我怎樣才能用較少的冗餘/複製粘貼來寫這個？

Question

我通常不會以最好的方式編寫我的Python代碼，因爲我相對較新，有人要求我更改Django應用程序，因爲代碼看起來不太好。

這裏是什麼樣子：

@login_required 
def submission_set_rank(request): 

    r1_obj_id = request.GET.get('rank1','') 
    r2_obj_id = request.GET.get('rank2','') 
    r3_obj_id = request.GET.get('rank3','') 
    r4_obj_id = request.GET.get('rank4','') 
    r5_obj_id = request.GET.get('rank5','') 

    #rate the first BallotStats object 
    ballot_1 = BallotStats.objects.get(object_id=r1_obj_id) 
    ballot_2 = BallotStats.objects.get(object_id=r2_obj_id) 
    ballot_3 = BallotStats.objects.get(object_id=r3_obj_id) 
    ballot_4 = BallotStats.objects.get(object_id=r4_obj_id) 
    ballot_5 = BallotStats.objects.get(object_id=r5_obj_id) 

    ballot_1.score += 5 
    ballot_2.score += 4 
    ballot_3.score += 3 
    ballot_4.score += 2 
    ballot_5.score += 1 

    ballot_1.save() 
    ballot_2.save() 
    ballot_3.save() 
    ballot_4.save() 
    ballot_5.save() 

    return HttpResponseRedirect('/submissions/results/film/') 

事實證明，我意識到，我一直在寫我的Python代碼這種方式，有沒有辦法讓它更好看，而不是佔用了21 +代碼行？

Answer 1

最大的問題是不是代碼的風格 - 它是你正在10個查詢：5用於獲取對象和5更新它們。使用__in一次

篩選出對象：

@login_required 
def submission_set_rank(request): 
    keys = {'rank1': 5, 'rank2': 4, 'rank3': 3, 'rank4': 2, 'rank5': 1} 
    ranks = [request.GET.get(key,'') for key in keys] 
    for ballot in BallotStats.objects.filter(object_id__in=ranks): 
     ballot.score += keys[ballot.object_id] 
     ballot.save() 

    return HttpResponseRedirect('/submissions/results/film/') 

這將至多使6個查詢：1獲取對象和5更新它們。

此外，您可以使用commit_manually修飾器「標記」視圖（commit_on_success也適用於您）。它應該speed up things significantly：

@login_required 
@transaction.commit_manually 
def submission_set_rank(request): 
    keys = {'rank1': 5, 'rank2': 4, 'rank3': 3, 'rank4': 2, 'rank5': 1} 
    ranks = [request.GET.get(key,'') for key in keys] 
    for ballot in BallotStats.objects.filter(object_id__in=ranks): 
     ballot.score += keys[ballot.object_id] 
     ballot.save() 
    transaction.commit() 

    return HttpResponseRedirect('/submissions/results/film/') 

和我有強烈的感覺，你可以在即使一個更新查詢，做到這一點。例如，通過使用connection.cursor()直接與executemany()幫助：

@login_required 
def submission_set_rank(request): 
    keys = {'rank1': 5, 'rank2': 4, 'rank3': 3, 'rank4': 2, 'rank5': 1} 
    ranks = [{'score': request.GET.get(key,''), 'id': key} for key in keys] 

    cursor = connection.cursor() 
    cursor.executemany(""" 
     UPDATE 
      ballot_stats 
     SET 
      score = score + %(score)s 
     WHERE 
      object_id = %(id)s 
    """, ranks) 

    return HttpResponseRedirect('/submissions/results/film/') 

確保字段和表名是正確的。

Answer 2

就你而言，一點循環根本不會傷害。事實上，一般來說，每當你不得不重複兩次以上時，試着讓它成爲一個循環。

n = 5 
for i in range(1, n+1): 
    obj_id = request.GET('rank' + str(i), '') 
    ballot = BallotStats.objects.get(object_id=obj_id) 
    ballot.score += n - i + 1 
    ballot.save() 

Answer 3

如果我們談論節約行代碼，你可以在4線通過用.update()更換您的.save()並使用F()表達照顧+=合併成一條線。另外，正如@alecxe所討論的，這會將查詢減少一半。它會是這樣的：

@login_required 
def submission_set_rank(request): 
    BallotStats.objects.filter(object_id=request.GET.get('rank1','')).update(score=‌​F('score') + 5) 
    BallotStats.objects.filter(object_id=request.GET.get('rank2','')).update(score=‌​F('score') + 4) 
    BallotStats.objects.filter(object_id=request.GET.get('rank3','')).update(score=‌​F('score') + 3) 
    BallotStats.objects.filter(object_id=request.GET.get('rank4','')).update(score=‌​F('score') + 2) 
    BallotStats.objects.filter(object_id=request.GET.get('rank5','')).update(score=‌​F('score') + 1) 

    return HttpResponseRedirect('/submissions/results/film/')