V2 rewrite

.github/workflows/check-style.yml

@@ -2,18 +2,26 @@ name: Check style
 on: [push, pull_request]
 jobs:
   style:
+    defaults:
+      run:
+        shell: bash -l {0}
     runs-on: ubuntu-latest
     steps:
     - uses: actions/checkout@v3
     - uses: psf/black@stable
-    - name: Set up Python ${{ matrix.python-version }}
-      uses: actions/setup-python@v3
+    - name: Set up Python 3.8
+      uses: actions/setup-python@v4
       with:
-        python-version: ${{ matrix.python-version }}
-    - name: Install Python dependencies
-      run: |
-        python -m pip install --upgrade pip
-        pip install -r requirements.txt
+        python-version: '3.8'
+    - name: Setup miniconda
+      uses: conda-incubator/setup-miniconda@v2
+      with:
+        python-version: 3.8
+        auto-update-conda: true
+        channels: conda-forge,defaults
+        mamba-version: "*"
+        activate-environment: AST
+        environment-file: environment.yml
     - name: Install shellcheck and checkbashisms
       run: sudo apt install shellcheck devscripts
     - name: Check style

.github/workflows/small-test.yml

@@ -8,7 +8,7 @@ jobs:
     runs-on: ubuntu-latest
     defaults:
       run:
-        shell: bash
+        shell: bash -l {0}
     steps:
     - uses: actions/setup-java@v3
       with:
@@ -30,18 +30,21 @@ jobs:
     - run: echo "${GITHUB_WORKSPACE}/src/scripts/merge_tools" >> $GITHUB_PATH
     - run: java -version
     - uses: actions/checkout@v3
-    - uses: fregante/setup-git-user@v2
-    - name: Set up Python ${{ matrix.python-version }}
-      uses: actions/setup-python@v3
+    - name: Set up Python 3.8
+      uses: actions/setup-python@v4
       with:
-        python-version: ${{ matrix.python-version }}
-    - run: python --version
+        python-version: '3.8'
+    - name: Setup miniconda
+      uses: conda-incubator/setup-miniconda@v2
+      with:
+        python-version: 3.8
+        auto-update-conda: true
+        channels: conda-forge,defaults
+        mamba-version: "*"
+        activate-environment: AST
+        environment-file: environment.yml
     - name: Install PdfLaTeX
       run: sudo apt update && sudo apt install texlive-latex-extra -y
-    - name: Install dependencies
-      run: |
-        python -m pip install --upgrade pip
-        pip install -r requirements.txt
     - name: Install maven
       uses: s4u/[email protected]
       with:

.gitignore

@@ -2,14 +2,15 @@
 .vscode
 venv
 **/*_explanation.txt
-test_cache/
+cache-small/
 apache-maven-3.9.2/
 results/local_repos.csv
 machines.txt
 .valid_merges
 cache/
 artifacts/
 artifacts.tar.gz
+*.hprof
 
 output/
 merge_repo/
@@ -108,3 +109,5 @@ docs/_build/
 
 # Emacs
 TAGS
+
+!JSCover/**

Makefile

@@ -29,14 +29,15 @@ clean:
 	rm -rf repos
 	rm -rf scratch
 	rm -rf results-small
+	rm -rf .valid_merges_counters
 
 # This target deletes files in the cache, which is commited to version control.
 clean-cache:
 	rm -rf cache
 
 # This target deletes files in the test cache.
 clean-test-cache:
-	rm -rf test_cache
+	rm -rf cache-small
 
 # This target deletes files that are committed to version control.
 clean-stored-hashes:
@@ -49,7 +50,7 @@ clean-everything: clean clean-cache clean-test-cache clean-stored-hashes
 # Compresses the cache.
 compress-cache:
 	rm -r cache.tar
-	tar --exclude="*explanation.txt" -czf cache.tar cache
+	tar --exclude="lock" -czf cache.tar cache
 
