[Ktechlab-devel] broke SVN. =(
P Zoltan
zoltan.padrah at gmail.com
Sun Nov 16 10:02:35 UTC 2008
On Sun, 16 Nov 2008 08:36:18 +0100, Alan Grimes <agrimes at speakeasy.net>
wrote:
> Sorry, I broke the SVN by recklessly moving circuit document into the
> same directory as its parent class...
>
> It broke a bunch of linking stuff, which exposed a __HUGE CAN OF WORMS__.
>
> Apparently a buch of classes in the *FLOWPARTS* branch of the tree was
> using CircuitICNDocument instead of FlowICNDocument! =(
Actually I've created those classes, to separate the functionality
related to flowcode from the electronics.
What is the 'policy' about where to put new classes? It seems I didn't
created it at the right place in the tree.
Do you refer to the errors like this:
trunk/src/flowparts/pinmapping.cpp: In member function 'void
PinMapDocument::init(const PinMapping&, MicroInfo*)':
trunk/src/flowparts/pinmapping.cpp:161: error: no matching function for
call to 'PinMapDocument::createConnector(Node*, Node*)'
trunk/src/flowparts/flowicndocument.h:47: note: candidates are: virtual
Connector* FlowICNDocument::createConnector(Node*, Connector*, const
QPoint&, QPointLi
This is not that a huge problem.
The PinMapDocument _is_ a CircuitDocument. All it does is to help the
user to connect the PIC's pins to a seven segment, a keypad, or something
similar. The class is used only in one place in ktechlab: new -> flowcode
-> on the pic: advanced -> Pin map definitions: Create -> enter some name;
ok -> you'll see a simple circuit with 2 elements: a pic and a keypad or a
seven-segment. You can make the connections between the pic and the other
component there, and to save it as a 'pin map'. That's it.
Maybe it's a good idea to move all the pinmapping's source in a separade
directory, so it won't mislead people because it uses CircuitICNDocument.
> I propose that we press ahead and fix these issues instead of
> reverting... (which we'll still be able to do if we really need to...)
> Last good version is 149 or 150.
>
> It's 2:30 AM, so I won't be able to fix much more tonight.
>
More information about the Ktechlab-devel
mailing list