Combine FileLoader/HttpLoader into a single BlockLoader which

delegates based on scheme

pywb/cdx/zipnum.py

@@ -7,7 +7,7 @@ from cStringIO import StringIO
 from cdxsource import CDXSource
 from cdxobject import IDXObject
 
-from pywb.utils.loaders import FileLoader, LimitReader
+from pywb.utils.loaders import BlockLoader
 from pywb.utils.loaders import SeekableTextFileReader
 from pywb.utils.bufferedreaders import gzip_decompressor
 from pywb.utils.binsearch import iter_range, linearsearch
@@ -101,7 +101,11 @@ class ZipNumCluster(CDXSource):
                 if blocks:
                     yield self.block_to_cdx_iter(blocks, ranges, params)
 
-                blocks = ZipBlocks(idx['part'], idx['offset'], idx['length'], 1)
+                blocks = ZipBlocks(idx['part'],
+                                   idx['offset'],
+                                   idx['length'],
+                                   1)
+
                 ranges = [blocks.length]
 
         if blocks:
@@ -130,7 +134,7 @@ class ZipNumCluster(CDXSource):
             msg = 'Loading {b.count} blocks from {loc}:{b.offset}+{b.length}'
             logging.debug(msg.format(b=blocks, loc=location))
 
-        reader = FileLoader().load(location, blocks.offset, blocks.length)
+        reader = BlockLoader().load(location, blocks.offset, blocks.length)
 
         def decompress_block(range_):
             decomp = gzip_decompressor()

pywb/utils/loaders.py

@@ -9,18 +9,50 @@ import urllib2
 import time
 
 
+def is_http(filename):
+    return any(filename.startswith(x) for x in ['http://', 'https://'])
+
+
 #=================================================================
-# load a reader from http
-#=================================================================
-class HttpLoader(object):
+class BlockLoader(object):
     """
-    Load a file-like reader over http using range requests
-    and an optional cookie created via a cookie_maker
+    a loader which can stream blocks of content
+    given a uri, offset and optional length.
+    Currently supports: http/https and file/local file system
     """
     def __init__(self, cookie_maker=None):
         self.cookie_maker = cookie_maker
 
     def load(self, url, offset, length):
+        """
+        Determine loading method based on uri
+        """
+        if is_http(url):
+            return self.load_http(url, offset, length)
+        else:
+            return self.load_file(url, offset, length)
+
+    def load_file(self, url, offset, length):
+        """
+        Load a file-like reader from the local file system
+        """
+
+        if url.startswith('file://'):
+            url = url[len('file://'):]
+
+        afile = open(url, 'rb')
+        afile.seek(offset)
+
+        if length > 0:
+            return LimitReader(afile, length)
+        else:
+            return afile
+
+    def load_http(self, url, offset, length):
+        """
+        Load a file-like reader over http using range requests
+        and an optional cookie created via a cookie_maker
+        """
         if length > 0:
             range_header = 'bytes={0}-{1}'.format(offset, offset + length - 1)
         else:
@@ -71,27 +103,6 @@ class HMACCookieMaker(object):
         return cookie
 
 
-#=================================================================
-# load a reader from local filesystem
-#=================================================================
-class FileLoader(object):
-    """
-    Load a file-like reader from the local file system
-    """
-
-    def load(self, url, offset, length):
-        if url.startswith('file://'):
-            url = url[len('file://'):]
-
-        afile = open(url, 'rb')
-        afile.seek(offset)
-
-        if length > 0:
-            return LimitReader(afile, length)
-        else:
-            return afile
-
-
 #=================================================================
 # Limit Reader
 #=================================================================

pywb/utils/test/loaders_test.py

@@ -10,9 +10,9 @@
 >>> read_multiple(LimitReader(StringIO.StringIO('abcdefghjiklmnopqrstuvwxyz'), 10), [2, 2, 20])
 'efghji'
 
-# FileLoader Tests (includes LimitReader)
+# BlockLoader Tests (includes LimitReader)
 # Ensure attempt to read more than 100 bytes, reads exactly 100 bytes
->>> len(FileLoader().load(test_cdx_dir + 'iana.cdx', 0, 100).read('400'))
+>>> len(BlockLoader().load(test_cdx_dir + 'iana.cdx', 0, 100).read('400'))
 100
 
 # SeekableTextFileReader Test
@@ -34,7 +34,7 @@
 >>> DecompressingBufferedReader(open(test_cdx_dir + 'iana.cdx.gz', 'rb'), decomp_type = 'gzip').readline()
 ' CDX N b a m s k r M S V g\\n'
 
->>> HttpLoader(HMACCookieMaker('test', 'test', 5)).load('http://example.com', 41, 14).read()
+>>> BlockLoader(HMACCookieMaker('test', 'test', 5)).load('http://example.com', 41, 14).read()
 'Example Domain'
 
 # test very small block size
@@ -53,7 +53,7 @@
 #=================================================================
 import os
 import StringIO
-from pywb.utils.loaders import FileLoader, HttpLoader, HMACCookieMaker
+from pywb.utils.loaders import BlockLoader, HMACCookieMaker
 from pywb.utils.loaders import LimitReader, SeekableTextFileReader
 from pywb.utils.bufferedreaders import DecompressingBufferedReader
 

pywb/warc/recordloader.py

@@ -6,7 +6,7 @@ from pywb.utils.statusandheaders import StatusAndHeaders
 from pywb.utils.statusandheaders import StatusAndHeadersParser
 from pywb.utils.statusandheaders import StatusAndHeadersParserException
 
-from pywb.utils.loaders import FileLoader, HttpLoader
+from pywb.utils.loaders import BlockLoader
 from pywb.utils.bufferedreaders import DecompressingBufferedReader
 
 #=================================================================
@@ -32,23 +32,11 @@ class ArcWarcRecordLoader:
     ARC_HEADERS = ["uri", "ip-address", "creation-date",
                    "content-type", "length"]
 
-    @staticmethod
-    def create_default_loaders(cookie_maker=None):
-        http = HttpLoader(cookie_maker)
-        file = FileLoader()
-        return {
-            'http': http,
-            'https': http,
-            'file': file,
-            '': file
-            }
-
-    def __init__(self, loaders={}, cookie_maker=None, block_size=8192):
-        self.loaders = loaders
-
-        if not self.loaders:
-            self.loaders = self.create_default_loaders(cookie_maker)
+    def __init__(self, loader=None, cookie_maker=None, block_size=8192):
+        if not loader:
+            loader = BlockLoader(cookie_maker)
 
+        self.loader = loader
         self.block_size = block_size
 
         self.arc_parser = ARCHeadersParser(self.ARC_HEADERS)
@@ -60,16 +48,16 @@ class ArcWarcRecordLoader:
     def load(self, url, offset, length):
         url_parts = urlparse.urlsplit(url)
 
-        loader = self.loaders.get(url_parts.scheme)
-        if not loader:
-            raise ArchiveLoadFailed('Unknown Protocol', url)
+        #loader = self.loaders.get(url_parts.scheme)
+        #if not loader:
+        #    raise ArchiveLoadFailed('Unknown Protocol', url)
 
         try:
             length = int(length)
         except:
             length = -1
 
-        raw = loader.load(url, long(offset), length)
+        raw = self.loader.load(url, long(offset), length)
 
         decomp_type = 'gzip'