In the development process of a software project some source code gets designed and a compiler (or interpreter) transforms that source code to executable assembler commands for the CPU. In a similar way a documentation tool scans the source code and transforms data to generate documentation for the user. Simple tools extract information from special comments marked by (one or more) magic character. All context for the documentation is in the comments, created by the programmer. Advanced tools like Doxygen also scan the source code (and the special comments), in order to read information like symbol names, class declarations, ... Scanning the source code avoids redundant information and helps to keep the documentation up to date.
In order to get familiar with the command line interface here are some exercises, operating fbdoc on its own source code. It's presumed that you compiled and installed the executable in a system path, so that you can call it from any folder (see section Build for details). Otherwise you have to prepend the complete path to execute the binary.
src/bas
.$
character are the commands executed at the terminal. The other lines are output.src/bas/fbdoc.bas -->
and done
, written to STDERR
) and the transformed output (written to STDOUT
). Both get mixed in to one stream by the terminal. The default mode is designed to be used as a Doxygen filter. In that use case the STDERR
stream goes as user information to the Doxygen log and the STDOUT
stream is the internal data flow (invisible)./*!
and ending with */
) and the include statements (removed keyword ONCE
).STDERR
and STDOUT
streams by writing the STDOUT
output to a file, executing out.c
.fbdoc.c
) and the file gets generated in the (newly created) default path ../c_src
, so check file src/c_src/fbdoc.c
. In order to overridde the default path and generate the file in the current directory use option –outpath (-o), like *.bas
and *.bi
and operates on all of those files. In order to select specific files you can customize the patterns like