Because the ifeq was evaluated during the first phase but the eval was deferred to the second.
https://www.gnu.org/software/make/manual/html_node/Reading-Makefiles.html ________________________________ From: Ed L Wolf <e.l.w...@cummins.com> Sent: Monday, September 25, 2023 03:32 To: Martin Dorey <martin.do...@hitachivantara.com>; bug-make@gnu.org <bug-make@gnu.org> Subject: RE: Problem setting a variable inside a target ***** EXTERNAL EMAIL ***** One more comment. For some reason SUPLLIER_A2l is blank Ed L Wolf Technical Advisor - Embedded Software e.l.w...@cummins.com Cummins Inc. Mail Code: C7004 1460 National Road Columbus, Indiana 47201 United States From: Ed L Wolf Sent: Monday, September 25, 2023 6:10 AM To: Martin Dorey <martin.do...@hitachivantara.com>; bug-make@gnu.org Subject: RE: Problem setting a variable inside a target The following code response was Supplier folder detected Supplier folder detected but no a2l files present but the response should have been Supplier folder detected Supplier a2l files found since the folder DIR_SUPPLIER did exist and DIR_A2LGEN_SUPPLIER did have a file called Rte.a2l Ed L Wolf Technical Advisor - Embedded Software e.l.w...@cummins.com<mailto:e.l.w...@cummins.com> Cummins Inc. Mail Code: C7004 1460 National Road Columbus, Indiana 47201 United States From: Martin Dorey <martin.do...@hitachivantara.com<mailto:martin.do...@hitachivantara.com>> Sent: Sunday, September 24, 2023 6:02 PM To: Ed L Wolf <e.l.w...@cummins.com<mailto:e.l.w...@cummins.com>>; bug-make@gnu.org<mailto:bug-make@gnu.org> Subject: Re: Problem setting a variable inside a target EXTERNAL SENDER: This email originated outside of Cummins. Do not click links or open attachments unless you verify the sender and know the content is safe. > The eval statement below highlighted in red is not working Perhaps you'd expect to see "BADGER is wombat" from: mad@shuttle:~/tmp/wolf-2023-09-24$ cat Makefile default: $(eval BADGER = wombat) ifeq ($(BADGER),wombat) echo BADGER is wombat else echo BADGER is not wombat endif mad@shuttle:~/tmp/wolf-2023-09-24$ make echo BADGER is not wombat BADGER is not wombat mad@shuttle:~/tmp/wolf-2023-09-24$ If that wasn't what you were trying to convey, then perhaps you could concoct a similarly small example, including what you typed, what the computer said in response and what your expectation was. If we look at the results of the first phase of Make described in: https://www.gnu.org/software/make/manual/html_node/Reading-Makefiles.html<https://nam04.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.gnu.org%2Fsoftware%2Fmake%2Fmanual%2Fhtml_node%2FReading-Makefiles.html&data=05%7C01%7CMartin.Dorey%40hitachivantara.com%7Ce8982d30476347672db208dbbdb2bebd%7C18791e1761594f52a8d4de814ca8284a%7C0%7C0%7C638312347658379205%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=lV3uCgBhF%2BT3fMjMkullyq2paKRJ8hRKvDigE4htKWo%3D&reserved=0> mad@shuttle:~/tmp/wolf-2023-09-24$ make -p nothing 2>&1 | grep -B2 -A1 BADGER # File has not been updated. # recipe to execute (from 'Makefile', line 2): $(eval BADGER = wombat) echo BADGER is not wombat mad@shuttle:~/tmp/wolf-2023-09-24$ ... then we see that the "BADGER is wombat" arm of the conditional is not even there to be run when it comes to executing the recipe in the second phase. The ifeq was "immediate" but the $(eval) was "deferred". It's a separate matter, but I fear your second red line might be expecting this behavior: mad@shuttle:~/tmp/wolf-2023-09-24$ echo *.a2l *.a2l mad@shuttle:~/tmp/wolf-2023-09-24$ ... which I doubt you're going to get from find(1): mad@shuttle:~/tmp/wolf-2023-09-24$ find -name "*.a2l" mad@shuttle:~/tmp/wolf-2023-09-24$ ________________________________ From: bug-make-bounces+martin.dorey=hds....