169 lines
4.6 KiB
Python
169 lines
4.6 KiB
Python
|
#!/usr/bin/python3
|
||
|
|
||
|
import sys
|
||
|
import glob
|
||
|
import re
|
||
|
import matplotlib.pyplot as plt
|
||
|
import numpy as np
|
||
|
|
||
|
def main(argv):
|
||
|
|
||
|
# Load file
|
||
|
assert len(argv) >= 2, "No arguments given. -h for help"
|
||
|
|
||
|
if argv[1] == "-h":
|
||
|
print("""Specify the path to prsim.out either by giving the full path,
|
||
|
or the folder name like 'buf_15'.
|
||
|
Use -include='regex' to specify signals to include (or -in).
|
||
|
Use -exclude='regex' to specify signals to exclude (or -ex).""")
|
||
|
return
|
||
|
|
||
|
|
||
|
file_path = argv[1]
|
||
|
if not ".out" in file_path:
|
||
|
file_path = f"./unit_tests/{file_path}/run/prsim.out"
|
||
|
assert len(glob.glob(file_path)) >= 1, "prsim.out file not found!"
|
||
|
print(f"Loading {file_path}")
|
||
|
f = open(file_path,'r').read()
|
||
|
|
||
|
# Start regexxing
|
||
|
entries = re.findall(r"\t *(\d+) ([^:]+) : (\d)( \[by.+\])?[\n\r]", f)
|
||
|
assert len(entries) >= 1, "Could not find signal info in prsim.out!"
|
||
|
|
||
|
# Check if user gave an include filter
|
||
|
include_given = False
|
||
|
include_re = None
|
||
|
for arg in argv:
|
||
|
r = re.findall(r'(-include|-in)=(.+)', arg)
|
||
|
if len(r) >= 1:
|
||
|
include_given = True
|
||
|
include_re = r[0][1]
|
||
|
|
||
|
# Check if user gave an exclude filter
|
||
|
exclude_given = False
|
||
|
exclude_re = None
|
||
|
for arg in argv:
|
||
|
r = re.findall(r'(-exclude|-ex)=(.+)', arg)
|
||
|
if len(r) >= 1:
|
||
|
exclude_given = True
|
||
|
exclude_re = r[0][1]
|
||
|
|
||
|
assert not (exclude_given and include_given), "Can't give include and exclude re simultaneously."
|
||
|
if include_given: print(f"Including signals that match regex {include_re}")
|
||
|
if exclude_given: print(f"Excluding signals that match regex {exclude_re}")
|
||
|
|
||
|
if include_given:
|
||
|
entries = [e for e in entries if not re.search(include_re, e[1]) == None]
|
||
|
|
||
|
if exclude_given:
|
||
|
entries = [e for e in entries if re.search(exclude_re, e[1]) == None]
|
||
|
|
||
|
assert len(entries) >= 1, "No valid entries in prsim.out!"
|
||
|
|
||
|
# Get list of all sigs and times
|
||
|
times = np.array([int(e[0]) for e in entries])
|
||
|
unique_times = np.unique(times)
|
||
|
num_times = unique_times.shape[0]
|
||
|
|
||
|
sigs = np.array([e[1] for e in entries])
|
||
|
unique_sigs = np.unique(sigs)
|
||
|
num_sigs = unique_sigs.shape[0]
|
||
|
|
||
|
print(f"Plotting signals:")
|
||
|
print(unique_sigs)
|
||
|
|
||
|
|
||
|
# Some functions to order everything nicely
|
||
|
# Should probably put these outside but whatever.
|
||
|
def time_to_index(time):
|
||
|
'''
|
||
|
Since times are random, need to convert them to an index.
|
||
|
'''
|
||
|
out = np.argwhere(unique_times == int(time))
|
||
|
return out[0][0]
|
||
|
|
||
|
def sig_to_index(sig):
|
||
|
'''
|
||
|
Handles signal name ordering.
