Claude merge 2

.gitattributes

@@ -13,6 +13,7 @@ README.*	conflict-marker-size=32
 *.data						-whitespace
 contrib/pgcrypto/sql/pgp-armor.sql		whitespace=-blank-at-eol
 src/backend/catalog/sql_features.txt		whitespace=space-before-tab,blank-at-eof,-blank-at-eol
+src/backend/utils/Gen_dummy_probes.pl.prolog	whitespace=-blank-at-eof
 
 # Test output files that contain extra whitespace
 *.out					-whitespace

CLAUDE.md

@@ -0,0 +1,107 @@
+# CLAUDE.md
+
+This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
+
+## Project overview
+
+This is **ArenaDatabaseDB (ADB)** — an MPP (Massively Parallel Processing) database fork of Greenplum Database (GPDB), itself built on PostgreSQL. The repo tracks upstream PostgreSQL and periodically merges new major PostgreSQL versions into GPDB-specific branches.
+
+**Key branches:**
+- `adb-6.x` — production ADB 6.x line (main branch for PRs)
+- `ai-merge-stage1` — staging branch for the current PG 8.0.0-alpha.0 → GPDB merge
+- `gg_upgrade` — tracks upstream PostgreSQL code
+
+## Build
+
+Requires: GNU make, autoconf 2.69, Bison, Flex, Perl, and standard C toolchain.
+
+```bash
+# Configure (first time or after configure.in changes)
+./configure --prefix=/usr/local/pgsql
+
+# Build everything
+make world
+
+# Install
+make install
+
+# Clean
+make distclean
+```
+
+`make world` builds `src/` and `contrib/`. Plain `make` builds `src/` only.
+
+## Running tests
+
+```bash
+# Run regression tests against a temporary installation (no running server needed)
+make check
+
+# Run against an already-running server
+make installcheck
+
+# Run all test suites (regress, isolation, pl, contrib, bin)
+make check-world
+make installcheck-world
+
+# Run parallel regression tests
+make installcheck-parallel
+
+# Run a specific test file by name against a running server
+cd src/test/regress && ./pg_regress --inputdir=. --schedule=serial_schedule <testname>
+
+# Run isolation (concurrent transaction) tests
+cd src/test/isolation && make installcheck
+# Or a specific spec: ./pg_isolation_regress <specname>
+```
+
+Set `MAX_CONNECTIONS=N` to cap parallelism: `make check MAX_CONNECTIONS=4`.
+
+## Code style
+
+- **C/Perl**: tabs, 4-space indent. Run `pgindent` (see `src/tools/pgindent/README.gpdb`) before submitting.
+- **Python**: spaces, 4-space indent. Must pass `pylint`.
+- **Go**: formatted with `gofmt`.
+- Formatting config is in `.editorconfig`.
+- Follow [PostgreSQL Coding Conventions](https://www.postgresql.org/docs/current/source.html).
+
+## Architecture
+
+```
+src/backend/         Main server process
+  access/            Table and index access methods (heap, nbtree, gin, gist, brin, hash, spgist)
+  catalog/           System catalog management
+  commands/          SQL command execution (DDL)
+  executor/          Query execution engine
+  nodes/             Node type definitions, copy/equal/out functions
+  optimizer/         Query planner (geqo, path, plan, prep, util)
+  parser/            SQL parser
+  replication/       WAL streaming, logical replication
+  storage/           Buffer manager, file I/O, lock manager, page layout
+  utils/             Memory management, error handling, type system, caching
+
+src/include/         Header files (mirrors backend/ structure)
+src/bin/             Client utilities: psql, pg_dump, pg_ctl, pg_basebackup, pg_upgrade, etc.
+src/pl/              Procedural languages: plpgsql, plperl, plpython, tcl
+src/test/            Test suites: regress, isolation, authentication, subscription, ssl
+contrib/             Optional extensions (pg_stat_statements, pageinspect, postgres_fdw, etc.)
+```
+
+GPDB/ADB-specific distributed execution concepts used throughout the codebase:
+- **Motion nodes** — data movement operators between MPP segments
+- **Slices** — independent units of parallel execution
+- **ORCA** — the Greenplum cost-based optimizer (referenced in optimizer/ and JIT-related code)
+- **arenadata_toolkit** — ADB-specific monitoring extension (tested in isolation2 tests)
+
+## PostgreSQL major-version merge workflow
+
+The primary ongoing task on `ai-merge-stage1` is merging upstream PostgreSQL into GPDB. The full decision matrix and workflow are documented in [`GG_PG_MERGE_SKILL.md`](./GG_PG_MERGE_SKILL.md). Key points:
+
+1. `git merge --no-commit --no-ff <upstream-tag>`
+2. Record conflicts: `git diff --name-only --diff-filter=U`
+3. Resolve semantically — never blindly take `ours` or `theirs`
+4. Adopt upstream API shapes first; re-graft GPDB-specific logic into the new shape
+5. Verify: `rg "^(<<<<<<<|=======|>>>>>>>)"` must return nothing
+6. Build and run targeted regression tests before finalizing
+
+Reference commits for resolution style: `1e11aaff762`, `f2b03841`, `1fa092913d2`, `3e9744465db`, `ed7a5095716ee`, `4dbcb3f844ec`, `a91e2fa94180`, `55a1954da16`, `80831bcdbe`, `eb57bd9c1`.

COPYRIGHT

@@ -14,7 +14,7 @@ the PostgreSQL License is provided below:
 PostgreSQL Database Management System
 (formerly known as Postgres, then as Postgres95)
 
-Portions Copyright (c) 1996-2020, PostgreSQL Global Development Group
+Portions Copyright (c) 1996-2021, PostgreSQL Global Development Group
 
 Portions Copyright (c) 1994, The Regents of the University of California