fix: parsing code blocks & custom separators (st3v3nmw#1081)

* fix: parsing code blocks & custom separators This PR: 1. Adds test cases for parsing {{curly}} clozes 2. Fixes st3v3nmw#1072 - Ignore 'cards' in ```block`` & `inline` codeblocks - Adds test cases for both 3. Fixes st3v3nmw#1077 - When all the cloze options aren't being used, `cloze_rules` in `parser.ts` becomes empty which leads to the empty grammar rule `cloze_text` which prevents the grammar from being generated. This change adds a tombstone \0 to address this. - Adds test cases for this * Fixed inline code in clozes * - Fixed issue raised at st3v3nmw#1072 (comment) where inline code was not properly hanlded by clozes - Fixed a new issue related to the case when the user provides empty markers for the inline and multiline cards. With the new version, the rule is disabled when the marker is empty. * add parser test case * add more parser test cases --------- Co-authored-by: Andrea Alberti <a.alberti82@gmail.com>
open-spaced-repetition · Sep 25, 2024 · 5772c1e · 5772c1e
1 parent c45d09d
commit 5772c1e
Show file tree

Hide file tree

Showing 7 changed files with 373 additions and 111 deletions.
diff --git a/CODE_OF_CONDUCT.md b/CODE_OF_CONDUCT.md
@@ -60,7 +60,7 @@ representative at an online or offline event.
 
 Instances of abusive, harassing, or otherwise unacceptable behavior may be
 reported to the community leaders responsible for enforcement at
-st3v3n.mw@gmail.com.
+mail@stephenmwangi.com.
 All complaints will be reviewed and investigated promptly and fairly.
 
 All community leaders are obligated to respect the privacy and security of the

diff --git a/src/algorithms/base/srs-algorithm.ts b/src/algorithms/base/srs-algorithm.ts
@@ -5,7 +5,7 @@ export class SrsAlgorithm {
 
     public static getInstance(): ISrsAlgorithm {
         if (!SrsAlgorithm.instance) {
-            throw Error("there is no SrsAlgorithm instance.");
+            throw new Error("there is no SrsAlgorithm instance.");
         }
         return SrsAlgorithm.instance;
     }

diff --git a/src/constants.ts b/src/constants.ts
@@ -3,7 +3,6 @@
 export const SCHEDULING_INFO_REGEX =
     /^---\r?\n((?:.*\r?\n)*)sr-due: (.+)\r?\nsr-interval: (\d+)\r?\nsr-ease: (\d+)\r?\n((?:.*\r?\n)?)---/;
 export const YAML_FRONT_MATTER_REGEX = /^---\r?\n((?:.*\r?\n)*?)---/;
-export const NON_LETTER_SYMBOLS_REGEX = /[!-/:-@[-`{-~}\s]/g;
 
 export const MULTI_SCHEDULING_EXTRACTOR = /!([\d-]+),(\d+),(\d+)/gm;
 export const LEGACY_SCHEDULING_EXTRACTOR = /<!--SR:([\d-]+),(\d+),(\d+)-->/gm;

diff --git a/src/data-store-algorithm/data-store-algorithm.ts b/src/data-store-algorithm/data-store-algorithm.ts
@@ -5,7 +5,7 @@ export class DataStoreAlgorithm {
 
     public static getInstance(): IDataStoreAlgorithm {
         if (!DataStoreAlgorithm.instance) {
-            throw Error("there is no DataStoreAlgorithm instance.");
+            throw new Error("there is no DataStoreAlgorithm instance.");
         }
         return DataStoreAlgorithm.instance;
     }

diff --git a/src/data-stores/base/data-store.ts b/src/data-stores/base/data-store.ts
@@ -17,7 +17,7 @@ export class DataStore {
 
     public static getInstance(): IDataStore {
         if (!DataStore.instance) {
-            throw Error("there is no DataStore instance.");
+            throw new Error("there is no DataStore instance.");
         }
         return DataStore.instance;
     }

diff --git a/src/parser.ts b/src/parser.ts
@@ -33,6 +33,20 @@ function areParserOptionsEqual(options1: ParserOptions, options2: ParserOptions)
 export function generateParser(options: ParserOptions): Parser {
     let grammar: string | null = null;
 
