Initiate a new release.

[youtubedl] / youtube_dl / downloader / common.py

diff --git a/youtube_dl/downloader/common.py b/youtube_dl/downloader/common.py
index f85f0c94e7544b6eabad3d1f4158f88b9e3c981d..7bb3a948d2ebd0eaca46ca72f72dfaa2e7ffd1bd 100644 (file)
--- a/youtube_dl/downloader/common.py
+++ b/youtube_dl/downloader/common.py
@@ -1,10 +1,12 @@
+from __future__ import unicode_literals
+

 import os
 import re
 import sys
 import time
 
 import os
 import re
 import sys
 import time
 

+from ..compat import compat_str

 from ..utils import (
 from ..utils import (

-    compat_str,

     encodeFilename,
     format_bytes,
     timeconvert,
     encodeFilename,
     format_bytes,
     timeconvert,


@@ -23,21 +25,23 @@ class FileDownloader(object):
 
     Available options:
 
 
     Available options:
 

-    verbose:           Print additional info to stdout.
-    quiet:             Do not print messages to stdout.
-    ratelimit:         Download speed limit, in bytes/sec.
-    retries:           Number of times to retry for HTTP error 5xx
-    buffersize:        Size of download buffer in bytes.
-    noresizebuffer:    Do not automatically resize the download buffer.
-    continuedl:        Try to continue downloads if possible.
-    noprogress:        Do not print the progress bar.
-    logtostderr:       Log messages to stderr instead of stdout.
-    consoletitle:      Display progress in console window's titlebar.
-    nopart:            Do not use temporary .part files.
-    updatetime:        Use the Last-modified header to set output file timestamps.
-    test:              Download only first bytes to test the downloader.
-    min_filesize:      Skip files smaller than this size
-    max_filesize:      Skip files larger than this size
+    verbose:            Print additional info to stdout.
+    quiet:              Do not print messages to stdout.
+    ratelimit:          Download speed limit, in bytes/sec.
+    retries:            Number of times to retry for HTTP error 5xx
+    buffersize:         Size of download buffer in bytes.
+    noresizebuffer:     Do not automatically resize the download buffer.
+    continuedl:         Try to continue downloads if possible.
+    noprogress:         Do not print the progress bar.
+    logtostderr:        Log messages to stderr instead of stdout.
+    consoletitle:       Display progress in console window's titlebar.
+    nopart:             Do not use temporary .part files.
+    updatetime:         Use the Last-modified header to set output file timestamps.
+    test:               Download only first bytes to test the downloader.
+    min_filesize:       Skip files smaller than this size
+    max_filesize:       Skip files larger than this size
+    xattr_set_filesize: Set ytdl.filesize user xattribute with expected size.
+                        (experimenatal)

 
     Subclasses of this one must re-define the real_download method.
     """
 
     Subclasses of this one must re-define the real_download method.
     """


@@ -78,8 +82,10 @@ class FileDownloader(object):
     def calc_eta(start, now, total, current):
         if total is None:
             return None
     def calc_eta(start, now, total, current):
         if total is None:
             return None

+        if now is None:
+            now = time.time()

         dif = now - start
         dif = now - start

-        if current == 0 or dif < 0.001: # One millisecond
+        if current == 0 or dif < 0.001:  # One millisecond

             return None
         rate = float(current) / dif
         return int((float(total) - float(current)) / rate)
             return None
         rate = float(current) / dif
         return int((float(total) - float(current)) / rate)


@@ -93,7 +99,7 @@ class FileDownloader(object):
     @staticmethod
     def calc_speed(start, now, bytes):
         dif = now - start
     @staticmethod
     def calc_speed(start, now, bytes):
         dif = now - start

-        if bytes == 0 or dif < 0.001: # One millisecond
+        if bytes == 0 or dif < 0.001:  # One millisecond

             return None
         return float(bytes) / dif
 
             return None
         return float(bytes) / dif
 


@@ -106,7 +112,7 @@ class FileDownloader(object):
     @staticmethod
     def best_block_size(elapsed_time, bytes):
         new_min = max(bytes / 2.0, 1.0)
     @staticmethod
     def best_block_size(elapsed_time, bytes):
         new_min = max(bytes / 2.0, 1.0)

