indexing: for minimal index, use a single -m flag to create a 6 field index.

minimal index also skips parsing contents of warc/arc records altogether
add cli docs for minimal index, tracked via #75

pywb/warc/archiveiterator.py

@@ -40,7 +40,7 @@ class ArchiveIterator(object):
     warc2warc -Z myfile.{0} > myfile.{0}.gz
     """
 
-    def __init__(self, fileobj):
+    def __init__(self, fileobj, no_record_parse=False):
         self.fh = fileobj
 
         self.loader = ArcWarcRecordLoader()
@@ -50,6 +50,7 @@ class ArchiveIterator(object):
         self.known_format = None
 
         self.member_info = None
+        self.no_record_parse = no_record_parse
 
     def iter_records(self, block_size=16384):
         """ iterate over each record
@@ -176,7 +177,8 @@ class ArchiveIterator(object):
         """
         record = self.loader.parse_record_stream(self.reader,
                                                  next_line,
-                                                 self.known_format)
+                                                 self.known_format,
+                                                 self.no_record_parse)
 
         self.member_info = None
 
@@ -254,6 +256,10 @@ class DefaultRecordIter(object):
         minimal = self.options.get('minimal')
         append_post = self.options.get('append_post')
 
+        if append_post and minimal:
+            raise Exception('Sorry, minimal index option and ' +
+                            'append POST options can not be used together')
+
         for record in arcv_iter.iter_records(block_size):
             entry = None
 
@@ -423,7 +429,7 @@ class DefaultRecordIter(object):
         return entry
 
     def __call__(self, fh):
-        aiter = ArchiveIterator(fh)
+        aiter = ArchiveIterator(fh, self.options.get('minimal', False))
 
         entry_iter = self.create_record_iter(aiter)
 

pywb/warc/cdxindexer.py

@@ -173,7 +173,7 @@ def get_cdx_writer_cls(options):
 
     if options.get('cdx09'):
         format_mixin = CDX09
-    elif options.get('cdx06'):
+    elif options.get('minimal'):
         format_mixin = CDX06
     else:
         format_mixin = CDX11
@@ -269,6 +269,15 @@ Not-recommended for new cdx, use only for backwards-compatibility.
 
     cdx09_help = """
 Use older 9-field cdx format, default is 11-cdx field
+"""
+    minimal_help = """
+Use a minimal 6-field cdx format, outputing only the basic fields
+needed to identiyfy record:
+canonicalized url, timestamp, original url, archive offset, archive length
+and archive filename.
+
+This option skips record parsing and will not work with
+POST append (-p) option
 """
 
     output_help = """output file or directory.
@@ -320,9 +329,9 @@ if input is a directory"""
                         action='store_true',
                         help=cdx09_help)
 
-    group.add_argument('-6', '--cdx06',
+    group.add_argument('-m', '--minimal',
                         action='store_true',
-                        help=cdx09_help)
+                        help=minimal_help)
 
     parser.add_argument('output', nargs='?', default='-', help=output_help)
     parser.add_argument('inputs', nargs='+', help=input_help)
@@ -336,8 +345,7 @@ if input is a directory"""
                           append_post=cmd.postappend,
                           recurse=cmd.recurse,
                           cdx09=cmd.cdx09,
-                          cdx06=cmd.cdx06,
-                          minimal=cmd.cdx06)
+                          minimal=cmd.minimal)
 
 
 if __name__ == '__main__':

pywb/warc/recordloader.py

@@ -83,7 +83,9 @@ class ArcWarcRecordLoader:
         return self.parse_record_stream(stream)
 
     def parse_record_stream(self, stream,
-                            statusline=None, known_format=None):
+                            statusline=None,
+                            known_format=None,
+                            no_record_parse=False):
         """ Parse file-like stream and return an ArcWarcRecord
         encapsulating the record headers, http headers (if any),
         and a stream limited to the remainder of the record.
@@ -128,8 +130,12 @@ class ArcWarcRecordLoader:
         # limit stream to the length for all valid records
         stream = LimitReader.wrap_stream(stream, length)
 
+        # don't parse the http record at all
+        if no_record_parse:
+            status_headers = StatusAndHeaders('', [])
+
         # if empty record (error or otherwise) set status to 204
-        if length == 0:
+        elif length == 0:
             if is_err:
                 msg = '204 Possible Error'
             else: