020-12-30 10:51 [PATCH v4] popcount David Fetter <[email protected]>
2024-04-04 21:24 IPC::Run::time[r|out] vs our TAP tests Tom Lane <[email protected]>

This inbox is served by agora; see mirroring instructions
for how to clone and mirror all data and code used for this inbox