Support external algorithms and versions

This release adds the ability to use external algorithms and versions. This is accomplished using the ExternalStagingFileDataProvider. This change did not affect the internal algorithms (CS and TNM) and they are still accessed the same way.

Internal cleanup

Remove StagingAlgorithm and StagingVersion classes. They were not used by the library.

Minor internal changes

Changes were made to the library to help support its usage in SEER*API. None of these changes affect the usage of the library directly.

TNM 1.4

This release adds TNM 1.4 which includes the following changes:

1. Oral Cavity schemas and Larynx Glottic: Path Stage Group notes regarding combinations only found in SEER*RSA were corrected. Stage Group IVA should have referenced Tis in addition to T0.

2. NasoPharynx: Path Stage Group notes regarding combinations only found in SEER*RSA were corrected. Stage Group IV is only in SEER*RSA; Stage Group II and III Tis combination is only SEER*RSA.

3. Corpus Carcinoma: Clin N and Path N: N1 definition modified to replace text for FIGO IIIC with 'Statement of FIGO IIIC only' because it was confusing to have metastasis of para-aortic lymph nodes in both N1 and N2.

4. Rectum: The combined T table was replaced with the table used by Colon. The Colon table does NOT generate an error when both Clin T and Path T are pTis and Behavior = 3.

TNM 1.3 was removed from the library.

TNM v1.3

TNM 1.3 was released with the following differences since 1.2:

• Added 88 to the Derived Stage Group calculation tables in order to correctly process 88 in the T, N, M and Assigned Stage fields for Lymphoma, LymphomaOcularAdnexa, MelanomaSkin and the 4 MerkelCell schemas. 88 is the required value for a DCO case.
• Registrar Notes for Melanoma Skin Clin T and Path T were revised to clarify the definitions of T1a and T1b.

TNM 1.2 is not longer available in the library.

Context references

Context references are values that reference another value in the context. For example, the common way to set a value in the context is:

"outputs" : [ {
  "key" : "derived_version",
  "name" : "Derived Version",
  "default" : "1.2"
}

Will result in a a new key, derived_version being added to the context with a value of 1.2.

However there are system context values added automatically at the start of every staging execution. One of them is ctx_alg_version. It will be set to the version of the algorithm that is being run. So same example above can be accomplished with context references now:

"outputs" : [ {
  "key" : "derived_version",
  "name" : "Derived Version",
  "default" : "{{ctx_alg_version}}"
}

The final value will be 1.2 just like before but the default will not have to be changed every time a new version is released.

Also note that context references can reference any value in the context, just just system context values like ctx_alg_version.

Context references already were being using in processing table input and output values. Now they can be used for the following as well:

• initial context
• input defaults
• output defaults

TNM 1.2

TNM 1.2 was released with only a couple of minor differences since 1.1:

• Added 88 as a possible choice for the clin and path stage groups of Lymphoma, LymphomaOcularAdnexa, MelanomaSkin and the 4 MerkelCell schemas. 88 is the required value for a DCO case.
• Modified Nasopharynx schema selection for Site = C111 to remove SSF25 = 981. 981 should only be selected if Site = C110, C112, C113, C118, C119

TNM 1.1 is not longer available in the library.

CS 02.05.50 AJCC 6 Cleanup

There were extra tables in some of the CS 02.05.50 AJCC 6 mappings for certain schemas. They did not affect the final staging values but could sometimes produce a false error during staging. The error could be ignored but removing the tables is a better solution.

Here are the AJCC 6 mapping tables that were removed.

Schema                   Mapping          Table Removed
--------------------     -------------    -------------------------------------
cervix               ->  mapping_ajcc6 -> extension_size_ajcc7_xgi
cervix               ->  mapping_ajcc6 -> csextension_size_tablefor_ajcc6th_xba
gist_colon           ->  mapping_ajcc6 -> mitotic_rate_calculation_xpi
gist_esophagus       ->  mapping_ajcc6 -> mitotic_rate_calculation_xqe
gist_rectum          ->  mapping_ajcc6 -> mitotic_rate_calculation_xpi
gist_small_intestine ->  mapping_ajcc6 -> mitotic_rate_calculation_xqe
gist_stomach         ->  mapping_ajcc6 -> mitotic_rate_calculation_xqe
mycosis_fungoides    ->  mapping_ajcc6 -> ssf1_jaj

New loading cache implementation

Switched the loading cache implementation from Guava to Caffeine. Caffeine showed roughly a 20% performance increase in our heavily threaded integration tests.

Caffeine requires Java 8 so that will become the new minimum JDK requirement. In addition, since the loading cache was the main thing being used from Guava that dependency was removed from the library.

Encoding fix

The code which updated the schemas and tables from SEER*API was not handling encoding correctly. This caused issues with certain characters being converted incorrectly in some table notes. The code was corrected and the tables were also updated to fix the bad characters.