Assigned myself for now, support is pretty much welcome. We have these modules available in the HAW, so if anyone from here volunteers to do some testing, go ahead ;-). @thomaseichinger you've been involved in #2692. Do you find time to take care of this PR?

@lebrush There seem to be a couple of functions that are never used. Are those meant to improve functionality at some later point?
Otherwise I think this is a great clean and slim driver! Congratulations.
@PeterKietzmann I won't find extensive time to test this driver, if any, can you take on that part?

@thomaseichinger @PeterKietzmann we have only this [1] PN532 module at the HAW.
Unfortunately it only provides UART communication and does not support for switching the communication modes to I2C or SPI :(

[1] http://www.dfrobot.com/index.php?route=product/product&product_id=892&search=nfc&description=true

@lebrush please provide a Makefile in RIOT/drivers/pn532

@BytesGalore thx! true story: I forgot the Makefile 👍

@thomaseichinger @BytesGalore, I just addressed your comments

@thomaseichinger @BytesGalore ping

@thomaseichinger, do you think you could find some time during the next days for testing?

Let's hope Murdock is now happy

Murdock is happy. @PeterKietzmann should this a candidate for 2017.01 ?

Murdock is happy. @PeterKietzmann should this a candidate for 2017.01 ?

If it's merged it's merged ;-)

Tested I2C and SPI mode successfully. Code is also in good shape. @lebrush can you squash your commits?

@cgundogan thanks for testing it! 🎉 I just squashed and left the meaningful ones :-)

@lebrush thanks! (: let's wait for Murdock now and merge as soon as it turns green.

cppcheck found some issues and some of the Cortex-M0 ports fail.

@OlegHahm why tries to compile it if the periph_i2c requirement is not satisfied?

Compiling it manually returns: "There are unsatisfied feature requirements: periph_i2c. Expect Errors"

The error of cppcheck:

Running './dist/tools/cppcheck/check.sh master --diff-filter=AC' x
Command output:

	drivers/pn532/pn532.c:135: style (knownConditionTrueFalse): Condition 'ret==0' is always false

Refers to the code:

    int ret = -1;
(...)
    if (mode == PN532_I2C) {
#ifdef PN532_SUPPORT_I2C
        ret = i2c_init_master(dev->conf->i2c, I2C_SPEED_FAST);
#endif
    }
    else {
#ifdef PN532_SUPPORT_SPI
        ret = spi_init_master(dev->conf->spi, SPI_CONF_FIRST_RISING,
                              SPI_SPEED_1MHZ);
        gpio_init(dev->conf->nss, GPIO_OUT);
        gpio_set(dev->conf->nss);
#endif
    }

--> if (ret == 0) {
        pn532_reset(dev);
    }

Which does only apply when PN532_SUPPORT_I2C or PN532_SUPPORT_SPI are set. At least one of them must be set, otherwise the compilation fails and an informative error is displayed (pn532.h:31):

#if !defined(PN532_SUPPORT_I2C) && !defined(PN532_SUPPORT_SPI)
#error Please define PN532_SUPPORT_I2C and/or PN532_SUPPORT_SPI to enable \
    the functionality on this device
#endif

I would ignore this one otherwise please suggest how to fix it

I would ignore this one otherwise please suggest how to fix it

If so, please suppress the cppcheck warning with a corresponding reasoning (take #6024 as a reference about how to do this).

@miri64 thanks!

    /* cppcheck-suppress knownConditionTrueFalse
     * (reason: variable set when PN532_SUPPORT_{I2C,SPI} defined) */
    if (ret == 0) {
        pn532_reset(dev);
    }

Should do the trick right? What about the compilation on platforms which don't have periph_i2c?

This you should do in the build system AFAIK, using the FEATURE_REQUIRED macro. But better ask someone more knowledgable about drivers like @haukepetersen or @PeterKietzmann.

Oh and please squash all commits to one.