-        new_max = min(max(bytes * 2.0, 1.0), 4194304) # Do not surpass 4 MB
+        new_max = min(max(bytes * 2.0, 1.0), 4194304)  # Do not surpass 4 MB

         if elapsed_time < 0.001:
             return int(new_max)
         rate = bytes / elapsed_time
         if elapsed_time < 0.001:
             return int(new_max)
         rate = bytes / elapsed_time


@@ -144,29 +150,30 @@ class FileDownloader(object):
     def report_error(self, *args, **kargs):
         self.ydl.report_error(*args, **kargs)
 
     def report_error(self, *args, **kargs):
         self.ydl.report_error(*args, **kargs)
 

-    def slow_down(self, start_time, byte_counter):
+    def slow_down(self, start_time, now, byte_counter):

         """Sleep if the download speed is over the rate limit."""
         rate_limit = self.params.get('ratelimit', None)
         if rate_limit is None or byte_counter == 0:
             return
         """Sleep if the download speed is over the rate limit."""
         rate_limit = self.params.get('ratelimit', None)
         if rate_limit is None or byte_counter == 0:
             return

-        now = time.time()
+        if now is None:
+            now = time.time()

         elapsed = now - start_time
         if elapsed <= 0.0:
             return
         speed = float(byte_counter) / elapsed
         if speed > rate_limit:
         elapsed = now - start_time
         if elapsed <= 0.0:
             return
         speed = float(byte_counter) / elapsed
         if speed > rate_limit:

-            time.sleep((byte_counter - rate_limit * (now - start_time)) / rate_limit)
+            time.sleep(max((byte_counter // rate_limit) - elapsed, 0))

 
     def temp_name(self, filename):
         """Returns a temporary filename for the given filename."""
 
     def temp_name(self, filename):
         """Returns a temporary filename for the given filename."""

-        if self.params.get('nopart', False) or filename == u'-' or \
+        if self.params.get('nopart', False) or filename == '-' or \

                 (os.path.exists(encodeFilename(filename)) and not os.path.isfile(encodeFilename(filename))):
             return filename
                 (os.path.exists(encodeFilename(filename)) and not os.path.isfile(encodeFilename(filename))):
             return filename

-        return filename + u'.part'
+        return filename + '.part'

 
     def undo_temp_name(self, filename):
 
     def undo_temp_name(self, filename):

-        if filename.endswith(u'.part'):
-            return filename[:-len(u'.part')]
+        if filename.endswith('.part'):
+            return filename[:-len('.part')]

         return filename
 
     def try_rename(self, old_filename, new_filename):
         return filename
 
     def try_rename(self, old_filename, new_filename):


@@ -175,7 +182,7 @@ class FileDownloader(object):
                 return
             os.rename(encodeFilename(old_filename), encodeFilename(new_filename))
         except (IOError, OSError) as err:
                 return
             os.rename(encodeFilename(old_filename), encodeFilename(new_filename))
         except (IOError, OSError) as err:

-            self.report_error(u'unable to rename file: %s' % compat_str(err))
+            self.report_error('unable to rename file: %s' % compat_str(err))

 
     def try_utime(self, filename, last_modified_hdr):
         """Try to set the last-modified time of the given file."""
 
     def try_utime(self, filename, last_modified_hdr):
         """Try to set the last-modified time of the given file."""


@@ -200,10 +207,10 @@ class FileDownloader(object):
 
     def report_destination(self, filename):
         """Report destination filename."""
 
     def report_destination(self, filename):
         """Report destination filename."""

