fix: escape terms containing characters with special meaning in regExp

chore(): Count term occurrences based on AST nodes. Don't count occurrences in term's own definition. This change is necessary because with introducing linkification of term occurrences in glossaries, too, we also counted occurrences in the term's own definition. This way we were no longer able to detect orphan terms and term definitions used no longer anywhere else.
about-code · Dec 23, 2019 · 1815304 · 1815304
1 parent cc199a3
commit 1815304
Show file tree

Hide file tree

Showing 29 changed files with 714 additions and 126 deletions.
diff --git a/lib/counter.js b/lib/counter.js
@@ -0,0 +1,28 @@
+const uVisit = require('unist-util-visit');
+const {getLinkUrl} = require('./ast-tools');
+/**
+ * Unified plug-in to count occurrences and mentions of a term.
+ * Won't count occurrences in the terms own definition.
+ */
+function counter(context) {
+    return () => (tree, vFile) => {
+        uVisit(tree, 'term-occurrence', (node, idx, parent) => {
+            node.termDefs.forEach(term => {
+                if (vFile.path === term.glossary.vFile.path) {
+                    // current file is the glossary in which the term has been
+                    // defined...
+                    if (getLinkUrl(node.headingNode) !== term.anchor) {
+                        // ...count term occurence only, if it is not in the
+                        // terms own definition.
+                        term.countOccurrence();
+                    }
+                } else {
+                    term.countOccurrence();
+                }
+            });
+        });
+        return tree;
+    };
+}
+
+module.exports = { counter };
diff --git a/lib/indexer.js b/lib/indexer.js
@@ -29,11 +29,12 @@ const index = {}
 function indexer(context) {
     const indexFilename = getIndexFilename(context);
     if (! indexFilename) {
-        return () => (tree, vFile) => {};
+        return () => (tree, vFile) => tree;
     } else {
         return () => (tree, vFile) => {
             const currentDocFilename = `${vFile.dirname}/${vFile.basename}`;
             uVisit(tree, 'term-occurrence', getNodeVisitor(context, indexFilename, currentDocFilename));
+            return tree;
         };
     }
 }

diff --git a/lib/linker.js b/lib/linker.js
@@ -1,6 +1,7 @@
 const Term = require("./term.js");
 const linkifyRegex = require("./linkify");
 const {indexer} = require("./indexer");
+const {counter} = require("./counter");
 const {printAst, noopCompiler} = require("./ast-tools.js");
 const {toForwardSlash} = require("./pathplus");
 const path = require("path");
@@ -44,6 +45,7 @@ api.linkTermOccurrences = function(context) {
                 .use(remark_ref_links)
                 .use(remark_link_headings, {behavior: 'wrap'})
                 .use(indexer(context))
+                .use(counter(context))
                 .use(printAst(context.opts.dev.printOutputAst))
                 .use(noopCompiler)
                 .use(remark_stringify)
@@ -142,7 +144,6 @@ function linkify(txtNode, headingNode, termDefs, context, vFile) {
             headingNode: headingNode
         });
         if (! hasMultipleDefs) {
-            term.countOccurrence();
             if (term.hint) {
                 if (/\$\{term\}/.test(term.hint)) {
                     linkNode.children[0].value = term.hint.replace("${term}", linkNode.children[0].value);
@@ -169,7 +170,6 @@ function linkify(txtNode, headingNode, termDefs, context, vFile) {
         paragraph.children = childr
             .slice(0, linkIndex + 1)
             .concat(termDefs.map((t, i) => {
-                t.countOccurrence();
                 return {
                     type: "link",
                     title: t.getShortDescription(),

diff --git a/lib/term.js b/lib/term.js
@@ -76,8 +76,9 @@ function updateRegExp(term) {
     // sort by length descending to create a regExp which tests for the longest
     // term first.
     termAndAliases
+        .map((term) => escapeRegExp(term))
         .sort((term1, term2) => term2.length - term1.length)
-        .forEach((alias, idx)  => regExp += (idx > 0 ? "|" : "") + alias);
+        .forEach((term, idx) => regExp += (idx > 0 ? "|" : "") + term)
     regExp += `)`;
 