+    // Debug the grammar before generating the parser `generate(grammar)` from the grammar.
+    if (debugParser) {
+        if (grammar === null) {
+            grammar = generateGrammar(options);
+        }
+        console.log(
+            "The parsers grammar is provided below. You can test it with https://peggyjs.org/online.html.",
+        );
+        console.log({
+            info: "Copy the grammar by right-clicking on the property grammar and copying it as a string. Then, paste it in https://peggyjs.org/online.html.",
+            grammar: grammar,
+        });
+    }
+
     // If the parser did not already exist or if the parser options changed since last the last
     // parser was generated, we generate a new parser. Otherwise, we skip the block to save
     // some execution time.
@@ -53,56 +67,111 @@ export function generateParser(options: ParserOptions): Parser {
         }
     } else {
         if (debugParser) {
-            console.log("Parser already existed. No need to generate a new parser.");
-        }
-    }
-
-    if (debugParser) {
-        if (grammar === null) {
-            grammar = generateGrammar(options);
+            console.log("Parser already exists. No need to generate a new parser.");
         }
-        console.log(
-            "The parsers grammar is provided below. You can test it with https://peggyjs.org/online.html.",
-        );
-        console.log({
-            info: "Copy the grammar by right-clicking on the property grammar and copying it as a string. Then, paste it in https://peggyjs.org/online.html.",
-            grammar: grammar,
-        });
     }
 
     return parser;
 }
 
 function generateGrammar(options: ParserOptions): string {
-    const close_rules_list: string[] = [];
+    // Contains the grammar for cloze cards
+    let clozes_grammar = "";
 
-    if (options.convertHighlightsToClozes) close_rules_list.push("close_equal");
-    if (options.convertBoldTextToClozes) close_rules_list.push("close_star");
-    if (options.convertCurlyBracketsToClozes) close_rules_list.push("close_bracket");
+    // An array contianing the types of cards enabled by the user
+    const card_rules_list: string[] = ["html_comment", "tilde_code", "backprime_code"];
 
-    const close_rules = close_rules_list.join(" / ");
+    // Include reversed inline flashcards rule only if the user provided a non-empty marker for reversed inline flashcards
+    if (options.singleLineCardSeparator.trim() !== "") card_rules_list.push("inline_rev_card");
 
-    return `{
-// The fallback case is important if we want to test the rules with https://peggyjs.org/online.html
-const CardTypeFallBack = {
-  SingleLineBasic: 0,
-  SingleLineReversed: 1,
-  MultiLineBasic: 2,
-  MultiLineReversed: 3,
-  Cloze: 4,
-};
-
-// The fallback case is important if we want to test the rules with https://peggyjs.org/online.html
-const createParsedQuestionInfoFallBack = (cardType, text, firstLineNum, lastLineNum) => {
-  return {cardType, text, firstLineNum, lastLineNum};
-};
-
-const CardType = options.CardType ? options.CardType : CardTypeFallBack;
-const createParsedQuestionInfo = options.createParsedQuestionInfo ? options.createParsedQuestionInfo : createParsedQuestionInfoFallBack;
-
-function filterBlocks(b) {
-  return b.filter( (d) => d !== null )
+    // Include inline flashcards rule only if the user provided a non-empty marker for inline flashcards
+    if (options.singleLineCardSeparator.trim() !== "") card_rules_list.push("inline_card");
+
+    // Include reversed multiline flashcards rule only if the user provided a non-empty marker for reversed multiline flashcards
+    if (options.multilineReversedCardSeparator.trim() !== "")
+        card_rules_list.push("multiline_rev_card");
+
+    // Include multiline flashcards rule only if the user provided a non-empty marker for multiline flashcards
+    if (options.multilineCardSeparator.trim() !== "") card_rules_list.push("multiline_card");
+
+    const cloze_rules_list: string[] = [];
+    if (options.convertHighlightsToClozes) cloze_rules_list.push("cloze_equal");
+    if (options.convertBoldTextToClozes) cloze_rules_list.push("cloze_star");
+    if (options.convertCurlyBracketsToClozes) cloze_rules_list.push("cloze_bracket");
+
+    // Include cloze cards only if the user enabled at least one type of cloze cards
+    if (cloze_rules_list.length > 0) {
+        card_rules_list.push("cloze_card");
+        const cloze_rules = cloze_rules_list.join(" / ");
+        clozes_grammar = `
+cloze_card
+= $(multiline_before_cloze? cloze_line (multiline_after_cloze)? (newline annotation)?) {
+  return createParsedQuestionInfo(CardType.Cloze,text().trimEnd(),location().start.line-1,location().end.line-1);
 }
+
+cloze_line
+= ((!cloze_text (inline_code / non_newline))* cloze_text) text_line_nonterminated?
+
+multiline_before_cloze
+= (!cloze_line nonempty_text_line)+
+
+multiline_after_cloze
+= e:(!(newline separator_line) text_line1)+
+
+cloze_text
+= ${cloze_rules}
+
+cloze_equal
+= cloze_mark_equal (!cloze_mark_equal non_newline)+  cloze_mark_equal
+
+cloze_mark_equal
+= "=="
+
+cloze_star
+= cloze_mark_star (!cloze_mark_star non_newline)+  cloze_mark_star
+
+cloze_mark_star
+= "**"
+
+cloze_bracket
+= cloze_mark_bracket_open (!cloze_mark_bracket_close non_newline)+  cloze_mark_bracket_close
+
+cloze_mark_bracket_open
+= "{{"
+
+cloze_mark_bracket_close
+= "}}"
+`;
+    }
+
+    // Important: we need to include `loose_line` rule to detect any other loose line.
+    // Otherwise, we get a syntax error because the parser is likely not able to reach the end
+    // of the file, as it may encounter loose lines, which it would not know how to handle.
+    card_rules_list.push("loose_line");
+
+    const card_rules = card_rules_list.join(" / ");
+
+    return `{
+    // The fallback case is important if we want to test the rules with https://peggyjs.org/online.html
+    const CardTypeFallBack = {
+      SingleLineBasic: 0,
+      SingleLineReversed: 1,
+      MultiLineBasic: 2,
+      MultiLineReversed: 3,
+      Cloze: 4,
+    };
+
+    // The fallback case is important if we want to test the rules with https://peggyjs.org/online.html
+    const createParsedQuestionInfoFallBack = (cardType, text, firstLineNum, lastLineNum) => {
+      return {cardType, text, firstLineNum, lastLineNum};
+    };
+
+    const CardType = options.CardType ? options.CardType : CardTypeFallBack;
+    const createParsedQuestionInfo = options.createParsedQuestionInfo ? options.createParsedQuestionInfo : createParsedQuestionInfoFallBack;
+
+    function filterBlocks(b) {
+      return b.filter( (d) => d !== null )
+    }
 }
 
