Abstract: The present invention relates to the exchange of data between a server and a receiving device. The exchange method comprises receiving, at the receiving device, a push message comprising pushed data from the server; storing received pushed data in a cache memory of the receiving device, the stored data being identified as being of push type; transmitting, from the receiving device to the server, a request for data comprising information about pushed data stored in the cache memory of the receiving device; and receiving, from the server, at the receiving device, a response to said request comprising requested data.

Abstract: A method for providing resources to a client, the method being implemented at a server and comprising: making available an indication to the client, indicating that the server is configured to activate a persistent push feature upon authorization of said client, said persistent push feature allowing the server to push resources to the client at any time; and receiving a first message from the client, said first message being based on said indication and allowing the server to continuously push resources to the client.

Abstract: The present invention relates to optimizing pushing of at least one resource associated with a main resource, in response to a request for this main resource, in an intermediary component comprising a cache memory. After having sent a request for a main resource to a second component different from the intermediary component, at least one identifier of at least one resource to be pushed by the second component is received from the second component in response to the sent request. Then, at least one of said at least one received identifier is stored in a list associated with the main resource, the list being stored in the cache memory.

Abstract: The invention concerns a method of exchanging information items, e.g. HTTP headers, between a server device and a plurality of clients, and also concerns such a server device. The server device establishes connections with clients, wherein each connection involves a server-initiated indexing table for the server device to encode information items to be sent over the connection in the server-to-client direction. The method comprises the following steps performed at the server device: obtaining a same single indexing table as the server-initiated indexing table of the connections; in response to receiving a request for data from a client, pushing entries of the table to the client to configure the latter for item exchange in the server-to-client direction, encoding information items associated with the requested data using item indexing based on the table and sending the encoded information items to the client over the connection established with it.

Abstract: There is provided methods for managing streaming over communication networks. Server and client devices share a push policy so that the client device may anticipate data pushes by the server. Anticipation makes it possible to cancel early the sending of some pushed data, therefore reducing bandwidth consumption. The shared push policy may be implicit to both server and client. In embodiments, it is explicitly specified by the server to the client, for example embedded in the Media Presentation Description file or included in a specific HTTP header. The client may also request an update of the shared push policy to meet its own requirements.

Abstract: The invention relates to generating a web feed composed of content elements. A number of content elements to be added to the web feed is determined, according to a number of available content elements to be published and according to a predefined visibility period associated with the web feed and defining a minimal time period for which any content element has to remain present in the web feed, wherein the determining of the number of elements comprises estimating a predicted number of new content elements liable to arrive during said visibility period, from historical data of said web feed representing statistics of arrivals, during past visibility periods, of content elements to be published. From the number of available content elements to be published, the determined number of content elements are selected, and entries corresponding to the selected content elements are created in the web feed.

Abstract: Methods and network devices for communicating data packets in a set of transmissions between a transmitter and a receiver of a communications network are described. The method includes prior exchanging transmitter-provided packet information and receiver-provided packet information between the transmitter and the receiver, generating an initial compression dictionary before the first transmission, based on the exchanged transmitter-provided packet information and the exchanged receiver-provided packet information. At the transmitter, the method further includes compressing the data packets of the first and subsequent transmissions of the set using the initial compression dictionary, and sending the compressed data packets over the communications network to the receiver.

Abstract: The invention relates to processing a service request by a web runtime environment in a processing device, the processing of the service request enabling a service provider to provide a service requested in the service request. After having selected a specific interface based on the service request, a web driver application associated with the service requested in the service request is executed and the selected specific interface is implemented. Then, it is possible to interact with the web driver application, via said specific interface, for providing the service by the service provider.

Abstract: A processing method and an associated device for accessing a spatio-temporal part of a compressed video sequence. The method includes the following steps: obtaining a request for access to the part by identifying a temporal section defined between initial and final times of the initial and final spatial regions—different from the initial region of the sequence at the initial and final times; decomposing the access request into a plurality of elementary requests for access to a video fragment, each elementary request identifying a fixed spatial region to extract in a temporal sub-interval of the temporal section; and exploiting at least one elementary request to access video fragments constituting the part to access.

Abstract: The present invention relates generally to network quality estimation using HTTP/2 push feature. When network quality estimation is required, a client web application requests a server to enter a probing mode. During the probing mode, the server pushes useful data to the application for the latter to perform the estimation. This avoids wasting bandwidth since the useful data are likely to be used by the application. Furthermore, during the probing mode, the server notifies the application of the push start and the push end of each pushed data. This is one using a web socket established with the application. This makes it possible to overcome the limitations of the application in accessing HTTP/2 features of a client browser. Based on the notifications, the application is able to perform accurate estimations.

Abstract: The invention relates to a device and a method for processing requests. In particular, the invention concerns building a dependency tree between the requests received from the client device for prioritizing the processing of the requests by the server device; maintaining a list of requests of the dependency tree in their removal order from the dependency tree while the server device is prioritizing the processing of the requests; and removing a selection of requests from the dependency tree wherein the selection is based on the order of requests in the list.

