句子的正则表达

Question

I'm trying to write a regular expression to represent a sentence with the following conditions: starts with a capital letter, ends with a period (and only one period can appear), and is allowed to contain a comma or semi-colon, but when it does, it must appear as (letter)(semicolon)(space) or (letter)(comma)(space).

I've got the capital letter and period down. I have the idea for the code but I think I'm not getting the syntax completely right...

In English, my expression for a sentence looks like this:

(capital letter) ((lowercase letter)(space) ((lowercase letter)(comma)(space))* 
((lowercase letter)(semicolon)(space)* )* (period)

I realize this ignores the case where the first letter of the sentence is immediately followed by a comma or semicolon, but it's safe to ignore that case.

Now when I try to code this in Python, I try the following (I've added whitespace to make things easier to read):

sentence = re.compile("^[A-Z]  [a-z\\s  (^[a-z];\\s$)* (^[a-z],\\s$)*]*  \.$")

I feel like it's a syntax issue... I'm not sure if I'm allowed to have the semicolon and comma portions inside of parentheses.

Sample inputs that match the definition:

"This is a sentence."
"Hello, world."
"Hi there; hi there."

Sample inputs that do not match the definition:

"i ate breakfast."
"This is , a sentence."
"What time is it?"

Answer 1

^(?!.*[;,]\S)(?!.* [;,])[A-Z][a-z\s,;]+\.$

Its easier to use lookaheads to remove invalid sentences.See demo.

https://regex101.com/r/vV1wW6/36#python

Answer 2

This would match what you said above.

^"[AZ][az]*(\\s*|[az]*|(?<!\\s)[;,](?=\\s))*[.]"$ ? => demo

This would match:

"This is a sentence."
"Hello, world."
"Hi there; hi there."

This won't match:

"i ate breakfast."
"This is , a sentence."
"What time is it?"
"I a ,d am."
"I a,d am."

If you don't need the " just remove it from the regex.

---

If you need the regex in python, try this

re.compile(r'^[AZ][az]*(\\s*|[az]*|(?<!\\s)[;,](?=\\s))*[.]$')

Python demo

import re
tests = ["This is a sentence."
,"Hello, world."
,"Hi there; hi there."
,"i ate breakfast."
,"This is , a sentence."
,"What time is it?"]
rex = re.compile(r'^[A-Z][a-z]*(\s*|[a-z]*|(?<![\s])[;,])*[.]$')
for test in tests:
    print rex.match(test)

output

<_sre.SRE_Match object at 0x7f31225afb70>
<_sre.SRE_Match object at 0x7f31225afb70>
<_sre.SRE_Match object at 0x7f31225afb70>
None
None
None

Answer 3

I ended up modifying my regular expression to

"^[A-Z][a-z\s (a-z,\s)* (a-z;\s)*]*\.$"

and it ended up working just fine. Thanks for everyone's help!