[3.14] reinstate _OPTIMIZED & LTO_CFLAGS for correct flag handling in install_base.sh

[h-vetinari]

Fix-up for #894 (noticed while working on #913), because the following code actually requires _OPTIMIZED to be set

python-feedstock/recipe/install_base.sh

Lines 40 to 59 in ec8a1a1

	declare -a _FLAGS_REPLACE=()
	if [[ ${_OPTIMIZED} == yes ]]; then
	_FLAGS_REPLACE+=(-O3)
	_FLAGS_REPLACE+=(-O2)
	_FLAGS_REPLACE+=("-fprofile-use")
	_FLAGS_REPLACE+=("")
	_FLAGS_REPLACE+=("-fprofile-correction")
	_FLAGS_REPLACE+=("")
	_FLAGS_REPLACE+=("-L.")
	_FLAGS_REPLACE+=("")
	for _LTO_CFLAG in "${LTO_CFLAGS[@]}"; do
	_FLAGS_REPLACE+=(${_LTO_CFLAG})
	_FLAGS_REPLACE+=("")
	done
	fi
	SYSCONFIG=$(find ${_buildd_static}/$(cat ${_buildd_static}/pybuilddir.txt) -name "_sysconfigdata*.py" -print0)
	cat ${SYSCONFIG} | ${SYS_PYTHON} "${RECIPE_DIR}"/replace-word-pairs.py \
	"${_FLAGS_REPLACE[@]}" \
	> ${PREFIX}/lib/python${VERABI_NO_DBG}/$(basename ${SYSCONFIG})

[conda-forge-admin]

Hi! This is the friendly automated conda-forge-linting service.

I just wanted to let you know that I linted all conda-recipes in your PR (recipe/meta.yaml) and found it was in an excellent condition.

I do have some suggestions for making it better though...

For recipe/meta.yaml:

• ℹ️ The recipe is not parsable by parser conda-souschef (grayskull). This parser is not currently used by conda-forge, but may be in the future. We are collecting information to see which recipes are compatible with grayskull.
• ℹ️ The recipe is not parsable by parser conda-recipe-manager. The recipe can only be automatically migrated to the new v1 format if it is parseable by conda-recipe-manager.

_{This message was generated by GitHub Actions workflow run https://github.com/conda-forge/conda-forge-webservices/actions/runs/33570485145. Examine the logs at this URL for more detail.}

[h-vetinari]

6d161cc is a glimpse (though for 3.12) of how 636474b should have looked.

[h-vetinari]

@isuruf, this is an attempt at a minimal fix for the sysconfig data, which is not getting populated as before due to missing variables after splitting the build scripts (as an aside, we could add a check for undefined variables if you'd like).

However, we may want to improve the situation beyond the minimal fix. When everything was part of a single build_base.sh this wasn't an issue, but after the split it's IMO not very attractive to duplicate the LTO_CFLAGS & _FLAGS_REPLACE logic on both the build & install side. Perhaps we should move it to CBC?

[isuruf]

I think it's fine to duplicate for now

[itamarst]

I'm seeing errors compiling Python extensions because of gcc being called with a non-existent -partition=none argument.

It kinda seems like -flto-partition=none got turned into -partition=none in Python source code (e.g. in lib/python3.14/_sysconfigdata__linux_x86_64-linux-gnu.py).

Would that be fixed by this PR, or caused by this PR? This just started happening recently in CI, I reproduced locally using:

libpython                   3.14.7           hdc7f604_105_cp314    conda-forge
python                      3.14.7           hcd007b5_105_cp314    conda-forge

Looking at the PR it kinda seems like it might be caused by this but I don't understand enough bash so before I go further figured I'd ask.

[isuruf]

Fixed in #919