[path-] fix undercounted progress for multibyte chars #2323

Co-authored-by: @midichef

visidata/path.py

@@ -134,13 +134,13 @@ def read(self, size=-1):
  r = self.fp_orig_read(size)
  if self.prog:
  if r:
- self.prog.addProgress(len(r))
+ self.prog.addProgress(r)
  return r
 
  def readline(self, size=-1):
  r = self.fp_orig_readline(size)
  if self.prog:
- self.prog.addProgress(len(r))
+ self.prog.addProgress(r)
  return r
 
  def __getattr__(self, k):
@@ -152,15 +152,15 @@ def __enter__(self):
 
  def __next__(self):
  r = next(self.fp)
- self.prog.addProgress(len(r))
+ self.prog.addProgress(r)
  return r
 
  def __iter__(self):
  if not self.prog:
  yield from self.fp
  else:
  for line in self.fp:
- self.prog.addProgress(len(line))
+ self.prog.addProgress(line)
  yield line
 
  def __exit__(self, type, value, tb):
@@ -332,7 +332,7 @@ def __iter__(self):
  with Progress(total=filesize(self)) as prog:
  with self.open(encoding=vd.options.encoding) as fd:
  for i, line in enumerate(fd):
- prog.addProgress(len(line))
+ prog.addProgress(line)
  yield line.rstrip('\n')
 
  def open_bytes(self, mode='rb'):

visidata/threads.py

@@ -62,8 +62,10 @@ def __enter__(self):
  self.sheet.progresses.insert(0, self)
  return self
 
- def addProgress(self, n):
+ def addProgress(self, n:'int|str|bytes'):
  'Increase the progress count by *n*.'
+ if isinstance(n, (str, bytes)):
+ n = len(n)
  self.made += n
  return True
 
@@ -83,9 +85,13 @@ def __init__(self, encoding='utf-8', **kwargs):
  super().__init__(**kwargs)
  self.est_sample = ''
  self.est_charbytes = 1
+ self.encoding = encoding
 
- def addProgress(self, n:int):
+ def addProgress(self, n:'int|str|bytes'):
  if self.made < self.total:
+ if isinstance(n, (str, bytes)):
+ self.addSample(n)
+ n = len(n)
  return super().addProgress(n * self.est_charbytes)
 
  def addSample(self, s:str):