Began exploratory work on storybooks

Started taking code from PR sillsdev#645. Discovered that storybook files
seem to be modified before the code in convertBooks runs.

Author: TheNonPirate

convert/convertBooks.ts

@@ -11,6 +11,7 @@ import { convertMarkdownsToMilestones } from './convertMarkdown';
 import { hasAudioExtension, hasImageExtension } from './stringUtils';
 import { Promisable, Task, TaskOutput } from './Task';
 import { verifyGlossaryEntries } from './verifyGlossaryEntries';
+import { convertStorybookElements } from './storybook';
 
 const base = process.env.BUILD_BASE_PATH || '';
 
@@ -291,12 +292,19 @@ function applyFilters(
     text: string,
     filterFunctions: FilterFunction[],
     bcId: string,
-    bookId: string
+    bookId: string, bookType?: string
 ): string {
     let filteredText = text;
+    if (bookType === 'story') {
+        console.log("Before filters:");
+        console.log(text);
+    }
     for (const filterFn of filterFunctions) {
         filteredText = filterFn(filteredText, bcId, bookId);
     }
+    if (bookType === 'story') {
+        filteredText = convertStorybookElements(filteredText);
+    }
     return filteredText;
 }
 
@@ -394,7 +402,6 @@ export async function convertBooks(
         for (const book of collection.books) {
             let bookConverted = false;
             switch (book.type) {
-                case 'story':
                 case 'songs':
                 case 'audio-only':
                 case 'bloom-player':
@@ -702,7 +709,7 @@ function convertScriptureBook(
     function processBookContent(resolve: () => void, err: any, content: string) {
         //process.stdout.write(`processBookContent: bookId:${book.id}, error:${err}\n`);
         if (err) throw err;
-        content = applyFilters(content, usfmFilterFunctions, context.bcId, book.id);
+        content = applyFilters(content, usfmFilterFunctions, context.bcId, book.id, book.type);
         if (context.scriptureConfig.traits['has-glossary']) {
             content = verifyGlossaryEntries(content, bcGlossary);
         }
@@ -727,10 +734,10 @@ function convertScriptureBook(
                 if (context.verbose) {
                     console.log(
                         context.docSet +
-                            ' <- ' +
-                            book.name +
-                            ': ' +
-                            path.join(context.dataDir, 'books', context.bcId, book.file)
+                        ' <- ' +
+                        book.name +
+                        ': ' +
+                        path.join(context.dataDir, 'books', context.bcId, book.file)
                     );
                 }
                 displayBookId(context.bcId, book.id);
@@ -780,8 +787,22 @@ function convertScriptureBook(
                     const filePath = path.join(context.dataDir, 'books', context.bcId, file);
                     fileContents.push(fs.readFileSync(filePath, 'utf-8'));
                 });
+                // Collect the file contents into a single document
+                let usfm: string;
+
+                if (book.type == 'story') {
+                    // The first file contains meta-content (id, title, etc)
+                    usfm = fileContents[0];
 
-                processBookContent(resolve, null, fileContents.join(''));
+                    // Subsequent files represent storybook pages.
+                    // SAB deletes the \page tags. Replace them with chapter tags.
+                    for (let i = 1; i < fileContents.length; i++) {
+                        usfm += `\\c ${i} ${fileContents[i]}`;
+                    }
+                } else {
+                    usfm = fileContents.join('');
+                }
+                processBookContent(resolve, null, usfm);
             }
         })
     );

convert/storybook.ts

@@ -0,0 +1,50 @@
+//It seems that even before the code in convertBooks, the storybook already has had its img tags removed. I need to figure out how to prevent this from happening.
+/*By the time that applyFilters (In convertBooks) is called, which is at the beginning of convertScriptureBook and thus the beginning of where I would expect
+anything to alter the file, the storybook has already had some changes made to it compared to what is displayed in the Scripture App Builder in the "Book File"
+tab of the Unmerciful Servant book. These changes include the removal of img tags. The Unmerciful Servant book has its source in a .docx file, so I am guessing
+that the changes are made in the conversion of the .docx to the .sfm files.
+*/
+
+/**
+ * Replace page tags with chapters
+ */
+export function replacePageTags(usfm: string): string {
+    return usfm.replace(/\\page\s+(\d+)/g, '\\c $1');
+}
+
+/**
+ * Replace img tags with figs
+ */
+export function replaceImageTags(usfm: string): string {
+    return usfm.replace(/\\img\s+(\S+)/g, '\\fig caption |src="$1"\\fig*');
+}//"caption" is just a placeholder for now.
+//\fig Jesus Birth|src="JesusBirth.jpg" size="span"\fig*
+//\img image1.jpeg
+
+
+
+/**
+ * Convert list tags to milestones (I need to test this to see if this is needed or if the list tags can be parsed as they currently are)
+ */
+export function transformLists(usfm: string): string {
+    return usfm
+        .replace(/\\zuli1\s+([^\\]*)/g, '\\zuli1-s\\* $1 \\zuli1-e\\* ')
+        .replace(/\\zon1 (\d+)(([^\\]|\\zoli)*)/g, '\\zon1-s |start="$1"\\* $2 \\zon1-e\\* ')
+        .replace(/\\zoli1\s+([^\\]*)/g, '\\zoli1-s\\* $1 \\zoli1-e\\* ');
+}//Maybe keep, maybe not
+
+export function transformHeadings(usfm: string): string {
+    return usfm.replace(/\\(m?s\d?)([^\\]*)/g, '\\m \\zusfm-s |class="$1"\\* $2 \\zusfm-e ');
+}//Also maybe keep, maybe not
+
+export function convertStorybookElements(usfm: string) {
+    console.log("Storybook start:");
+    console.log(usfm);
+    usfm = replacePageTags(usfm);
+    usfm = replaceImageTags(usfm);
+    //usfm = transformLists(usfm);
+    //usfm = transformHeadings(usfm);
+    console.log("Storybook end:");
+    console.log(usfm);
+    return usfm;
+}