| ASRouter.jsm |
|
66649 |
| ASRouterFeed.jsm |
@class ASRouterFeed - Connects ASRouter singleton (see above) to Activity Stream's
store so that it can use the RemotePageManager.
|
1222 |
| ASRouterPreferences.jsm |
|
7099 |
| ASRouterTargeting.jsm |
CachedTargetingGetter
@param property {string} Name of the method called on ActivityStreamProvider
@param options {{}?} Options object passsed to ActivityStreamProvider method
@param updateInterval {number?} Update interval for query. Defaults to FRECENT_SITES_UPDATE_INTERVAL
|
21823 |
| ASRouterTriggerListeners.jsm |
Check current location against the list of whitelisted hosts
Additionally verify for redirects and check original request URL against
the whitelist.
@returns {object} - {host, url} pair that matched the whitelist
|
16833 |
| AboutPreferences.jsm |
Render preferences to an about:preferences content window with the provided
preferences structure.
|
9527 |
| ActivityStream.jsm |
|
23256 |
| ActivityStreamMessageChannel.jsm |
ActivityStreamMessageChannel - This module connects a Redux store to a RemotePageManager in Firefox.
Call .createChannel to start the connection, and .destroyChannel to destroy it.
You should use the BroadcastToContent, AlsoToOneContent, and AlsoToMain action creators
in common/Actions.jsm to help you create actions that will be automatically routed
to the correct location.
@param {object} options
@param {function} options.dispatch The dispatch method from a Redux store
@param {string} options.pageURL The URL to which a RemotePageManager should be attached.
Note that if it is about:newtab, the existing RemotePageManager
for about:newtab will also be disabled
@param {string} options.outgoingMessageName The name of the message sent to child processes
@param {string} options.incomingMessageName The name of the message received from child processes
@return {ActivityStreamMessageChannel}
|
10612 |
| ActivityStreamPrefs.jsm |
Prefs - A wrapper around Preferences that always sets the branch to
ACTIVITY_STREAM_PREF_BRANCH
|
3065 |
| ActivityStreamStorage.jsm |
@param storeNames Array of strings used to create all the required stores
|
3307 |
| BookmarkPanelHub.jsm |
@param {function} handleMessageRequest
@param {function} addImpression
@param {function} dispatch - Used for sending user telemetry information
|
10814 |
| CFRMessageProvider.jsm |
|
25820 |
| CFRPageActions.jsm |
A WeakMap from browsers to {host, recommendation} pairs. Recommendations are
defined in the ExtensionDoorhanger.schema.json.
A recommendation is specific to a browser and host and is active until the
given browser is closed or the user navigates (within that browser) away from
the host.
|
37241 |
| DiscoveryStreamFeed.jsm |
Send SPOCS Fill telemetry.
@param {object} filteredItems An object keyed on filter reasons, and the value
is a list of SPOCS.
reasons: blocked_by_user, frequency_cap, below_min_score, flight_duplicate
@param {boolean} fullRecalc A boolean indicating if it's a full recalculation.
Calling `loadSpocs` will be treated as a full recalculation.
Whereas responding the action "DISCOVERY_STREAM_SPOC_IMPRESSION"
is not a full recalculation.
|
53153 |
| DownloadsManager.jsm |
null for non-private downloads |
5689 |
| FaviconFeed.jsm |
Get favicon info (uri and size) for a uri from Places.
@param uri {nsIURI} Page to check for favicon data
@returns A promise of an object (possibly null) containing the data
|
5965 |
| FilterAdult.jsm |
Run some text through md5 and return the base64 result.
|
89163 |
| HighlightsFeed.jsm |
Chronologically sort highlights of all types except 'visited'. Then just append
the rest at the end of highlights.
@param {Array} pages The full list of links to order.
@return {Array} A sorted array of highlights
|
11136 |
| LinksCache.jsm |
Cache link results from a provided object property and refresh after some
amount of time has passed. Allows for migrating data from previously cached
links to the new links with the same url.
|
4801 |
| NaiveBayesTextTagger.jsm |
Determines if the tokenized text belongs to class according to binary naive Bayes
classifier. Returns an object containing the class label ("label"), and
the log probability ("logProb") that the text belongs to that class. If
the positive class is more likely, then "label" is the positive class
label. If the negative class is matched, then "label" is set to null.
|
2222 |
| NewTabInit.jsm |
NewTabInit - A placeholder for now. This will send a copy of the state to all
newly opened tabs.
|
1718 |
| NmfTextTagger.jsm |
A multiclass classifier that scores tokenized text against several classes through
inference of a nonnegative matrix factorization of TF-IDF vectors and
class labels. Returns a map of class labels as string keys to scores.
(Higher is more confident.) All classes get scored, so it is up to
consumer of this data determine what classes are most valuable.
|
2082 |
| OnboardingMessageProvider.jsm |
globals Localization |
20087 |
| PanelTestProvider.jsm |
|
15783 |
| PersistentCache.jsm |
A file (disk) based persistent cache of a JSON serializable object.
|
2801 |
| PersonalityProvider.jsm |
V2 provider builds and ranks an interest profile (also called an “interest vector”) off the browse history.
This allows Firefox to classify pages into topics, by examining the text found on the page.
