timing: user selected scale and sample count for terse display

Optionally let users pick the scale for terse timing annotation text.
Which potentially makes numbers show up earlier (at zoom levels of a
further distance). And drops the unit to present mere numbers, which
could speed up navigation during inspection. Keep providing automatic
scaling which then includes the unit text, as it did before.

Extend the automatic scaling to include picoseconds. Which avoids the
fallback to unit-less floating point with uncertain decimals when the
samplerate was 1GHz or higher.

Optionally present distances in terms of sample counts. This supports
decoder development, and can help users spot and judge glitches.

All "terse" presentation so far exclusively affects the 'time' row. It
remains an option for later to migrate averages and deltas as well. For
now it's assumed that high(er) precision and fine grained details are
more important for these rows.

1 files changed, 54 insertions, 21 deletions

diff --git a/decoders/timing/pd.py b/decoders/timing/pd.py
index b179ed6..95c5332 100644
--- a/decoders/timing/pd.py
+++ b/decoders/timing/pd.py
@@ -45,20 +45,49 @@ def normalize_time(t):
     else:
         return '%f' % t
 
-def terse_times(t):
-    if abs(t) >= 1e0:
-        t *= 1e0
-        return ['{:.0f}s'.format(t), '{:.0f}'.format(t)]
-    if abs(t) >= 1e-3:
-        t *= 1e3
-        return ['{:.0f}ms'.format(t), '{:.0f}'.format(t)]
-    if abs(t) >= 1e-6:
-        t *= 1e6
-        return ['{:.0f}us'.format(t), '{:.0f}'.format(t)]
-    if abs(t) >= 1e-9:
-        t *= 1e9
-        return ['{:.0f}ns'.format(t), '{:.0f}'.format(t)]
-    return ['{:f}'.format(t = t)]
+def terse_times(t, fmt):
+    # Strictly speaking these variants are not used in the current
+    # implementation, but can reduce diffs during future maintenance.
+    if fmt == 'full':
+        return [normalize_time(t)]
+    # End of "forward compatibility".
+
+    if fmt == 'samples':
+        # See below. No unit text, on purpose.
+        return ['{:d}'.format(t)]
+
+    # Use caller specified scale, or automatically find one.
+    scale, unit = None, None
+    if fmt == 'terse-auto':
+        if abs(t) >= 1e0:
+            scale, unit = 1e0, 's'
+        elif abs(t) >= 1e-3:
+            scale, unit = 1e3, 'ms'
+        elif abs(t) >= 1e-6:
+            scale, unit = 1e6, 'us'
+        elif abs(t) >= 1e-9:
+            scale, unit = 1e9, 'ns'
+        elif abs(t) >= 1e-12:
+            scale, unit = 1e12, 'ps'
+    # Beware! Uses unit-less text when the user picked the scale. For
+    # more consistent output with less clutter, thus faster navigation
+    # by humans. Can also un-hide text at higher distance zoom levels.
+    elif fmt == 'terse-s':
+        scale, unit = 1e0, ''
+    elif fmt == 'terse-ms':
+        scale, unit = 1e3, ''
+    elif fmt == 'terse-us':
+        scale, unit = 1e6, ''
+    elif fmt == 'terse-ns':
+        scale, unit = 1e9, ''
+    elif fmt == 'terse-ps':
+        scale, unit = 1e12, ''
+    if scale:
+        t *= scale
+        return ['{:.0f}{}'.format(t, unit), '{:.0f}'.format(t)]
+
+    # Unspecified format, and nothing auto-detected.
+    return ['{:f}'.format(t)]
 
 class Pin:
     (DATA,) = range(1)
@@ -96,8 +125,10 @@ class Decoder(srd.Decoder):
           'default': 'any', 'values': ('any', 'rising', 'falling') },
         { 'id': 'delta', 'desc': 'Show delta from last',
           'default': 'no', 'values': ('yes', 'no') },
-        { 'id': 'terse', 'desc': 'Show periods in terse format',
-          'default': 'no', 'values': ('yes', 'no') },
+        { 'id': 'format', 'desc': 'Format of \'time\' annotation',
+          'default': 'full', 'values': ('full', 'terse-auto',
+          'terse-s', 'terse-ms', 'terse-us', 'terse-ns', 'terse-ps',
+          'samples') },
     )
 
     def __init__(self):
@@ -119,7 +150,7 @@ class Decoder(srd.Decoder):
         edge = self.options['edge']
         avg_period = self.options['avg_period']
         delta = self.options['delta'] == 'yes'
-        terse = self.options['terse'] == 'yes'
+        fmt = self.options['format']
         ss = None
         last_n = deque()
         last_t = None
@@ -138,11 +169,13 @@ class Decoder(srd.Decoder):
             sa = es - ss
             t = sa / self.samplerate
 
-            if terse:
-                cls, txt = Ann.TERSE, terse_times(t)
-                self.put(ss, es, self.out_ann, [cls, txt])
-            else:
+            if fmt == 'full':
                 cls, txt = Ann.TIME, [normalize_time(t)]
+            elif fmt == 'samples':
+                cls, txt = Ann.TERSE, terse_times(sa, fmt)
+            else:
+                cls, txt = Ann.TERSE, terse_times(t, fmt)
+            if txt:
                 self.put(ss, es, self.out_ann, [cls, txt])
 
             if avg_period > 0: