From 8c6b63219d8b3ffbd96ce3cebb999005f93f9a6c Mon Sep 17 00:00:00 2001 From: Martin Olveyra Date: Fri, 21 Dec 2012 13:33:48 +0000 Subject: [PATCH] allow to disable application of extra required attributes for testing purposes --- scrapely/extraction/__init__.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/scrapely/extraction/__init__.py b/scrapely/extraction/__init__.py index dd3bab5..9f942d8 100644 --- a/scrapely/extraction/__init__.py +++ b/scrapely/extraction/__init__.py @@ -26,7 +26,7 @@ class InstanceBasedLearningExtractor(object): extract data from web pages. """ - def __init__(self, td_pairs, trace=False): + def __init__(self, td_pairs, trace=False, apply_extrarequired=True): """Initialise this extractor td_pairs is a list of (template, item descriptor) pairs. @@ -54,7 +54,7 @@ def __init__(self, td_pairs, trace=False): modified_parsed_tdpairs = [] # apply extra required attributes for parsed, (t, descriptor) in parsed_tdpairs: - if descriptor is not None: + if descriptor is not None and apply_extrarequired: descriptor = descriptor.copy() for attr in parsed.extra_required_attrs: descriptor._required_attributes.append(attr)