     if (term.ignoreCase) {

diff --git a/test/input/config-shared/glossarify-md.conf.json b/test/input/config-shared/glossarify-md.conf.json
@@ -7,10 +7,9 @@
     "experimentalFootnotes": true,
     "keepRawFiles": [],
     "glossaries": [
-        { "file": "./markdown/glossary.md", "termHint": "↴" },
-        { "file": "./substring_behavior/glossary.md", "termHint": ""},
-        { "file": "./term_descriptions/glossary.md", "termHint": "ⓕ"},
-        { "file": "./aliases_and_synonyms/glossary.md", "termHint": ""}
+        { "file": "./markdown/glossary.md", "termHint": "↴" }
+        ,{ "file": "./term_descriptions/glossary.md", "termHint": "ⓕ" }
+        ,{ "file": "./aliases_and_synonyms/glossary.md", "termHint": "" }
     ],
     "linking": "relative",
     "ignoreCase": false,

diff --git a/test/input/config-tailored/count-term-occurrences/document.md b/test/input/config-tailored/count-term-occurrences/document.md
@@ -0,0 +1,15 @@
+# Count Term Occurrences
+
+## Once
+
+GIVEN a document
+AND it mentions the term "Mentioned-in-document-once" once
+AND noohere else
+THEN the term's term occurrence count MUST be 1.
+
+## Twice
+
+GIVEN a document
+AND it mentions the term "Mentioned-in-document-twice" once
+AND it mentions the term "Mentioned-in-document-twice" twice
+THEN the term's term occurrence count MUST be 2.
diff --git a/test/input/config-tailored/count-term-occurrences/glossarify-md.conf.json b/test/input/config-tailored/count-term-occurrences/glossarify-md.conf.json
@@ -0,0 +1,19 @@
+{
+    "$schema": "../../../../conf.schema.json",
+    "baseDir": ".",
+    "outDir": "../../../output-actual/config-tailored/count-term-occurrences",
+    "includeFiles": ["."],
+    "excludeFiles": [],
+    "keepRawFiles": [],
+    "glossaries": [
+        { "file": "./glossary.md", "termHint": "" }
+    ],
+    "linking": "relative",
+    "ignoreCase": false,
+    "dev": {
+        "termsFile": "../../../output-actual/config-tailored/count-term-occurrences/terms.json",
+        "printInputAst": false,
+        "reproducablePaths": true,
+        "printOutputAst": false
+    }
+}
diff --git a/test/input/config-tailored/count-term-occurrences/glossary.md b/test/input/config-tailored/count-term-occurrences/glossary.md
@@ -0,0 +1,52 @@
+# Glossary
+
+## Zero
+
+GIVEN a term "Zero"
+AND the term "Zero" is mentioned twice in its own definition
+AND nowhere else
+THEN this term's occurrence count MUST be 0.
+
+## One
+
+GIVEN a term "One"
+AND the term is mentioned only once in a subsequent term definition
+AND nowhere else
+THEN this term's occurrence count MUST be 1.
+
+#### mention-one-once
+
+GIVEN this definition
+AND it mentions another term "One" once
+AND itself is being mentioned nowhere else
+THEN the other term's occurrence count MUST be 1
+AND this term's occurrence count MUST be 0.
+
+## Two
+
+GIVEN a term "Two"
+AND the term is mentioned term two times in a subsequent term definition
+AND nowhere else
+THEN this term's occurrence count MUST be 2.
+
+#### mention-two-twice
+
+GIVEN this definition
+AND it mentions another term "Two" Two times
+AND itself is being mentioned nowhere else
+THEN the other term's occurrence count MUST be 2
+AND this term's occurrence count MUST be 0.
+
+## Mentioned-in-document-once
+
+GIVEN a term
+AND it is mentioned once in a document
+AND not in the glossary
+THEN this term's occurrence count MUST be 1.
+
+## Mentioned-in-document-twice
+
+GIVEN a term
+AND it is mentioned twice in a document
+AND not in the glossary
+THEN this term's occurrence count MUST be 2.
diff --git a/test/input/config-tailored/glossary-crosslinks/glossarify-md.conf.json b/test/input/config-tailored/glossary-crosslinks/glossarify-md.conf.json
@@ -0,0 +1,21 @@
+{
+    "$schema": "../../../../conf.schema.json",
+    "baseDir": ".",
