英文:
Writing a proper regular expression that handles multiple spaces
问题
以下是您要提取的信息的翻译：
NOWAKOWSKA
NOWAK_S6_2
KOWALSKA_S6_
NOWACKI S6_
您之前使用的正则表达式是%^Numer właściciela:[[:space:]](?<owner_id>.+)$%imu，但出现了问题，因为源文本中的空格可能会在模式内部，并且这些空格必须包含在结果中，但如果有多个重复的空格，则不应包含在结果中。您尝试使用条件和负向预查来排除多个空格，但未成功。希望这有所帮助。
英文:
I'm struggling to write a proper regexp to use with PHP7.4 to extract required information from a string.
Here are the sample strings:
Numer właściciela: NOWAKOWSKA 01-234 Warsaw
Numer właściciela: NOWAK_S6_2
Numer właściciela: KOWALSKA_S6_ 01-234 Warsaw
Numer właściciela: NOWACKI S6_ 01-234 Warsaw
What I want to extract is accordingly:
NOWAKOWSKA
NOWAK_S6_2
KOWALSKA_S6_
NOWACKI S6_
So far I was using the %^Numer właściciela:[[:space:]](?<owner_id>.+)$%imu which worked fine (with example from row#2). However, turns out that the other cases (#1, #3, #4) appeared during a roll-out phase and our text extraction is not accurate enough.
The problem here is with spaces, the source text may contain space inside the pattern and this space must be included in the result. However, if there are repeating spaces, they must not be included.
Tried playing around with some conditionals and negative lookaheads to exclude multiple spaces, but failed to do so.
Would really appreciate any help here.
答案1
得分: 1
在一般情况下，当您想要匹配由单个空格分隔的字符序列时，您可以使用以下正则表达式：
/^Numer właściciela:\h*(?<owner_id>\S+(?:\h\S+)*)/imu
请查看正则表达式演示。在这里，\h 优于 \s，因为您是从更长的文本中提取数据而不是独立的文本。
如果您要提取的字符串都很短，您还可以使用以下正则表达式：
/^Numer właściciela:\h*(?<owner_id>.*?)(?:\h{2}|$)/imu
这样的话，效率可能会更高，但前提是它们与问题中的字符串一样短。.*? 通常与任意长度的字符串一样昂贵。
模式详细信息：
^- 行的开头（由于m标志）Numer właściciela:- 一个字面字符串（用\h替换\h*- 零个或多个水平空格(?<owner_id>\S+(?:\h\S+)*)- "owner_id" 组：一个或多个非空格字符，后跟零个或多个单个水平空格，然后跟一个或多个非空格字符。(?<owner_id>.*?)(?:\h{2}|$)- "owner_id" 组，捕获尽可能少的零个或多个非换行字符，然后是两个水平空格或行尾。
英文:
In a general case, when you want to match sequences of chars separated with a single whitespace, you can use
/^Numer właściciela:\h*(?<owner_id>\S+(?:\h\S+)*)/imu
See the regex demo. \h is preferred to \s since you are extracting data from lines in a longer text, not standalone texts.
If the strings you extract are all short, you may also use
/^Numer właściciela:\h*(?<owner_id>.*?)(?:\h{2}|$)/imu
Then, it should be even more efficient, but only if they are that short as in the question. The .*? is usually as expensive as .* in strings of arbitrary length.
Pattern details:
^- start of a line (due tomflag)Numer właściciela:- a literal string (replace\hto match any horizontal whitespace)\h*- zero or more horizontal whitespaces(?<owner_id>\S+(?:\h\S+)*)- Group "owner_id": one or more non-whitespace chars followed with zero or more sequences of a single horizontal whitespace followed with one or more non-whitespace chars.(?<owner_id>.*?)(?:\h{2}|$)- Group "owner_id" that captures any zero or more chars other than line break chars as few as possible, and then either two horizontal whitespaces or end of a line.
答案2
得分: 1
这个正则表达式：
/^Numer właściciela:\s+(?<owner_id>.*?)(?=\s{20,}|$)/imu
通过集体智慧和协作来改善编程学习和解决问题的方式。致力于成为全球开发者共同参与的知识库，让每个人都能够通过互相帮助和分享经验来进步。
评论