Match subgroup with same name on separate conditions in Python

I'm trying to create a regex to parse a cookie header I'm getting back that the cookies package complains about. I've come up with the following regex that works:

((?P<name>\w+)(=(\"(?P<value1>[\w;,#\.\/\\=]+)\"|(?P<value2>[\w#\.\/\\]+)))?)[;,]?

For an example, see this link to regex101. Essentially, this works but I'd like to use value in place of both value1 and value2. This shouldn't be an issue because it's impossible to generate a match on both subgroups at the same time (one is quoted and the other isn't). Is there a way I can do this?

Yes, you can match precisely what you need with a lookbehind:

(
  (?P<name>\w+)
  (
    =
    (?P<quote>"?)            # Match an optional quote, which we capture,
    (?P<value>               # then either
      (?<=")[\w;,#./\\=]+    # a quoted value, which follows a quote,
    |                        # or
      [\w#./\\]+             # an unquoted value, which is the fallback,
    )                        # and
    (?P=quote)               # the quote we captured or an empty string.
  )?
)
[;,]?

A value always follows =, so (?<=") matches iff it's a quoted value.

Note that you don't have to escape / (it doesn't mean anything special in Python flavor), . (inside a character class), nor " (if you use ', ''' or """ as string delimiters).

Try it on regex101.com.

Alternatively, you can use the conditional group syntax:

(
  (?P<name>\w+)
  (
    =
    (
      (?P<quote>")?          # Match an optional group which contains a quote
      (?P<value>             # then
        (?(quote)            #                if group <quote> was captured
          [\w;,#./\\=]+      # a quoted value
        |                    # or
          [\w#./\\]+         # an unquoted value
        )                    #
      )                      # followed by
      (?(quote)(?P=quote))   # the same quote, if the group presents.
    )
  )?
)
[;,]?

Try it on regex101.com.

Just place the optional quantifier, ?, after each of the quotation marks.

(?P<name>\w+)(=(\"?(?P<value>[\w;,#\.\/\\=]+)\"?|)?)[;,]?

Furthermore, I don't believe the additional capture groups are necessary.
And, you could decrease the character class requirement, using a lazy quantifier, ?.

(?P<name>[^ ]+?)=\"?(?P<value>.+?)\"?[;,]

An alternate method would be to preserve the quotation marks, and remove them from within the program.

(?P<name>[^ ]+?)=(?P<value>\".+?\"|.+?)[;,]

For reference, there is an RFC for cookies, which provides a definition for its syntax.
RFC 6265 – HTTP State Management Mechanism.

It appears that the delimiting value is a semi-colon, ;.

cookie-header     = "Cookie:" OWS cookie-string OWS
cookie-string     = cookie-pair *( ";" SP cookie-pair )

cookie-pair       = cookie-name "=" cookie-value
cookie-name       = token
cookie-value      = *cookie-octet / ( DQUOTE *cookie-octet DQUOTE )

token             = <token, defined in [RFC2616], Section 2.2>

token             = 1*<any CHAR except CTLs or separators>
separators        = "(" | ")" | "<" | ">" | "@"
                  | "," | ";" | ":" | "\" | <">
                  | "/" | "[" | "]" | "?" | "="
                  | "{" | "}" | SP | HT