tcpdump/CONTRIBUTING
Denis Ovsienko be1e19f50c refine the feedback guidelines
Provide better detailed guidelines in CONTRIBUTING and update a number
of other files to refer to that file so that the directions are now
more uniform.
2017-02-02 11:38:05 +00:00

152 lines
5.9 KiB
Plaintext

Some Information for Contributors
---------------------------------
You want to contribute to Tcpdump, Thanks!
Please, read these lines.
How to report bugs and other problems
-------------------------------------
To report a security issue (segfault, buffer overflow, infinite loop, arbitrary
code execution etc) please send an e-mail to security@tcpdump.org, do not use
the bug tracker!
To report a non-security problem (failure to compile, incorrect output in the
protocol printout, missing support for a particular protocol etc) please check
first that it reproduces with the latest stable release of tcpdump and the latest
stable release of libpcap. If it does, please check that the problem reproduces
with the current git master branch of tcpdump and the current git master branch of
libpcap. If it does (and it is not a security-related problem, otherwise see
above), please navigate to https://github.com/the-tcpdump-group/tcpdump/issues
and check if the problem has already been reported. If it has not, please open
a new issue and provide the following details:
* tcpdump and libpcap version (tcpdump --version)
* operating system name and version and any other details that may be relevant
(uname -a, compiler name and version, CPU type etc.)
* configure flags if any were used
* statement of the problem
* steps to reproduce
Please note that if you know exactly how to solve the problem and the solution
would not be too intrusive, it would be best to contribute some development time
and open a pull request instead as discussed below.
Still not sure how to do? Feel free to [subscribe](http://www.tcpdump.org/#mailing-lists)
to the mailing list tcpdump-workers@lists.tcpdump.org and ask!
How to add new code and to update existing code
-----------------------------------------------
0) Check that there isn't a pull request already opened for the changes you
intend to make.
1) Fork the Tcpdump repository on GitHub from
https://github.com/the-tcpdump-group/tcpdump
(See https://help.github.com/articles/fork-a-repo/)
2) Setup an optional Travis-CI build
You can setup a travis build for your fork. So, you can test your changes
on Linux and OSX before sending pull requests.
(See http://docs.travis-ci.com/user/getting-started/)
3) Setup your git working copy
git clone https://github.com/<username>/tcpdump.git
cd tcpdump
git remote add upstream https://github.com/the-tcpdump-group/tcpdump
git fetch upstream
4) Do a 'touch .devel' in your working directory.
Currently, the effect is
a) add (via configure, in Makefile) some warnings options ( -Wall
-Wmissing-prototypes -Wstrict-prototypes, ...) to the compiler if it
supports these options,
b) have the Makefile support "make depend" and the configure script run it.
5) Configure and build
./configure && make -s && make check
6) Add/update sample.pcap files
We use tests directory to do regression tests on the dissection of captured
packets, by running tcpdump against a savefile sample.pcap, created with -w
option and comparing the results with a text file sample.out giving the
expected results.
Any new/updated fields in a dissector must be present in a sample.pcap file
and the corresponding output file.
Configuration is set in tests/TESTLIST.
Each line in this file has the following format:
test-name sample.pcap sample.out tcpdump-options
the sample.out file can be build by:
(cd tests && ../tcpdump -n -r sample.pcap tcpdump-options > sample.out)
It is often useful to have test outputs with different verbosity levels
(none, -v, -vv, -vvv, etc.) depending on the code.
7) Test with 'make check'
Don't send a pull request if 'make check' gives failed tests.
8) Try to rebase your commits to keep the history simple.
git rebase upstream/master
(If the rebase fails and you cannot resolve, issue "git rebase --abort"
and ask for help in the pull request comment.)
9) Once 100% happy, put your work into your forked repository.
git push
10) Initiate and send a pull request
(See https://help.github.com/articles/using-pull-requests/)
Code style and generic remarks
------------------------------
a) A thorough reading of some other printers code is useful.
b) Put the normative reference if any as comments (RFC, etc.).
c) Put the format of packets/headers/options as comments if there is no
published normative reference.
d) The printer may receive incomplete packet in the buffer, truncated at any
random position, for example by capturing with '-s size' option.
Thus use ND_TTEST, ND_TTEST2, ND_TCHECK or ND_TCHECK2 for bound checking.
For ND_TCHECK2:
Define : static const char tstr[] = " [|protocol]";
Define a label: trunc
Print with: ND_PRINT((ndo, "%s", tstr));
You can test the code via:
sudo ./tcpdump -s snaplen [-v][v][...] -i lo # in a terminal
sudo tcpreplay -i lo sample.pcap # in another terminal
You should try several values for snaplen to do various truncation.
e) Do invalid packet checks in code: Think that your code can receive in input
not only a valid packet but any arbitrary random sequence of octets (packet
- built malformed originally by the sender or by a fuzz tester,
- became corrupted in transit).
Print with: ND_PRINT((ndo, "%s", istr)); /* to print " (invalid)" */
f) Use 'struct tok' for indexed strings and print them with
tok2str() or bittok2str() (for flags).
g) Avoid empty lines in output of printers.
h) A commit message must have:
First line: Capitalized short summary in the imperative (70 chars or less)
Body: Detailed explanatory text, if necessary. Fold it to approximately
72 characters. There must be an empty line separating the summary from
the body.
i) Avoid non-ASCII characters in code and commit messages.
j) Use the style of the modified sources.
k) Don't mix declarations and code
l) Don't use // for comments
Not all C compilers accept C++/C99 comments by default.
m) Avoid trailing tabs/spaces