diff --git a/RELEASE_NOTES.md b/RELEASE_NOTES.md new file mode 100644 index 00000000..87dd4614 --- /dev/null +++ b/RELEASE_NOTES.md @@ -0,0 +1,246 @@ +# Phileas Release Notes + +Issues whose identifiers start with `PHL-` were previously tracked in Jira before the project's issues were managed in GitHub. + +## Version 2.6.0 (not yet released) + +* PHL-313 - Remove OWNER project fork for managing application properties +* PHL-312 - Add death date detection similar to the birthdate detection +* PHL-311 - Return the probabilities of sentiment and offensiveness in addition to the predicted categories + +## Version 2.5.0 + +* PHL-309 - Load OpenNLP models from the classpath by default +* PHL-308 - Add sentiment as a condition + +## Version 2.4.0 + +* PHL-306 - Rename "filter profile" to "policy" +* PHL-305 - Allow dates to be shifted by a random value +* PHL-304 - Skip onnx runtime tests on OSX + +## Version 2.3.0 +No issues found Refresh + +## Version 2.2.1 + +* PHL-277 - Combine PDFs image into a new PDF +* PHL-264 - Bring back Flair as PersonsV1 filter + +## Version 2.2.0 + +* PHL-277 - Combine PDFs image into a new PDF - **05/Sep/23 12:27** - Today 1:37 +* PHL-264 - Bring back Flair as PersonsV1 filter + +## Version 2.1.0 + +* PHL-261 - Incorporate a sentence detector into the NER +* PHL-260 - Switch to Apache OpenNLP 2.0 instead of using ONNX Runtime directly +* PHL-259 - Allow for reading encryption values from environment variables +* PHL-258 - Implement format-preserving encryption + +## Version 2.0.0 + +* PHL-254 - Remove splitting capability +* PHL-251 - Replace phileas-ner with Java ONNX service + +## Version 1.12.2 + +* PHL-249 - Upgrade log4j to 2.16.0 + +## Version 1.12.1 + +* PHL-249 - Upgrade log4j to 2.16.0 + +## Version 1.12.0 + +* PHL-248 - Upgrade log4j to 2.15.0 +* PHL-246 - Remove store from FilterService +* PHL-245 - Add S3 key to S3FilterProfileService +* PHL-244 - Allow user to set bounding boxes for PDF redaction +* PHL-243 - Allow for combining filter profiles per request +* PHL-242 - A date filter test is failing due to mid-month +* PHL-239 - Support dates like Aug. 31, 2020 +* PHL-238 - Support ages like: 61 y/o +* PHL-233 - Filter profiles reference environment variable values +* PHL-227 - Support currency such as $.50 +* PHL-173 - 9 digit zip codes without a delimiter are not found - Improvement + +## Version 1.11.0 + +* PHL-226 - Add currency filter +* PHL-222 - Add bank routing number filter +* PHL-218 - Add document analysis prior to filtering +* PHL-211 - Encapsulate the constructor arguments to the filters + +## Version 1.10.1 +No issues found Refresh + +## Version 1.10.0 + +* PHL-217 - Change redisson delete to delete by index +* PHL-216 - Update project dependencies +* PHL-208 - Redact multiple occurrences of same span in line in PDF - Bug +* PHL-207 - Improve date year validation +* PHL-204 - Date "July 3, 2012" is not being identified - Bug +* PHL-201 - Add domain property to filter profile +* PHL-200 - Add properties to the filter profile to enable/disable post filters +* PHL-198 - Add property to NER filter to set a confidence threshold +* PHL-197 - Add function to return Span as a CSV +* PHL-196 - Create spans from LAPPS JSON +* PHL-195 - Include JPEG processing libraries for PDF extraction +* PHL-187 - PostFilter classes with empty constructors should be singletons +* PHL-183 - Adjust phone number confidence based on pattern +* PHL-162 - Fix RELATIVE date filter strategy for "October 2009" date + +## Version 1.9.0 + +* PHL-188 - Add street address filter +* PHL-184 - Remove new line characters from spans +* PHL-179 - Lowercase names in first name index +* PHL-178 - Surname filter tests aren't finding anything +* PHL-177 - Identify physician names +* PHL-176 - IDENTIFIER filter is finding just capitalized words +* PHL-175 - Add support for age format "64-year-old" +* PHL-174 - Ages should not identify "10 years" +* PHL-171 - Add redaction color for PDF +* PHL-169 - Add PDF document processing +* PHL-166 - Add redaction method to leave the last four digits +* PHL-165 - Redact dates to just the 4 digit year +* PHL-163 - Handle dates like 09-2021 +* PHL-161 - Fix logger class names in date filter and identifier filter +* PHL-160 - Add optional period in ages +* PHL-159 - Increase ner.timeout.sec to a large value +* PHL-158 - Set -1 to split threshold to set no limit +* PHL-157 - Resolve NPEs with dictionary filter +* PHL-4 - Shift dates by some time period + +## Version 1.8.0 + +* PHL-155 - Maintain metrics on individual filter performance times +* PHL-148 - Read list of filter ignored terms from a file + +## Version 1.7.0 + +* PHL-154 - Add Prometheus monitoring endpoint +* PHL-152 - Add detect() method to filters +* PHL-151 - Make filter ignore lists be not case-sensitive +* PHL-150 - BloomDictionaryFilter needs to support phrases and not just words +* PHL-149 - Fix IBAN code validation - Bug +* PHL-147 - Allow for ignoring based on patterns +* PHL-146 - Allow timeout between Philter and Philter NER to be customizable +* PHL-145 - Split large text into multiple pieces +* PHL-141 - Read