-Werror breaks the build process

[gvanuxem]

Apparently, it is the first time I compile Aldor, by default -Werror is used as STRICTCFLAGS or something like that. I had to remove it from config.status to build Aldor since there are some warnings during Aldor compilation.

What I did:

git clone --depth=1 https://github.com/aldorlang/aldor.git
cd aldor/aldor
./autogen.sh
./configure
make

And the build process stops with:

../../../aldor/aldor/src/macex.c: In function 'pushMacDef':
../../../aldor/aldor/src/macex.c:184:32: error: the comparison will always evaluate as 'true' for the address of 'val' will never be NULL [-Werror=address]
  184 |         if (! symInfo(name) || ! symCoInfo(name))
      |                                ^
In file included from ../../../aldor/aldor/src/macex.c:30:
../../../aldor/aldor/src/symcoinfo.h:19:11: note: 'val' declared here
   19 |         } val;
      |           ^~~
cc1: all warnings being treated as errors
make[4]: *** [Makefile:1729: macex.o] Error 1
make[4]: *** Waiting for unfinished jobs....

Regards

[gvanuxem]

Modifying debian/rules to add --disable-error-on-warning after --enable-silent-rules (dh_auto_configure) solved the problem but a spurious file was missing with debuild -b. So,

cd aldor && ./autogen.sh && ./configure --disable-error-on-warning

does the trick.

[pbroadbery]

What compiler version are you using?
The error looks to be due to a new version doing more checks.

I want to keep the error on warn as a default - warnings do in general indicate issues, but we do ignore a subset.
The build instructions should mention disable-error-on-warning I guess.

Ignored warnings are listed in aldor/m4/strict_compile.m4.. Adding '-Wno-error=address' will work around your issue, but this one looks to be worth investigating.

Peter
(at work, so I can't look at this too deeply at the moment)

[gvanuxem]

gcc (Debian 12.2.0-14) 12.2.0 Kali Linux on WSL2. Le ven. 2 juin 2023, 11:15, pbroadbery ***@***.***> a écrit :

…

What compiler version are you using? The error looks to be due to a new version doing more checks. I want to keep the error on warn as a default - warnings do in general indicate issues, but we do ignore a subset. The build instructions should mention disable-error-on-warning I guess. Ignored warnings are listed in aldor/m4/strict_compile.m4.. Adding '-Wno-error=address' will work around your issue, but this one looks to be worth investigating. Peter (at work, so I can't look at this too deeply at the moment) — Reply to this email directly, view it on GitHub <#158 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AKGLAMCSI4C35XKUXH45HD3XJGVKFANCNFSM6AAAAAAYXPZ3KI> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[gvanuxem]

Hello, Just Git cloned Aldor and FriCAS and all seems fine for me. FriCAS with Aldor tests also pass. Aldor is now even buildable on a Debian-like system with 'dpkg-buildpackage -b', many thanks for your work! - Greg Le dim. 24 nov. 2024 à 16:12, pbroadbery ***@***.***> a écrit :

…

I think I've fixed it with the latest couple of commits - could you give it a test? — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: ***@***.***>

[pbroadbery]

Hi, Thanks for testing it - it's reassuring that it works in a couple of environments. I've a few more changes in the pipeline and hoping to get them released soon. Peter

…

On Wed, 27 Nov 2024, 11:00 Grégory Vanuxem, ***@***.***> wrote: Hello, Just Git cloned Aldor and FriCAS and all seems fine for me. FriCAS with Aldor tests also pass. Aldor is now even buildable on a Debian-like system with 'dpkg-buildpackage -b', many thanks for your work! - Greg Le dim. 24 nov. 2024 à 16:12, pbroadbery ***@***.***> a écrit : > > I think I've fixed it with the latest couple of commits - could you give it a test? > > — > Reply to this email directly, view it on GitHub, or unsubscribe. > You are receiving this because you authored the thread.Message ID: ***@***.***> — Reply to this email directly, view it on GitHub <#158 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AABWHLMF7LV4XZY6YLKHZ2D2CWQ3NAVCNFSM6AAAAABSMKK5JGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDKMBTGU3TCNJTHE> . You are receiving this because you commented.Message ID: ***@***.***>