pgsql: Use heap_inplace_update() to unset pg_database.dathasloginevt

2024-02-11 Thread Alexander Korotkov
Use heap_inplace_update() to unset pg_database.dathasloginevt Doing this instead of regular updates serves two purposes. First, that avoids possible waiting on the row-level lock. Second, that avoids dealing with TOAST. It's known that changes made by heap_inplace_update() may be lost due to

pgsql: Fix some typos in event trigger docs

2024-02-11 Thread Alexander Korotkov
Fix some typos in event trigger docs Discussion: https://postgr.es/m/CALj2ACWFUW4jX9EW7CLxbzSS%2Bb7b0Z%3DxKYrqzj2Rstc9MCEx7g%40mail.gmail.com Author: Bharath Rupireddy Branch -- master Details --- https://git.postgresql.org/pg/commitdiff/5fce30e77fe133e1404167c13caaf7e0d1369295

pgsql: Remove obsolete script related to MSVC build system

2024-02-11 Thread Peter Eisentraut
Remove obsolete script related to MSVC build system Branch -- master Details --- https://git.postgresql.org/pg/commitdiff/428e2de1b895079bbf50a128f8b3824679f4bb65 Modified Files -- src/tools/ci/windows_build_config.pl | 16 1 file changed, 16 deletions(-)

pgsql: Fix gai_strerror() thread-safety on Windows.

2024-02-11 Thread Thomas Munro
Fix gai_strerror() thread-safety on Windows. Commit 5579388d removed code that supplied a fallback implementation of getaddrinfo(), which was dead code on modern systems. One tiny piece of the removed code was still doing something useful on Windows, though: that OS's own

pgsql: Fix gai_strerror() thread-safety on Windows.

2024-02-11 Thread Thomas Munro
Fix gai_strerror() thread-safety on Windows. Commit 5579388d removed code that supplied a fallback implementation of getaddrinfo(), which was dead code on modern systems. One tiny piece of the removed code was still doing something useful on Windows, though: that OS's own

pgsql: Use extensible buffers to assemble command lines

2024-02-11 Thread Peter Eisentraut
Use extensible buffers to assemble command lines This makes use of StringInfo to assemble command lines, instead of using fixed-size buffers and the (remote) possibility of "command too long" errors. Also makes the code a bit simpler. This covers the test driver programs pg_regress and