@gnu.org<mailto:bug-make-bounces+martin.dorey=hds....@gnu.org> <bug-make-bounces+martin.dorey=hds....@gnu.org<mailto:bug-make-bounces+martin.dorey=hds....@gnu.org>> on behalf of Ed L Wolf <e.l.w...@cummins.com<mailto:e.l.w...@cummins.com>> Sent: Sunday, September 24, 2023 10:23 To: bug-make@gnu.org<mailto:bug-make@gnu.org> <bug-make@gnu.org<mailto:bug-make@gnu.org>> Subject: Problem setting a variable inside a target ***** EXTERNAL EMAIL ***** Hello I am having some trouble trying to set a variable inside a target. When reading the variable, it comes up blank. Below is my make file code. The eval statement below highlighted in red is not working. Can you provide some assistance please? I am using GNU Make 4.1 $(DIR_A2LGEN_SETUP)/McData-setup.a2l: $(BSW_A2LS) $(DIR_SCRIPTS)/update_mcdata_a2l.py $(DIR_SCRIPTS)/update_copyright_info.py $(DIR_SCRIPTS)/merge_a2l_files.py $(DIR_TARGET)/$(PROJECT).odx-f $(DIR_A2L)/Merger.ini @echo "A2L $@" @rm -rf $(DIR_A2LGEN) @rm -rf $(DIR_A2LLOGS) @rm -rf $(DIR_A2LGEN_SETUP) @rm -rf $(DIR_A2LGEN_UPDATE) @rm -rf $(DIR_A2LGEN_MODIFY) @rm -rf $(DIR_A2LGEN_CHECK) @rm -rf $(DIR_A2LGEN_SUPPLIER) @mkdir -p $(DIR_A2LGEN) @mkdir -p $(DIR_A2LLOGS) @mkdir -p $(DIR_A2LGEN_SETUP) @mkdir -p $(DIR_A2LGEN_UPDATE) @mkdir -p $(DIR_A2LGEN_MODIFY) @mkdir -p $(DIR_A2LGEN_CHECK) @mkdir -p $(DIR_A2LGEN_SUPPLIER) @find $(DIR_BSWPROJECT) -type f -name "*.a2l" | xargs -i cp {} $(DIR_A2LGEN_SETUP) #TODO: IS this still needed @cp $(DIR_A2L)/_Master*.a2l $(DIR_A2LGEN_SETUP) @python $(DIR_SCRIPTS)/update_mcdata_a2l.py $(DIR_A2LGEN_SETUP) @python $(DIR_SCRIPTS)/update_copyright_info.py \ -a $(DIR_A2LGEN_SETUP)/McData.a2l.patched \ -o $(DIR_A2LGEN_SETUP)/McData-copyright.a2l @python $(DIR_SCRIPTS)/merge_two_files.py \ -f1 $(CAN_A2L) \ -f2 $(DIR_A2LGEN_SETUP)/McData-copyright.a2l \ -sp '/end DAQ' \ -ep '/end IF_DATA' \ -op $(DIR_A2LGEN_SETUP)/McData-copyright-can.a2l # When there are no supplier a2l files then the merge process doesn't need to be done. ifneq ($(wildcard $(DIR_SUPPLIER)),) @find $(DIR_SUPPLIER) -type f -name "*.a2l" | xargs -i cp {} $(DIR_A2LGEN_SUPPLIER) $(eval SUPPLIER_A2l := $(sort $(shell find $(DIR_A2LGEN_SUPPLIER) -type f -name "*.a2l"))) @echo "A2L Supplier folder detected" @echo "Supplier a2l $(SUPPLIER_A2l)" # Just becuase there is a supplier folder does not mean it has a2l files. ifeq ($(SUPPLIER_A2l),"*.a2l") @echo "A2L Supplier a2l files found" @python $(DIR_SCRIPTS)/merge_a2l_files.py \ -s $(DIR_A2LGEN_SUPPLIER) \ -b a2l_file_list_supplier.opt \ -P $(DIR_A2L)/Merger.ini \ -M $(DIR_A2LGEN_SETUP)/McData-copyright-can.a2l \ -O $(DIR_A2LGEN_SETUP)/McData-setup.a2l \ -L $(DIR_A2LLOGS)/Log0-A2LMerger.log \ -g $(DIR_BUILDGEN) @ASAP2Merger.exe \ @$(DIR_BUILDGEN)/a2l_file_list_supplier.opt \ > $(subst $(DIR_A2LGEN_SETUP),$(DIR_A2LLOGS),$(@:.a2l=.log)) \ || echo "A2L $(DIR_A2LLOGS)/Log0-A2L-Supplier-Merger.log - Check for warnings/errors!" else @cp $(DIR_A2LGEN_SETUP)/McData-copyright-can.a2l $(DIR_A2LGEN_SETUP)/McData-setup.a2l @echo "A2L Supplier folder detected but no a2l files present" endif else @cp $(DIR_A2LGEN_SETUP)/McData-copyright-can.a2l $(DIR_A2LGEN_SETUP)/McData-setup.a2l @echo "A2L Supplier folder not available" endif Ed L Wolf Technical Advisor - Embedded Software e.l.w...@cummins.com<mailto:e.l.w...@cummins.com> Cummins Inc. Mail Code: C7004 1460 National Road Columbus, Indiana 47201 United States