list of globally ignored words from a file +* PHL-140 - Add classification condition to IdentifierFilterStrategy +* PHL-139 - Allow spaces in IBAN codes + +## Version 1.6.0 + +* PHL-143 - Fix spans not going to the store - Bug +* PHL-142 - Add PhileasSystemLoader to OWNER +* PHL-137 - Create filter for mailing tracking numbers +* PHL-136 - Replace DropWizard metrics with micrometer +* PHL-135 - Change date parsing to use year instead of year of era - Improvement +* PHL-134 - Fix filter type for Mac address - Bug +* PHL-133 - Fix case sensitivity check for ignored terms - Bug +* PHL-132 - Fix credit card validation - Bug +* PHL-131 - Upgrade to JUnit 5 +* PHL-129 - Change "label" to "classification" - Improvement +* PHL-128 - Fix ignore check for LuceneDictionaryFilter - Bug +* PHL-127 - Add fuzzy property to custom dictionary filter to trigger a bloom filter +* PHL-126 - Allow custom dictionaries to read terms from a file +* PHL-125 - Add tests for all FilterStrategy classes +* PHL-124 - Add condition for checking classification of a span +* PHL-123 - Expand US passport regex to only include valid first two digits +* PHL-122 - Fix mac address filter strategy assignment +* PHL-121 - Add option to disable metrics printed to the console +* PHL-120 - Set contextual words for each regex filter +* PHL-118 - Wrap regular expression patterns into an object + +## Version 1.5.0 + +* PHL-98 - Make cache generic and use cache for filter profiles +* PHL-97 - Incorporate the S3FilterProfileService into Phileas +* PHL-69 - Create "Section" filter that removes between start and end tags + +## Version 1.4.0 + +* PHL-96 - Upgrade libphonenumber to 8.12.1 +* PHL-95 - Add token condition to NerFilterStrategy +* PHL-93 - Presence of a confidence condition on a phone number filter strategy causes the text to not be redacted +* PHL-92 - The token condition specifies quotes twice +* PHL-89 - Add detection framework +* PHL-87 - Drop ignored spans before overlapping spans +* PHL-86 - Use "input" and "token" consistently +* PHL-85 - StateAbbreviationFilter is making a span with the input and not the state +* PHL-84 - Capture a span's window when creating a span + +## Version 1.3.1 + +* PHL-82 - Allow client to set document ID +* PHL-78 - Switch to Java 11 +* PHL-76 - Support credit card numbers with dashes +* PHL-75 - Add support for TIN to SsnFilter +* PHL-74 - Add filter for MAC addresses +* PHL-73 - Add a redaction option for a crypto value +* PHL-71 - Add replacement template value for the token +* PHL-70 - Add filter condition based on context + +## Version 1.3.0 + +* PHL-68 - When there are no strategies just redact +* PHL-67 - Don't identify "may" as a date +* PHL-66 - Check filter profile delete status +* PHL-65 - Span text should not end with a period or a space +* PHL-64 - Add regex for URLs containing IP addresses instead of domain name +* PHL-62 - Add option to URL filter to require http(s) protocol or www +* PHL-59 - Improve performance by tuning PyTorch filter client +* PHL-58 - Generate document ID based on input hash +* PHL-56 - Don't throw NPE when Philter-NER throws an error + +## Version 1.2.0 + +* PHL-55 - Fix issue where NER conditions are not being applied to NER spans +* PHL-54 - Add filter profile option to remove punctuation +* PHL-53 - Release 1.2.0 +* PHL-52 - Move Status class from Phileas to Philter +* PHL-50 - Offer encrypted connections to Redis +* PHL-47 - Add enabled property to filter types in filter profile +* PHL-46 - Add option to require dates to be valid dates +* PHL-42 - Add option to find invalid credit cards +* PHL-35 - Add ignore lists specific to individual filters +* PHL-33 - Test multiple identifiers in a filter profile + +## Version 1.1.0 + +* PHL-41 - Fix replacement checks in strategies +* PHL-39 - Allow for the filters to be reloaded at runtime +* PHL-38 - Don't use spring boot for dependency management +* PHL-36 - Integrate Philter Profile Registry into Phileas +* PHL-34 - Make store implementation for Elasticsearch +* PHL-32 - In IdentifierFilter fix the name/label assignment +* PHL-31 - Automatically set Lucene distance based on the string length +* PHL-30 - Add ignore lists to filter profile schema +* PHL-29 - Include item text in Span +* PHL-24 - Allow for custom dictionary lookup in filters +* PHL-23 - Get philter-ner endpoint from Philter +* PHL-21 - Add filter profile schema +* PHL-19 - Provide an enhanced response to describe how the spans were found/removed +* PHL-18 - Set default values for filter profile fields +* PHL-16 - Report metrics per individual filter type +* PHL-15 - Add metrics prefix property +* PHL-13 - Do not require filter profile name for StaticFilterProfileService +* PHL-3 - Apply sensitivity level to NER entities + +## Version 1.0.1 + +* PHL-11 - Fix issue where spans aren't getting applied - Bug +* PHL-10 - Remove unneeded guava dependency +* PHL-9 - Add license information to poms + +## Version 1.0.0 + +Initial release. + +* PHL-7 - RedisAnonymizationCacheService needs containsValue implemented +* PHL-5 - Allow Identifier filters to specify regex \ No newline at end of file