需要从字典列表中删除重复项并修改剩余重复项的数据（python）

Question

考虑一下这个简单的 Python 字典列表（第一个字典项是一个字符串，第二个项是一个 Widget 对象）：

raw_results =  
     [{'src': 'tag', 'widget': <Widget: to complete a form today>},   # dupe 1a
      {'src': 'tag', 'widget': <Widget: a newspaper>},                # dupe 2a
      {'src': 'zip', 'widget': <Widget: to complete a form today>},   # dupe 1b
      {'src': 'zip', 'widget': <Widget: the new Jack Johnson album>},
      {'src': 'zip', 'widget': <Widget: a newspaper>},                # dupe 2b
      {'src': 'zip', 'widget': <Widget: premium dog food >}]

我想遍历这个列表，去掉重复的项，这个 StackOverflow 的问题给了我答案：

在保持顺序的情况下去掉列表中的重复项（Python）

    known_widgets= set()
    processed_results = []

    for x in raw_results:
        widget = x['widget']
        if widget in known_widgets: 
            continue
        else:
            processed_results.append(x)
            known_widgets.add(widget)

不过，在我去掉重复的行（比如说 dupe 1b）后，我想要修改剩下的重复项（比如说 dupe 1a）的 "src" 数据。我想把被去掉的重复项的 "src" 添加到原来的项中。这是我想要的结果：

processed_results =  
    [{'src': 'tag-zip', 'widget': <Widget: to complete a form today>},  # dupe 1a
     {'src': 'tag-zip', 'widget': <Widget: a newspaper>},               # dupe 2a
     {'src': 'zip', 'widget': <Widget: the new Jack Johnson album>},
     {'src': 'zip', 'widget': <Widget: premium dog food >}]

我相信这很简单，但我喝了太多咖啡，思路有点混乱，已经花了很多小时在这个问题上打转。我非常希望能得到专家的帮助，感谢你们！

Answer 1

假设你想要根据重复的src值来整理一个小部件的列表，这就是你需要的做法：

class Widget(object):
    def __init__(self, desc):
        self.desc = desc
    def __str__(self):
        return "Widget(%s)" % self.desc

raw_results = [
    {'src':'tag-zip', 'widget':Widget('to complete a form today')},
    {'src':'tag-zip', 'widget':Widget('a newspaper')},
    {'src':'zip', 'widget':Widget('the new Jack Johnson album')},
    {'src':'zip', 'widget':Widget('premium dog food')}
]

from collections import defaultdict
known_widgets = defaultdict(list)
for x in raw_results:
    k, v = x['src'], x['widget']
    known_widgets[k].append(v)

for k, v in known_widgets.iteritems():
    print "%s: %s" % (k, ",".join(str(w) for w in v))

如果你想要去掉重复的widget5，那就这样做：

class Widget(object):
    def __init__(self, desc):
        self.desc = desc
    def __str__(self):
        return "Widget(%s)" % self.desc
    def __hash__(self):
        return hash(self.desc)
    def __cmp__(self, other):
        return cmp(self.desc, other.desc)

raw_results = [
    {'src':'tag-zip', 'widget':Widget('to complete a form today')},
    {'src':'tag-zip', 'widget':Widget('a newspaper')},
    {'src':'zip', 'widget':Widget('the new Jack Johnson album')},
    {'src':'zip', 'widget':Widget('premium dog food')},
    {'src':'tag-zip', 'widget':Widget('to complete a form today')},
    {'src':'tag-zip', 'widget':Widget('a newspaper')},
    {'src':'zip', 'widget':Widget('the new Jack Johnson album')},
    {'src':'zip', 'widget':Widget('premium dog food')},
]

from collections import defaultdict
known_widgets = defaultdict(set)
for x in raw_results:
    k, v = x['src'], x['widget']
    known_widgets[k].add(v)

for k, v in known_widgets.iteritems():
    print "%s: %s" % (k, ",".join(str(w) for w in v))

Answer 2

def find_widget(widget, L):
    for i, v in enumerate(L):
      if v[widget] == widget:
          return i

known_widgets= set()
processed_results = []

for x in raw_results:
    widget = x['widget']
    if widget in known_widgets:
        processed_widgets[find_widget(widget, processed_results)]['src'] += '-%s' % x['tag']
        continue
    else:
        processed_results.append(x)
        known_widgets.add(widget)

这个方法可能还有更好的做法，因为它对每个重复的组件进行了两次处理。