Abstract: The present invention concerns coding an XML-type structured document. The structured document includes values to be coded. The coding uses at least one indexing dictionary which includes entries. At least one entry value of the dictionary linked to the value to be coded is identified, by matching between at least part of one of the values and a subpart of the other value. The value to be coded is coded by reference, according to the matching, to at least the identified entry of the dictionary.

Abstract: The invention relates to a method for controlling communication between web applications in a web runtime environment such as a web browser. The method comprises the step of: executing code of a service provider application by the web runtime environment thereby providing a data object; obtaining a list of one or more service requester applications, each service requester application being adapted to generate a service request and implementing an interface for receiving a data object in response to the generated service request; enabling the selection, among the obtained list, of one or more requester applications adapted to receive data objects of the same type as the provided data object; and communicating the provided data object to the selected one or more requester applications via their corresponding interface. The invention also relates to a computer program and a processing device adapted to execute the method.

Abstract: The invention relates to a server and a method for processing requests when several requests compete within the same connection having fixed resources. Conventionally, a priority scheme provides priority values for a priority parameter that express priorities relatively to the priorities of other requests, meaning that a relative allocation of resources is obtained for each request. To achieve a better control on the server by the client, the invention provides specific possible values for the priority parameter that correspond to respective processing profiles. Each processing profile represents specific server behaviour for resources allocation, by defining a predefined amount of resources. Thus, the client may choose between using the specific possible values to control the exact amount of resources that will be allocated by the server to process some specific requests, and using the other possible values that correspond to a relative allocation of resources as already done in known techniques.

Abstract: A method and system for coding/decoding structured documents having structural items that include one or more elements, by generating configuration data including a unified model representative of the structure of a type of element on the basis of at least one structured configuration document. The unified model is modified to improve compression efficiency by grouping at least two transitions between a same starting item and a group of following items into a single transition from the same starting item in a first priority level and assigning a second priority level for one or more transitions within the grouping. A coding/decoding table using the modified unified model associates first and second priority coding values.

Abstract: The present invention relates to the notification of a server device with the availability of resources in cache memories of a client device and to the serving of digital resources in such a client-server communication system. The notifying method comprises: obtaining a first list of resources available in the cache memories of the client device; filtering the first list according to filtering criteria relating to a resource parameter, to obtain a filtered list of fewer resources available in the client device or splitting the first list according to splitting criteria relating to a resource parameter, to obtain a plurality of sub-lists of resources available in the client device; and notifying the server device with data structures representing the filtered list or sub-lists of resources.

Abstract: The present invention concerns a method for processing a structured document to render, such as XML or HTML files. The method comprises the steps of: parsing the structured document into parsed tokens TK of structured data; constructing a first tree structure DTree storing the structured data of parsed tokens; constructing a render tree structure RTree storing the document content to render, said render tree structure being synchronized with the first tree structure; rendering the structured document based on the render tree structure RTree; wherein constructing the first tree structure DTree comprises deciding, for each parsed token TK, whether or not the parsed token is to be stored in said first tree structure, and storing the structured data of the parsed token therein only in case of positive decision. Partial first (DOM) tree DTree is therefore stored in memory, reducing memory use and processing time before rendering the structured document.

Abstract: The invention concerns a method of exchanging information items, e.g. HTTP headers, between a server device and a plurality of clients, and also concerns such a server device. The server device establishes connections with clients, wherein each connection involves a server-initiated indexing table for the server device to encode information items to be sent over the connection in the server-to-client direction. The method comprises the following steps performed at the server device: obtaining a same single indexing table as the server-initiated indexing table of the connections; in response to receiving a request for data from a client, pushing entries of the table to the client to configure the latter for item exchange in the server-to-client direction, encoding information items associated with the requested data using item indexing based on the table and sending the encoded information items to the client over the connection established with it.

Abstract: The present invention relates to the exchange of data between a server and a receiving device. The exchange method comprises receiving, at the receiving device, a push message comprising pushed data from the server; storing received pushed data in a cache memory of the receiving device, the stored data being identified as being of push type; transmitting, from the receiving device to the server, a request for data comprising information about pushed data stored in the cache memory of the receiving device; and receiving, from the server, at the receiving device, a response to said request comprising requested data.

Abstract: This invention relates to a method of encoding a path in layered form so as to allow partial decoding and progressive rendering while decoding. The path is made of an original sequence of original draw commands, each corresponding to an original portion of a drawing. The method of encoding includes transforming at least one original draw command into an approximating draw command according to a geometrical approximation of the corresponding original portion of a drawing, to obtain a reduced sequence, encoding the reduced sequence into an encoded reduced sequence, and encoding a reconstruction information set representing the difference between the reduced sequence and the original sequence.