Allow mfultralight and mfc tools to work also with specific UID's #350
Conversation
Currently the check on argc is done twice, once in each if branch. This is silly and we can just check once and fail right away. Signed-off-by: Olliver Schinagl <[email protected]>
The nfc-mfclassic utility will pick a seemingly random (the libnfc default which seems to be the lowest UID). With the new (u|U) options it is now possible to force a UID and thus write a specific tag, which can be very useful if there are more then one tag visible. Signed-off-by: Olliver Schinagl <[email protected]>
For some reason, 0cece94 changed the argc count check to only show the help if argc is 0. Obviously, argc is never zero, as the first argument in argv is always the binary itself. Revert that and show usage if there is no arguments supplied to the binary. Signed-off-by: Olliver Schinagl <[email protected]>
When several tags are in range, nfc-mfultralight uses the lowest ID it finds by default. This patch adds some code from nfc-list that lists the tags in range whenever an operation is performed (r/w). Further more it adds the --with-uid <UID> option to force reading/writing of a specific tag. The UID can be up to 10 bytes long and can be optionally separated by colons or hyphens (MAC address style). Signed-off-by: Olliver Schinagl <[email protected]>
Clean up some stray unintended whitespaces. This patch does not introduce any binary changes. Signed-off-by: Olliver Schinagl <[email protected]>
We use a variable, uiBlocks, to determine how many blocks to read/write. Reading is actually done via a hardcoded 0xF value however. Additionally, make uiblocks a const, as we use it as a constant and change the page variable to uint32_t for consistency sake. Signed-off-by: Olliver Schinagl <[email protected]>
Currently, we return false, and after turning set the bFailure state. This is of course not possible. Signed-off-by: Olliver Schinagl <[email protected]>
|
@oliv3r : Good job, thanks ! 👍 |
|
@neomilium This is probably way too late, but it is worth noting that this patch breaks backward compatibility by adding the (unnecessary) u switch. I propose that u/U is unnecessary, defaulting to 'standard' behavious but can be provided if required. It seems like this patch is a 'convenience' patch by @oliv3r - while it's a useful contribution, it shouldn't be enforced on everyone else. |
|
@quantum-x we try to be careful regarding the libnfc itself and its API, but in our opinion the provided tools are just (useful) examples based on the libnfc and their CLI usage must not be preserved at all cost. E.g. if you want to provide a more future-proof interface with single-letter Unix-style options + their GNU-style long-named equivalents, you're more than welcome to submit a patch. |
|
@doegox No arguments from me based off your points. If we don't need to break CLI compatibility, why do it? |
|
The current scheme to pass arguments to |
|
Seems like a great idea. @oliv3r is this something that you would consider implementing? |
This patch series does some code cleanup/fixups on a small scale, but most importantly adds the ability to specify UID's to mfc-classic and mfultralight tools to r/w a specific tag when multiple tags are insight.