Streamline anachronistic Ptr2Long union usage#533
Merged
jcflack merged 2 commits intoREL1_6_STABLEfrom May 29, 2025
Merged
Conversation
PostgreSQL itself has been requiring a C compiler that has uintptr_t since 9.0 (postgres/postgres@85d02a6). The PL/Java 1.6.x branch only supports back to PG 9.5, so there is no question the type is supported. An existing static assert in DualState.c is hereby changed to assert that sizeof(uintptr_t) <= sizeof(jlong); this is also assured for the present by a comment upstream in postgres.h section 1: "we require: sizeof(Datum) == sizeof(void *) == 4 or 8" (where Datum is uintptr_t). In passing, specifically type as SPIPlanPtr some variables that had been typed void *. That was once their actual type in the SPI docs, but not since PG 8.2. The few sites that ended up as PointerGetDatum(JLongGet(Pointer,...)) are effectively a longwinded way of doing nothing, but I left them that way to clearly reflect that a pointer is involved, and avoid assuming PointerGetDatum will always be as simple as it is.
While void * works, mistakes are more easily caught with the intended pointer type.
jcflack
added a commit
that referenced
this pull request
May 30, 2025
The merged work includes PR #533, which does away with the old Ptr2Long union in favor of new PointerGetJLong and JlongGet(... conversions. In merging, also convert the uses of Ptr2Long that were added on this branch.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
PostgreSQL itself has been requiring a C compiler that has
uintptr_tsince 9.0 (postgres/postgres@85d02a6). The PL/Java 1.6.x branch only supports back to PG 9.5, so there is no question the type is supported.An existing static assert in
DualState.cis hereby changed to assert thatsizeof(uintptr_t) <= sizeof(jlong); this is also assured for the present by a comment upstream inpostgres.hsection 1: "we require: sizeof(Datum) == sizeof(void *) == 4 or 8" (whereDatumisuintptr_t).In passing, specifically type as
SPIPlanPtrsome variables that had been typedvoid *. That was once their actual type in the SPI docs, but not since PG 8.2.The few sites that ended up as
PointerGetDatum(JLongGet(Pointer,...))are effectively a longwinded way of doing nothing, but I left them that way to clearly reflect that a pointer is involved, and avoid assumingPointerGetDatumwill always be as simple as it currently is.