"podman cp" copies directories with a name ending with a "." differently

[eriksjolund]

Is this a BUG REPORT or FEATURE REQUEST? (leave only one on its own line)

/kind bug

Description

Steps to reproduce the issue:

On a Fedora 36 computer

1. Create a new directory and cd into it

   $ mkdir src
   $ cd src
   

2. In this directory create a file named Dockerfile with the file contents:

   FROM docker.io/library/alpine
   RUN mkdir -p /test/test1.
   RUN mkdir -p /test/test2
   RUN touch /test/test1./file1
   RUN touch /test/test2/file2
   

3. Create a file named /tmp/cp_test.sh with the file contents:

   #!/bin/bash
   
   engine=$1
   path=$2
   
   img=$($engine build -q --no-cache .)
   ctr=$($engine run --rm -d $img sleep inf)
   tmpdir=$(mktemp -d)
   cd $tmpdir
   $engine cp $ctr:$path dest
   find .
   $engine container rm -f $ctr > /dev/null
   

4. Run these commands

   $ bash /tmp/cp_test.sh podman /test/test1.
   .
   ./dest
   ./dest/test2
   ./dest/test2/file2
   ./dest/test1.
   ./dest/test1./file1
   WARN[0010] StopSignal SIGTERM failed to stop container eager_shockley in 10 seconds, resorting to SIGKILL 
   $ bash /tmp/cp_test.sh podman /test/test2
   .
   ./dest
   ./dest/file2
   WARN[0010] StopSignal SIGTERM failed to stop container quizzical_herschel in 10 seconds, resorting to SIGKILL
   $
   

On a macOS laptop having docker installed

1. Same as above
2. Same as above
3. Same as above
4. Run these commands

   $ bash /tmp/cp_test.sh docker /test/test1.
   .
   ./dest
   ./dest/file1
   $ bash /tmp/cp_test.sh docker /test/test2 
   .
   ./dest
   ./dest/file2
   $
   

Describe the results you received:

The commands

bash /tmp/cp_test.sh podman /test/test1.

and

bash /tmp/cp_test.sh docker /test/test1.

produce different output. I'm only referring to the paths being printed on the terminal as I guess this warning

    WARN[0010] StopSignal SIGTERM failed to stop container eager_shockley in 10 seconds, resorting to SIGKILL 

can be ignored.

Describe the results you expected:

The commands should produce identical output.

Additional information you deem important (e.g. issue happens only occasionally):

Output of podman version:

Client:       Podman Engine
Version:      4.2.1
API Version:  4.2.1
Go Version:   go1.18.5
Built:        Wed Sep  7 21:58:19 2022
OS/Arch:      linux/amd64

Output of podman info:

host:
  arch: amd64
  buildahVersion: 1.27.0
  cgroupControllers:
  - cpu
  - io
  - memory
  - pids
  cgroupManager: systemd
  cgroupVersion: v2
  conmon:
    package: conmon-2.1.4-3.fc36.x86_64
    path: /usr/bin/conmon
    version: 'conmon version 2.1.4, commit: '
  cpuUtilization:
    idlePercent: 99.95
    systemPercent: 0.03
    userPercent: 0.02
  cpus: 16
  distribution:
    distribution: fedora
    variant: workstation
    version: "36"
  eventLogger: journald
  hostname: asus
  idMappings:
    gidmap:
    - container_id: 0
      host_id: 1000
      size: 1
    - container_id: 1
      host_id: 100000
      size: 65536
    uidmap:
    - container_id: 0
      host_id: 1000
      size: 1
    - container_id: 1
      host_id: 100000
      size: 65536
  kernel: 6.0.5-200.fc36.x86_64
  linkmode: dynamic
  logDriver: journald
  memFree: 366977024
  memTotal: 7691595776
  networkBackend: netavark
  ociRuntime:
    name: crun
    package: crun-1.6-2.fc36.x86_64
    path: /usr/bin/crun
    version: |-
      crun version 1.6
      commit: 18cf2efbb8feb2b2f20e316520e0fd0b6c41ef4d
      spec: 1.0.0
      +SYSTEMD +SELINUX +APPARMOR +CAP +SECCOMP +EBPF +CRIU +YAJL
  os: linux
  remoteSocket:
    path: /run/user/1000/podman/podman.sock
  security:
    apparmorEnabled: false
    capabilities: CAP_CHOWN,CAP_DAC_OVERRIDE,CAP_FOWNER,CAP_FSETID,CAP_KILL,CAP_NET_BIND_SERVICE,CAP_SETFCAP,CAP_SETGID,CAP_SETPCAP,CAP_SETUID,CAP_SYS_CHROOT
    rootless: true
    seccompEnabled: true
    seccompProfilePath: /usr/share/containers/seccomp.json
    selinuxEnabled: true
  serviceIsRemote: false
  slirp4netns:
    executable: /usr/bin/slirp4netns
    package: slirp4netns-1.2.0-0.2.beta.0.fc36.x86_64
    version: |-
      slirp4netns version 1.2.0-beta.0
      commit: 477db14a24ff1a3de3a705e51ca2c4c1fe3dda64
      libslirp: 4.6.1
      SLIRP_CONFIG_VERSION_MAX: 3
      libseccomp: 2.5.3
  swapFree: 5284818944
  swapTotal: 7691300864
  uptime: 49h 40m 12.00s (Approximately 2.04 days)
plugins:
  authorization: null
  log:
  - k8s-file
  - none
  - passthrough
  - journald
  network:
  - bridge
  - macvlan
  volume:
  - local
registries:
  search:
  - registry.fedoraproject.org
  - registry.access.redhat.com
  - docker.io
  - quay.io
store:
  configFile: /home/esjolund/.config/containers/storage.conf
  containerStore:
    number: 30
    paused: 0
    running: 0
    stopped: 30
  graphDriverName: overlay
  graphOptions: {}
  graphRoot: /home/esjolund/.local/share/containers/storage
  graphRootAllocated: 407822663680
  graphRootUsed: 355461595136
  graphStatus:
    Backing Filesystem: xfs
    Native Overlay Diff: "true"
    Supports d_type: "true"
    Using metacopy: "false"
  imageCopyTmpDir: /var/tmp
  imageStore:
    number: 147
  runRoot: /run/user/1000/containers
  volumePath: /home/esjolund/.local/share/containers/storage/volumes
version:
  APIVersion: 4.2.1
  Built: 1662580699
  BuiltTime: Wed Sep  7 21:58:19 2022
  GitCommit: ""
  GoVersion: go1.18.5
  Os: linux
  OsArch: linux/amd64
  Version: 4.2.1

Package info (e.g. output of rpm -q podman or apt list podman or brew info podman):

podman-4.2.1-2.fc36.x86_64

Have you tested with the latest version of Podman and have you checked the Podman Troubleshooting Guide? (https://github.com/containers/podman/blob/main/troubleshooting.md)

Yes

Additional environment details (AWS, VirtualBox, physical, etc.):
The test with Podman was done on a x86_64 computer (Fedora 36).

The test with Docker was done on a macOS laptop with macOS Monterey 12.6.1
and the docker version:

$ docker --version
Docker version 20.10.17, build 100c701

[mheon]

@vrothberg PTAL

[vrothberg]

I am flooded at the moment. If others have cycles, please take a look.

[vrothberg]

Found some time and the bug. Will wrap up fixes. Thanks for the great reproducer, @eriksjolund !