来自SonarQube的SonarJava的静态分析警告自动修复系统_Java_Python_下载.zip

# Sorald support scripts This directory contains a Python package with support scripts for Sorald. ## Requirements Python 3.8+ and Git 2.x. Some scripts also require Sorald to be available from the command line, which in that case is detailed in the individual script's description. ## Install Assuming that you are currently in this directory, install the support scripts package like so. ```bash $ python3 -m pip install . ``` It is however strongly recommended that you install the scripts inside of a [virtual environment (link to a good article about it)](https://realpython.com/effective-python-environment/#virtual-environments), to avoid polluting your system-wide or user packages. One way to do that is to use the `venv` module, which normally ships with Python. ```bash $ python3 -m venv env # creates a virtual environment in directory "env" $ source env/bin/activate # activates the virtual environment [ ... work on your stuff ... ] $ deactivate # exits the virtual environment ``` Once the virtual environment is active, any package installed with `pip install` will end up in the local `env` directory. > **Note:** On some Linux distributions, such as Debian-based ones, `venv` is > not always shipped with `python3`, and you may need to install a separate > `python3-venv` or `python3-dev` system package. For development, add the `-e` flag to make an editable install. ```bash $ python3 -m pip install -e . ``` An editable install lets any changes you make to your source code be visible in the installed package immediately, whereas a normal install requires you to reinstall the package for each change. ## Uninstall ```bash $ python3 -m pip uninstall soraldscripts ``` ## Running tests To run the tests for the helper scripts, first install the test dependencies. ```bash $ python3 -m pip install -e .[TEST] ``` Then execute `pytest`. ```bash $ python3 -m pytest tests/ ``` ## Adding a new script Simply put a Python script into the `sorald` directory. That makes it part of the `sorald` package. Assuming that the script is called `script.py`, you can run the script (after installing the support scripts package according to the instructions in this README) like so: ```bash $ python3 -m sorald.script ``` To test the script, you can add a new file in the `tests` directory, called `tests/test_script.py`. Those tests can be executed as described in [Running tests](#running-tests). > **Important:** Don't forget to add a description of the new script in the > [Scripts](#scripts) section. ## Scripts This section details usage of the scripts in the support package. ### `sorald.achievements` The achievements script can be used to generate an "achievements" file for Sorald, detailing the pull requests performed with it. It's essentially a human-readable version of the PRs JSON file. The script is only usable in a single way: given a `prs.json` file, it generates a human-readable Markdown file with some of the interesting content. Use like so: ```bash $ python3 -m sorald.achievements --prs-json-file path/to/prs.json --output ACHIEVEMENTS.md ``` ### `sorald.handled_rules` The handled_rules script generates a markdown file for listing all the Sonar rules which are currently repaired by sorald. Each rule is accompanied by the corresponding processor summary and example repairs. The script outputs a markdown file whose name is given as parameter to the command like so: ```bash python3 -m sorald.handled_rules --output HANDLED_RULES.md ``` ### `sorald.prmessage` The PR message script can be used to easily generate a PR message with information about the repairs that have been made. The current version is tailored for repair of a single rule, as that's what we have been targeting with our PRs thus far. It could fairly easily be extended to encompass multi-rule repairs, for example by parsing the JSON stats output from Sorald. The script currently just prints the message to standard output, with the first line being the title (make sure to remove the `PULL REQUEST TITLE: ` prefix). Usage is currently like so: ``` $ python3 -m sorald.prmessage --rule-key <RULE_KEY> --num-repairs <NUM_REPAIRS> ``` For example, generating a PR message for a patch in which Sorald has repaired 2 violations of Sonar Rule 2111, one would call the script like so. ``` $ python3 -m sorald.prmessage --rule-key 2111 --num-repairs 2 ``` This generates a PR message with some appropriate text and links to both Sorald's and Sonar's documentation for the given rule. ### `sorald.prrecorder` The PR recorder script can be used to collect data from pull requests made with sorald. It is the JSON file generated by this script that's used by `sorald.achievements`. One can run the PR recorder by executing the `sorald.prrecorder` script with Python. ```bash $ python3 -m sorald.prrecorder [options] ``` To show the available commands, run with the `--help` flag. ``` $ python3 sorald.prrecorder --help usage: sorald.prrecorder [-h] {record-initial,record-final,add-manual-edit} ... positional arguments: {record-initial,record-final,add-manual-edit} record-initial make the initial record for a PR record-final make the final record for a PR add-manual-edit add a diff representing a manual edit of a PR optional arguments: -h, --help show this help message and exit ``` In general, one must always make an initial and final record using the `record-initial` and `record-final` commands. Running `add-manual-edit` is optional and allows one to record manual edits performed on the PR, i.e. edits that were not made with Sorald itself. The rest of this section contains an example use case of recording the PR [redhat-developer/rsp-server#619](https://github.com/redhat-developer/rsp-server/pulls/619). #### Step 1: Make the initial record (`record-initial`) When the PR has been opened, one should make the initial record. It can be done like so. ```bash $ python3 -m sorald.prrecorder record-initial \ --owner redhat-developer \ --repo-name rsp-server \ --pr-number 619 \ --sorald-stats-file stats.json ``` Note that `stats.json` is the statistics file generated by Sorald when running with statistics collection. The command executed above creates a file `prs.json` with an initial record like this: ```json { "redhat-developer/rsp-server#619": { "repoSlug": "redhat-developer/rsp-server", "prMetadata": { "url": "https://github.com/redhat-developer/rsp-server/pull/619", "createdAt": "2020-11-25 12:01:06", "closedAt": null, "mergedAt": null", "state": "open", "isMerged": false, "number": 619 }, "soraldStatistics": { "repairs": [ { "nbPerformedRepairs": 4, "ruleName": "XxeProcessing", "crashedRepairsLocations": [], "nbCrashedRepairs": 0, "ruleKey": "2755", "nbViolationsBefore": 4, "performedRepairsLocations": [ { "endLine": 85, "endColumn": 52, "startColumn": 41, "startLine": 85, "filePath": "framework/bundles/org.jboss.tools.rsp.launching.java/src/main/java/org/jboss/tools/rsp/internal/launching/java/util/LaunchingSupportUtils.java", "violationSpecifier": "2755:framework/bundles/org.jboss.tools.rsp.launching.java/src/main/java/org/jboss/tools/rsp/internal/launching/java/util/LaunchingSupportUtils.java:85:41:85:52" }, { "endLine": 92, "endColumn": 70, "startColumn": 59, "startLine": 92, "filePa