🚧 Improved name parsing logic (WIP)

Author: pmonks

src/lice_comb/impl/id_detection.clj

@@ -371,7 +371,7 @@
        :fn         (constantly ["Zlib" :high])}
       ])))
 
-(defn- parse-id
+(defn- detect-id
   "If a match occured for the given regex element when tested against string s,
   returns a map containing the following keys:
   * :id         The SPDX license or exception identifier that was determined
@@ -395,7 +395,7 @@
               :start      (:start match)}
              (when (seq confidence-explanations) {:confidence-explanations confidence-explanations})))))
 
-(defn parse-ids
+(defn detect-ids
   "Returns a sequence (NOT A SET!) of maps where each key is a SPDX license or
   exception identifier (a String) that was found in s, and the value is a
   sequence containing a single map describing how the identifier was determined.
@@ -411,7 +411,7 @@
   Results are in the order in which they appear in the string, and the function
   returns nil if there were no matches."
   [s]
-  (when-let [matches (seq (filter identity (e/pmap* (partial parse-id s) @license-name-matching-d)))]
+  (when-let [matches (seq (filter identity (e/pmap* (partial detect-id s) @license-name-matching-d)))]
     (some->> matches
              (med/distinct-by :id)    ;####TODO: THINK ABOUT MERGING INSTEAD OF DROPPING
              (sort-by :start)

src/lice_comb/impl/parsing.clj

@@ -207,8 +207,8 @@
                 (when-let [ids (parse-uri s)]
                   (map #(hash-map (key %) (val %)) ids))
 
-                ; 5. Attempt to parse ids from the name
-                (lciid/parse-ids s)
+                ; 5. Attempt to detect ids in the string
+                (lciid/detect-ids s)
 
                 ; 6. No clue, so return a single info map, but with a made up "UNIDENTIFIED-" value (NOT A LICENSEREF!) instead of an SPDX license or exception identifier
                 (let [id (str "UNIDENTIFIED-" s)]

test/lice_comb/impl/id_parsing_test.clj test/lice_comb/impl/id_detection_test.clj

@@ -16,13 +16,13 @@
 ; SPDX-License-Identifier: Apache-2.0
 ;
 
-(ns lice-comb.impl.id-parsing-test
-  (:require [clojure.test               :refer [deftest testing is use-fixtures]]
-            [clojure.set                :as set]
-            [rencg.api                  :as rencg]
-            [lice-comb.impl.utils       :as lcu]
-            [lice-comb.test-boilerplate :refer [fixture testing-with-data]]
-            [lice-comb.impl.id-parsing  :refer [init! version-re only-or-later-re agpl-re lgpl-re gpl-re gnu-re parse-ids]]))
+(ns lice-comb.impl.id-detection-test
+  (:require [clojure.test                :refer [deftest testing is use-fixtures]]
+            [clojure.set                 :as set]
+            [rencg.api                   :as rencg]
+            [lice-comb.impl.utils        :as lcu]
+            [lice-comb.test-boilerplate  :refer [fixture testing-with-data]]
+            [lice-comb.impl.id-detection :refer [init! version-re only-or-later-re agpl-re lgpl-re gpl-re gnu-re detect-ids]]))
 
 (use-fixtures :once fixture)
 
@@ -248,5 +248,5 @@
     (is (every? not-nil? (map (partial test-regex gnu-re) gnu-licenses)))))
 
 (deftest match-regexes-tests
-  (testing-with-data "GNU Family Regexes - correct identifier results" #(mapcat keys (parse-ids %)) gnu-licenses-and-ids)
-  (testing-with-data "CC Family Regexes - correct identifier results"  #(mapcat keys (parse-ids %)) cc-by-licenses-and-ids))
+  (testing-with-data "GNU Family Regexes - correct identifier results" #(mapcat keys (detect-ids %)) gnu-licenses-and-ids)
+  (testing-with-data "CC Family Regexes - correct identifier results"  #(mapcat keys (detect-ids %)) cc-by-licenses-and-ids))