monster/yourovo
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix(urovo): encode before geometry — the print-setup burst silences the UHF module

Browse Source 
Every print-encode today failed its read-back with 'before: null' while a
standalone /read-tag a second earlier read the chip every time, at the same
media position. Bisected to the pre-read sequence itself: after the
SIZE/GAP/TEAR/DIRECTION/CLS burst, UHF READ and QUERY return no bytes at all
until a power-cycle. SET RIBBON OFF + FORMFEED provably do not do this.

UHF WRITE commits immediately and needs no geometry, and printing works after
geometry (print-test proved it), so read/write/verify now happen first on the
untouched module and the print setup follows. Nothing feeds between, so the
label at the antenna is the one printed.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
This commit is contained in:
type-two 2026-09-04 16:29:42 +10:00
parent c1801c8c34
commit f87804de0a
1 changed files with 23 additions and 14 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

37
mac/daemon.py
View File

@ -67,20 +67,17 @@ def do_print(body, encode):
return {'ok': False, 'error': str(e)}
with LOCK, Urovo() as p:
# 1) direct thermal. Wait out transient busy states before feeding to clear,
# so we don't waste a label on a momentary post-print code.
p.ribbon_off()
time.sleep(0.4)
if p.wait_ready(3.0) != 0x00:
st = p.recover()
if st != 0x00:
return {'ok': False, 'error': 'printer not ready',
'status': st, 'statusText': status_text(st)}
# 2) geometry, then the artwork
geometry(p, prof)
p.cls()
# ENCODE BEFORE GEOMETRY. Order matters on this firmware, and it cost most of a day:
# the SIZE/GAP/TEAR/DIRECTION/CLS burst that sets up a print silences the UHF module
# -- every UHF READ/QUERY afterwards returns no bytes at all until a power-cycle --
# so an encode placed after it always failed its own read-back with "before: null",
# which looks exactly like a chip that never reached the antenna. It had. The module
# was fine right up until the geometry went out (a standalone /read-tag a second
# earlier read the chip every time; SET RIBBON OFF + FORMFEED provably do not kill
# it). UHF WRITE commits immediately and needs no geometry, and printing works fine
# AFTER geometry, so: read, write, verify on the untouched module first; only then
# set up the print. Nothing feeds in between, so the label at the antenna is the one
# that prints.
before = verified = None
if encode:
# Deliberately NO `SET RFID` here. It was tried as an "arm the radio first" step and
@ -100,6 +97,18 @@ def do_print(body, encode):
return {'ok': False, 'error': 'RFID encode failed verification',
'epc': epc, 'epcBefore': before, 'readBack': p.read_uhf(),
'note': 'nothing printed -- no label wasted'}
# 1) direct thermal. Wait out transient busy states before feeding to clear,
# so we don't waste a label on a momentary post-print code.
p.ribbon_off()
time.sleep(0.4)
if p.wait_ready(3.0) != 0x00:
st = p.recover()
if st != 0x00:
return {'ok': False, 'error': 'printer not ready',
'status': st, 'statusText': status_text(st)}
# 2) geometry, then the artwork -- AFTER the encode, see above
geometry(p, prof)
p.cls()
p.write(label.to_tspl(label.render(body, prof)))
# 4) re-assert ribbon-off immediately before firing