Write a test case for empty NFC tag because of bug 997094

Please this into backlog.

defer to sprint3.

Comment on attachment 8450816 [details] [diff] [review] NFC add tag clear command v1 Review of attachment 8450816 [details] [diff] [review]: ----------------------------------------------------------------- This looks very good already. Thanks for the patch. ::: hw/nfc-tag.c @@ +79,5 @@ > { > ssize_t offset = 0; > uint8_t* data; > > assert(tag); Please use 'assert(ndef_msg || !len);'. This tests for either an NDEF message, or no message. @@ +91,5 @@ > > data[offset++] = NDEF_MESSAGE_TLV; > data[offset++] = len; > > + if (len) { I'm not sure, but I think you can just use '0' as length argument for memcpy. @@ +92,5 @@ > data[offset++] = NDEF_MESSAGE_TLV; > data[offset++] = len; > > + if (len) { > + assert(ndef_msg); Should be removed with comment above addressed. @@ +103,5 @@ > memset(data + offset, 0, sizeof(tag->t.t1.format.data) - offset); > } > > static void > set_t2t_data(struct nfc_tag* tag, const uint8_t* ndef_msg, ssize_t len) See my comments for |set_t1t_data|. @@ +128,5 @@ > memset(data + offset, 0, sizeof(tag->t.t2.format.data) - offset); > } > > static void > set_t3t_data(struct nfc_tag* tag, const uint8_t* ndef_msg, ssize_t len) See my comments for |set_t1t_data|. @@ +158,4 @@ > } > > static void > set_t4t_data(struct nfc_tag* tag, const uint8_t* ndef_msg, ssize_t len) See my comments for |set_t1t_data|.

Comment on attachment 8450819 [details] [diff] [review] Testcase v1 Review of attachment 8450819 [details] [diff] [review]: ----------------------------------------------------------------- This looks good. Thank you. ::: dom/nfc/tests/marionette/test_nfc_read_tag.js @@ +9,5 @@ > +// TODO : Get this from emulator console command. > +const T1T_RE_INDEX = 2; > +const T2T_RE_INDEX = 3; > +const T3T_RE_INDEX = 4; > +const T4T_RE_INDEX = 5; Would it make sense to move these constants to head.js in another cleanup bug? There could also be P2P_RE_INDEX_*. @@ +103,5 @@ > + testUrlT1TDiscover, > + testUrlT2TDiscover, > + testUrlT3TDiscover, > + testUrlT4TDiscover > +]; Would it make sense to first run the Url test and afterwards run the Empty tests? This way, the 'clear' operation would really *clear* an NDEF message from the tag, and not just set an empty message.

(In reply to Thomas Zimmermann [:tzimmermann] [:tdz] from comment #7) > Comment on attachment 8450819 [details] [diff] [review] > Testcase v1 > > Review of attachment 8450819 [details] [diff] [review]: > ----------------------------------------------------------------- > > This looks good. Thank you. > > ::: dom/nfc/tests/marionette/test_nfc_read_tag.js > @@ +9,5 @@ > > +// TODO : Get this from emulator console command. > > +const T1T_RE_INDEX = 2; > > +const T2T_RE_INDEX = 3; > > +const T3T_RE_INDEX = 4; > > +const T4T_RE_INDEX = 5; > > Would it make sense to move these constants to head.js in another cleanup > bug? There could also be P2P_RE_INDEX_*. > I will create another cleanup bug for it.

Address Thomas's comment

merged: https://github.com/mozilla-b2g/platform_external_qemu/commit/c61e5f15fd62888f2c33d7d542b5b65c38102e8b