Интерактивная система просмотра системных руководств (man-ов)
>> ccomps (1) ( Linux man: Команды и прикладные программы пользовательского уровня )
ccomps - connected components filter for graphs
decomposes graphs into their connected components,
printing the components to standard output.
The following options are supported:
No output graph is printed. Implies the
Only the connected components are printed, as separate graphs.
Counts of nodes, edges and connected components are printed.
Use clusters in computing components in addition to normal edge
connectivity. In essence, this gives the connected components of the
derived graph in which nodes top-level clusters and nodes in the
original graph. This maintains all subgraph structure within a
component, even if a subgraph does not contain any nodes.
Do not project subgraph structure. Normally, if
produces components as graphs distinct from the input graph, it will
define subgraphs which are projections of subgraphs of the input graph
onto the component. (If the projection is empty, no subgraph is produced.)
If this flag is set, the component contains only the relevant nodes and
Prints only the component containing the node node_name,
Prints only component number index, if any, starting at 0.
If specified, each graph will be written to a different file
with the names derived from outfile. In particular,
if both -o and -x flags are used, then each connected
component is written to a different file. If outfile does
not have a suffix, the first file will have the name outfile;
then next outfile_1, then next outfile_2, and so on.
If outfile has a suffix, i.e., has the form base.sfx,
then the files will be named base.sfx, base_1.sfx,
By default, each input graph is printed, with each connected
component given as a subgraph whose name is a concatenation of
the name of the input graph, the string "_component_" and the
number of the component.
The following operand is supported:
Names of files containing 1 or more graphs in dot format.
operand is specified,
the standard input will be used.
Unless used to extract a single connected component,
if all the input graphs are connected; and
non-zero if any graph has multiple components, or any error occurred.
If just extracting a single component,
on success and non-zero if an error occurred.
It is possible, though unlikely, that the names used for connected
components and their subgraphs may conflict with existing subgraph names.