+    "outDir": "../../../output-actual/config-tailored/glossary-crosslinks",
+    "includeFiles": ["."],
+    "excludeFiles": [],
+    "experimentalFootnotes": true,
+    "keepRawFiles": [],
+    "glossaries": [
+         { "file": "./glossary-g1.md", "termHint": "" }
+        ,{ "file": "./glossary-g2.md", "termHint": "" }
+    ],
+    "linking": "relative",
+    "ignoreCase": false,
+    "dev": {
+        "termsFile": "../../../output-actual/config-tailored/glossary-crosslinks/terms.json",
+        "printInputAst": false,
+        "reproducablePaths": true,
+        "printOutputAst": false
+    }
+}
diff --git a/test/input/config-tailored/glossary-crosslinks/glossary-g1.md b/test/input/config-tailored/glossary-crosslinks/glossary-g1.md
@@ -0,0 +1,21 @@
+# Glossary G1
+
+## Term A
+
+"Term A" only exists for the sake of being linked to from within this glossary.
+
+## Term B
+
+GIVEN a term "Term A" AND a definition of "Term B" in the same glossary G1
+THEN any occurrence of "Term A" in the definition of "Term B" MUST be linked
+to the definition of "Term A".
+
+## Term C
+
+GIVEN a term "Term C" in glossary G1 AND a term "Term D" in another glossary G2
+THEN any occurrence of "Term D" in glossary G1 MUST be linked to the definition
+of "Term D".
+
+## Term E
+
+"Term E" only exists for the sake of being linked to from glossary G2.
diff --git a/test/input/config-tailored/glossary-crosslinks/glossary-g2.md b/test/input/config-tailored/glossary-crosslinks/glossary-g2.md
@@ -0,0 +1,12 @@
+# Glossary G2
+
+## Term D
+
+"Term D" only exists for the sake of being linked to by the definition of
+"Term C" in Glossary G1.
+
+## Term F
+
+GIVEN a term "Term F" in glossary G2 AND a term "Term E" in another glossary G1
+THEN any occurrence of "Term E" in glossary G2 MUST be linked to the definition
+of "Term E".
diff --git a/...fig-shared/substring_behavior/document.md → ...g-tailored/substring_behavior/document.md b/...fig-shared/substring_behavior/document.md → ...g-tailored/substring_behavior/document.md
diff --git a/test/input/config-tailored/substring_behavior/glossarify-md.conf.json b/test/input/config-tailored/substring_behavior/glossarify-md.conf.json
@@ -0,0 +1,20 @@
+{
+    "$schema": "../../../../conf.schema.json",
+    "baseDir": ".",
+    "outDir": "../../../output-actual/config-tailored/substring_behavior",
+    "includeFiles": ["."],
+    "excludeFiles": [],
+    "experimentalFootnotes": true,
+    "keepRawFiles": [],
+    "glossaries": [
+         { "file": "./glossary.md", "termHint": "" }
+    ],
+    "linking": "relative",
+    "ignoreCase": false,
+    "dev": {
+        "termsFile": "../../../output-actual/config-tailored/substring_behavior/terms.json",
+        "printInputAst": false,
+        "reproducablePaths": true,
+        "printOutputAst": false
+    }
+}
diff --git a/...fig-shared/substring_behavior/glossary.md → ...g-tailored/substring_behavior/glossary.md b/...fig-shared/substring_behavior/glossary.md → ...g-tailored/substring_behavior/glossary.md
diff --git a/test/output-expected/config-shared/glossarify-md.conf.json b/test/output-expected/config-shared/glossarify-md.conf.json
@@ -7,10 +7,9 @@
     "experimentalFootnotes": true,
     "keepRawFiles": [],
     "glossaries": [
-        { "file": "./markdown/glossary.md", "termHint": "↴" },
-        { "file": "./substring_behavior/glossary.md", "termHint": ""},
-        { "file": "./term_descriptions/glossary.md", "termHint": "ⓕ"},
-        { "file": "./aliases_and_synonyms/glossary.md", "termHint": ""}
+        { "file": "./markdown/glossary.md", "termHint": "↴" }
+        ,{ "file": "./term_descriptions/glossary.md", "termHint": "ⓕ" }
+        ,{ "file": "./aliases_and_synonyms/glossary.md", "termHint": "" }
     ],
     "linking": "relative",
     "ignoreCase": false,

