C2DF

Introduction

Synthesis of data flow diagrams from annotated C programs (C2DF) is a command line program for the automated generation of data flow diagrams. It use KAOM for generation and the automatic layout provided by KIML.

Manual

To work with C2DF you must pass as the first source and a destination directory as an argument. C2DF generates then in the target directory for each annotated source file from the source directory or a subdirectory thereof a .KAOT-file. Also, a subdirectory called xml will be created in the target directory. This have to be deleted.

General description of annotations:

 To declare annotations, you need to put in a C-comment before the annotated object. This comment will be initiated by /* and completed with */. Before the start  of each content must be written the keyword @kaom. Then follow the individual contents.

IDs must consist only of a particular character set. These are all signs of the Latin alphabet in uppercase and lowercase letters (a-z,A-Z), as well as Hindu-Arabic numerals in European notation(0-9). In addition, spaces and underscores are allowed. Other characters are replaced internally by underscores _. This also applies to the diacritical marks ä,ö,ü , and the ligature ß. Similarly,it is not allowed to write the Keywords input:, output:, link:, content: and toplevel: in IDs. According to the ASCII are the characters  32, 48-57, 65-90, 95 and 97-122 allowed in ids.

Labels can hold all characters, except quotation marks, commas, semicolons, at sign, asterisks with a backslash and backslashes with a asterisk.

The object names must be identical with the object names in the source code.

There is no check if IDs are used or defined.

/*!  @kaom
input:    ID_1 "label", ID_2 "label", ... ; 
output:   ID_1 "label", ID_2 "label", ... ;
link:     source -> context.target, context.source -> context.target, source -> target, ... ;
content:  function_1:ID "label", function_2:ID "label", ... ;
toplevel: ID "label";
*/

Specification of inputs:

The specification of the inputs of an actor must be initiated by the keyword input:. Then follows the list of inputs. Each entry must have a unique ID in the actor. Optional behind the ID can be any label in quotation marks. Other entries are separated with commas. The list of all inputs must be terminated with a semicolon.

input: ID_1 "label", ID_2 "label", ... ;

Specification of outputs:

The specification of the outputs  of an actor must be initiated by the keyword output:. Then follows the list of outputs. Each entry must have a unique ID in the actor. Optional behind the ID can be any label in quotation marks. Other entries are separated with commas. The list of all outputs must be terminated with a semicolon.

output: ID_1 "label", ID_2 "label", ... ;

Specification of data flows:

The specification of the data flows of an actor must be initiated by the keyword link:.  Then follows the list of the data flows. The ID of the source and the ID of the target must be specified for each entry. Source and destination are separated by a right-leaning arrow indicating the direction of data flow. If source or destination are defined in another actor as the data flow, then the ID of the actor of the source or destination must be specified. The ID of the actor is placed directly in front of the ID of the source or target, separated by a dot. Actors, in which source and destination are defined, may not differ by more than one hierarchical level. Other entries are separated with commas. The list of all data flows must be terminated with a semicolon.

link: source -> context.target, context.source -> context.target, source -> target, ... ;

Specification of included actors:

The specification of the included actors of an actor must be initiated by the keyword content:.   Followed by the list of the included actors. Each entry must be specified with the name of its represented function in the source code. Followed by a unique ID within the actor, which is separated from the function name with a colon. Optional behind the ID can be any label in quotation marks. Other entries are separated with commas. The list of all included actors must be terminated with a semicolon.

content: function_1:ID "label", function_2:ID "label", ... ;

toplevel:ID "label";

link: source -> ID0_ID1_ID2_..._ID:target, ... ;