 main
@@ -111,7 +180,7 @@ main
 /* The input text to the parser contains arbitrary text, not just card definitions.
 Hence we fallback to matching on loose_line. The result from loose_line is filtered out by filterBlocks() */
 block
-= html_comment / tilde_code / backprime_code / inline_rev_card / inline_card / multiline_rev_card / multiline_card / close_card / loose_line
+= ${card_rules}
 
 html_comment
 = $("<!--" (!"-->" (html_comment / .))* "-->" newline?) {
@@ -139,7 +208,7 @@ inline_rev_card
 inline_rev
 = left:(!inline_rev_mark (inline_code / non_newline))+ inline_rev_mark right:text_till_newline (newline annotation)? {
     return createParsedQuestionInfo(CardType.SingleLineReversed,text(),location().start.line-1,location().end.line-1);
-  }
+}
 
 multiline_card
 = c:multiline separator_line {
@@ -196,43 +265,7 @@ multiline_rev_before
 multiline_rev_after
 = $(!separator_line text_line)+
 
-close_card
-= $(multiline_before_close? close_line (multiline_after_close)? (newline annotation)?) {
-  return createParsedQuestionInfo(CardType.Cloze,text().trimEnd(),location().start.line-1,location().end.line-1);
-}
-
-close_line
-= ((!close_text non_newline)* close_text) text_line_nonterminated?
-
-multiline_before_close
-= (!close_line nonempty_text_line)+
-
-multiline_after_close
-= e:(!(newline separator_line) text_line1)+
-
-close_text
-= ${close_rules}
-
-close_equal
-= close_mark_equal (!close_mark_equal non_newline)+  close_mark_equal
-
-close_mark_equal
-= "=="
-
-close_star
-= close_mark_star (!close_mark_star non_newline)+  close_mark_star
-
-close_mark_star
-= "**"
-
-close_bracket
-= close_mark_bracket_open (!close_mark_bracket_close non_newline)+  close_mark_bracket_close
-
-close_mark_bracket_open
-= "{{"
-
-close_mark_bracket_close
-= "}}"
+${clozes_grammar}
 
 inline_mark
 = "${options.singleLineCardSeparator}"
@@ -343,7 +376,7 @@ export function parseEx(text: string, options: ParserOptions): ParsedQuestionInf
 
     let cards: ParsedQuestionInfo[] = [];
     try {
-        if (!options) throw Error("No parser options provided.");
+        if (!options) throw new Error("No parser options provided.");
 
         const parser: Parser = generateParser(options);
 
@@ -370,8 +403,7 @@ export function parseEx(text: string, options: ParserOptions): ParsedQuestionInf
     }
 
     if (debugParser) {
-        console.log("Parsed cards:");
-        console.log(cards);
+        console.log("Parsed cards:\n", cards);
     }
 
     return cards;