-        self.to_screen(u'[download] Destination: ' + filename)
+        self.to_screen('[download] Destination: ' + filename)

 
     def _report_progress_status(self, msg, is_last_line=False):
 
     def _report_progress_status(self, msg, is_last_line=False):

-        fullmsg = u'[download] ' + msg
+        fullmsg = '[download] ' + msg

         if self.params.get('progress_with_newline', False):
             self.to_screen(fullmsg)
         else:
         if self.params.get('progress_with_newline', False):
             self.to_screen(fullmsg)
         else:


@@ -211,13 +218,13 @@ class FileDownloader(object):
                 prev_len = getattr(self, '_report_progress_prev_line_length',
                                    0)
                 if prev_len > len(fullmsg):
                 prev_len = getattr(self, '_report_progress_prev_line_length',
                                    0)
                 if prev_len > len(fullmsg):

-                    fullmsg += u' ' * (prev_len - len(fullmsg))
+                    fullmsg += ' ' * (prev_len - len(fullmsg))

                 self._report_progress_prev_line_length = len(fullmsg)
                 self._report_progress_prev_line_length = len(fullmsg)

-                clear_line = u'\r'
+                clear_line = '\r'

             else:
             else:

-                clear_line = (u'\r\x1b[K' if sys.stderr.isatty() else u'\r')
+                clear_line = ('\r\x1b[K' if sys.stderr.isatty() else '\r')

             self.to_screen(clear_line + fullmsg, skip_eol=not is_last_line)
             self.to_screen(clear_line + fullmsg, skip_eol=not is_last_line)

-        self.to_console_title(u'youtube-dl ' + msg)
+        self.to_console_title('youtube-dl ' + msg)

 
     def report_progress(self, percent, data_len_str, speed, eta):
         """Report download progress."""
 
     def report_progress(self, percent, data_len_str, speed, eta):
         """Report download progress."""


@@ -233,7 +240,7 @@ class FileDownloader(object):
             percent_str = 'Unknown %'
         speed_str = self.format_speed(speed)
 
             percent_str = 'Unknown %'
         speed_str = self.format_speed(speed)
 

-        msg = (u'%s of %s at %s ETA %s' %
+        msg = ('%s of %s at %s ETA %s' %

                (percent_str, data_len_str, speed_str, eta_str))
         self._report_progress_status(msg)
 
                (percent_str, data_len_str, speed_str, eta_str))
         self._report_progress_status(msg)
 


@@ -243,44 +250,56 @@ class FileDownloader(object):
         downloaded_str = format_bytes(downloaded_data_len)
         speed_str = self.format_speed(speed)
         elapsed_str = FileDownloader.format_seconds(elapsed)
         downloaded_str = format_bytes(downloaded_data_len)
         speed_str = self.format_speed(speed)
         elapsed_str = FileDownloader.format_seconds(elapsed)

-        msg = u'%s at %s (%s)' % (downloaded_str, speed_str, elapsed_str)
+        msg = '%s at %s (%s)' % (downloaded_str, speed_str, elapsed_str)

         self._report_progress_status(msg)
 
     def report_finish(self, data_len_str, tot_time):
         """Report download finished."""
         if self.params.get('noprogress', False):
         self._report_progress_status(msg)
 
     def report_finish(self, data_len_str, tot_time):
         """Report download finished."""
         if self.params.get('noprogress', False):

-            self.to_screen(u'[download] Download completed')
+            self.to_screen('[download] Download completed')

         else:
             self._report_progress_status(
         else:
             self._report_progress_status(

-                (u'100%% of %s in %s' %
+                ('100%% of %s in %s' %

                  (data_len_str, self.format_seconds(tot_time))),
                 is_last_line=True)
 
     def report_resuming_byte(self, resume_len):
         """Report attempt to resume at given byte."""
                  (data_len_str, self.format_seconds(tot_time))),
                 is_last_line=True)
 
     def report_resuming_byte(self, resume_len):
         """Report attempt to resume at given byte."""

-        self.to_screen(u'[download] Resuming download at byte %s' % resume_len)
+        self.to_screen('[download] Resuming download at byte %s' % resume_len)

 
     def report_retry(self, count, retries):
         """Report retry in case of HTTP error 5xx"""
 
     def report_retry(self, count, retries):
         """Report retry in case of HTTP error 5xx"""

-        self.to_screen(u'[download] Got server HTTP error. Retrying (attempt %d of %d)...' % (count, retries))
+        self.to_screen('[download] Got server HTTP error. Retrying (attempt %d of %d)...' % (count, retries))

 
     def report_file_already_downloaded(self, file_name):
         """Report file has already been fully downloaded."""
         try:
 
     def report_file_already_downloaded(self, file_name):
         """Report file has already been fully downloaded."""
         try:

-            self.to_screen(u'[download] %s has already been downloaded' % file_name)
+            self.to_screen('[download] %s has already been downloaded' % file_name)

         except UnicodeEncodeError:
         except UnicodeEncodeError:

-            self.to_screen(u'[download] The file has already been downloaded')
+            self.to_screen('[download] The file has already been downloaded')

 
     def report_unable_to_resume(self):
         """Report it was impossible to resume download."""
 
     def report_unable_to_resume(self):
         """Report it was impossible to resume download."""

