SPI_SAVEPLAN(3)

💾 SPI_SAVEPLAN(3) 💾

okay bestie, welcome to the PostgreSQL 18.6 Documentation page for SPI_saveplan — the little function that makes your prepared statement immortal (well, immortal-ish, we'll get to that) 🫶

---

✨ NAME ✨

SPI_saveplan — save a prepared statement 💅

---

🌸 SYNOPSIS 🌸

SPIPlanPtr SPI_saveplan(SPIPlanPtr plan)

like that's it. one argument. she's lowkey elegant 💜

---

🦋 DESCRIPTION 🦋

so here's the tea ☕ SPI_saveplan copies a statement that you passed in (and that was prepared by SPI_prepare) into memory that will NOT be freed by SPI_finish, and also NOT by the transaction manager. then it hands you back a pointer to that copied statement ✨

why do you care?? because this gives you the ability to reuse prepared statements in the subsequent invocations of your C function, all within the current session. main character energy fr fr 👑

---

🎀 ARGUMENTS 🎀

• SPIPlanPtr plan — the prepared statement that's getting saved 🥺 she's going into the vault

---

💖 RETURN VALUE 💖

a pointer to the copied statement 🫶 or NULL if it didn't work out 😭

and if there's an error, SPI_result gets set like this:

• SPI_ERROR_ARGUMENT — if plan is NULL or invalid 🙈
• SPI_ERROR_UNCONNECTED — if you called this from an unconnected C function 💀

---

⭐ NOTES ⭐

the statement you originally passed in does not get freed, so you might wanna run SPI_freeplan on it to avoid leaking memory until SPI_finish 👀 like, please don't leak, bestie.

also, real talk: in most cases, SPI_keepplan is preferred over this function, since it accomplishes largely the same result without needing to physically copy the prepared statement's data structures. so like... SPI_keepplan is the it-girl here, no cap 💅

not me telling you the function you're reading about is usually not the one you want 😭 but the docs said it, not me!

---

that's the whole page, you ate that 🍓 go forth and prepare those statements responsibly, queen 👑

with love from the PostgreSQL 18.6 docs, dated 2026, page SPI_SAVEPLAN(3) — mwah 💋