 # Decompresses the cache.
 decompress-cache:
@@ -74,21 +75,17 @@ update-cache-results:
 # As of 2023-07-31, this takes 5-20 minutes to run, depending on your machine.
 small-test:
 	${MAKE} clean-test-cache clean
-	./run_small.sh -d
+	./run_small.sh --include_trivial_merges
 	${MAKE} small-test-diff
 
 small-test-diff:
 	@echo
 	@echo "Here is the file content, in case a diff fails."
-	more results-small/*.csv results-small/merges/*.csv results-small/merges_valid/*.csv | cat
+	more results-small/*.csv results-small/merges/*.csv results-small/merges_compared/*.csv results-small/merges_tested/*.csv | cat
 	@echo
-	if grep -Fqvf results-small/merges/ez-vcard.csv test/small-goal-files/merges/ez-vcard.csv; then exit 1; fi
-	if grep -Fqvf results-small/merges/Algorithms.csv test/small-goal-files/merges/Algorithms.csv; then exit 1; fi
-	python3 test/remove-run_time-columns.py --input results-small/result.csv --output results-small/result-without-times.csv
-	python3 test/remove-run_time-columns.py --input results-small/filtered_result.csv --output results-small/filtered_result-without-times.csv
+	python3 test/check_equal_csv.py --actual_folder results-small/ --goal_folder test/small-goal-files/
 	@echo
-	diff -x tools -x defs.tex -x git -x merges -x .gitignore -x git -x result.csv -x plots -x filtered_result.csv -x table_run_time.tex -x .DS_Store -x '*~' -x '#*#' -r -U3 test/small-goal-files results-small
-	rm -f test/small-goal-files/result-without-times.txt results-small/result-without-times.txt
+	diff -x table_run_time.tex -r -U3 test/small-goal-files/tables/all test/small-goal-files/tables/all
 
 gradle-assemble:
 	./gradlew assemble -g ../.gradle/

README.md

@@ -8,24 +8,22 @@
 
 ### Python
 
-To install all the python requirements:
+To install all the python requirements create a conda environment:
+
+With mamba (faster https://github.com/mamba-org/mamba):
 
 ```bash
-pip install -r requirements.txt
+mamba env create -f environment.yml
+mamba activate AST
 ```
 
-### Alternative Python installation
-
-If you don't want to mess with your local python installation you can create a python virtual environment to install all dependencies with the following commands:
+With conda:
 
 ```bash
-pip3 install virtualenv
-python3 -m venv venv
-source venv/bin/activate
+conda env create -f environment.yml
+conda activate AST
 ```
 
-If you did the previous step make sure the virtual environemnt is activated when you use the repo (`source venv/bin/activate`).
-
 ### Maven
 
 Make sure you use maven version 3.9.*.
@@ -68,9 +66,16 @@ make small-test
 
 This runs the entire code on two small repos.
 The output data appears in `results-small/`.
- * `results-small/result.csv`: the final result
- * `results-small/merges_small/` contains all the merges.
- * `results-small/merges_small_valid/` contains all the merges and also records whether the parents of a merge pass tests.
+
+* `results-small/result.csv`: the final result
+
+* `results-small/merges/` contains all the merges.
+
+* `results-small/merges_compared/` contains all merges and indicates whether the merge results are different and thus need to be analyzed.
+
+* `results-small/merges_tested/` contains all merges that have been tested.
+
+* `results-small/result.csv` contains the final result.
 
 ### Perform full analysis
 
@@ -90,13 +95,7 @@ This will run the entire code on all the repos and automatically decompress the
 All the output data can be found in `results/`.
 The final result is found in `results/result.csv`.
 Directory `results/merges` contains all the merges for each repo.
-Directory `results/merges_valid` contains all the merges and also stores if the parents of a merge pass tests.
-
-To delete cache entries on failed merges, inconsistent merges, failed trivial merges and reexecute the stack multiple times over and over:
-
-```bash
-./run_full_restart.sh <n_repeat>
-```
+Directory `results/merges_tested` contains all the merges that have been tested.
 
 To execute `run_full.sh` on multiple machines in parallel create a machine address list in `machines.txt` and run:
 
@@ -126,81 +125,82 @@ To run style checking run `make style`.
 
 ---
 
+## Code structure
+
+![alt text](illustrations/Architecture.drawio.png "Title")
+
 ## Directory structure
 
-### Commited files
+### Committed files
 
 * run.sh -> This file executes each step of the stack.
 
 * run_small.sh -> This file executes the stack on two repositories.
 
 * run_full.sh -> This file executes the stack on all the repositories.
 
-* run_full_restart.sh -> This file executes the stack and repeats failed merges, inconsistent merges and failed trivial multiple times.
-
 * src/ -> contains the following scripts:
 
   * python/ -> contains the following scripts:
 
     * merge_tester.py -> Main file which performs merges and evaluates all the results across all projects.
 
-    * validate_repos.py -> Checks out all repos and removes all repos that fail their tests on main branch.
+    * test_repo_heads.py -> Checks out all repos and removes all repos that fail their tests on main branch.
 
     * latex_output.py -> Output latex code for the resulting plots and table.
 
