From bffbd5f03806836d5a55f65542ad6061a7a03ccb Mon Sep 17 00:00:00 2001
From: Philipp Hagemeister <phihag@phihag.de>
Date: Sat, 12 Jan 2013 20:34:50 +0100
Subject: [PATCH] Download progress hooks

---
 test/test_download.py        |  6 ++++
 youtube_dl/FileDownloader.py | 53 +++++++++++++++++++++++++++++++++++-
 2 files changed, 58 insertions(+), 1 deletion(-)

diff --git a/test/test_download.py b/test/test_download.py
index e36774de9..5877c42b3 100644
--- a/test/test_download.py
+++ b/test/test_download.py
@@ -82,6 +82,11 @@ def test_template(self):
         fd.add_info_extractor(ie())
         for ien in test_case.get('add_ie', []):
             fd.add_info_extractor(getattr(youtube_dl.InfoExtractors, ien + 'IE')())
+        finished_hook_called = set()
+        def _hook(status):
+            if status['status'] == 'finished':
+                finished_hook_called.add(status['filename'])
+        fd.add_progress_hook(_hook)
 
         test_cases = test_case.get('playlist', [test_case])
         for tc in test_cases:
@@ -94,6 +99,7 @@ def test_template(self):
             for tc in test_cases:
                 if not test_case.get('params', {}).get('skip_download', False):
                     self.assertTrue(os.path.exists(tc['file']))
+                    self.assertTrue(tc['file'] in finished_hook_called)
                 self.assertTrue(os.path.exists(tc['file'] + '.info.json'))
                 if 'md5' in tc:
                     md5_for_file = _file_md5(tc['file'])
diff --git a/youtube_dl/FileDownloader.py b/youtube_dl/FileDownloader.py
index 04ecd1ac5..e3131bbe6 100644
--- a/youtube_dl/FileDownloader.py
+++ b/youtube_dl/FileDownloader.py
@@ -95,6 +95,7 @@ def __init__(self, params):
         """Create a FileDownloader object with the given options."""
         self._ies = []
         self._pps = []
+        self._progress_hooks = []
         self._download_retcode = 0
         self._num_downloads = 0
         self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)]
@@ -594,8 +595,15 @@ def _download_with_rtmpdump(self, filename, url, player_url, page_url):
                 retval = 0
                 break
         if retval == 0:
-            self.to_screen(u'\r[rtmpdump] %s bytes' % os.path.getsize(encodeFilename(tmpfilename)))
+            fsize = os.path.getsize(encodeFilename(tmpfilename))
+            self.to_screen(u'\r[rtmpdump] %s bytes' % fsize)
             self.try_rename(tmpfilename, filename)
+            self._hook_progress({
+                'downloaded_bytes': fsize,
+                'total_bytes': fsize,
+                'filename': filename,
+                'status': 'finished',
+            })
             return True
         else:
             self.trouble(u'\nERROR: rtmpdump exited with code %d' % retval)
@@ -607,6 +615,10 @@ def _do_download(self, filename, info_dict):
         # Check file already present
         if self.params.get('continuedl', False) and os.path.isfile(encodeFilename(filename)) and not self.params.get('nopart', False):
             self.report_file_already_downloaded(filename)
+            self._hook_progress({
+                'filename': filename,
+                'status': 'finished',
+            })
             return True
 
         # Attempt to download using rtmpdump
@@ -678,6 +690,10 @@ def _do_download(self, filename, info_dict):
                             # the one in the hard drive.
                             self.report_file_already_downloaded(filename)
                             self.try_rename(tmpfilename, filename)
+                            self._hook_progress({
+                                'filename': filename,
+                                'status': 'finished',
+                            })
                             return True
                         else:
                             # The length does not match, we start the download over
@@ -736,6 +752,14 @@ def _do_download(self, filename, info_dict):
                 eta_str = self.calc_eta(start, time.time(), data_len - resume_len, byte_counter - resume_len)
                 self.report_progress(percent_str, data_len_str, speed_str, eta_str)
 
+            self._hook_progress({
+                'downloaded_bytes': byte_counter,
+                'total_bytes': data_len,
+                'tmpfilename': tmpfilename,
+                'filename': filename,
+                'status': 'downloading',
+            })
+
             # Apply rate limit
             self.slow_down(start, byte_counter - resume_len)
 
@@ -752,4 +776,31 @@ def _do_download(self, filename, info_dict):
         if self.params.get('updatetime', True):
             info_dict['filetime'] = self.try_utime(filename, data.info().get('last-modified', None))
 
+        self._hook_progress({
+            'downloaded_bytes': byte_counter,
+            'total_bytes': byte_counter,
+            'filename': filename,
+            'status': 'finished',
+        })
+
         return True
+
+    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"
+
+        It can also have some of the following entries:
+
+        * downloaded_bytes: Bytes on disks
+        * total_bytes: Total bytes, None if unknown
+        * tmpfilename: The filename we're currently writing to
+
+        Hooks are guaranteed to be called at least once (with status "finished")
+        if the download is successful.
+        """
+        self._progress_hooks.append(ph)