Lizard is an extensible Cyclomatic Complexity Analyzer for many programming languages including C/C++ (doesn't require all the header files or Java imports). It also does copy-paste detection (code clone detection/code duplicate detection) and many other forms of static code analysis.
A list of supported languages:
- C/C++ (works with C++14)
- Java
- C# (C Sharp)
- JavaScript (With ES6 and JSX)
- Objective-C
- Swift
- Python
- Ruby
- TTCN-3
- PHP
- Scala
- GDScript
- Golang
- Lua
- Rust
By default lizard will search for any source code that it knows and mix all the results together. This might not be what you want. You can use the "-l" option to select language(s).
It counts
- the nloc (lines of code without comments),
- CCN (cyclomatic complexity number),
- token count of functions.
- parameter count of functions.
You can set limitation for CCN (-C), the number of parameters (-a). Functions that exceed these limitations will generate warnings. The exit code of lizard will be none-Zero if there are warnings.
This tool actually calculates how complex the code 'looks' rather than how complex the code really 'is'. People will need this tool because it's often very hard to get all the included folders and files right when they are complicated. But we don't really need that kind of accuracy for cyclomatic complexity.
It requires python2.7 or above (early versions are not verified).
lizard.py can be used as a stand alone Python script, most functionalities are there. You can always use it without any installation. To acquire all the functionalities of lizard, you will need a proper install.
python lizard.py
If you want a proper install:
[sudo] pip install lizard
Or if you've got the source:
[sudo] python setup.py install --install-dir=/path/to/installation/directory/
lizard [options] [PATH or FILE] [PATH] ...
Run for the code under current folder (recursively):
lizard
Exclude anything in the tests folder:
lizard mySource/ -x"./tests/*"
-h, --help show this help message and exit --version show program's version number and exit -l LANGUAGES, --languages LANGUAGES List the programming languages you want to analyze. if left empty, it'll search for all languages it knows. lizard -l cpp -l java searches for C++ and Java code. The available languages are: cpp, csharp, java, javascript, objectivec, php, python, ruby, swift, ttcn -V, --verbose Output in verbose mode (long function name) -C CCN, --CCN CCN Threshold for cyclomatic complexity number warning. The default value is 15. Functions with CCN bigger than it will generate warning -L LENGTH, --length LENGTH Threshold for maximum function length warning. The default value is 1000. Functions length bigger than it will generate warning -a ARGUMENTS, --arguments ARGUMENTS Limit for number of parameters -w, --warnings_only Show warnings only, using clang/gcc's warning format for printing warnings. https://clang.llvm.org/docs/UsersManual.html#cmdoption- fdiagnostics-format -i NUMBER, --ignore_warnings NUMBER If the number of warnings is equal or less than the number, the tool will exit normally, otherwise it will generate error. Useful in makefile for legacy code. -x EXCLUDE, --exclude EXCLUDE Exclude files that match this pattern. * matches everything, ? matches any single character, "./folder/*" exclude everything in the folder recursively. Multiple patterns can be specified. Don't forget to add "" around the pattern. --csv Generate CSV output as a transform of the default output -X, --xml Generate XML in cppncss style instead of the tabular output. Useful to generate report in Jenkins server -t WORKING_THREADS, --working_threads WORKING_THREADS number of working threads. The default value is 1. Using a bigger number can fully utilize the CPU and often faster. -m, --modified Calculate modified cyclomatic complexity number, which count a switch/case with multiple cases as one CCN. -E EXTENSIONS, --extension EXTENSIONS User the extensions. The available extensions are: -Ecpre: it will ignore code in the #else branch. -Ewordcount: count word frequencies and generate tag cloud. -Eoutside: include the global code as one function. -s SORTING, --sort SORTING Sort the warning with field. The field can be nloc, cyclomatic_complexity, token_count, parameter_count, etc. Or an customized file. -W WHITELIST, --whitelist WHITELIST The path and file name to the whitelist file. It's './whitelizard.txt' by default.
