-
Notifications
You must be signed in to change notification settings - Fork 51
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Build failure with 2.3.10: prototypes.h not found #107
Comments
On 02/05/2022 15:11, Michael Vetter wrote:
With 2.3.10 I get:
|[ 33s] In file included from SQLBrowseConnect.c:14: [ 33s]
driver.h:34:10: fatal error: prototypes.h: No such file or directory [
33s] 34 | #include "prototypes.h" [ 33s] | ^~~~~~~~~~~~~~ [ 33s]
compilation terminated. [ 33s] make[3]: *** [Makefile:743:
SQLBrowseConnect.lo] Error 1 [ 33s] make[3]: Leaving directory
'/home/abuild/rpmbuild/BUILD/unixODBC-2.3.10/Drivers/template' [ 33s]
make[3]: Entering directory
'/home/abuild/rpmbuild/BUILD/unixODBC-2.3.10/Drivers/template' [ 33s]
/usr/bin/bash ../../libtool --tag=CC --mode=compile gcc
-DHAVE_CONFIG_H -I. -I../.. -I../../include -I. -O2 -Wall
-D_FORTIFY_SOURCE=2 -fstack-protector-strong -funwind-tables
-fasynchronous-unwind-tables -fstack-clash-protection
-Werror=return-type -flto=auto -g -pthread -c -o SQLAllocStmt.lo
SQLAllocStmt.c [ 33s] libtool: compile: gcc -DHAVE_CONFIG_H -I.
-I../.. -I../../include -I. -O2 -Wall -D_FORTIFY_SOURCE=2
-fstack-protector-strong -funwind-tables -fasynchronous-unwind-tables
-fstack-clash-protection -Werror=return-type -flto=auto -g -pthread -c
SQLAllocStmt.c -fPIC -DPIC -o .libs/SQLAllocStmt.o [ 33s] In file
included from SQLAllocStmt.c:14: [ 33s] driver.h:34:10: fatal error:
prototypes.h: No such file or directory [ 33s] 34 | #include
"prototypes.h" [ 33s] | ^~~~~~~~~~~~~~ [ 33s] compilation terminated.
[ 33s] make[3]: *** [Makefile:743: SQLAllocStmt.lo] Error 1 |
—
Reply to this email directly, view it on GitHub
<#107>, or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ABYK62NDBOHQBXLWKT7UWLDVH7PBZANCNFSM5U4DEF2Q>.
You are receiving this because you are subscribed to this
thread.Message ID: ***@***.***>
That what comes of not paying attention to changes.
|
I can confirm that dc160b3 fixes this. But now we have one problem :) It seems you took this commit and added it to the tarball at: ftp://ftp.unixodbc.org/pub/unixODBC/unixODBC-2.3.10.tar.gz Which will result in the autogenerated tarball of 2.3.10 that is available on GH https://github.com/lurcher/unixODBC/archive/refs/tags/2.3.10.tar.gz to have different content that the one on the official website. Now we could argue that the official website is the goto location for the tarball anyways. Which I can agree on. But still the git tag points to another commit thatn the tarball actually contains. Personally I don't care. And I think/hope other users can find this information here. |
On 02/05/2022 16:09, Michael Vetter wrote:
I can confirm that dc160b3
<dc160b3>
fixes this.
But now we have one problem :)
It seems you took this commit and added it to the tarball at:
ftp://ftp.unixodbc.org/pub/unixODBC/unixODBC-2.3.10.tar.gz
But the GitHub tag of 2.3.10 points to commit f9e77f8
<f9e77f8>
Which will result in the autogenerated tarball of 2.3.10 that is
available on GH
https://github.com/lurcher/unixODBC/archive/refs/tags/2.3.10.tar.gz to
have different content that the one on the official website.
Now we could argue that the official website is the goto location for
the tarball anyways. Which I can agree on.
But still the git tag points to another commit thatn the tarball
actually contains.
Personally I don't care. And I think/hope other users can find this
information here.
