Bump version to v0.6.3 for release

src/auto_archiver/core/media.py

@@ -89,7 +89,7 @@ class Media:
         try:
             streams = ffmpeg.probe(self.filename, select_streams='v')['streams']
             logger.warning(f"STREAMS FOR {self.filename} {streams}")
-            return any(s.get("duration_ts") > 0 for s in streams)
+            return any(s.get("duration_ts", 0) > 0 for s in streams)
         except Error: return False # ffmpeg errors when reading bad files
         except Exception as e:
             logger.error(e)

src/auto_archiver/core/metadata.py

@@ -139,7 +139,7 @@ class Metadata:
         new_media = []
         for m in self.media:
             h = m.get("hash")
-            if not h: h = calculate_hash_in_chunks(hashlib.sha256(), 1.6e7, m.filename)
+            if not h: h = calculate_hash_in_chunks(hashlib.sha256(), int(1.6e7), m.filename)
             if len(h) and h in media_hashes: continue
             media_hashes.add(h)
             new_media.append(m)

src/auto_archiver/enrichers/hash_enricher.py

@@ -23,7 +23,7 @@ class HashEnricher(Enricher):
     def configs() -> dict:
         return {
             "algorithm": {"default": "SHA-256", "help": "hash algorithm to use", "choices": ["SHA-256", "SHA3-512"]},
-            "chunksize": {"default": 1.6e7, "help": "number of bytes to use when reading files in chunks (if this value is too large you will run out of RAM), default is 16MB"},
+            "chunksize": {"default": int(1.6e7), "help": "number of bytes to use when reading files in chunks (if this value is too large you will run out of RAM), default is 16MB"},
         }
 
     def enrich(self, to_enrich: Metadata) -> None:

src/auto_archiver/enrichers/pdq_hash_enricher.py

@@ -26,11 +26,16 @@ class PdqHashEnricher(Enricher):
     def enrich(self, to_enrich: Metadata) -> None:
         url = to_enrich.get_url()
         logger.debug(f"calculating perceptual hashes for {url=}")
+        media_with_hashes = []
 
         for m in to_enrich.media:
             for media in m.all_inner_media(True):
-                if media.is_image() and "screenshot" not in media.get("id") and "warc-file-" not in media.get("id") and len(hd := self.calculate_pdq_hash(media.filename)):
+                media_id = media.get("id", "")
+                if media.is_image() and "screenshot" not in media_id and "warc-file-" not in media_id and len(hd := self.calculate_pdq_hash(media.filename)):
                     media.set("pdq_hash", hd)
+                    media_with_hashes.append(media.filename)
+
+        logger.debug(f"calculated '{len(media_with_hashes)}' perceptual hashes for {url=}: {media_with_hashes}")
 
     def calculate_pdq_hash(self, filename):
         # returns a hexadecimal string with the perceptual hash for the given filename

src/auto_archiver/enrichers/whisper_enricher.py

@@ -18,17 +18,18 @@ class WhisperEnricher(Enricher):
     def __init__(self, config: dict) -> None:
         # without this STEP.__init__ is not called
         super().__init__(config)
+        assert type(self.api_endpoint) == str and len(self.api_endpoint) > 0, "please provide a value for the whisper_enricher api_endpoint"
         assert type(self.api_key) == str and len(self.api_key) > 0, "please provide a value for the whisper_enricher api_key"
         self.timeout = int(self.timeout)
 
     @staticmethod
     def configs() -> dict:
         return {
-            "api_endpoint": {"default": "https://whisper.spoettel.dev/api/v1", "help": "WhisperApi api endpoint"},
+            "api_endpoint": {"default": None, "help": "WhisperApi api endpoint, eg: https://whisperbox-api.com/api/v1, a deployment of https://github.com/bellingcat/whisperbox-transcribe."},
             "api_key": {"default": None, "help": "WhisperApi api key for authentication"},
             "include_srt": {"default": False, "help": "Whether to include a subtitle SRT (SubRip Subtitle file) for the video (can be used in video players)."},
             "timeout": {"default": 90, "help": "How many seconds to wait at most for a successful job completion."},
-            "action": {"default": "translation", "help": "which Whisper operation to execute", "choices": ["transcript", "translation", "language_detection"]},
+            "action": {"default": "translate", "help": "which Whisper operation to execute", "choices": ["transcribe", "translate", "language_detection"]},
 
         }
 
@@ -76,6 +77,7 @@ class WhisperEnricher(Enricher):
             "type": self.action,
             # "language": "string" # may be a config
         }
+        logger.debug(f"calling API with {payload=}")
         response = requests.post(f'{self.api_endpoint}/jobs', json=payload, headers={'Authorization': f'Bearer {self.api_key}'})
         assert response.status_code == 201, f"calling the whisper api {self.api_endpoint} returned a non-success code: {response.status_code}"
         logger.debug(response.json())

src/auto_archiver/formatters/templates/macros.html

@@ -16,7 +16,7 @@ No URL available for {{ m.key }}.
         <a href="https://lens.google.com/uploadbyurl?url={{ url | quote }}">Google Lens</a>,&nbsp;
         <a href="https://yandex.ru/images/touch/search?rpt=imageview&url={{ url | quote }}">Yandex</a>,&nbsp;
         <a href="https://www.bing.com/images/search?view=detailv2&iss=sbi&form=SBIVSP&sbisrc=UrlPaste&q=imgurl:{{ url | quote }}">Bing</a>,&nbsp;
-        <a href="https://www.tineye.com/search/?url={{ url | quote }}">Tineye</a>,&nbsp;
+        <a href="https://www.tineye.com/search/?url={{ url | quote }}">Tineye</a>
     </div>
     <p></p>
 </div>

src/auto_archiver/utils/url.py

@@ -49,6 +49,14 @@ class UrlUtil:
         # instagram recurring images
         if "https://static.cdninstagram.com/rsrc.php/" in url: return False
 
+        # telegram
+        if "https://telegram.org/img/emoji/" in url: return False
+
+        # youtube
+        if "https://www.youtube.com/s/gaming/emoji/" in url: return False
+        if "https://yt3.ggpht.com" in url and "default-user=" in url: return False
+        if "https://www.youtube.com/s/search/audio/" in url: return False
+
         return True
 
     @staticmethod

src/auto_archiver/version.py

@@ -3,7 +3,7 @@ _MAJOR = "0"
 _MINOR = "6"
 # On main and in a nightly release the patch should be one ahead of the last
 # released build.
-_PATCH = "1"
+_PATCH = "3"
 # This is mainly for nightly builds which have the suffix ".dev$DATE". See
 # https://semver.org/#is-v123-a-semantic-version for the semantics.
 _SUFFIX = ""