audiogen http: download factored in fsutils

larigira/audiogen_http.py

@@ -1,31 +1,4 @@
-import os
-import logging
-import posixpath
-from tempfile import mkstemp
-import urllib.request
-from urllib.parse import urlparse
-
-log = logging.getLogger(__name__)
-
-
-def put(url, destdir=None, copy=False):
-    if url.split(":")[0] not in ("http", "https"):
-        log.warning("Not a valid URL: %s", url)
-        return None
-    ext = url.split(".")[-1]
-    if ext.lower() not in ("mp3", "ogg", "oga", "wma", "m4a"):
-        log.warning('Invalid format (%s) for "%s"', ext, url)
-        return None
-    if not copy:
-        return url
-    fname = posixpath.basename(urlparse(url).path)
-    # sanitize
-    fname = "".join(c for c in fname if c.isalnum() or c in list("._-")).rstrip()
-    tmp = mkstemp(suffix="." + ext, prefix="http-%s-" % fname, dir=destdir)
-    os.close(tmp[0])
-    log.info("downloading %s -> %s", url, tmp[1])
-    fname, headers = urllib.request.urlretrieve(url, tmp[1])
-    return "file://%s" % os.path.realpath(tmp[1])
+from larigira.fsutils import download_http
 
 
 def generate(spec):
@@ -35,10 +8,10 @@ def generate(spec):
     Recognized argument is  "paths" (list of static paths)
     """
     if "urls" not in spec:
-        raise ValueError("Malformed audiospec: missing 'paths'")
+        raise ValueError("Malformed audiospec: missing 'urls'")
 
     for url in spec["urls"]:
-        ret = put(url, copy=True)
+        ret = download_http(url, copy=True, prefix="http")
         if ret is None:
             continue
         yield ret

larigira/fsutils.py

@@ -1,6 +1,13 @@
-import os
 import fnmatch
+import logging
 import mimetypes
+import os
+import posixpath
+import urllib.request
+from tempfile import mkstemp
+from urllib.parse import urlparse
+
+log = logging.getLogger(__name__)
 
 
 def scan_dir(dirname, extension=None):
@@ -37,3 +44,27 @@ def shortname(path):
     name = name.rsplit(".", 1)[0]  # no extension
     name = "".join(c for c in name if c.isalnum())  # no strange chars
     return name
+
+
+def download_http(url, destdir=None, copy=False, prefix="httpdl"):
+    if url.split(":")[0] not in ("http", "https"):
+        log.warning("Not a valid URL: %s", url)
+        return None
+    ext = url.split(".")[-1]
+    if ext.lower() not in ("mp3", "ogg", "oga", "wma", "m4a"):
+        log.warning('Invalid format (%s) for "%s"', ext, url)
+        return None
+    if not copy:
+        return url
+    fname = posixpath.basename(urlparse(url).path)
+    # sanitize
+    fname = "".join(
+        c for c in fname if c.isalnum() or c in list("._-")
+    ).rstrip()
+    tmp = mkstemp(
+        suffix="." + ext, prefix="%s-%s-" % (prefix, fname), dir=destdir
+    )
+    os.close(tmp[0])
+    log.info("downloading %s -> %s", url, tmp[1])
+    fname, headers = urllib.request.urlretrieve(url, tmp[1])
+    return "file://%s" % os.path.realpath(tmp[1])