|
||
|
Assume ordered like unique_sigs for now
|
||
|
'''
|
||
|
out = np.argwhere(unique_sigs == sig)
|
||
|
return out[0][0]
|
||
|
|
||
|
# Create matrix of signals over time and populate
|
||
|
signals_matrix = np.zeros((num_sigs, num_times), dtype = int)
|
||
|
|
||
|
for sig in unique_sigs:
|
||
|
entries_filtered = [e for e in entries if e[1] == sig]
|
||
|
# make sure sorted
|
||
|
entries_filtered = sorted(entries_filtered, key = lambda e: int(e[0]))
|
||
|
for e in entries_filtered:
|
||
|
val = int(e[2])
|
||
|
val = 2*val -1
|
||
|
signals_matrix[sig_to_index(sig),time_to_index(e[0]):] = val
|
||
|
|
||
|
# Plot
|
||
|
|
||
|
colour_undefined = (255,0,0)
|
||
|
colour_high = (252, 186, 3)
|
||
|
colour_low = (20, 184, 186)
|
||
|
|
||
|
fig = plt.figure(figsize = (num_sigs/3,num_times/3))
|
||
|
|
||
|
image = np.zeros((num_sigs, num_times, 3), dtype = int)
|
||
|
image[signals_matrix == 0] = colour_undefined
|
||
|
image[signals_matrix == 1] = colour_high
|
||
|
image[signals_matrix == -1] = colour_low
|
||
|
|
||
|
plt.imshow(image)
|
||
|
ax = fig.gca()
|
||
|
ax.set_xlabel("Time")
|
||
|
# ax.set_ylabel("Signal")
|
||
|
ax.set_yticks([])
|
||
|
|
||
|
for sig in unique_sigs:
|
||
|
ax.text(-1, sig_to_index(sig), sig, ha = "right", va = "center", size = 10)
|
||
|
|
||
|
for i in range(num_sigs-1):
|
||
|
ax.axhline(i+0.5, c = "white", lw = 2)
|
||
|
|
||
|
for i in range(num_times-1):
|
||
|
ax.axvline(i+0.5, c = "white", lw = 2)
|
||
|
|
||
|
ax.axis("off")
|
||
|
|
||
|
# Draw arrows
|
||
|
for e in entries:
|
||
|
# check if has a causal signal
|
||
|
by = re.findall(r"\[by (.+):=(\d)",e[3])
|
||
|
if len(by) == 0: continue
|
||
|
sig = e[1]
|
||
|
time = e[0]
|
||
|
t_index = time_to_index(time)
|
||
|
by_sig = by[0][0]
|
||
|
by_val = int(by[0][1])
|
||
|
t0,t1 = (t_index, t_index)
|
||
|
|
||
|
# The sig that caused the change might have been excluded from plotting
|
||
|
s0 = sig_to_index(sig)
|
||
|
if by_sig in unique_sigs:
|
||
|
s1 = sig_to_index(by_sig)
|
||
|
else:
|
||
|
s1 = s0
|
||
|
|
||
|
|
||
|
arrow_c = "black" if by_val == 1 else "grey"
|
||
|
if by_val == 1:
|
||
|
plt.arrow(t0, s1, 0, s0-s1 + 0.2*np.sign(s0-s1), head_width = 0.5, width = 0.2, ec = "none", lw = 0, fc = "black", length_includes_head = True)
|
||
|
else:
|
||
|
plt.arrow(t0, s1, 0, s0-s1 + 0.2*np.sign(s0-s1), head_width = 0, width = 0.2, ec = "none", lw = 0, fc = "black", length_includes_head = True)
|
||
|
plt.scatter((t0),(s0), c = "black", s = 30)
|
||
|
|
||
|
file_out_path = file_path.replace(".out",".pdf")
|
||
|
plt.savefig(file_out_path, bbox_inches = "tight")
|
||
|
|
||
|
|
||
|
if __name__ == "__main__":
|
||
|
# print(sys.argv[0:])
|
||
|
main(sys.argv)
|
||
|
# main(sys.argv[1:])
|