From 64c464a144e2a96ec21a717d191217edda9107a4 Mon Sep 17 00:00:00 2001
From: pukkandan <pukkandan.ytdlp@gmail.com>
Date: Thu, 17 Nov 2022 08:40:34 +0530
Subject: [PATCH] [utils] Move `FileDownloader.parse_bytes` into utils

---
 yt_dlp/__init__.py          | 18 +++++++++---------
 yt_dlp/downloader/common.py |  9 ++-------
 yt_dlp/utils.py             | 19 ++++++++++++++-----
 3 files changed, 25 insertions(+), 21 deletions(-)

diff --git a/yt_dlp/__init__.py b/yt_dlp/__init__.py
index 726fb0685c..c03e6e6915 100644
--- a/yt_dlp/__init__.py
+++ b/yt_dlp/__init__.py
@@ -16,7 +16,6 @@ import sys
 
 from .compat import compat_shlex_quote
 from .cookies import SUPPORTED_BROWSERS, SUPPORTED_KEYRINGS
-from .downloader import FileDownloader
 from .downloader.external import get_external_downloader
 from .extractor import list_extractor_classes
 from .extractor.adobepass import MSO_INFO
@@ -50,6 +49,7 @@ from .utils import (
     format_field,
     int_or_none,
     match_filter_func,
+    parse_bytes,
     parse_duration,
     preferredencoding,
     read_batch_urls,
@@ -281,19 +281,19 @@ def validate_options(opts):
             raise ValueError(f'invalid {key} retry sleep expression {expr!r}')
 
     # Bytes
-    def parse_bytes(name, value):
+    def validate_bytes(name, value):
         if value is None:
             return None
-        numeric_limit = FileDownloader.parse_bytes(value)
+        numeric_limit = parse_bytes(value)
         validate(numeric_limit is not None, 'rate limit', value)
         return numeric_limit
 
-    opts.ratelimit = parse_bytes('rate limit', opts.ratelimit)
-    opts.throttledratelimit = parse_bytes('throttled rate limit', opts.throttledratelimit)
-    opts.min_filesize = parse_bytes('min filesize', opts.min_filesize)
-    opts.max_filesize = parse_bytes('max filesize', opts.max_filesize)
-    opts.buffersize = parse_bytes('buffer size', opts.buffersize)
-    opts.http_chunk_size = parse_bytes('http chunk size', opts.http_chunk_size)
+    opts.ratelimit = validate_bytes('rate limit', opts.ratelimit)
+    opts.throttledratelimit = validate_bytes('throttled rate limit', opts.throttledratelimit)
+    opts.min_filesize = validate_bytes('min filesize', opts.min_filesize)
+    opts.max_filesize = validate_bytes('max filesize', opts.max_filesize)
+    opts.buffersize = validate_bytes('buffer size', opts.buffersize)
+    opts.http_chunk_size = validate_bytes('http chunk size', opts.http_chunk_size)
 
     # Output templates
     def validate_outtmpl(tmpl, msg):
diff --git a/yt_dlp/downloader/common.py b/yt_dlp/downloader/common.py
index 8d110c3747..fe36332506 100644
--- a/yt_dlp/downloader/common.py
+++ b/yt_dlp/downloader/common.py
@@ -15,7 +15,6 @@ from ..minicurses import (
 from ..utils import (
     IDENTITY,
     NO_DEFAULT,
-    NUMBER_RE,
     LockingUnsupportedError,
     Namespace,
     RetryManager,
@@ -24,6 +23,7 @@ from ..utils import (
     encodeFilename,
     format_bytes,
     join_nonempty,
+    parse_bytes,
     remove_start,
     sanitize_open,
     shell_quote,
@@ -180,12 +180,7 @@ class FileDownloader:
     @staticmethod
     def parse_bytes(bytestr):
         """Parse a string indicating a byte quantity into an integer."""
-        matchobj = re.match(rf'(?i)^({NUMBER_RE})([kMGTPEZY]?)$', bytestr)
-        if matchobj is None:
-            return None
-        number = float(matchobj.group(1))
-        multiplier = 1024.0 ** 'bkmgtpezy'.index(matchobj.group(2).lower())
-        return int(round(number * multiplier))
+        parse_bytes(bytestr)
 
     def slow_down(self, start_time, now, byte_counter):
         """Sleep if the download speed is over the rate limit."""
diff --git a/yt_dlp/utils.py b/yt_dlp/utils.py
index 7cba13678f..9b6977b6df 100644
--- a/yt_dlp/utils.py
+++ b/yt_dlp/utils.py
@@ -2289,15 +2289,24 @@ def format_bytes(bytes):
     return format_decimal_suffix(bytes, '%.2f%sB', factor=1024) or 'N/A'
 
 
-def lookup_unit_table(unit_table, s):
+def lookup_unit_table(unit_table, s, strict=False):
+    num_re = NUMBER_RE if strict else NUMBER_RE.replace(R'\.', '[,.]')
     units_re = '|'.join(re.escape(u) for u in unit_table)
-    m = re.match(
-        r'(?P<num>[0-9]+(?:[,.][0-9]*)?)\s*(?P<unit>%s)\b' % units_re, s)
+    m = (re.fullmatch if strict else re.match)(
+        rf'(?P<num>{num_re})\s*(?P<unit>{units_re})\b', s)
     if not m:
         return None
-    num_str = m.group('num').replace(',', '.')
+
+    num = float(m.group('num').replace(',', '.'))
     mult = unit_table[m.group('unit')]
-    return int(float(num_str) * mult)
+    return round(num * mult)
+
+
+def parse_bytes(s):
+    """Parse a string indicating a byte quantity into an integer"""
+    return lookup_unit_table(
+        {u: 1024**i for i, u in enumerate(['', *'KMGTPEZY'])},
+        s.upper(), strict=True)
 
 
 def parse_filesize(s):