Re: remaining sql/json patches

On Fri, Nov 24, 2023 at 9:28 PM Alvaro Herrera <alvherre(at)alvh(dot)no-ip(dot)org> wrote:
> Some quick grepping gave me this table,
>
> YYLAST YYNTOKENS YYNNTS YYNRULES YYNSTATES YYMAXUTOK
> REL9_1_STABLE 69680 429 546 2218 4179 666
> REL9_2_STABLE 73834 432 546 2261 4301 669
> REL9_3_STABLE 77969 437 558 2322 4471 674
> REL9_4_STABLE 79419 442 576 2369 4591 679
> REL9_5_STABLE 92495 456 612 2490 4946 693
> REL9_6_STABLE 92660 459 618 2515 5006 696
> REL_10_STABLE 99601 472 653 2663 5323 709
> REL_11_STABLE 102007 480 668 2728 5477 717
> REL_12_STABLE 103948 482 667 2724 5488 719
> REL_13_STABLE 104224 492 673 2760 5558 729
> REL_14_STABLE 108111 503 676 3159 5980 740
> REL_15_STABLE 111091 506 688 3206 6090 743
> REL_16_STABLE 115435 519 706 3283 6221 756
> master 117115 521 707 3300 6255 758
> master+v26 121817 537 738 3415 6470 774
>
> and the attached chart. (v26 is with all patches applied, including the
> JSON_TABLE one whose grammar has not yet been fully tweaked.)

Thanks for the chart.

> So, while the jump from v26 is not a trivial one, it seems within
> reasonable bounds.

Agreed.

> For example, the jump between 13 and 14 looks worse.
> (I do wonder what happened there.)

The following commit sounds like it might be related?

commit 06a7c3154f5bfad65549810cc84f0e3a77b408bf
Author: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
Date: Fri Sep 18 16:46:26 2020 -0400

Allow most keywords to be used as column labels without requiring AS.

--
Thanks, Amit Langote
EDB: http://www.enterprisedb.com