如何使这个庞大的 Ruby if/elsif 语句更紧凑、更简洁？

Question

下面的 if/elsif 语句显然是一个庞然大物。它的目的是根据用户是否填写了某些数据来更改某些文本的措辞。我觉得必须有一种更好的方法来做到这一点，而无需占用 30 多行代码，但我只是不确定如何，因为我正在尝试根据可用数据显着地自定义文本。

if !birthdate.blank? && !location.blank? && !joined.blank? && !death.blank?
  "<p class='birthinfo'>#{name} was born on #{birthdate.strftime("%A, %B %e, %Y")} in #{location}. #{sex} passed away on #{death.strftime("%B %e, %Y")} at the age of #{calculate_age(birthdate, death)}. #{sex} was a member of #{link_to user.login, profile_path(user.permalink)}'s family for #{distance_of_time_in_words(joined,death)}.</p>"
elsif !birthdate.blank? && !location.blank? && !joined.blank? && death.blank?
  "<p class='birthinfo'>#{name} was born on #{birthdate.strftime("%A, %B %e, %Y")} in #{location} and is #{time_ago_in_words(birthdate)} old. #{sex} has been a member of #{link_to user.login, profile_path(user.permalink)}'s family for #{time_ago_in_words(joined)}.</p>"
elsif birthdate.blank? && !location.blank? && !joined.blank? && !death.blank?
  "<p class='birthinfo'>#{name} was born in #{location}. #{sex} passed away on #{death.strftime("%B %e, %Y")}. #{sex} was a member of #{link_to user.login, profile_path(user.permalink)}'s family for #{distance_of_time_in_words(joined,death)}.</p>"
elsif birthdate.blank? && !location.blank? && !joined.blank? && death.blank?
  "<p class='birthinfo'>#{name} was born in #{location}. #{sex} has been a member of #{link_to user.login, profile_path(user.permalink)}'s family for #{time_ago_in_words(joined)}.</p>"
elsif birthdate.blank? && location.blank? && !joined.blank? && !death.blank?
  "<p class='birthinfo'>#{name} was a member of #{link_to user.login, profile_path(user.permalink)}'s family for #{distance_of_time_in_words(joined,death)}. #{sex} passed away on #{death.strftime("%B %e, %Y")}.</p>"
elsif birthdate.blank? && location.blank? && !joined.blank? && death.blank?
  "<p class='birthinfo'>#{name} has been a member of #{link_to user.login, profile_path(user.permalink)}'s family for #{time_ago_in_words(joined)}.</p>"
elsif !birthdate.blank? && location.blank? && !joined.blank? && !death.blank?
  "<p class='birthinfo'>#{name} was born on #{birthdate.strftime("%A, %B %e, %Y")}. #{sex} passed away on #{death.strftime("%B %e, %Y")} at the age of #{calculate_age(birthdate, death)}. #{sex} was a member of #{link_to user.login, profile_path(user.permalink)}'s family for #{distance_of_time_in_words(joined,death)}.</p>"
elsif !birthdate.blank? && location.blank? && !joined.blank? && death.blank?
  "<p class='birthinfo'>#{name} was born on #{birthdate.strftime("%A, %B %e, %Y")} and is #{time_ago_in_words(birthdate)} old. #{sex} has been a member of #{link_to user.login, profile_path(user.permalink)}'s family for #{time_ago_in_words(joined)}.</p>"
elsif !birthdate.blank? && !location.blank? && joined.blank? && !death.blank?
  "<p class='birthinfo'>#{name} was born on #{birthdate.strftime("%A, %B %e, %Y")} in #{location}. #{sex} passed away on #{death.strftime("%B %e, %Y")} at the age of #{calculate_age(birthdate, death)}. #{sex} was a member of #{link_to user.login, profile_path(user.permalink)}'s family.</p>"
elsif !birthdate.blank? && !location.blank? && joined.blank? && death.blank?
  "<p class='birthinfo'>#{name} was born on #{birthdate.strftime("%A, %B %e, %Y")} in #{location} and is #{time_ago_in_words(birthdate)} old. #{sex} is a member of #{link_to user.login, profile_path(user.permalink)}'s family.</p>"
elsif !birthdate.blank? && location.blank? && joined.blank? && !death.blank?
  "<p class='birthinfo'>#{name} was born on #{birthdate.strftime("%A, %B %e, %Y")}. #{sex} passed away on #{death.strftime("%B %e, %Y")} at the age of #{calculate_age(birthdate, death)}. #{sex} was a member of #{link_to user.login, profile_path(user.permalink)}'s family.</p>"
elsif !birthdate.blank? && location.blank? && joined.blank? && death.blank?
  "<p class='birthinfo'>#{name} was born on #{birthdate.strftime("%A, %B %e, %Y")} and is #{time_ago_in_words(birthdate)} old. #{sex} is a member of #{link_to user.login, profile_path(user.permalink)}'s family.</p>"
elsif birthdate.blank? && !location.blank? && joined.blank? && !death.blank?
  "<p class='birthinfo'>#{name} was born in #{location}. #{sex} passed away on #{death.strftime("%B %e, %Y")}. #{sex} was a member of #{link_to user.login, profile_path(user.permalink)}'s family.</p>"
elsif birthdate.blank? && !location.blank? && joined.blank? && death.blank?
  "<p class='birthinfo'>#{name} was born in #{location}. #{sex} is a member of #{link_to user.login, profile_path(user.permalink)}'s family.</p>"
else
  "<p class='birthinfo'>#{name} is a member of #{link_to user.login, profile_path(user.permalink)}'s family.</p>"
end