It does this by looking at the history text content, title, and description.
|
13849 |
| PlacesFeed.jsm |
Observer - a wrapper around history/bookmark observers to add the QueryInterface.
|
15595 |
| PrefsFeed.jsm |
|
5594 |
| RecipeExecutor.jsm |
RecipeExecutor is the core feature engineering pipeline for the in-browser
personalization work. These pipelines are called "recipes". A recipe is an
array of objects that define a "step" in the recipe. A step is simply an
object with a field "function" that specifies what is being done in the step
along with other fields that are semantically defined for that step.
There are two types of recipes "builder" recipes and "combiner" recipes. Builder
recipes mutate an object until it matches some set of critera. Combiner
recipes take two objects, (a "left" and a "right"), and specify the steps
to merge the right object into the left object.
A short nonsense example recipe is:
[ {"function": "get_url_domain", "path_length": 1, "field": "url", "dest": "url_domain"},
{"function": "nb_tag", "fields": ["title", "description"]},
{"function": "conditionally_nmf_tag", "fields": ["title", "description"]} ]
Recipes are sandboxed by the fact that the step functions must be explicitly
whitelisted. Functions whitelisted for builder recipes are specifed in the
RecipeExecutor.ITEM_BUILDER_REGISTRY, while combiner functions are whitelisted
in RecipeExecutor.ITEM_COMBINER_REGISTRY .
|
32945 |
| RemoteL10n.jsm |
The downloaded Fluent file is located in this sub-directory of the local
profile directory.
|
2988 |
| Screenshots.jsm |
Get a screenshot / thumbnail for a url. Either returns the disk cached
image or initiates a background request for the url.
@param url {string} The url to get a thumbnail
@return {Promise} Resolves a custom object or null if failed
|
4115 |
| SearchShortcuts.jsm |
|
2617 |
| SectionsManager.jsm |
Generators for built in sections, keyed by the pref name for their feed.
Built in sections may depend on options stored as serialised JSON in the pref
`${feed_pref_name}.options`.
|
19026 |
| ShortURL.jsm |
Properly convert internationalized domain names.
@param {string} host Domain hostname.
@returns {string} Hostname suitable to be displayed.
|
2347 |
| SiteClassifier.jsm |
classifySite
Classifies a given URL into a category based on classification data from RemoteSettings.
The data from remote settings can match a category by one of the following:
- match the exact URL
- match the hostname or second level domain (sld)
- match query parameter(s), and optionally their values or prefixes
- match both (hostname or sld) and query parameter(s)
The data looks like:
[{
"type": "hostname-and-params-match",
"criteria": [
{
"url": "https://matchurl.com",
"hostname": "matchhostname.com",
"sld": "secondleveldomain",
"params": [
{
"key": "matchparam",
"value": "matchvalue",
"prefix": "matchpPrefix",
},
],
},
],
"weight": 300,
},...]
|
2977 |
| SnippetsTestMessageProvider.jsm |
|
23442 |
| Store.jsm |
Store - This has a similar structure to a redux store, but includes some extra
functionality to allow for routing of actions between the Main processes
and child processes via a ActivityStreamMessageChannel.
It also accepts an array of "Feeds" on inititalization, which
can listen for any action that is dispatched through the store.
|
6324 |
| SystemTickFeed.jsm |
|
1118 |
| TelemetryFeed.jsm |
|
32680 |
| TippyTopProvider.jsm |
|
1789 |
| Tokenize.jsm |
|
12657 |
| ToolbarBadgeHub.jsm |
Pref is set via Remote Settings message. We want to continously
monitor new messages that come in to ensure the one with the
highest priority is set.
|
12450 |
| ToolbarPanelHub.jsm |
|
19814 |
| TopSitesFeed.jsm |
|
26385 |
| TopStoriesFeed.jsm |
|
30684 |
| UTEventReporting.jsm |
Note: the schema can be found in
https://searchfox.org/mozilla-central/source/toolkit/components/telemetry/Events.yaml
|
2150 |
| UserDomainAffinityProvider.jsm |
Provides functionality to personalize content recommendations by calculating
user domain affinity scores. These scores are used to calculate relevance
scores for items/recs/stories that have domain affinities.
The algorithm works as follows:
- The recommendation endpoint returns a settings object containing
timeSegments and parametersets.
- For every time segment we calculate the corresponding domain visit counts,
yielding result objects of the following structure: {"mozilla.org": 12,
"mozilla.com": 34} (see UserDomainAffinityProvider#queryVisits)
- These visit counts are transformed to domain affinity scores for all
provided parameter sets: {"mozilla.org": {"paramSet1": 0.8,
"paramSet2": 0.9}, "mozilla.org": {"paramSet1": 1, "paramSet2": 0.9}}
(see UserDomainAffinityProvider#calculateScoresForParameterSets)
- The parameter sets provide factors for weighting which allows for
flexible targeting. The functionality to calculate final scores can
be seen in UserDomainAffinityProvider#calculateScores
- The user domain affinity scores are summed up across all time segments
see UserDomainAffinityProvider#calculateAllUserDomainAffinityScores
- An item's domain affinities are matched to the user's domain affinity
scores by calculating an item relevance score
(see UserDomainAffinityProvider#calculateItemRelevanceScore)
- The item relevance scores are used to sort items (see TopStoriesFeed for
more details)
- The data structure was chosen to allow for fast cache lookups during
relevance score calculation. While user domain affinities are calculated
infrequently (i.e. only once a day), the item relevance score (potentially)
needs to be calculated every time the feed updates. Therefore allowing cache
lookups of scores[domain][parameterSet] is beneficial
|
13556 |