diff --git a/test/output-expected/config-shared/terms.json b/test/output-expected/config-shared/terms.json
@@ -1,106 +1,4 @@
 [
-  {
-    "shortDesc": "GIVEN a term 'ÄÖ' WITH term 'Ä' being a substring of it and itself being substring of 'ÄÖÜ'",
-    "term": "'ÄÖ'",
-    "hint": "",
-    "longDesc": "GIVEN a term 'ÄÖ' WITH term 'Ä' being a substring of it and itself being substring of 'ÄÖÜ'",
-    "anchor": "#äö",
-    "glossary": {
-      "file": "./substring_behavior/glossary.md",
-      "termHint": "",
-      "basePath": "/{redacted}/input/config-shared/substring_behavior/glossary.md",
-      "outPath": "/{redacted}/output-actual/config-shared/substring_behavior/glossary.md"
-    },
-    "regex": {},
-    "aliases": [],
-    "ignoreCase": false,
-    "countOccurrenceTotal": 7
-  },
-  {
-    "shortDesc": "GIVEN a term 'ÄÖÜ' WITH terms 'ÄÖ' and 'Ä' being substrings of it",
-    "term": "'ÄÖÜ'",
-    "hint": "",
-    "longDesc": "GIVEN a term 'ÄÖÜ' WITH terms 'ÄÖ' and 'Ä' being substrings of it",
-    "anchor": "#äöü",
-    "glossary": {
-      "file": "./substring_behavior/glossary.md",
-      "termHint": "",
-      "basePath": "/{redacted}/input/config-shared/substring_behavior/glossary.md",
-      "outPath": "/{redacted}/output-actual/config-shared/substring_behavior/glossary.md"
-    },
-    "regex": {},
-    "aliases": [],
-    "ignoreCase": false,
-    "countOccurrenceTotal": 7
-  },
-  {
-    "shortDesc": "GIVEN an atomic term 'A' WITH term 'A' being a substring of 'AB' and 'ABC'",
-    "term": "A",
-    "hint": "",
-    "longDesc": "GIVEN an atomic term 'A' WITH term 'A' being a substring of 'AB' and 'ABC'",
-    "anchor": "#a",
-    "glossary": {
-      "file": "./substring_behavior/glossary.md",
-      "termHint": "",
-      "basePath": "/{redacted}/input/config-shared/substring_behavior/glossary.md",
-      "outPath": "/{redacted}/output-actual/config-shared/substring_behavior/glossary.md"
-    },
-    "regex": {},
-    "aliases": [],
-    "ignoreCase": false,
-    "countOccurrenceTotal": 9
-  },
-  {
-    "shortDesc": "GIVEN an atomic non-ASCII term 'Ä' WITH term 'Ä' being a substring of 'ÄÖ' and 'ÄÖÜ'",
-    "term": "Ä",
-    "hint": "",
-    "longDesc": "GIVEN an atomic non-ASCII term 'Ä' WITH term 'Ä' being a substring of 'ÄÖ' and 'ÄÖÜ'",
-    "anchor": "#ä",
-    "glossary": {
-      "file": "./substring_behavior/glossary.md",
-      "termHint": "",
-      "basePath": "/{redacted}/input/config-shared/substring_behavior/glossary.md",
-      "outPath": "/{redacted}/output-actual/config-shared/substring_behavior/glossary.md"
-    },
-    "regex": {},
-    "aliases": [],
-    "ignoreCase": false,
-    "countOccurrenceTotal": 9
-  },
-  {
-    "shortDesc": "GIVEN a term 'AB' WITH term A being a substring of it and itself being substring of 'ABC'",
-    "term": "AB",
-    "hint": "",
-    "longDesc": "GIVEN a term 'AB' WITH term A being a substring of it and itself being substring of 'ABC'",
-    "anchor": "#ab",
-    "glossary": {
-      "file": "./substring_behavior/glossary.md",
-      "termHint": "",
-      "basePath": "/{redacted}/input/config-shared/substring_behavior/glossary.md",
-      "outPath": "/{redacted}/output-actual/config-shared/substring_behavior/glossary.md"
-    },
-    "regex": {},
-    "aliases": [],
-    "ignoreCase": false,
-    "countOccurrenceTotal": 7
-  },
-  {
-    "shortDesc": "GIVEN a term 'ABC' WITH terms 'AB' and A being substrings of it",
-    "term": "ABC",
-    "hint": "",
-    "longDesc": "GIVEN a term 'ABC' WITH terms 'AB' and A being substrings of it",
-    "anchor": "#abc",
-    "glossary": {
-      "file": "./substring_behavior/glossary.md",
-      "termHint": "",
-      "basePath": "/{redacted}/input/config-shared/substring_behavior/glossary.md",
-      "outPath": "/{redacted}/output-actual/config-shared/substring_behavior/glossary.md"
-    },
-    "regex": {},
-    "aliases": [],
-    "ignoreCase": false,
-    "countOccurrenceTotal": 7
-  },
   {
     "shortDesc": "",
     "term": "Alias substring behavior",