============================================================== NLOC CCN token param function@line@file -------------------------------------------------------------- 10 2 29 2 start_new_player@26@./html_game.c ... 6 1 3 0 set_shutdown_flag@449@./httpd.c 24 3 61 1 server_main@454@./httpd.c -------------------------------------------------------------- 2 file analyzed. ============================================================== LOC Avg.NLOC AvgCCN Avg.ttoken function_cnt file -------------------------------------------------------------- 191 15 3 51 12 ./html_game.c 363 24 4 86 15 ./httpd.c ====================================== !!!! Warnings (CCN > 15) !!!! ====================================== 66 19 247 1 accept_request@64@./httpd.c ================================================================================= Total NLOC Avg.NLOC Avg CCN Avg token Fun Cnt Warning cnt Fun Rt NLOC Rt -------------------------------------------------------------------------------- 554 20 4.07 71.15 27 1 0.04 0.12
./src/html_ui/httpd.c:64: warning: accept_request has 19 CCN and 1 params (66 NLOC, 247 tokens) ./src/mahjong_game/mj_table.c:109: warning: mj_table_update_state has 20 CCN and 1 params (72 NLOC, 255 tokens)
The option -Tcyclomatic_complexity=10 is equal to -C10. The option -Tlength=10 is equal to -L10. The option -Tparameter_count=10 is equal to -a10.
You can also do -Tnloc=10 to set the limit of the NLOC. Any function that has NLOC greater than 10 will generate a warning.
Lizard has a simple solution with generated code. Any code in a source file that is following a comment containing "GENERATED CODE" will be ignored completely. The ignored code will not generate any data, except the file counting.
lizard -Eduplicate <path to your code>
You can generate a "Tag cloud" of your code by the following command. It counts the identifiers in your code (ignoring the comments).
lizard -EWordCount <path to your code>
You can also use lizard as a Python module in your code:
>>> import lizard
>>> i = lizard.analyze_file("../cpputest/tests/AllTests.cpp")
>>> print i.__dict__
{'nloc': 9, 'function_list': [<lizard.FunctionInfo object at 0x10bf7af10>], 'filename': '../cpputest/tests/AllTests.cpp'}
>>> print i.function_list[0].__dict__
{'cyclomatic_complexity': 1, 'token_count': 22, 'name': 'main', 'parameter_count': 2, 'nloc': 3, 'long_name': 'main( int ac , const char ** av )', 'start_line': 30}
You can also use source code string instead of file. But you need to provide a file name (to identify the language).
>>> i = lizard.analyze_file.analyze_source_code("AllTests.cpp", "int foo(){}")
If for some reason you would like to ignore the warnings, you can use the whitelist. Add 'whitelizard.txt' to the current folder (or use -W to point to the whitelist file), then the functions defined in the file will be ignored. Please notice that if you assign the file pathname, it needs to be exactly the same relative path as Lizard to find the file. An easy way to get the file pathname is to copy it from the Lizard warning output. This is an example whitelist:
#whitelizard.txt #The file name can only be whitelizard.txt and put it in the current folder. #You may have commented lines begin with #. function_name1, function_name2 # list function names in multiple lines or split with comma. file/path/name:function1, function2 # you can also specify the filename
You can use options in the comments of the source code to change the behavior of lizard. By putting "#lizard forgives" inside a function or before a function it will suppress the warning for that function.
int foo() { // #lizard forgives the complexity ... }
Lizard requires syntactically correct code. Upon processing input with incorrect or unknown syntax:
Lizard guarantees to terminate eventually (i.e., no forever loops, hangs) without hard failures (e.g., exit, crash, exceptions).
There is a chance of a combination of the following soft failures:
- omission
- misinterpretation
- improper analysis / tally
- success (the code under consideration is not relevant, e.g., global macros in C)
This approach makes the Lizard implementation simpler and more focused with partial parsers for various languages. Developers of Lizard attempt to minimize the possibility of soft failures. Hard failures are bugs in Lizard code, while soft failures are trade-offs or potential bugs.
In addition to asserting the correct code, Lizard may choose not to deal with some advanced or complicated language features:
- C/C++ digraphs and trigraphs are not recognized.
- C/C++ preprocessing or macro expansion is not performed. For example, using macro instead of parentheses (or partial statements in macros) can confuse Lizard's bracket stacks.
- Some C++ complicated templates may cause confusion with matching angle brackets
and processing less-than
<
or more-than>
operators inside of template arguments.
Lizard is often used in software related researches. If you used it to support your work, you may contact the lizard author to add your work in the following list.
Software Quality in the ATLAS experiment at CERN, which refers to Lizard as one of the tools, has been published in the Journal of Physics: https://iopscience.iop.org/article/10.1088/1742-6596/898/7/072011
- S Martin-Haugh et al 2017 J. Phys.: Conf. Ser. 898 072011
Lizard is also used as a plugin for fastlane to help check code complexity and submit xml report to sonar.