feat: support for UR tags, including multi-lines UR tags

customcommander · Jun 24, 2020 · b098ca6 · b098ca6
1 parent 9c7aece
commit b098ca6
Show file tree

Hide file tree

Showing 8 changed files with 96 additions and 21 deletions.
diff --git a/README.md b/README.md
@@ -37,9 +37,11 @@ parse
 
 ## Conversion Table
 
-| Tag | Key  | Description    |
-|:----|:-----|:---------------|
-| TY  | type | Reference type |
+| Tag | Key     | Description    |
+|:----|:--------|:---------------|
+| TY  | type    | Reference type |
+| KW  | keyword | Keywords       |
+| UR  | url     | Web/URL        |
 
 ## Development
 

diff --git a/grammar.js b/grammar.js
@@ -19,10 +19,25 @@ function id(x) { return x[0]; }
     {"name": "RECORD$ebnf$1", "symbols": ["RECORD$ebnf$1", "OTHER_TAG"], "postprocess": function arrpush(d) {return d[0].concat([d[1]]);}},
     {"name": "RECORD", "symbols": ["RTYPE", "RECORD$ebnf$1", "EOR"], "postprocess": ast => ast.filter(x => x !== null)},
     {"name": "RTYPE", "symbols": [(lexer.has("TY") ? {type: "TY"} : TY), (lexer.has("SEP") ? {type: "SEP"} : SEP), (lexer.has("TY_VAL") ? {type: "TY_VAL"} : TY_VAL), "__"], "postprocess": ([,,{value}]) => ({key: 'type', value})},
-    {"name": "OTHER_TAG", "symbols": ["KEYWORD"], "postprocess": ([d]) => d},
+    {"name": "OTHER_TAG$subexpression$1", "symbols": ["KEYWORD"]},
+    {"name": "OTHER_TAG$subexpression$1", "symbols": ["URL"]},
+    {"name": "OTHER_TAG", "symbols": ["OTHER_TAG$subexpression$1"], "postprocess": ([[d]]) => d},
     {"name": "KEYWORD$ebnf$1", "symbols": ["LINE"]},
     {"name": "KEYWORD$ebnf$1", "symbols": ["KEYWORD$ebnf$1", "LINE"], "postprocess": function arrpush(d) {return d[0].concat([d[1]]);}},
-    {"name": "KEYWORD", "symbols": [(lexer.has("KW") ? {type: "KW"} : KW), (lexer.has("SEP") ? {type: "SEP"} : SEP), "KEYWORD$ebnf$1"], "postprocess": ([,,lines]) => ({key: 'keyword', value: lines.join(' ')})},
+    {"name": "KEYWORD", "symbols": [(lexer.has("KW") ? {type: "KW"} : KW), (lexer.has("SEP") ? {type: "SEP"} : SEP), "KEYWORD$ebnf$1"], "postprocess":  ([,,lines]) =>
+        ( { key: 'keyword'
+          , value: lines.join(' ')
+          }
+        )
+                        },
+    {"name": "URL$ebnf$1", "symbols": ["LINE"]},
+    {"name": "URL$ebnf$1", "symbols": ["URL$ebnf$1", "LINE"], "postprocess": function arrpush(d) {return d[0].concat([d[1]]);}},
+    {"name": "URL", "symbols": [(lexer.has("UR") ? {type: "UR"} : UR), (lexer.has("SEP") ? {type: "SEP"} : SEP), "URL$ebnf$1"], "postprocess":  ([,,lines]) =>
+        ( { key: 'url'
+          , value: lines.flatMap(line => line.split(';').map(url => url.trim()).filter(Boolean))
+          }
+        )
+                        },
     {"name": "LINE", "symbols": [(lexer.has("CONTENT") ? {type: "CONTENT"} : CONTENT), "__"], "postprocess": ([{value}]) => value},
     {"name": "EOR", "symbols": [(lexer.has("ER") ? {type: "ER"} : ER), (lexer.has("SEP") ? {type: "SEP"} : SEP), "_"], "postprocess": () => null}
 ]

diff --git a/grammar.ne b/grammar.ne
@@ -13,11 +13,24 @@ RECORD      ->  RTYPE OTHER_TAG:* EOR
 RTYPE       ->  %TY %SEP %TY_VAL __
                 {% ([,,{value}]) => ({key: 'type', value}) %}
 
-OTHER_TAG   ->  KEYWORD
-                {% ([d]) => d %}
+OTHER_TAG   ->  (KEYWORD | URL)
+                {% ([[d]]) => d %}
 
 KEYWORD     ->  %KW %SEP LINE:+
-                {% ([,,lines]) => ({key: 'keyword', value: lines.join(' ')}) %}
+                {% ([,,lines]) =>
+                        ( { key: 'keyword'
+                          , value: lines.join(' ')
+                          }
+                        )
+                %}
+
+URL         ->  %UR %SEP LINE:+
+                {% ([,,lines]) =>
+                        ( { key: 'url'
+                          , value: lines.flatMap(line => line.split(';').map(url => url.trim()).filter(Boolean))
+                          }
+                        )
+                %}
 
 LINE        ->  %CONTENT __
                 {% ([{value}]) => value %}

diff --git a/index.js b/index.js
@@ -14,6 +14,10 @@ const to_record =
                     , acc.keyword.push(value)
                     , acc
                     )
+    , 'url'    , (acc, {value}) =>
+                    ( acc.url = (acc.url || []).concat(value)
+                    , acc
+                    )
     , acc => acc
     );
 

diff --git a/lexer.js b/lexer.js
@@ -10,6 +10,7 @@ const lexer =
       , SEP: "  - "
       , TY: "TY"
       , KW: "KW"
+      , UR: "UR"
       , ER: "ER"
       , TY_VAL: [ "ABST"   , "ADVS"  , "AGGR"
                 , "ANCIENT", "ART"   , "BILL"
@@ -31,7 +32,7 @@ const lexer =
                 , "STAT"   , "THES"  , "UNBILL"
                 , "UNPD"   , "VIDEO"
                 ]
-      , CONTENT: /[a-zA-Z0-9 \-\.']+/
+      , CONTENT: /[a-zA-Z0-9 \-\.':/;]+/
       }
     );
 

diff --git a/ris-parser.feature b/ris-parser.feature
@@ -21,12 +21,39 @@ Scenario Outline: Field mapping
     TY  - JOUR
     KW  - foo
     KW  - bar
+    UR  - http://example.com
     ER  - 
     """
   When I parse the file
   Then I will find a reference where '<field>' is set to '<value>'
 
   Examples:
-    | field    | value          |
-    | type     | JOUR           |
-    | keyword  | ["foo", "bar"] |
+    | field    | value                  |
+    | type     | JOUR                   |
+    | keyword  | ["foo", "bar"]         |
+    | url      | ["http://example.com"] |
+
+Scenario: URLs
+  Given I have this RIS file
+    """
+    TY  - JOUR
+    KW  - foo
+    UR  - url1
+    UR  - url2; url3
+    KW  - bar
+    UR  - url4;
+    url5;
+    url6;
+    ER  - 
+    """
+  When I parse the file
+  Then I will find a reference where 'url' is set to
+    """
+    [ "url1"
+    , "url2"
+    , "url3"
+    , "url4"
+    , "url5"
+    , "url6"
+    ]
+    """
diff --git a/sample.ris b/sample.ris
@@ -1,17 +1,20 @@
 TY  - JOUR
-KW  - cortical contusion
-KW  - blood-brain barrier
-KW  - horseradish peroxidase
-KW  - hippocampus
-KW  - rat
+UR  - http://example.com
+UR  - http://example.net
+KW  - foo
+KW  - bar
+KW  - baz
 ER  - 
 
 TY  - BOOK
-KW  - Epidemiology
-KW  - U.S. Gov't
+KW  - foo
+KW  - bar
+UR  - http://example.com; http://example.org;
+http://example.net
 ER  - 
 
 TY  - CHAP
-KW  - HIV
-KW  - AIDS
+UR  - http://example.net
+KW  - foo
+UR  - http://example.com; http://example.org
 ER  - 
diff --git a/steps.js b/steps.js
@@ -22,4 +22,14 @@ defineStep('I will find a reference where {string} is set to {string}', function
     expected_value = value; // it's a string
   }
   assert.deepStrictEqual(this.list[0][field], expected_value);
+});
+
+defineStep('I will find a reference where {string} is set to', function (field, value) {
+  let expected_value;
+  try {
+    expected_value = JSON.parse(value);
+  } catch (e) {
+    expected_value = value; // it's a string
+  }
+  assert.deepStrictEqual(this.list[0][field], expected_value);
 });