gecko-dev/caps/nsIPrincipal.idl
J. Ryan Stinnett 2a55d065b7 Bug 1238160 - Rename OriginAttributes.mInBrowser and associated methods. r=bz,mayhemer
This change renames OriginAttributes.mInBrowser to mInIsolatedMozBrowser and
nsIPrincipal::GetIsInBrowserElement to GetIsInIsolatedMozBrowserElement.  Other
methods that pass these values around also have name changes.

Tokens such as "inBrowser" have previously been serialized into cache keys, used
as DB column names, stored in app registries, etc.  No changes are made to any
serialization formats.  Only runtime method and variable names are updated.

No behavior changes are made in this patch, so some renamed methods may have
nonsensical implementations.  These are corrected in subsequent patches
focused on behavior.

MozReview-Commit-ID: 66HfMlsXFLs
2016-03-02 10:35:56 -06:00

373 lines
15 KiB
Plaintext

/* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4 -*- */
/* This Source Code Form is subject to the terms of the Mozilla Public
* License, v. 2.0. If a copy of the MPL was not distributed with this
* file, You can obtain one at http://mozilla.org/MPL/2.0/. */
/* Defines the abstract interface for a principal. */
#include "nsISerializable.idl"
%{C++
struct JSPrincipals;
#include "nsCOMPtr.h"
#include "nsTArray.h"
%}
interface nsIURI;
interface nsIContentSecurityPolicy;
interface nsIDOMDocument;
[ptr] native JSContext(JSContext);
[ptr] native JSPrincipals(JSPrincipals);
[ptr] native PrincipalArray(nsTArray<nsCOMPtr<nsIPrincipal> >);
[scriptable, builtinclass, uuid(d0391e86-1ad7-4ab0-bb7c-14d6d9967369)]
interface nsIPrincipal : nsISerializable
{
/**
* Returns whether the other principal is equivalent to this principal.
* Principals are considered equal if they are the same principal, or
* they have the same origin.
*/
boolean equals(in nsIPrincipal other);
/**
* Like equals, but takes document.domain changes into account.
*/
boolean equalsConsideringDomain(in nsIPrincipal other);
%{C++
inline bool Equals(nsIPrincipal* aOther) {
bool equal = false;
return NS_SUCCEEDED(Equals(aOther, &equal)) && equal;
}
inline bool EqualsConsideringDomain(nsIPrincipal* aOther) {
bool equal = false;
return NS_SUCCEEDED(EqualsConsideringDomain(aOther, &equal)) && equal;
}
%}
/**
* Returns a hash value for the principal.
*/
[noscript] readonly attribute unsigned long hashValue;
/**
* The codebase URI to which this principal pertains. This is
* generally the document URI.
*/
readonly attribute nsIURI URI;
/**
* The domain URI to which this principal pertains.
* This is null unless script successfully sets document.domain to our URI
* or a superdomain of our URI.
* Setting this has no effect on the URI.
* See https://developer.mozilla.org/en-US/docs/Web/Security/Same-origin_policy#Changing_origin
*/
[noscript] attribute nsIURI domain;
/**
* Returns whether the other principal is equal to or weaker than this
* principal. Principals are equal if they are the same object or they
* have the same origin.
*
* Thus a principal always subsumes itself.
*
* The system principal subsumes itself and all other principals.
*
* A null principal (corresponding to an unknown, hence assumed minimally
* privileged, security context) is not equal to any other principal
* (including other null principals), and therefore does not subsume
* anything but itself.
*/
boolean subsumes(in nsIPrincipal other);
/**
* Same as the previous method, subsumes(), but takes document.domain into
* account.
*/
boolean subsumesConsideringDomain(in nsIPrincipal other);
%{C++
inline bool Subsumes(nsIPrincipal* aOther) {
bool subsumes = false;
return NS_SUCCEEDED(Subsumes(aOther, &subsumes)) && subsumes;
}
inline bool SubsumesConsideringDomain(nsIPrincipal* aOther) {
bool subsumes = false;
return NS_SUCCEEDED(SubsumesConsideringDomain(aOther, &subsumes)) && subsumes;
}
%}
/**
* Checks whether this principal is allowed to load the network resource
* located at the given URI under the same-origin policy. This means that
* codebase principals are only allowed to load resources from the same
* domain, the system principal is allowed to load anything, and null
* principals can only load URIs where they are the principal. This is
* changed by the optional flag allowIfInheritsPrincipal (which defaults to
* false) which allows URIs that inherit their loader's principal.
*
* If the load is allowed this function does nothing. If the load is not
* allowed the function throws NS_ERROR_DOM_BAD_URI.
*
* NOTE: Other policies might override this, such as the Access-Control
* specification.
* NOTE: The 'domain' attribute has no effect on the behaviour of this
* function.
*
*
* @param uri The URI about to be loaded.
* @param report If true, will report a warning to the console service
