Techniques for recognizing multiple patterns within a string
View Patent ↗Techniques for recognizing multiple patterns within a string of characters are presented. A dictionary is hierarchically organized, such that leaf nodes within the dictionary represents words defined in the dictionary. A string of characters are received. Each character within the string is traversed by attempting to match it with a character defined in the dictionary. As long as a match continues with the dictionary the characters within the string are traversed. Once a longest possible match to a word within the dictionary is found, the next character following the last matched character for the string is processed.
1. A machine-implemented method, comprising:
organizing characters from words into a hierarchy, the words are obtained from a dictionary and organized into the hierarchy, each node of the hierarchy representing a single character representing a single letter and each first character of a particular word appears as a particular node within the hierarchy, and each leaf node of the hierarchy represents a particular one of the words, each word representing a common word, a domain specific word, an acronym for that word, a colloquialism for that word, an abbreviation for that word, a proper name, or a common misspelling for that word, and representing the dictionary as a directory where each character from each word is stored as its own independent subdirectory and a folder within the directory, each word also including an annotation indicating the origin of that word;
receiving a target string;
iterating each target character of the string and attempting to assemble a substring of the target characters that match to a particular leaf node of the hierarchy; and
identifying target words from the target string in response matched substrings found.
2. The method of claim 1 , wherein iterating further includes considering each target character as a potential start to a new word within the dictionary and persisting each target character that continues to match a potential word within the dictionary until it is determined that a match occurs or does not occur.
3. The method of claim 1 , wherein iterating further includes maintaining, during each iteration, a first pointer to a current target character and a processing pointer to a last potential matching target character to assist in assembling the substring.
4. The method of claim 3 , wherein iterating further includes advancing the first pointer to a next target character when a match is not found with the current target character.
5. The method of claim 3 , wherein iterating further includes advancing the first pointer to a location within the target string beyond a length of the substring when the substring is a match unless the location is beyond a total length of the target string in which case iterating ceasing with the match of the substring.
6. A machine-implemented method, comprising:
receiving characters associated with a string each character representing a single letter;
iteratively traversing the characters of the string finding substrings that match words of a hierarchically organized dictionary until each character of the string has been traversed, the hierarchically organized dictionary is a hierarchy where each node is a single character and where each leaf node represents a particular one of the words, and each word representing a common word, a domain specific word, an acronym for that word, a colloquialism for that word, an abbreviation for that word, a proper name, or a common misspelling for that word, the hierarchically organized dictionary stored as a directory where each character of each word is stored as its own independent sub-directory and a folder within the directory and each word including an annotation indicating an origin of that word; and
returning each substring, which is found from the string and that is also located in the hierarchically organized dictionary, to a requesting service for subsequent processing by that requesting service.
7. The method of claim 6 , wherein iteratively traversing further includes allowing a configurable number of garbage characters to be ignored in a middle of a particular one of the found substrings.
8. The method of claim 6 , wherein iteratively traversing further includes allowing a configurable number of replacement characters to be included in a middle of a particular one of the found substrings.
9. The method of claim 8 , wherein allowing further includes recognizing the replacement characters as special substitute characters permissible for dictionary words defined in the hierarchically organized dictionary.
10. The method of claim 6 , wherein iteratively traversing further includes traversing each character for a longest matching length for a word found in the hierarchically organized dictionary before advancing to a next character in the string for processing.
11. The method of claim 10 , wherein traversing further includes advancing to the next character by the longest matching length when a word is found, the longest matching length representing a word length for the word found.
12. The method of claim 6 , wherein iteratively traversing further includes advancing to an immediate next character in the string when a prior character being traversed is not a match to words defined in the hierarchically organized dictionary.
13. A system, comprising:
a hierarchically organized dictionary of words residing in a machine-accessible and computer-readable medium and accessible to a substring matching service that executes on a machine, the hierarchically organized dictionary of words stored in the machine-accessible and computer-readable medium as a directory where each character of each word is stored as its own independent sub-directory and a folder of the directory, and each word including an annotation indicating an origin of that word;
the substring matching service implemented in a machine-accessible and computer-readable medium and to process on the machine;
the substring matching service parses each character of a target string looking for a longest successive match of a substring of those characters within the hierarchically organized dictionary, each character representing a letter, the hierarchically organized dictionary of words is assembled so that each node of a hierarchy is a single character and each leaf node presents a particular word from the hierarchically organized dictionary of words and the substring matching service returns each of the substrings found to a requesting service for subsequent processing and manipulation by that requesting service, and each word representing a common word, a domain specific word, an acronym for that word, a colloquialism for that word, an abbreviation for that word, a proper name, or a common misspelling for that word.
14. The system of claim 13 , wherein the hierarchy is organized as a tree data structure where the leaf nodes of the tree data structure represents the words defined in the hierarchically organized dictionary of words.
15. The system of claim 14 , wherein the tree data structure is dynamically created as in memory or in cache data on behalf of the substring matching service.
16. The system of claim 13 , wherein the substring matching service permits a configurable number of garbage or replacement characters to appear or to be ignored for some of the found substrings.
17. The system of claim 13 , wherein substring matching service finds at least two or more substrings in the hierarchically organized dictionary for the target string being processed.
18. A system, comprising:
a password service implemented in a machine-accessible and computer-readable medium and is to process on a machine; and
a substring matching service implemented in a machine-accessible and computer-readable medium and is to process on the machine or a different machine;
wherein the password service passes passwords received as strings to the substring matching service, and the substring matching service identifies the substrings included within those passwords by searching a hierarchically organized dictionary of words and when matching substrings are found returning the matching substrings to the password service for evaluation against a policy and for subsequent action by the password service and the hierarchically organized dictionary of words is organized as a hierarchy, each node representing a single character and each leaf node represent a particular word from the hierarchically organized dictionary of words, each character representing a letter and, each word representing a common word, a domain specific word, an acronym for that word, a colloquialism for that word, an abbreviation for that word, a proper name, or a common misspelling for that word, and the hierarchically organized dictionary of words stored in the machine-accessible and computer-readable medium as a directory where each character of each word is stored as its own independent sub-directory and a folder of the directory and each word including an annotation indicating an origin of that word.
19. The system of claim 18 , wherein the substring matching service permits a configurable number of garbage or replacement characters to be ignored when attempting to match the substrings to the words of the hierarchically organized dictionary.
20. The system of claim 19 , wherein the policy precludes a particular password from being used when a particular found substring matches a particular word of the hierarchically organized dictionary.
21. The system of claim 20 , wherein the policy precludes a particular password from being used when a configurable number of found substrings matches for multiple words within the hierarchically organized dictionary.