Representation of an element in a page via an identifier

A method performed by a compiler of the processor comprising:

• encountering a first element in a web page for a first time;
• computing, by the compiler of the processor coupled to a memory, a first identifier for the first element using an algorithm responsive to the encountering of the first element,
• wherein the first identifier represents a subtree;
• causing, by the compiler, the first identifier to be stored in a storage device using a hash representation of the first identifier to minimize a density of the storage device that is used when a history is unavailable for the first element in the storage device,
• wherein the first identifier is available without re-computing the first identifier when the first identifier exists as the hash representation in the history, and
• wherein a re-computing of the first identifier for the first element using the algorithm responsive to the encountering of the first element is performed and the re-computed first identified is stored as the hash representation in the history for future reference when the first identifier does not exists as the hash representation in the history;
• subsequently encountering, by the compiler, the first element two or more times in a different location of the web page and in a second web page;
• representing, by the compiler, the first element two or more times time of the subsequent encounters by retrieving the first identifier from the storage device,
• wherein the algorithm computes the first identifier in a bottom-up fashion starting with leaf nodes and continuing to a root node, in a deterministic way to yield the first identifier when computing the first identifier, and
• wherein the algorithm computes the first identifier in screening operation that comprises a first stripping out of the first element when it is deemed unimportant, a grouping of similar leaf nodes under the root node, a second stripping out of the first element when the first element is repeated two or more times within the similar leaf nodes under the root node, and repeating the grouping and the second stripping until there are no more duplicates to group;
• and
• recursively computing respective unique identifiers for each of one or more subsequent elements derived from the first element,
• wherein each of the one or more subsequent elements derived are from the first identifier and represented in the subtree by the respective unique identifiers,
• wherein the recursive computing includes:
• encountering a second element in a web page for a second time,
• wherein the second element is derived from the first identifier;
• computing a second identifier for the second element using the algorithm,
• wherein second identifier is based on having encountered the second element, is different than the first identifier, and is stored within the subtree represented by the first identifier;
• causing the second identifier to be stored in the storage device using a hash representation of the second identifier to minimize the density of the storage device that is used;
• subsequently encountering the second element two or more times in a second different location of the web page and in the second web page;
• and
• representing the second element by the second identifier at the time of the subsequent encounters by retrieving the second identifier from the storage device.