def separate_regex_simple(voteddir, regexShr, regexDiff):
ballots = []
for dirpath, dirnames, filenames in os.walk(voteddir):
imgnames = [f for f in filenames if util.is_image_ext(f)]
shrPat = re.compile(regexShr)
diffPat = re.compile(regexDiff)
curmats = {} # maps {str sim_pat: [(str imgpath, str diff_pat), ...]}
for imgname in imgnames:
imgpath = pathjoin(dirpath, imgname)
sim_match = shrPat.match(imgname)
diff_match = diffPat.match(imgname)
if sim_match is None or diff_match is None:
warn(
"Ballot {0} was skipped because it didn't "
"match the regular expressions.", imgpath)
continue
sim_part = sim_match.groups()[0]
diff_part = diff_match.groups()[0]
curmats.setdefault(sim_part, []).append((imgpath, diff_part))
for sim_pat, tuples in curmats.iteritems():
# sort by diffPart
tuples_sorted = sorted(tuples, key=lambda t: t[1])
imgpaths_sorted = [t[0] for t in tuples_sorted]
ballots.append(imgpaths_sorted)
return ballots
def separate_regex_ctr(voteddir, regexShr):
""" Separates ballots whose filenames start with a shared prefix
REGEXSHR, but then contain two incrementing counters (very-much
Hart-specific), i.e. for the following images:
Ballot A:
339_1436_5_211_1.png
339_1436_5_212_2.png
339_1436_5_213_3.png
Ballot B:
339_1436_5_214_1.png
339_1436_5_215_2.png
"""
ballots = []
shrPat = re.compile(regexShr)
for dirpath, dirnames, filenames in os.walk(voteddir):
imgnames = [f for f in filenames if util.is_image_ext(f)]
curmats = {} # maps {str sim_pat: [(str imgpath, tuple ctr_vals), ...]}
for imgname in imgnames:
imgpath = pathjoin(dirpath, imgname)
sim_match = shrPat.match(imgname)
if sim_match == None:
print "Warning: ballot {0} was skipped because it didn't \
match the regular expressions.".format(imgpath)
continue
sim_part = sim_match.groups()[0]
# Assumes filename is := <SIM_PART>_N1_N2.png
ctr_vals = [int(n) for n in os.path.splitext(imgname)[0][len(sim_part):].split("_")]
curmats.setdefault(sim_part, []).append((imgpath, ctr_vals))
for sim_pat, tuples in curmats.iteritems():
# tuple TUPLES := [(str imgpath, (int N1, int N2)), ...]
consecs = get_consecutives(tuples)
for imgpaths in consecs:
ballots.append(imgpaths)
return ballots
def separate_singlesided(voteddir):
ballots = []
for dirpath, dirnames, filenames in os.walk(voteddir):
imgnames = [f for f in filenames if util.is_image_ext(f)]
for imgname in imgnames:
imgpath = pathjoin(dirpath, imgname)
ballots.append([imgpath])
return ballots
def get_all_images(directory):
'''
Walk a given directory and return a list of all the filenames
that are images.
'''
images = []
for _, _, filenames in os.walk(directory):
images.extend(f for f in filenames if util.is_image_ext(f))
return images
def separate_alternating(voteddir, num_pages):
ballots = []
for dirpath, dirnames, filenames in os.walk(voteddir):
imgnames = [f for f in filenames if util.is_image_ext(f)]
imgnames_ordered = util.sorted_nicely(imgnames)
if len(imgnames_ordered) % num_pages != 0:
print "Uh oh -- there are {0} images in directory {1}, \
which isn't divisible by num_pages {2}".format(len(imgnames_ordered), dirpath, num_pages)
pdb.set_trace()
raise RuntimeError
i = 0
while imgnames_ordered:
curballot = []
for j in xrange(num_pages):
imgpath = pathjoin(dirpath, imgnames_ordered.pop(0))
curballot.append(imgpath)
ballots.append(curballot)
return ballots
def separate_regex_simple(voteddir, regexShr, regexDiff):
ballots = []
for dirpath, dirnames, filenames in os.walk(voteddir):
imgnames = [f for f in filenames if util.is_image_ext(f)]
shrPat = re.compile(regexShr)
diffPat = re.compile(regexDiff)
curmats = {} # maps {str sim_pat: [(str imgpath, str diff_pat), ...]}
for imgname in imgnames:
imgpath = pathjoin(dirpath, imgname)
sim_match = shrPat.match(imgname)
diff_match = diffPat.match(imgname)
if sim_match == None or diff_match == None:
print "Warning: ballot {0} was skipped because it didn't \
match the regular expressions.".format(imgpath)
continue
sim_part = sim_match.groups()[0]
diff_part = diff_match.groups()[0]
curmats.setdefault(sim_part, []).append((imgpath, diff_part))
for sim_pat, tuples in curmats.iteritems():
# sort by diffPart
tuples_sorted = sorted(tuples, key=lambda t: t[1])
imgpaths_sorted = [t[0] for t in tuples_sorted]
ballots.append(imgpaths_sorted)
return ballots
def separate_alternating(voteddir, num_pages):
ballots = []
for dirpath, dirnames, filenames in os.walk(voteddir):
imgnames = [f for f in filenames if util.is_image_ext(f)]
imgnames_ordered = util.sorted_nicely(imgnames)
if len(imgnames_ordered) % num_pages != 0:
raise ffwx.Panel.StepNotFinished(
('The selected directory:\n\n'
' {0}\n\n'
'contains {1} images, but the '
'selected options indicate that pages should '
'come in sets of {2}, and {1} isn\'t divisible '
'by {2}.\n\n'
'Try double-checking your chosen settings and your '
'selected ballot directory.'.format(dirpath,
len(imgnames_ordered),
num_pages)))
while imgnames_ordered:
curballot = []
for j in xrange(num_pages):
imgpath = pathjoin(dirpath, imgnames_ordered.pop(0))
curballot.append(imgpath)
ballots.append(curballot)
return ballots