1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
|
##
## This file is part of the libsigrokdecode project.
##
## Copyright (C) 2018 Stefan Brüns <stefan.bruens@rwth-aachen.de>
##
## This program is free software; you can redistribute it and/or modify
## it under the terms of the GNU General Public License as published by
## the Free Software Foundation; either version 2 of the License, or
## (at your option) any later version.
##
## This program is distributed in the hope that it will be useful,
## but WITHOUT ANY WARRANTY; without even the implied warranty of
## MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
## GNU General Public License for more details.
##
## You should have received a copy of the GNU General Public License
## along with this program; if not, see <http://www.gnu.org/licenses/>.
##
import sigrokdecode as srd
PIN_DATA, PIN_RESET = range(2)
ROW_EDGE, ROW_WORD, ROW_RESET = range(3)
class Decoder(srd.Decoder):
api_version = 3
id = 'counter'
name = 'Counter'
longname = 'Edge counter'
desc = 'Count number of edges.'
license = 'gplv2+'
inputs = ['logic']
outputs = []
channels = (
{'id': 'data', 'name': 'Data', 'desc': 'Data line'},
)
optional_channels = (
{'id': 'reset', 'name': 'Reset', 'desc': 'Reset line'},
)
annotations = (
('edge_count', 'Edge count'),
('word_count', 'Word count'),
('word_reset', 'Word reset'),
)
annotation_rows = (
('edge_counts', 'Edges', (ROW_EDGE,)),
('word_counts', 'Words', (ROW_WORD,)),
('word_resets', 'Word resets', (ROW_RESET,)),
)
options = (
{'id': 'data_edge', 'desc': 'Edges to count (data)', 'default': 'any',
'values': ('any', 'rising', 'falling')},
{'id': 'divider', 'desc': 'Count divider (word width)', 'default': 0},
{'id': 'reset_edge', 'desc': 'Edge which clears counters (reset)',
'default': 'falling', 'values': ('rising', 'falling')},
)
def __init__(self):
self.reset()
def reset(self):
pass
def metadata(self, key, value):
if key == srd.SRD_CONF_SAMPLERATE:
self.samplerate = value
def start(self):
self.out_ann = self.register(srd.OUTPUT_ANN)
def putc(self, cls, ss, annlist):
self.put(ss, self.samplenum, self.out_ann, [cls, annlist])
def decode(self):
opt_edge_map = {'rising': 'r', 'falling': 'f', 'any': 'e'}
data_edge = self.options['data_edge']
divider = self.options['divider']
if divider < 0:
divider = 0
reset_edge = self.options['reset_edge']
condition = [{PIN_DATA: opt_edge_map[data_edge]}]
have_reset = self.has_channel(PIN_RESET)
if have_reset:
cond_reset = len(condition)
condition.append({PIN_RESET: opt_edge_map[reset_edge]})
edge_count = 0
edge_start = None
word_count = 0
word_start = None
while True:
self.wait(condition)
now = self.samplenum
if have_reset and self.matched[cond_reset]:
edge_count = 0
edge_start = now
word_count = 0
word_start = now
self.putc(ROW_RESET, now, ['Word reset', 'Reset', 'Rst', 'R'])
continue
# Implementation note: In the absence of a RESET condition
# before the first data edge, any arbitrary choice of where
# to start the annotation is valid. One may choose to emit a
# narrow annotation (where ss=es), or assume that the cycle
# which corresponds to the counter value started at sample
# number 0. We decided to go with the latter here, to avoid
# narrow annotations (see bug #1210). None of this matters in
# the presence of a RESET condition in the input stream.
if edge_start is None:
edge_start = 0
if word_start is None:
word_start = 0
edge_count += 1
self.putc(ROW_EDGE, edge_start, ["{:d}".format(edge_count)])
edge_start = now
if divider and (edge_count % divider) == 0:
word_count += 1
self.putc(ROW_WORD, word_start, ["{:d}".format(word_count)])
word_start = now
|