URL pattern matching

One of the powerful condition types in Userflow is the Current page conditions where you can target and control content based on URLs.

Userflow supports an easy-to-use, yet powerful URL pattern for matching pages in your app. URL pattern matching is useful when auto-starting flows, progressing to the next step, or marking a checklist task as completed.

Note that the following URL pattern matching rules only apply to Current page conditions and not to Navigate to page actions. For Navigate to page actions you will have to put the full url and insert user attributes (sent via userflow.js) for dynamic values.

Quickstart

• Only supply parts of the URL you care about. Parts left out will match anything. E.g. if you don’t care about the domain, then leave it out.  

• * is a wildcard, matching anything.  

• :segment (a colon followed by any word) matches a single dynamic segment.  

• You can supply multiple patterns to include or exclude. The current page URL must match at least one of the included patterns, and must not match any of the excluded patterns.  

Exact path (no subpaths) on any domain

/app/dashboard

Matches any domain, with exactly the given path.

• Matches:    

  • https://example.com/app/dashboard      

  • http://example.com/app/dashboard      

  • https://other-example.com/app/dashboard      

• Does not match:    

  • https://example.com/app      

  • https://example.com/app/dashboard/sub      

  • https://example.com/app/dashboard/sub/page      

Including subpaths

/app*

* matches everything (including nothing).

• Matches:    

  • https://example.com/app      

  • https://example.com/app/projects      

  • https://example.com/app/projects/1      

  • https://other-example.com/app      

  • https://example.com/applications      

• Does not match:    

  • https://example.com/settings      

Only subpaths

/app/*

Note the / before the *.

• Matches:    

  • https://example.com/app/projects      

  • https://example.com/app/projects/1      

  • https://other-example.com/app/projects      

• Does not match:    

  • https://example.com/app      

  • https://example.com/applications      

Dynamic path segment

/projects/:project/details

:project matches a single path segment (anything but /). You can use any word (consisting of letters a-z, numbers and underscores) in place of project - it’s not used for anything.

• Matches:    

  • https://example.com/projects/1/details      

  • https://example.com/projects/2/details      

• Does not match:    

  • https://example.com/projects/1/details/edit      

  • https://example.com/projects/1/history      

Mix dynamic path segment and subpaths

/app/:company/projects/*

• Matches:    

  • https://example.com/app/apple/projects/1      

  • https://example.com/app/banana/projects/1      

  • https://example.com/app/banana/projects/1/details      

• Does not match:    

  • https://example.com/app/apple/projects      

Specific domain on any path

app.com

• Matches:    

  • https://app.com/      

  • https://app.com/any/path      

• Does not match:    

  • https://www.app.com/      

  • https://not-app.com/      

Dynamic subdomain

:subdomain.app.com

• Matches:    

  • https://www.app.com/      

  • https://www.app.com/any/path      

  • https://dashboard.app.com/      

• Does not match:    

  • https://app.com/      

  • https://multiple.levels.app.com/      

  • https://www.not-app.com/      

Mix domain and path

:subdomain.example.com/app/:company/projects\*

• Matches:    

  • https://dashboard.example.com/app/apple/projects      

  • https://dashboard.example.com/app/apple/projects/1      

URL parts

URLs consist of the following parts:

https://my.app.com/acme-inc/dashboard?tab=analytics#charts
\___/   \________/\_________________/ \___________/ \____/
  |         |              |                |          |
scheme    domain          path            query     fragment

When writing URL patterns in Userflow, all parts are optional. If a part is left out, it’ll match anything.

Each part has slightly different rules.

Scheme

The scheme is what comes before :// in a URL.

https://my.app.com/acme-inc/dashboard?tab=analytics#charts
\___/
  |
scheme

The scheme is almost always either http or https. You only need to specify scheme in your URL patterns if you care about whether the user is on http or https. Usually your server takes care of redirecting users to one or the other, meaning in most cases you can leave it out of your pattern.