-        self.to_screen(u'[download] Unable to resume')
+        self.to_screen('[download] Unable to resume')

 
     def download(self, filename, info_dict):
         """Download to a filename using the info from info_dict
         Return True on success and False otherwise
         """
 
     def download(self, filename, info_dict):
         """Download to a filename using the info from info_dict
         Return True on success and False otherwise
         """

+
+        nooverwrites_and_exists = (
+            self.params.get('nooverwrites', False)
+            and os.path.exists(encodeFilename(filename))
+        )
+
+        continuedl_and_exists = (
+            self.params.get('continuedl', False)
+            and os.path.isfile(encodeFilename(filename))
+            and not self.params.get('nopart', False)
+        )
+

         # Check file already present
         # Check file already present

-        if self.params.get('continuedl', False) and os.path.isfile(encodeFilename(filename)) and not self.params.get('nopart', False):
+        if filename != '-' and nooverwrites_and_exists or continuedl_and_exists:

             self.report_file_already_downloaded(filename)
             self._hook_progress({
                 'filename': filename,
             self.report_file_already_downloaded(filename)
             self._hook_progress({
                 'filename': filename,


@@ -289,30 +308,43 @@ class FileDownloader(object):
             })
             return True
 
             })
             return True
 

+        sleep_interval = self.params.get('sleep_interval')
+        if sleep_interval:
+            self.to_screen('[download] Sleeping %s seconds...' % sleep_interval)
+            time.sleep(sleep_interval)
+

         return self.real_download(filename, info_dict)
 
     def real_download(self, filename, info_dict):
         """Real download process. Redefine in subclasses."""
         return self.real_download(filename, info_dict)
 
     def real_download(self, filename, info_dict):
         """Real download process. Redefine in subclasses."""

-        raise NotImplementedError(u'This method must be implemented by subclasses')
+        raise NotImplementedError('This method must be implemented by subclasses')

 
     def _hook_progress(self, status):
         for ph in self._progress_hooks:
             ph(status)
 
     def add_progress_hook(self, ph):
 
     def _hook_progress(self, status):
         for ph in self._progress_hooks:
             ph(status)
 
     def add_progress_hook(self, ph):

-        """ ph gets called on download progress, with a dictionary with the entries
-        * filename: The final filename
-        * status: One of "downloading" and "finished"
+        # See YoutubeDl.py (search for progress_hooks) for a description of
+        # this interface
+        self._progress_hooks.append(ph)

 
 

-        It can also have some of the following entries:
+    def _debug_cmd(self, args, subprocess_encoding, exe=None):
+        if not self.params.get('verbose', False):
+            return

 
 

-        * downloaded_bytes: Bytes on disks
-        * total_bytes: Total bytes, None if unknown
-        * tmpfilename: The filename we're currently writing to
-        * eta: The estimated time in seconds, None if unknown
-        * speed: The download speed in bytes/second, None if unknown
+        if exe is None:
+            exe = os.path.basename(args[0])

 
 

-        Hooks are guaranteed to be called at least once (with status "finished")
-        if the download is successful.
-        """
-        self._progress_hooks.append(ph)
+        if subprocess_encoding:
+            str_args = [
+                a.decode(subprocess_encoding) if isinstance(a, bytes) else a
+                for a in args]
+        else:
+            str_args = args
+        try:
+            import pipes
+            shell_quote = lambda args: ' '.join(map(pipes.quote, str_args))
+        except ImportError:
+            shell_quote = repr
+        self.to_screen('[debug] %s command line: %s' % (
+            exe, shell_quote(str_args)))