Skip to content

Commit

Permalink
Fix #118
Browse files Browse the repository at this point in the history
  • Loading branch information
@cesarsotovalero
cesarsotovalero committed Dec 21, 2022
1 parent 1a876e7 commit ab9cb8b
Show file tree
Hide file tree
Showing 2 changed files with 18 additions and 14 deletions.
24 changes: 12 additions & 12 deletions README.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -62,18 +62,18 @@ Let's see an example of running DepClean version 2.0.1 in the project [Apache Co
The Maven plugin can be configured with the following additional parameters.


| Name | Type | Description |
|:---------------------------|:-------------:|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| `<ignoreDependencies>` | `Set<String>` | Add a list of dependencies, identified by their coordinates, to be ignored by DepClean during the analysis and considered as used dependencies. Useful to override incomplete result caused by bytecode-level analysis. **Dependency format is:** `groupId:artifactId:version:scope`. |
| `<ignoreScopes>` | `Set<String>` | Add a list of scopes, to be ignored by DepClean during the analysis. Useful to not analyze dependencies with scopes that are not needed at runtime. **Valid scopes are:** `compile`, `provided`, `test`, `runtime`, `system`, `import`. An Empty string indicates no scopes (default). |
| `<ignoreTests>` | `boolean` | If this is true, DepClean will not analyze the test classes in the project, and, therefore, the dependencies that are only used for testing will be considered unused. This parameter is useful to detect dependencies that have `compile` scope but are only used for testing. **Default value is:** `false`. |
| `<createPomDebloated>` | `boolean` | If this is true, DepClean creates a debloated version of the pom without unused dependencies called `debloated-pom.xml`, in the root of the project. **Default value is:** `false`. |
| `<createResultJson>` | `boolean` | If this is true, DepClean creates a JSON file of the dependency tree along with metadata of each dependency. The file is called `depclean-results.json`, and is located in the `target` directory of the project. **Default value is:** `false`. |
| `<createCallGraphCsv>` | `boolean` | If this is true, DepClean creates a CSV file with the static call graph of the API members used in the project. The file is called `depclean-callgraph.csv`, and is located in the `target` directory of the project. **Default value is:** `false`. |
| `<failIfUnusedDirect>` | `boolean` | If this is true, and DepClean reported any unused direct dependency in the dependency tree, the build fails immediately after running DepClean. **Default value is:** `false`. |
| `<failIfUnusedTransitive>` | `boolean` | If this is true, and DepClean reported any unused transitive dependency in the dependency tree, the build fails immediately after running DepClean. **Default value is:** `false`. |
| `<failIfUnusedInherited>` | `boolean` | If this is true, and DepClean reported any unused inherited dependency in the dependency tree, the build fails immediately after running DepClean. **Default value is:** `false`. |
| `<skipDepClean>` | `boolean` | Skip plugin execution completely. **Default value is:** `false`. |
| Name | Type | Description |
|:---------------------------|:-------------:|:--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| `<ignoreDependencies>` | `Set<String>` | A regex expression matching dependencies to be ignored by DepClean during the analysis (i.e., considered as used). This is useful to bypass incomplete result caused by bytecode-level analysis. **For example:** `-DignoreDependencies="net.bytebuddy:byte-buddy:*.","com.google.guava*."` ignores dependencies with `groupId:artifactId` equals to `net.bytebuddy:byte-buddy` and `groupId` equals to `com.google.guava`. |
| `<ignoreScopes>` | `Set<String>` | Add a list of scopes, to be ignored by DepClean during the analysis. Useful to not analyze dependencies with scopes that are not needed at runtime. **Valid scopes are:** `compile`, `provided`, `test`, `runtime`, `system`, `import`. An Empty string indicates no scopes (default). |
| `<ignoreTests>` | `boolean` | If this is true, DepClean will not analyze the test classes in the project, and, therefore, the dependencies that are only used for testing will be considered unused. This parameter is useful to detect dependencies that have `compile` scope but are only used for testing. **Default value is:** `false`. |
| `<createPomDebloated>` | `boolean` | If this is true, DepClean creates a debloated version of the pom without unused dependencies called `debloated-pom.xml`, in the root of the project. **Default value is:** `false`. |
| `<createResultJson>` | `boolean` | If this is true, DepClean creates a JSON file of the dependency tree along with metadata of each dependency. The file is called `depclean-results.json`, and is located in the `target` directory of the project. **Default value is:** `false`. |
| `<createCallGraphCsv>` | `boolean` | If this is true, DepClean creates a CSV file with the static call graph of the API members used in the project. The file is called `depclean-callgraph.csv`, and is located in the `target` directory of the project. **Default value is:** `false`. |
| `<failIfUnusedDirect>` | `boolean` | If this is true, and DepClean reported any unused direct dependency in the dependency tree, the build fails immediately after running DepClean. **Default value is:** `false`. |
| `<failIfUnusedTransitive>` | `boolean` | If this is true, and DepClean reported any unused transitive dependency in the dependency tree, the build fails immediately after running DepClean. **Default value is:** `false`. |
| `<failIfUnusedInherited>` | `boolean` | If this is true, and DepClean reported any unused inherited dependency in the dependency tree, the build fails immediately after running DepClean. **Default value is:** `false`. |
| `<skipDepClean>` | `boolean` | Skip plugin execution completely. **Default value is:** `false`. |


You can integrate DepClean in your CI/CD pipeline.
Expand Down
8 changes: 6 additions & 2 deletions depclean-core/src/main/java/se/kth/depclean/core/DepCleanManager.java
View file
Original file line number Diff line number Diff line change
Expand Up @@ -7,9 +7,11 @@
import java.util.Objects;
import java.util.Set;
import java.util.concurrent.TimeUnit;
import java.util.regex.Pattern;
import java.util.stream.Collectors;
import lombok.AllArgsConstructor;
import lombok.SneakyThrows;
import lombok.extern.slf4j.Slf4j;
import org.apache.commons.io.FileUtils;
import se.kth.depclean.core.analysis.AnalysisFailureException;
import se.kth.depclean.core.analysis.DefaultProjectDependencyAnalyzer;
Expand All @@ -26,6 +28,7 @@
* Runs the depclean process, regardless of a specific dependency manager.
*/
@AllArgsConstructor
@Slf4j
public class DepCleanManager {

private static final String SEPARATOR = "-------------------------------------------------------";
Expand Down Expand Up @@ -224,14 +227,15 @@ private ProjectContext buildProjectContext() {
*/
private Set<Dependency> toDependency(Set<Dependency> allDependencies, Set<String> ignoreDependencies) {
return ignoreDependencies.stream()
.map(dependency -> findDependency(allDependencies, dependency))
.map(dependencyToIgnore -> findDependency(allDependencies, dependencyToIgnore))
.filter(Objects::nonNull)
.collect(Collectors.toSet());
}

private Dependency findDependency(Set<Dependency> allDependencies, String dependency) {
Pattern pattern = Pattern.compile(dependency, Pattern.CASE_INSENSITIVE);
return allDependencies.stream()
.filter(dep -> dep.toString().toLowerCase().contains(dependency.toLowerCase()))
.filter(dep -> pattern.matcher(dep.toString()).lookingAt())
.findFirst()
.orElse(null);
}
Expand Down

0 comments on commit ab9cb8b

Please sign in to comment.