Just wanted to mention this :)
—
Reply to this email directly, view it on GitHub
<#107 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ABYK62MEBIVRQQQ3OQZGBRTVH7V23ANCNFSM5U4DEF2Q>.
You are receiving this because you commented.Message ID:
***@***.***>
Yep, I should have checked the sample drivers build, but forgot as I
just build the standard parts. Forgot they had been changed. If its a
problem I will have to do a new release, but will see how it plays.
|
Alright. I'll close this issue since the reported problem is fixed. |
Well, it didn't take long. The two versions of 2.3.10 has bitten
someone. Unless there are any suggestions I would think 2.3.11 is the
only way to fix it.
|
On 04/05/2022 15:06, Nick Gorham wrote:
Well, it didn't take long. The two versions of 2.3.10 has bitten
someone. Unless there are any suggestions I would think 2.3.11 is the
only way to fix it.
As I try to get the build that github creates to include config.h.in,
can someone explain why including that file is a bad idea and why its in
.gitignore?
|
I don't think you need to include that. You won't have it in the GitHub created tarball. If people decide to use the autogenerated tarball they need to run Some projects upload an additional upload (like you do on your website) that is created (like you did, with probably As an example see: https://github.com/profanity-im/profanity/releases/tag/0.12.1 |
I would say 37c45a6 is not correct. What exactly are you trying to solve? If the problem was that in the GH tarball there wasn't a config.h then my previous comment should explain this. |
On 04/05/2022 15:52, Michael Vetter wrote:
I would say 37c45a6
<37c45a6>
is not correct.
What exactly are you trying to solve? If the problem was that in the
GH tarball there wasn't a config.h then my previous comment should
explain this.
The problem I am trying to solve is that the GH tarball is missing
config.h.in
|
And why do you think it should include it? I think I tried to explain in #107 (comment) what's the correct way. Feel free to ask away. |
On 04/05/2022 15:55, Michael Vetter wrote:
And why do you think it should include it?
I think I tried to explain in #107 (comment)
<#107 (comment)>
what's the correct way.
Is there something unclear where I should expand the explanation?
Feel free to ask away.
—
Reply to this email directly, view it on GitHub
<#107 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ABYK62MLDQBBECZRO6LFEHDVIKFX3ANCNFSM5U4DEF2Q>.
You are receiving this because you commented.Message ID:
***@***.***>
Because (and yes, please explain what I am missing) when I download the
2.3.11-beta, rebuild the missing files using Makefile.svn then
config.h.in was missing. However I now see that your instruction of
|autoreconf -fi| fixes that.
So now I have 2.3.11-beta that contains the prototypes.h that was
missing in 2.3.10 from the other day and I could now create a 2.3.11
release, or I guess rewind and recreate 2.3.10 again, and have the
source on the website match the source in git. Which is all good.
So I guess I need to know why 37c45a6
<37c45a6>
is wrong and what you suggest to fix it, and then I will try and get it
all to line up.
It still doesn't address the email that I was trying to sort out where
the checksums were different, but as you have said that the GH tarball
is not the same as the release TGZ I create containing a built configure
then I would expect the checksums will not match anyway.
|
Honestly I don't quite remember why all the projects omit this file to be tracked in git. AFAIK the main (only?) reason is that it's generated from something (either during What you need to do in my opinion is the following:
the result will be that your tarball will be available on both your website and on the github release page and that their checksum will be the same. But the tarball that you uploaded will be highlighted and people will use that on. |
On 04/05/2022 16:16, Michael Vetter wrote:
|git tag 2.3.11; git push; git push --tags|
Ok, thanks for that. Hopefully we are good now.
|
Everything looks perfect! |
https://build.opensuse.org/request/show/974978 by user jubalh + dimstar_suse - Update to 2.3.11: * Add missing files to 2.3.10 - See lurcher/unixODBC#107
With 2.3.10 I get:
The text was updated successfully, but these errors were encountered: