Bareos Developer Notes

This document is intended mostly for developers and describes how you can contribute to the Bareos project and the general framework of making Bareos source changes.

History

Bareos is a fork of the open source project Bacula version 5.2. In 2010 the Bacula community developer Marco van Wieringen started to collect rejected or neglected community contributions in his own branch. This branch was later on the base of Bareos and since then was enriched by a lot of new features.

This documentation also bases on the original Bacula documentation, it is technically also a fork of the documentation created following the rules of the GNU Free Documentation License.

Original author of Bacula and it’s documentation is Kern Sibbald. We thank Kern and all contributors to Bacula and its documentation. We maintain a list of contributors to Bacula (until the time we’ve started the fork) and Bareos in our AUTHORS file.

Contributions

Contributions to the Bareos project come in many forms: ideas, participation in helping people on the bareos-users email list, packaging Bareos binaries for the community, helping improve the documentation, and submitting code.

Patches

Patches should be sent as a pull-request to the master branch of the GitHub repository. A detailed description can be found in the chapter Git Workflow. If you don’t want to sign up to GitHub, you can also send us your patches via E-Mail in git format-patch format to the bareos-devel mailing list.

Please make sure to use the Bareos Automatic Sourcecode Formatting. Don’t forget any Copyrights and acknowledgments if it isn’t 100% your code. Also, include the Bareos copyright notice that can be found in every source file.

Issue Database

We use github issues as the Bareos issue tracker. Before 2024 we’ve used https://bugs.bareos.org/, which is now in read-only mode.

All new bugs are now collected using the github issues component. If you want to take over an issue, rather than just make a comment, you should assign the issue to yourself. This helps other developers know that you are the principal person to deal with the bug. You can do so by going into the issue and use assign yourself in Assignees section.

Generally, we set the Status field to either acknowledged, confirmed, or feedback when we first start working on the bug. Feedback is set when we expect that the user should give us more information.

Normally, once you are reasonably sure that the bug is fixed, and a patch is made and attached to the bug report, and/or in the Git, you can close the bug. If you want the user to test the patch, then leave the bug open, otherwise close it and set Resolution to Fixed. We generally close bug reports rather quickly, even without confirmation, especially if we have run tests and can see that for us the problem is fixed. However, in doing so, it avoids misunderstandings if you leave a note while you are closing the bug that says something to the following effect: We are closing this bug because… If for some reason, it does not fix your problem, please feel free to reopen it, or to open a new bug report describing the problem“.

We do not recommend that you attempt to edit any of the bug notes that have been submitted, nor to delete them or make them private. In fact, if someone accidentally makes a bug note private, you should ask the reason and if at all possible (with his agreement) make the bug note public.

If the user has not properly filled in most of the important fields (platform, OS, Product Version, …) please do not hesitate to politely ask him to do so. The same applies to a support request (we answer only bugs), you might give the user a tip, but please politely refer him to the manual, the bareos-users mailing list and maybe the commercial support.

Reporting security issues

If you want to report a security-related problem, please take a look at our security policy.

Memory Leaks

Use standard C++17 resource management (RAII and smart pointers) to prevent memory leaks in general. If you need to detect memory leaks, you can just use valgrind to do so.

We also use sanitizers to detect memory leaks. To enable them, ensure you install libasan, libubsan, and libtsan and then enable sanitizers in the cmake arguments with -DENABLE_SANITIZERS=YES.

Guiding Principles

All new code should be written in modern C++17 following the Google C++ Style Guide and the C++ Core Guidelines.

We like simple rather than complex code, but complex code is still better than complicated code.

Currently there is still a lot of old C++ and C code in the code base and a lot of existing code violates our do’s and don’ts. Therefore our long-term goal is to modernize the code-base to make it easier to understand, easier to maintain and better approachable for new developers.

Boy Scout Rule

Before submitting your pull request, please ensure that you have followed the Boy Scout Rule:

> “Leave the codebase better than you found it.”

The Boy Scout Rule encourages contributors to make small improvements or clean-ups while working on a task without being explicitly requested. By following this rule, we can collectively improve the quality, readability, and maintainability of the codebase over time.

To apply the Boy Scout Rule, consider the following guidelines:

  • Clean up code: Review the code you are modifying and check if there are any areas that could be improved, such as variable naming, code formatting, or removing unnecessary comments.

  • Refactor when appropriate: If you find a block of code that you can refactor to improve its clarity or efficiency, take the initiative to make those changes.

  • Fix adjacent issues: If you notice any related issues or bugs while working on your task, address them if it’s within the scope of your current changes. This proactive approach helps prevent future problems.

  • Update documentation: If you modify a part of the codebase that affects the existing documentation, ensure that relevant documentation is updated accordingly.

Remember, the goal of the Boy Scout Rule is to foster continuous improvement and create a more sustainable and maintainable codebase. By leaving the code better than you found it, you contribute to the overall health and longevity of the project.

Usage of C++ Exceptions

We encourage developers to use C++ exceptions for the reason of simplicity and readability of the code. In contrast to long if/else constructs C++ exceptions are the state-of-the-art error handling mechanism of this programming language. With exceptions it is easier to transport errors and information about errors from the lowest stack level to the uppermost function.

In order to avoid memory leaks it is very advisable to use RAII or smart pointers for memory allocation. With regard to legacy code it is important to understand how memory management in detail works before throwing exceptions across several stack levels and causing leaks by accident.

General advice, many examples and debunked myths about C++ Exceptions can be found here: https://isocpp.org/wiki/faq/exceptions.

Automatic Sourcecode Formatting

All C/C++ code should be formatted properly based on the principles mentioned above. Therefore we provide a configuration file for clang-format that contains all formatting rules. The filename is “.clang-format” and it is located in the root directory of the bareos repo.

The configuration file will be automatically found and used by clang-format:

Example shell script
#!/bin/sh

#format one sourcecode file in-place
clang-format -i ./core/src/dird/dird_conf.cc

The Bareos project has bundled its automatic sourcecode formatting into one tool: bareos-check-sources. https://github.com/bareos/bareos/blob/master/devtools/pip-tools/README.md describes how to use it. In short:

$ cd devtools/pip-tools
$ pipenv sync
$ pipenv shell
(pip-tools)$ bareos-check-sources --since-merge --diff
(pip-tools)$ bareos-check-sources --since-merge --modify

Formatting exceptions

For some parts of code it works best to hand-optimize the formatting. We sometimes do this for larger tables and deeply nested brace initialization. If you need to hand-optimize make sure you add clang-format off and clang-format on comments so applying clang-format on your source will not undo your manual optimization. Please apply common sense and use this exception sparingly.

Sourcecode Comments

Use /* */ for multi-line comments. Use // for single-line comments.

SQL queries

Developers will have to use SQL queries to get data from the database. When you navigate the current code you might get a bit confused as there are different ways to do it: First, there are direct queries written within the functions that need them. Second, there are functions within the cats library containing ready made queries that get called. And finally there are the generated SQL files within the cats/dml folder that get invoked in certain situations.

Until we decide on a unified way to handle sql queries, we advise the following:

  • If your queries are trivial, you can put them as a string within the code you are writing, make sure you wrap them in a function, and make sure it can be reused,

  • If you are dealing with long and convoluted queries, write them within the cats/dml folder and update the related files and enums.

Do’s

Don’ts

avoid new

Starting with C++11 there are smart pointers like shared_ptr and unique_ptr. To create a shared_ptr you should use make_shared() from the standard library. If possible use unique_ptr instead of shared_ptr.

avoid delete

You should use the RAII paradigm, so cleanup is handled automatically.

don’t transfer ownership of heap memory without move semantics

No returning of raw pointers where the caller is supposed to free the resource.

don’t use C string functions

If you can, use std::string and don’t rely on C string functions.

don’t use the bareos replacements for C string functions.

These are deprecated.

avoid the edit_*() functions from edit.cc

Just use the appropriate format string. This will also avoid the temporary buffer that is required otherwise.

avoid pool memory allocation

The whole allocation library with get_pool_memory() and friends do not mix with RAII, so we will try to remove them step by step in the future. Avoid in new code if possible.