• https://example.com matches any URL on https://example.com but not on http://example.com.  

• example.com matches any URL on both https://example.com and http://example.com.  

Domain

The domain part is what comes after :// and before the path’s first /.

https://my.app.com/acme-inc/dashboard?tab=analytics#charts
        \________/
            |
          domain

Dynamic subdomain

To match dynamic subdomains (e.g. apple.app.com and banana.app.com), use a colon and an arbitrary name such as :company.app.com. :company will match a non-empty subdomain, meaning it will not include .. The name after : does not matter and is currently not used for anything.

:company.app.com and :subdomain.app.com are equivalent and will match:

• apple.app.com  

• banana.app.com  

They will not match:

• app.com (missing subdomain)  

• more.apple.app.com (more subdomains)  

Wildcard domains

You can also use * as a wildcard character to match any number of subdomains.

*.app.com will match both one.app.com and one.two.app.com, but not app.com.

*app.com will match app.com, but will also match myapp.com.

Path

The path part is what comes after the domain and before the query (starting with ?), or the fragment (starting with #) if there’s no query.

https://my.app.com/acme-inc/dashboard?tab=analytics#charts
                  \_________________/
                           |
                          path

Important: Paths MUST start with a slash (/). The URL pattern app/projects will consider app as the domain, and will match full URLs such as https://app/projects (which doesn’t really make sense). The right pattern here would be /app/projects.

Dynamic path segments

To match dynamic path segments (e.g. /app/apple/projects and /app/banana/projects), use a colon and an arbitrary name such as /app/:company/projects. :company will match a non-empty path segment, meaning it will not include /. The name after : does not matter and is currently not used for anything.

/app/:company/projects and /app/:slug/projects are equivalent and will match:

• /app/apple/projects  

• /app/banana/projects  

They will not match:

• /app/projects (missing path segment)  

• /app/apple/projects/1 (deeper path)  

Wildcard paths

You can also use * as a wildcard character to match nested paths.

/app/* will match both /app/one and /app/one/two, but not /app.

/app* will match /app, but will also match /applications.

/*/projects will match /app/projects and /app/one/two/three/projects.

You can mix wildcards and dynamic path segments at will. Example: /*/customers/:customer/invoices/:invoice/*

Query

The query is what comes after ? and before the fragment (starting with #), if any.

https://my.app.com/acme-inc/dashboard?tab=analytics#charts
                                      \___________/
                                            |
                                          query

A URL can contain multiple query parameters separated by ampersands (&), as in ?fruit=apple&tab=analytics. Query parameters are matched one by one. Order of query parameters does not matter. Extra, non-matched query parameters, are not considered.

?fruit=apple&tab=analytics will match:

• ?fruit=apple&tab=analytics  

• ?tab=analytics&fruit=apple (different order is OK)  

• ?other=test&fruit=apple&another=test&tab=analytics (extra parameters are ignored)  

But will not match:

• ?fruit=apple (tab parameter is missing)  

• ?fruit=banana&tab=analytics (wrong value for fruit parameter)  

Wildcard query parameter values

You can use * to match any value in individual query parameters. It’s required that the query parameter is present though.

?param=a* will match:

• ?param=apple  

• ?param=alphabet  

It will not match:

• ?param=banana (does not start with a)  

• ?param (no value)  

• ?param2=apple (wrong parameter name)  

?param=* will match:

• ?param=apple  

• ?param (no value, but present)  

It will not match:

• ?param2=apple (wrong parameter name)  

Fragment

The fragment is what comes after #.

https://my.app.com/acme-inc/dashboard?tab=analytics#charts
                                                    \____/
                                                       |
                                                    fragment

Since the fragment is often used like a path in single-page applications, fragment obeys the same rules as Paths.

To match a dynamic fragment segment: #/projects/:project/dashboard.

Using wildcard: #/projects/:project/dashboard/*.