* if the load is not allowed.
* @param allowIfInheritsPrincipal If true, the load is allowed if the
* loadee inherits the principal of the
* loader.
* @throws NS_ERROR_DOM_BAD_URI if the load is not allowed.
*/
void checkMayLoad(in nsIURI uri, in boolean report,
in boolean allowIfInheritsPrincipal);
/**
* A Content Security Policy associated with this principal.
*
* Use this function to query the associated CSP with this principal.
*/
[noscript] readonly attribute nsIContentSecurityPolicy csp;
/*
* Use this function to query a CSP associated with this principal.
* If no CSP is associated with this principal then one is created
* internally and setRequestContext is called on the CSP using aDocument.
*
* Please note if aDocument is null, then setRequestContext on the
* CSP object is called using the current principal.
*/
[noscript] nsIContentSecurityPolicy ensureCSP(in nsIDOMDocument aDocument);
/**
* A speculative Content Security Policy associated with this
* principal. Set during speculative loading (preloading) and
* used *only* for preloads.
*
* If you want to query the CSP associated with that principal,
* then this is *not* what you want. Instead query 'csp'.
*/
[noscript] readonly attribute nsIContentSecurityPolicy preloadCsp;
/*
* Use this function to query a speculative CSP associated with this
* principal. If no speculative CSP is associated with this principal
* then one is created internally and setRequestContext is called on
* the CSP using aDocument.
*
* Please note if aDocument is null, then setRequestContext on the
* speculative CSP object is called using the current principal.
*/
[noscript] nsIContentSecurityPolicy ensurePreloadCSP(in nsIDOMDocument aDocument);
/**
* The CSP of the principal in JSON notation.
* Note, that the CSP itself is not exposed to JS, but script
* should be able to obtain a JSON representation of the CSP.
*/
readonly attribute AString cspJSON;
/**
* Returns the jar prefix of the principal.
* The jar prefix is a string that can be used to isolate data or
* permissions between different principals while taking into account
* parameters like the app id or the fact that the principal is embedded in
* a mozbrowser.
* Some principals will return an empty string.
* Some principals will assert if you try to access the jarPrefix.
*
* The jarPrefix is intended to be an opaque identifier. It is currently
* "human-readable" but no callers should assume it will stay as is and
* it might be crypto-hashed at some point.
*/
readonly attribute AUTF8String jarPrefix;
/**
* A dictionary of the non-default origin attributes associated with this
* nsIPrincipal.
*
* Attributes are tokens that are taken into account when determining whether
* two principals are same-origin - if any attributes differ, the principals
* are cross-origin, even if the scheme, host, and port are the same.
* Attributes should also be considered for all security and bucketing decisions,
* even those which make non-standard comparisons (like cookies, which ignore
* scheme, or quotas, which ignore subdomains).
*
* If you're looking for an easy-to-use canonical stringification of the origin
* attributes, see |originSuffix| below.
*/
[implicit_jscontext]
readonly attribute jsval originAttributes;
/**
* A canonical representation of the origin for this principal. This
* consists of a base string (which, for codebase principals, is of the
* format scheme://host:port), concatenated with |originAttributes| (see
* below).
*
* We maintain the invariant that principalA.equals(principalB) if and only
* if principalA.origin == principalB.origin.
*/
readonly attribute ACString origin;
/**
* The base part of |origin| without the concatenation with |originSuffix|.
* This doesn't have the important invariants described above with |origin|,
* and as such should only be used for legacy situations.
*/
readonly attribute ACString originNoSuffix;
/**
* A string of the form !key1=value1&key2=value2, where each pair represents
* an attribute with a non-default value. If all attributes have default
* values, this is the empty string.
*
* The value of .originSuffix is automatically serialized into .origin, so any
* consumers using that are automatically origin-attribute-aware. Consumers with
* special requirements must inspect and compare .originSuffix manually.
*
* originsuffix are intended to be a replacement for jarPrefix, which will
* eventually be removed.
*/
readonly attribute AUTF8String originSuffix;
/**
* The base domain of the codebase URI to which this principal pertains
* (generally the document URI), handling null principals and
* non-hierarchical schemes correctly.
*/
readonly attribute ACString baseDomain;
const short APP_STATUS_NOT_INSTALLED = 0;
const short APP_STATUS_INSTALLED = 1;