-    * test_parent_commits.py -> Tests if the parents of a commit pass their tests.
+    * merge_tools_comparator.py -> Compares merges that produce different output.
 
     * get_repos.py -> Downloads the repos list.
 
-    * cache_merger.py -> Merges the current cache with the cache.tar
+    * cache_utils.py -> Contains functions to store and load the cache.
 
-    * delete_cache_entries.py -> Delete specific cache entries.
+    * clean_cache_placeholders.py -> Removes all the cache placeholders.
 
-    * delete_inconsistent_merge_results.py -> Delete inconsistent merge results.
+    * repo.py -> Contains the Repo class which represents a repo.
 
-    * delete_failed_trivial_merge_results.py -> Delete failed trivial merge results.
+    * split_repos.py -> Splits the repos for parallel execution.
+
+    * write_head_hashes.py -> Writes the head hashes of all repos to a file.
 
   * scripts/ -> contains the following scripts:
 
     * run_repo_tests.sh -> Runs a repo's programmer provided tests.
 
-    * merge_tools/ -> contains the following scripts:
-      * gitmerge.sh -> Executes git merge on a specific merge.
-      * intellimerge.sh -> Executes intellimerge on a specific merge.
-      * spork.sh -> Executes spork on a specific merge.
+    * merge_tools/ -> Contains all the merge tools scripts.
 
     * utils/
+
       * run_remotely.sh -> Runs the full stack on a remote machine.
+
       * run_multiple_machine.sh -> Runs the full stack on multiple remote machines.
 
   * src/main/java/astmergeevaluation/FindMergeCommits.java -> Finds all merge commits in a repo.
 
 * input_data/ -> Input data, which is a list of repositories; see its README.md.
 
-### Uncommited Files
+### Uncommitted Files
 
 * cache/ -> This folder is a cache for each computation. contains:
 
   * test_result/ -> Caches the test results for a specific commit. Used for parent testing and repo validation.
 
-  * merge_test_results/ -> Caches the test results for specific merges. Used for merge testing. First line indicates the merge result, second line indicates the runtime.
+  * merge_test_results/ -> Caches the test results for specific merges. Used for merge testing. First line indicates the merge result, second line indicates the run time.
 
   * merge_diff_results/ -> Caches the diff results for specific merges.
 
-* test_cache/ -> This folder is a cache for each test computation. contains:
+* cache-small/ -> This folder is a cache for each test computation. contains:
 
   * test_result/ -> Caches the test results for a specific commit. Used for parent testing and repo validation.
 
-  * merge_test_results/ -> Caches the test results for specific merges. Used for merge testing. First line indicates the merge result, second line indicates the runtime.
+  * merge_test_results/ -> Caches the test results for specific merges. Used for merge testing. First line indicates the merge result, second line indicates the run time.
 
-* .workdir/ -> This folder is used for the local computations of each process and contaent is named by Unix process (using "$$").
+* .workdir/ -> This folder is used for the local computations of each process and content is named by Unix process (using "$$"). If `DELETE_WORKDIRS` is set to `false` in `src/python/repo.py` this folder is not deleted after the computation and can be inspectedx.
 
 * repos/ -> In this folder each repo is cloned.
 
 * results/ -> Contains all the results for the full analysis.
 
 * results-small/ -> Contains all the results for the small analysis.
 
-* jars/ -> Location for the Intellimerge and Spork jars.
-
-* scratch/ -> If STORE_SCRATCH is enabled in `merge_tester.py`, each merge will be stored in this location.
+* jars/ -> Location for the IntelliMerge and Spork jars.

environment.yml

@@ -0,0 +1,33 @@
+name: AST
+channels:
+  - defaults
+dependencies:
+  - bzip2
+  - ca-certificates
+  - ld_impl_linux-64
+  - libffi
+  - libuuid
+  - ncurses
+  - openssl
+  - pip
+  - readline
+  - setuptools
+  - sqlite
+  - tk
+  - tzdata
+  - wheel
+  - xz
+  - zlib
+  - python=3.8
+  - pip:
+      - fasteners==0.18
+      - matplotlib==3.6.3
+      - numpy==1.23.5
+      - pandas==2.0.2
+      - pylint==2.17.5
+      - GitPython==3.1.31
+      - prettytable==3.8.0
+      - seaborn==0.12.2
+      - tqdm==4.64.1
+      - black==23.3.0
+      - psutil==5.9.5

input_data/README.md

@@ -1,9 +1,9 @@
 This directory contains the input data, which is a list of repositories.
 To refresh the contents of this directory, run script `get_repos.py`.
 
