Path Constraints and Link Exclusions

Back to the main Table of Contents

Article Navigation

Path Constraints

Introduction
Instructions

Link Excludes

Introduction
Instructions

Additional Information

Path constraints

This chapter gives instructions and examples for how to set up Path Constraints for the Monsido scan.

Path constraints are regular expressions. A Regular Expression (regex) is an API that defines string patterns. Use a regex to search, manipulate, and edit a string in Java. Email validation and passwords are two examples of strings where Regex can define the constraints.

Return to Article Navigation

Introduction

Use Path Constraints to instruct the scan to only process parts of a domain. The scan compares URLs in the site with the pattern given. URLs that match the pattern are scanned and URLs that do not match the pattern are considered external links.

A Path Constraint can be a word or a regular expression. In most cases, users set up Path Constraints to:

Restrict the scanner to only recognize parts of a site with a pattern. For example:
```
^/en
```
This instructs the scan to handle any URL that does not begin with /en (for example, http://foo(dot)com/fr/bar) as an external link. For example, the crawler tests the link but does not follow any links on http://foo(dot)com/fr/bar.
Instruct the scan to ignore parts of the site with a pattern. For example:
```
!^/fr
```
This instructs the scan to handle any URL that begin with /fr (for example, http://foo(dot)com/fr/bar) as an external link. This means that the scan tests the link but does not follow any links on http://foo(dot)com/fr/bar.

The difference between the two is that in the first case ALL pages under /en are scanned and nothing else. In the second example, all pages EXCEPT /fr are scanned.

IMPORTANT! Make sure that the URL for the domain is set to a page that matches the constraint.

If this is not done, just one page is scanned - since the scanner cannot proceed to any other page than the page it starts on.

For example, with a constraint of:

^/en/booking

the crawler cannot start on http://foo(dot)com. The crawler will request http://foo(dot)com, receive the page, and find that no links match http://foo(dot)com/en/booking, and therefore only the first page is scanned.

Return to Article Navigation

Instructions

From the Monsido Domain Overview (globe icon), click Settings (gear icon) at the top of the page. The Admin Settings page opens.

Note: The Settings button is only available to site admins.
On the same row as the domain to scan, click Action.
In the drop-down list, select Edit Domain.

The Edit Domain page opens.
Scroll to the Advanced Domain Options section.
Locate the Path Constraints section:
- Search: Enter a search parameter for matching strings within the Constraint Patterns list.
- Constraint pattern: Enter a constraint pattern.
- + Add: Click + to add a new Constraint pattern. An empty row is added to the list.
  
  Note: The window only shows the first five items. With more than five list items, a paginate function sorts the list items.
- Delete: Click the trashcan icon to delete an item from the list.
Note: Some CMS selections have pre-filled parameters, as per the majority of users with those CMS operators. Click Default to remove them if needed. The Default button only appears for certain operators.

For more information and examples, see the external article:
- Class Patterns by Oracle.

Return to Article Navigation

Link Excludes

This chapter gives instructions and examples for how to set up the link exclusions. Link exclusions are regular expressions. A Regular Expression (regex) is an API to define String patterns. Use a regex to search, manipulate, and edit a string in Java. Email validation and passwords are two examples of strings where Regex can define the constraints.

Return to Article Navigation

Introduction

Choose to exclude a word or a regular expression. Use Link Excludes to instruct the crawler to completely ignore a link on the pages. Pages that match the pattern are not tested.

Use Link excludes to:

Filter out print pages with a pattern such as print=true

This instructs the scan to ignore (and not test) any URL with the pattern, for example:
http://foo.com/bar?print=true.
Filter out redirected login pages with a pattern such as:
login.aspx?return_url=zyx.

This instructs the scan to ignore all URLs with the pattern, for example: http://foo(dot)com/bar/login.aspx?return_url=zyx.

Tip! If "Scan subdomains" is turned on for the domain, use the § sign in front of the exclude pattern to match URLs that use the full string instead of the relative one. For example, to exclude the scan for the "blog" subdomain, enter as the pattern:

§http://blog.foo.bar

Note: Regular Expressions in source-code exclusions are not 100% compatible with Monsido Policies. The languages are different (Java and Ruby).

Return to Article Navigation

Instructions

From the Monsido Domain Overview (globe icon), click Settings (gear icon) at the top of the page. The Admin Settings page opens.

Note: The Settings button is only available to site admins.
On the same row as the domain to scan, click Action.
In the drop-down list, select Edit Domain.

The Edit Domain page opens.
Scroll to the Advanced Domain Options section.
In the Link Excludes section:
- Search: Enter a search parameter for matching strings within the Link excludes list.
- Exclude pattern: Enter a pattern to exclude from the scan.
  
  Note: The window only shows the first five items. With more than five list items, a paginate function begins to sort consecutive list items.
- + Add: Click + to add a new Exclude pattern. An empty row appears in the list.
- Delete: Click the trashcan icon to delete an item from the list.
- Internal URLs:
  - Operator: Click the drop-down arrow to select Contains, Starts with, or Regex.
  - URL: Type a URL in the field.
  - Delete: Click the trashcan icon to delete the row.
  - + Add: Click to add a new Input Selector. An empty row appears in the list.
Important! It is possible to do a link exclusion for a link that is attached to an image. The link is then excluded from the scan. However, the image itself could still appear on the SEO and QA pages as an issue to be fixed if it does not meet other requirements (for example, missing ALT text).