const short APP_STATUS_PRIVILEGED = 2;
const short APP_STATUS_CERTIFIED = 3;
/**
* Gets the principal's app status, which indicates whether the principal
* corresponds to "app code", and if it does, how privileged that code is.
* This method returns one of the APP_STATUS constants above.
*
* Note that a principal may have
*
* appId != nsIScriptSecurityManager::NO_APP_ID &&
* appId != nsIScriptSecurityManager::UNKNOWN_APP_ID
*
* and still have appStatus == APP_STATUS_NOT_INSTALLED. That's because
* appId identifies the app that contains this principal, but a window
* might be contained in an app and not be running code that the app has
* vouched for. For example, the window might be inside an <iframe
* mozbrowser>, or the window's origin might not match the app's origin.
*
* If you're doing a check to determine "does this principal correspond to
* app code?", you must check appStatus; checking appId != NO_APP_ID is not
* sufficient.
*/
[infallible] readonly attribute unsigned short appStatus;
/**
* Gets the id of the app this principal is inside. If this principal is
* not inside an app, returns nsIScriptSecurityManager::NO_APP_ID.
*
* Note that this principal does not necessarily have the permissions of
* the app identified by appId. For example, this principal might
* correspond to an iframe whose origin differs from that of the app frame
* containing it. In this case, the iframe will have the appId of its
* containing app frame, but the iframe must not run with the app's
* permissions.
*
* Similarly, this principal might correspond to an <iframe mozbrowser>
* inside an app frame; in this case, the content inside the iframe should
* not have any of the app's permissions, even if the iframe is at the same
* origin as the app.
*
* If you're doing a security check based on appId, you must check
* appStatus as well.
*/
[infallible] readonly attribute unsigned long appId;
/**
* Gets the id of the user context this principal is inside. If this
* principal is inside the default userContext, this returns
* nsIScriptSecurityManager::DEFAULT_USER_CONTEXT_ID.
*/
[infallible] readonly attribute unsigned long userContextId;
/**
* Returns true iff the principal is inside an isolated mozbrowser element.
* <iframe mozbrowser mozapp> and <xul:browser> are not considered to be
* mozbrowser elements. <iframe mozbrowser noisolation> does not count as
* isolated since isolation is disabled. Isolation can only be disabled if
* the containing document is chrome.
*/
[infallible] readonly attribute boolean isInIsolatedMozBrowserElement;
/**
* Returns true if this principal has an unknown appId. This shouldn't
* generally be used. We only expose it due to not providing the correct
* appId everywhere where we construct principals.
*/
[infallible] readonly attribute boolean unknownAppId;
/**
* Returns true iff this is a null principal (corresponding to an
* unknown, hence assumed minimally privileged, security context).
*/
[infallible] readonly attribute boolean isNullPrincipal;
/**
* Returns true iff this principal corresponds to a codebase origin.
*/
[infallible] readonly attribute boolean isCodebasePrincipal;
/**
* Returns true iff this is an expanded principal.
*/
[infallible] readonly attribute boolean isExpandedPrincipal;
/**
* Returns true iff this is the system principal.
*/
[infallible] readonly attribute boolean isSystemPrincipal;
/**
* Returns true if this principal's origin is recognized as being on the
* whitelist of sites that can use the CSS Unprefixing Service.
*
* (This interface provides a trivial implementation, just returning false;
* subclasses can implement something more complex as-needed.)
*/
[noscript,notxpcom,nostdcall] bool IsOnCSSUnprefixingWhitelist();
};
/**
* If nsSystemPrincipal is too risky to use, but we want a principal to access
* more than one origin, nsExpandedPrincipals letting us define an array of
* principals it subsumes. So script with an nsExpandedPrincipals will gain
* same origin access when at least one of its principals it contains gained
* sameorigin acccess. An nsExpandedPrincipal will be subsumed by the system
* principal, and by another nsExpandedPrincipal that has all its principals.
* It is added for jetpack content-scripts to let them interact with the
* content and a well defined set of other domains, without the risk of
* leaking out a system principal to the content. See: Bug 734891
*/
[uuid(f3e177Df-6a5e-489f-80a7-2dd1481471d8)]
interface nsIExpandedPrincipal : nsISupports
{
/**
* An array of principals that the expanded principal subsumes.
* Note: this list is not reference counted, it is shared, so
* should not be changed and should only be used ephemerally.
*/
[noscript] readonly attribute PrincipalArray whiteList;
};