Re: [Rpm-maint] [rpm-software-management/rpm] RFE: %exclude should not permit files to bypass check-files and be omitted from all packages built from spec (#994)

2020-01-20 Thread Kevin Kofler 
The current behavior of `%exclude` is a feature and should not be incompatibly 
changed.

-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/rpm-software-management/rpm/issues/994#issuecomment-576466597___
Rpm-maint mailing list
Rpm-maint@lists.rpm.org
http://lists.rpm.org/mailman/listinfo/rpm-maint

Re: [Rpm-maint] [rpm-software-management/rpm] Fix compatible-release, versions with dot-star, semantically versioned deps, and python version (#951)

2020-01-20 Thread Miro Hrončok 
`.*` is still broken with `!=`:


python-kubernetes:

python-10.0.1/requirements.txt
`websocket-client>=0.32.0,!=0.40.0,!=0.41.*,!=0.42.* `

`error: Illegal char '*' (0x2a) in: 0.42.*`


`.*` is still broken with `>=`:

python-fsleyes:

fsleyes-0.32.0/fsleyes.egg-info/requires.txt
`nibabel>=2.3.*`

`error: Illegal char '*' (0x2a) in: 2.3.*`


The code explicitly only work with `== ...*` but not other specs:

```python
elif spec[0] == '==' and spec[1].endswith('.*'):
```

-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/rpm-software-management/rpm/pull/951#issuecomment-576382832___
Rpm-maint mailing list
Rpm-maint@lists.rpm.org
http://lists.rpm.org/mailman/listinfo/rpm-maint

Re: [Rpm-maint] [rpm-software-management/rpm] %define macro value is not evaluated on rpm 4.11.3 (#1013)

2020-01-20 Thread Jun Aruga 
## How to reproduce this issue on your environment

1. Fork the repository: https://github.com/junaruga/report-rpm-macro-define
2. Enable Travis CI for the repository.
3. Check the Travis CI result modifying the `foo.spec` file.

-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/rpm-software-management/rpm/issues/1013#issuecomment-576319148___
Rpm-maint mailing list
Rpm-maint@lists.rpm.org
http://lists.rpm.org/mailman/listinfo/rpm-maint

[Rpm-maint] [rpm-software-management/rpm] %define macro value is not evaluated on rpm 4.11.3 (#1013)

2020-01-20 Thread Jun Aruga 
## Sumamry

`%define` macro is not evaluated, the value is not set when building a RPM spec 
file with rpm-4.11.3-40 and `%bcond_without git` is executed.

## Context

I was trying to build a `ruby.spec` file on RHEL7, that is based on the Fedora 
rawhide current latest `ruby.spec` 
[URL](https://src.fedoraproject.org/rpms/ruby/blob/345ce0e5b8a3a6d9c962807da12e80bdaee9fb90/f/ruby.spec#_25).

But the built RPM was `ruby-2.7.0-%{release_string}.x86_64.rpm`. The 
`%{release_string}` value was not set. On Fedora rawhide, the value was set.

The `ruby.spec` code is like this.

```
...
%global release 126
%{!?release_string:%define release_string 
%{?development_release:0.}%{release}%{?development_release:.%{development_release}}%{?dist}}
...
%bcond_without git
...
Release: %{release_string}
...
```

So, I prepared the simple example to reproduce this issue, with GitHub and 
Travis CI.

Here is [the reproducing 
example](https://github.com/junaruga/report-rpm-macro-define), and the [Travis 
CI 
result](https://travis-ci.org/junaruga/report-rpm-macro-define/builds/639514301).

Using very simple RPM spec file 
[foo.spec](https://github.com/junaruga/report-rpm-macro-define/blob/master/foo.spec),
 and building it by `rpmbuild -ba foo.spec`.

Here is the result. See above Travis CI result.

1. fedora:rawhide / rpm-4.15.1-2.fc32.x86_64 : OK
2. fedora:31 / rpm-4.15.1-1.fc31.x86_64 : OK
3. centos:8.1 / rpm-4.14.2-25.el8.x86_64 : OK
4. centos:7.7 / rpm-4.11.3-40.el7.x86_64 : NG
5. centos:7.7 / rpm-4.11.3-40.el7.x86_64 does not execute `%bcond_without git`: 
interestingly this case is OK

## Result

### OK case

https://travis-ci.org/junaruga/report-rpm-macro-define/jobs/639514306#L510

```
+ echo 'release_string: 2.el7.el7'
release_string: 2.el7.el7
+ echo 'bbb: 2'
bbb: 2
+ echo 'ddd: 2'
ddd: 2
+ echo 'fff: 2'
fff: 2
```

https://travis-ci.org/junaruga/report-rpm-macro-define/jobs/639514306#L556

```
+ find /root/rpmbuild/ -type f
/root/rpmbuild/SRPMS/foo-2.7.0-2.el7.src.rpm
/root/rpmbuild/RPMS/x86_64/foo-2.7.0-2.el7.x86_64.rpm
/root/rpmbuild/BUILD/foo
```

### NG case

https://travis-ci.org/junaruga/report-rpm-macro-define/jobs/639514305#L506

```
+ echo 'release_string: %{release_string}'
release_string: %{release_string}
+ echo 'bbb: %{bbb}'
bbb: %{bbb}
+ echo 'ddd: 2'
ddd: 2
+ echo 'fff: %{fff}'
fff: %{fff}
```

https://travis-ci.org/junaruga/report-rpm-macro-define/jobs/639514305#L552

```
+ find /root/rpmbuild/ -type f
/root/rpmbuild/SRPMS/foo-2.7.0-%{release_string}.src.rpm
/root/rpmbuild/RPMS/x86_64/foo-2.7.0-%{release_string}.x86_64.rpm
/root/rpmbuild/BUILD/foo
```

## Questions

1. Why this issue happens on rpm 4.11.3 when only `%bcond_without git` is 
executed?
2. When Fedora's `ruby.spec` has to consider backward ccompatibility on 
CentOS7, what is the best pracitce dealing with `%define` macro? We should not 
use `%define` macro?

Thank you.

-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/rpm-software-management/rpm/issues/1013___
Rpm-maint mailing list
Rpm-maint@lists.rpm.org
http://lists.rpm.org/mailman/listinfo/rpm-maint