Answer 1

我认为您希望使所有这些条件更具可读性，并消除逻辑检查和字符串创建中存在的重复。

我注意到的第一件事是你到处重复这个：

<p class='birthinfo'>#{name} was born in

我会尝试将这些因素分解为接受参数并返回格式化文本的函数，或计算为表达式的类。

您也根本没有利用嵌套。与其让每个条件都检查四个表达式的值，不如试试这样：

if birthdate.blank?
    # half of your expressions
else
    # other half of your expressions

它可能不会使您的代码更具可读性，但值得一试。我建议的最后一件事是，您可能能够巧妙地重新排列您的文本，以便它既易于分段构建，又能很好地被最终用户阅读。这是一个例子：

notice = "#{name} was born on #{birthdate.strftime("%A, %B %e, %Y")} in #{location}."

生成此代码的一个代码 sn-p 可能是：

notice = "#{name} was born on #{birthdate.strftime("%A, %B %e, %Y")}"
if !location.is_blank? 
    notice += " in #{location}"
end
notice += "."

您拥有的类似代码更容易阅读，它会分别检查每个条件并根据布尔变量的值生成完全不同的字符串。关于你现在拥有的逻辑最糟糕的部分是，如果你添加第五个变量，你必须在你的代码中添加更多的特殊情况。将您的代码构建成独立的部分会更容易阅读和维护。

Answer 2

我会把它分成几部分。 DRY。只生成一次给定的文本段。使用StringIO 保持字符串生成可分离。

sio = StringIO.new("")
know_birthdate, know_location, did_join, has_died = [ birthdate, location, joined, death ].map { |s| !s.blank? }

print_death = lambda do 
  sio.print ". #{sex} passed away on #{death.strftime("%B %e, %Y")}"
end
show_birth = know_birthdate or know_location

sio.print "<p class='birthinfo'>#{name} "
if show_birth
  sio.print "was born" 
  sio.print " on #{birthdate.strftime("%A, %B %e, %Y")}" if know_birthdate
  sio.print " in #{location}" if know_location
  if has_died
    print_death[]
    sio.print " at the age of #{calculate_age(birthdate, death)}" if know_birthdate
  elsif know_birthdate 
    sio.print " and is #{time_ago_in_words(birthdate)} old"
  end
  sio.print ". #{sex} "
end
sio.print "#{(has_died ? "was" : did_join ? "has been" : "is")} a member of #{link_to user.login, profile_path(user.permalink)}'s family"
sio.print " for #{distance_of_time_in_words(joined,death)}" if did_join and has_died
print_death[] if has_died and not show_birth
sio.print ".</p>"

