docker: revise setup to automatically determine user and group id

Author: AlexanderSchultheiss

Dockerfile

@@ -14,8 +14,12 @@ RUN mvn package || exit
 
 FROM openjdk:19-alpine
 
+ARG GROUP_ID
+ARG USER_ID
+
 # Create a user
-RUN adduser --disabled-password  --home /home/user --gecos '' user
+RUN addgroup -g $GROUP_ID user
+RUN adduser --disabled-password -G user -u $USER_ID --home /home/user --gecos '' user
 
 # Prepare the environment
 RUN apk update
@@ -69,6 +73,4 @@ RUN chmod +x entrypoint.sh
 RUN chmod +x fix-perms.sh
 RUN chmod +x extract.sh
 
-ENTRYPOINT ["./entrypoint.sh", "./extract.sh"]
-# calculon user
-USER 1015
+ENTRYPOINT ["./entrypoint.sh", "./extract.sh"]

build-docker-image.bat

@@ -1,2 +1,2 @@
-docker build -t extraction .
+docker build --build-arg USER_ID=1000 --build-arg GROUP_ID=1000 -t extraction .
 @pause

build-docker-image.sh

@@ -1,2 +1,3 @@
 #! /bin/bash
-docker build -t extraction .
+docker build --build-arg USER_ID=$(id -u) --build-arg GROUP_ID=$(id -g) -t extraction .
+

docker-resources/entrypoint.sh

@@ -2,6 +2,7 @@
 if [ "$(id -u)" = "0" ]; then
   # running on a developer laptop as root
   ls -l
+  echo "adjusting permissions for user"
   ./fix-perms.sh -r -u user -g user /home/user
   exec gosu user "$@"
 else

docker-resources/verification.md

@@ -1,8 +1,3 @@
 Project name       | Domain                  | Source code available (**y**es/**n**o)? | Is it a git repository (**y**es/**n**o)? | Repository URL                                               | Clone URL                                          | Estimated number of commits
 -------------------|-------------------------|-----------------------------------------|------------------------------------------|--------------------------------------------------------------|----------------------------------------------------|-----------------------------
 busybox            | embedded systems        | y                                       | y                                        | https://git.busybox.net/busybox                              | https://github.com/DiffDetective/busybox           | 17,447
-lynx               | web browser             | y                                       | y                                        | https://github.com/lynx/lynx                                 | https://github.com/DiffDetective/lynx.git          | 125
-sendmail           | mail transfer agent     | y                                       | y                                        | https://github.com/guileen/node-sendmail                     | https://github.com/DiffDetective/node-sendmail.git | 86
-xfig               | vector graphics editor  | y                                       | y                                        | https://github.com/hhoeflin/xfig                             | https://github.com/DiffDetective/xfig.git          | 9
-xine-lib           | media library           | y                                       | y                                        | https://github.com/rpmfusion/xine-lib                        | https://github.com/DiffDetective/xine-lib.git      | 114
-xterm              | terminal emulator       | y                                       | y                                        | https://github.com/Maximus5/xterm                            | https://github.com/DiffDetective/xterm.git         | 112

docker-resources/verification.properties

@@ -13,8 +13,8 @@ extraction.extract-code-matching=false
 # Path in which diff detective keeps its intermediate results. No need to change this.
 diff-detective.output-dir=ground-truth/dd
 # Path in which diff detective keeps the repos. No need to change this.
-diff-detective.repo-storage-dir=DiffDetectiveMining
+diff-detective.repo-storage-dir=ground-truth/REPOS
 # Path to which the ground truth is saved. Do NOT change this without knowing how this affects the Docker file system interface
 extraction.gt-save-dir=ground-truth
 # Number of commits to process in a single batch by one thread
-diff-detective.batch-size=64
+diff-detective.batch-size=8