-  * repos.csv.gz -> The reaper dataset.  (This is not committed to this repository; it is downloaded by `get_repos.py`.)
+  * repos.csv.gz -> The Reaper dataset.  (This is not committed to this repository; it is downloaded by `get_repos.py`.)
 
-  * repos.csv -> List of all repos that fulfill the reaper selection criterion.
+  * repos.csv -> List of all repos that fulfill the Reaper selection criterion.
 
   * repos_with_hashes.csv -> repos.csv but with the tested commit for repository validation
 

input_data/repos.csv

@@ -1,4 +1,4 @@
-,repository,language,architecture,community,continuous_integration,documentation,history,issues,license,size,unit_test,stars,scorebased_org,randomforest_org,scorebased_utl,randomforest_utl
+idx,repository,language,architecture,community,continuous_integration,documentation,history,issues,license,size,unit_test,stars,scorebased_org,randomforest_org,scorebased_utl,randomforest_utl
 329,Complexible/stardog-spring,Java,1.0,1,0,0.36999,0.0,0.0,1,2153,0.403618,12,0,0,1,1
 1502,sdeo/protobuf-socket-rpc,Java,1.0,1,0,0.233562,0.0,0.0,1,7472,0.323622,18,0,0,1,1
 2865,twitter/netty-http2,Java,1.0,2,1,0.22231,0.368421,0.0,1,5886,0.341296,41,1,0,1,1

input_data/repos_small.csv

@@ -1,4 +1,4 @@
-,repository,language,architecture,community,continuous_integration,documentation,history,issues,license,size,unit_test,stars,scorebased_org,randomforest_org,scorebased_utl,randomforest_utl
+idx,repository,language,architecture,community,continuous_integration,documentation,history,issues,license,size,unit_test,stars,scorebased_org,randomforest_org,scorebased_utl,randomforest_utl
 16594,mangstadt/ez-vcard,Java,1.0,1,0,0.433617,10.5,0.0,1,34146,0.448443,54,1,0,1,1
 296895,pedrovgs/Algorithms,Java,0.932927,1,1,0.390942,89.0,0.75,1,6616,0.556371,1377,1,1,1,1
 1624004,tntim96/JSCover,Java,0.985816,1,0,0.490264,25.0,6.448276,1,54612,0.627527,298,1,1,1,1

input_data/repos_small_with_hashes.csv

@@ -1,4 +1,4 @@
-idx,repository,language,architecture,community,continuous_integration,documentation,history,issues,license,size,unit_test,stars,scorebased_org,randomforest_org,scorebased_utl,randomforest_utl,Validation hash
+idx,repository,language,architecture,community,continuous_integration,documentation,history,issues,license,size,unit_test,stars,scorebased_org,randomforest_org,scorebased_utl,randomforest_utl,head hash
 0,mangstadt/ez-vcard,Java,1.0,1,0,0.433617,10.5,0.0,1,34146,0.448443,54,1,0,1,1,95298de036f1df38fd7ca7e3225c4fd5ad5c44e6
 1,pedrovgs/Algorithms,Java,0.932927,1,1,0.390942,89.0,0.75,1,6616,0.556371,1377,1,1,1,1,c108e3a4b4fa2556b8c88d7ab6803084b01a4364
 1783,tntim96/JSCover,Java,0.985816,1,0,0.490264,25.0,6.448276,1,54612,0.627527,298,1,1,1,1,fff7a84ce12a037518e3c58e1d941a4d3dabb7e2

input_data/repos_with_hashes.csv

@@ -1,4 +1,4 @@
-idx,repository,language,architecture,community,continuous_integration,documentation,history,issues,license,size,unit_test,stars,scorebased_org,randomforest_org,scorebased_utl,randomforest_utl,Validation hash
+idx,repository,language,architecture,community,continuous_integration,documentation,history,issues,license,size,unit_test,stars,scorebased_org,randomforest_org,scorebased_utl,randomforest_utl,head hash
 0,Complexible/stardog-spring,Java,1.0,1,0,0.36999,0.0,0.0,1,2153,0.403618,12,0,0,1,1,05ed950161bfa7a4cd23922eddd760a51cfc7089
 1,sdeo/protobuf-socket-rpc,Java,1.0,1,0,0.233562,0.0,0.0,1,7472,0.323622,18,0,0,1,1,b1cf847b9922c2bc146a9eabf7c0df52428bf076
 2,twitter/netty-http2,Java,1.0,2,1,0.22231,0.368421,0.0,1,5886,0.341296,41,1,0,1,1,e8df896db61045b118f3423196c41c11c162b1c0