sio.to_s

这使得逻辑更容易遵循，并且更容易进行更改。

Answer 3

您可以将字符串放入数组中，然后根据每个变量是否为空来创建数组索引：

strings = [
  "<p class='birthinfo'>#{name} was born on #{birthdate.strftime("%A, %B %e, %Y")} in #{location}. #{sex} passed away on #{death.strftime("%B %e, %Y")} at the age of #{calculate_age(birthdate, death)}. #{sex} was a member of #{link_to user.login, profile_path(user.permalink)}'s family for #{distance_of_time_in_words(joined,death)}.</p>",
  "<p class='birthinfo'>#{name} was born on #{birthdate.strftime("%A, %B %e, %Y")} in #{location} and is #{time_ago_in_words(birthdate)} old. #{sex} has been a member of #{link_to user.login, profile_path(user.permalink)}'s family for #{time_ago_in_words(joined)}.</p>",
  "<p class='birthinfo'>#{name} was born on #{birthdate.strftime("%A, %B %e, %Y")} in #{location}. #{sex} passed away on #{death.strftime("%B %e, %Y")} at the age of #{calculate_age(birthdate, death)}. #{sex} was a member of #{link_to user.login, profile_path(user.permalink)}'s family.</p>",
  "<p class='birthinfo'>#{name} was born on #{birthdate.strftime("%A, %B %e, %Y")} in #{location} and is #{time_ago_in_words(birthdate)} old. #{sex} is a member of #{link_to user.login, profile_path(user.permalink)}'s family.</p>",
  "<p class='birthinfo'>#{name} was born on #{birthdate.strftime("%A, %B %e, %Y")}. #{sex} passed away on #{death.strftime("%B %e, %Y")} at the age of #{calculate_age(birthdate, death)}. #{sex} was a member of #{link_to user.login, profile_path(user.permalink)}'s family for #{distance_of_time_in_words(joined,death)}.</p>",
  "<p class='birthinfo'>#{name} was born on #{birthdate.strftime("%A, %B %e, %Y")} and is #{time_ago_in_words(birthdate)} old. #{sex} has been a member of #{link_to user.login, profile_path(user.permalink)}'s family for #{time_ago_in_words(joined)}.</p>",
  "<p class='birthinfo'>#{name} was born on #{birthdate.strftime("%A, %B %e, %Y")}. #{sex} passed away on #{death.strftime("%B %e, %Y")} at the age of #{calculate_age(birthdate, death)}. #{sex} was a member of #{link_to user.login, profile_path(user.permalink)}'s family.</p>",
  "<p class='birthinfo'>#{name} was born on #{birthdate.strftime("%A, %B %e, %Y")} and is #{time_ago_in_words(birthdate)} old. #{sex} is a member of #{link_to user.login, profile_path(user.permalink)}'s family.</p>",
  "<p class='birthinfo'>#{name} was born in #{location}. #{sex} passed away on #{death.strftime("%B %e, %Y")}. #{sex} was a member of #{link_to user.login, profile_path(user.permalink)}'s family for #{distance_of_time_in_words(joined,death)}.</p>",
  "<p class='birthinfo'>#{name} was born in #{location}. #{sex} has been a member of #{link_to user.login, profile_path(user.permalink)}'s family for #{time_ago_in_words(joined)}.</p>",
  "<p class='birthinfo'>#{name} was born in #{location}. #{sex} passed away on #{death.strftime("%B %e, %Y")}. #{sex} was a member of #{link_to user.login, profile_path(user.permalink)}'s family.</p>",
  "<p class='birthinfo'>#{name} was born in #{location}. #{sex} is a member of #{link_to user.login, profile_path(user.permalink)}'s family.</p>",
  "<p class='birthinfo'>#{name} was a member of #{link_to user.login, profile_path(user.permalink)}'s family for #{distance_of_time_in_words(joined,death)}. #{sex} passed away on #{death.strftime("%B %e, %Y")}.</p>",
  "<p class='birthinfo'>#{name} has been a member of #{link_to user.login, profile_path(user.permalink)}'s family for #{time_ago_in_words(joined)}.</p>",
  "", # This case was missing from your code. (Where all are blank except 'death'.)
  "<p class='birthinfo'>#{name} is a member of #{link_to user.login, profile_path(user.permalink)}'s family.</p>"
]

index = 0
index += 1 unless death.blank?
index += 2 unless joined.blank?
index += 4 unless location.blank?
index += 8 unless birthdate.blank?

return strings[index]

Answer 4

这是一个片段，展示了我将如何做到这一点：

ret = []
ret << "<p class='birthinfo'>#{name}"
ret << "was born on #{birthdate.strftime("%A, %B %e, %Y")}" unless birthdate.blank?
ret << "in #{location}." unless location.blank?
ret << sex
ret << "passed away on #{death.strftime("%B %e, %Y")}" unless death.blank?
ret << "at the age of #{calculate_age(birthdate, death)}"
ret << "#{sex} was a member of #{link_to user.login, profile_path(user.permalink)}'s family"
ret << 'for #{distance_of_time_in_words(joined,death)}" unless joined.blank? || death.blank?
ret << '.</p>'

ret.join(' ')

Answer 5

如果有一个简单的描述性术语，我会很想编写描述每种状态的便捷方法。它不会导致更少的代码，但会导致更易读的代码。该代码几乎无法阅读。

不过，看着它，我不确定您能想出什么样的描述性便捷方法名称。鉴于您正在尝试做的事情，您可能只需要处理丑陋。

Answer 6

我能想到的一件事是，不要执行每个查询，而是执行所有查询并分配布尔变量的结果，然后测试布尔值。这不会减少 LOC，但确实会减少行长。忘记我之前所说的关于 switch 语句的内容，那至少没有用。您可能会尝试将每个条件分解为几种方法。在给定的 if 语句中检查的第一件事仍然存在，但这会为下一个条件调用两个方法之一，依此类推。这实际上会扩展 LOC（我觉得这并不重要），但可读性会增加。这不是最佳解决方案，但它会起作用。

Answer 7

您应该能够使用deMorgan's theorem 重写查询，以便更有效地评估它。
您还可以将公共子表达式分解到外部 if 语句中。
但是，这样做可能会牺牲意图的清晰性。
如今，写一个清晰的、如果很长的 if 语句而不是一个简短的难以理解的语句可能是更好的形式。