@paroicms/site-generator-plugin 0.5.1 → 0.6.0

package/gen-backend/dist/generator/fake-content-generator.ts/create-database-with-fake-content.js

@@ -33,6 +33,16 @@ function fillRoutingDocumentAndAddChildren(ctx, tasks, siteOptions, nodeOptions)
     const { routingIds, nodeType } = nodeOptions;
     const { siteSchema } = siteOptions;
     tasks.add(() => updateRoutingDocument(ctx, siteOptions, nodeOptions));
+    for (const listType of nodeType.lists ?? []) {
+        for (const typeName of listType.parts) {
+            const partType = getPartTypeByName(siteSchema, typeName);
+            tasks.add(() => addParts(ctx, siteOptions, {
+                parentNodeId: routingIds.nodeId,
+                nodeType: partType,
+                documentType: nodeType,
+            }));
+        }
+    }
     for (const typeName of nodeType.routingChildren ?? []) {
         const childType = getRoutingDocumentTypeByName(siteSchema, typeName);
         const childIds = routingIds.children?.[typeName];
@@ -48,17 +58,9 @@ function fillRoutingDocumentAndAddChildren(ctx, tasks, siteOptions, nodeOptions)
         tasks.add(() => addRegularDocuments(ctx, siteOptions, {
             parentNodeId: routingIds.nodeId,
             nodeType: childType,
+            documentType: childType,
         }));
     }
-    for (const listType of nodeType.lists ?? []) {
-        for (const typeName of listType.parts) {
-            const childType = getPartTypeByName(siteSchema, typeName);
-            tasks.add(() => addParts(ctx, siteOptions, {
-                parentNodeId: routingIds.nodeId,
-                nodeType: childType,
-            }));
-        }
-    }
 }
 async function updateSiteFields(ctx, options) {
     const { service, logger } = ctx;
@@ -127,6 +129,7 @@ async function updateRoutingDocument(ctx, siteOptions, nodeOptions) {
     const { fqdn, siteSchema, schemaI18n } = siteOptions;
     const content = await generateFieldSetContent(ctx, {
         nodeType,
+        documentType: nodeType,
         siteSchema,
         schemaI18n,
         withTitle: false,
@@ -139,7 +142,7 @@ async function updateRoutingDocument(ctx, siteOptions, nodeOptions) {
 }
 async function addRegularDocuments(ctx, siteOptions, nodeOptions) {
     ctx.logger.debug(`[TASK] Adding regular documents "${nodeOptions.nodeType.typeName}"…`);
-    const { parentNodeId, nodeType } = nodeOptions;
+    const { parentNodeId, nodeType, documentType } = nodeOptions;
     const { fqdn, siteSchema, schemaI18n } = siteOptions;
     const tolerateErrors = {
         errorMessages: [],
@@ -147,6 +150,7 @@ async function addRegularDocuments(ctx, siteOptions, nodeOptions) {
     const list = await generateMultipleFieldSetContents(ctx, {
         siteSchema,
         nodeType,
+        documentType,
         schemaI18n,
         count: getDefaultNodeContentCount(nodeType),
         withTitle: true,
@@ -163,7 +167,7 @@ async function addRegularDocuments(ctx, siteOptions, nodeOptions) {
 }
 async function addParts(ctx, siteOptions, nodeOptions) {
     ctx.logger.debug(`[TASK] Adding parts "${nodeOptions.nodeType.typeName}"…`);
-    const { parentNodeId, nodeType } = nodeOptions;
+    const { parentNodeId, nodeType, documentType } = nodeOptions;
     const { fqdn, siteSchema, schemaI18n } = siteOptions;
     const tolerateErrors = {
         errorMessages: [],
@@ -171,6 +175,7 @@ async function addParts(ctx, siteOptions, nodeOptions) {
     const list = await generateMultipleFieldSetContents(ctx, {
         siteSchema,
         nodeType,
+        documentType,
         schemaI18n,
         count: getDefaultNodeContentCount(nodeType),
         withTitle: true,

package/gen-backend/dist/generator/fake-content-generator.ts/generate-fake-content.js

@@ -11,10 +11,12 @@ export async function generateFieldSetContent(ctx, options) {
     return list[0];
 }
 export async function generateMultipleFieldSetContents(ctx, options) {
-    const { siteSchema, nodeType, schemaI18n, count, withTitle, tolerateErrors, debugName } = options;
+    const { siteSchema, nodeType, documentType, schemaI18n, count, withTitle, tolerateErrors, debugName, } = options;
     if (nodeType.kind === "site")
         throw new Error("Cannot generate content for site node type");
-    const outputTags = withTitle
+    // for a document, the LLM is best at generating the title, so we ask for it and remove it later
+    const skipTitle = nodeType.kind === "document" && !withTitle;
+    const outputTags = withTitle || skipTitle
         ? [{ tagName: "title", key: "title", format: "text", tagDescription: "Write the title here" }]
         : [];
     if (nodeType.fields) {
@@ -29,20 +31,32 @@ export async function generateMultipleFieldSetContents(ctx, options) {
         defaultValue: "",
         defaultLanguage,
     });
+    const documentDescription = translateText(schemaI18n, `nodeTypes.${documentType.typeName}.description`, {
+        defaultValue: "",
+        defaultLanguage,
+    });
     const siteTheme = translateText(schemaI18n, "siteTheme", {
         defaultValue: "",
         defaultLanguage,
     });
     const language = defaultLanguage ?? "en";
-    const generatedContents = outputTags.length > 0
+    let generatedContents = outputTags.length > 0
         ? await invokeGenerateFakeContent(ctx, {
             count,
+            typeKind: nodeType.kind,
             typeLabel,
             typeDescription,
+            documentDescription,
             siteTheme,
             language,
         }, outputTags, { tolerateErrors, debugName })
         : undefined;
+    if (skipTitle && generatedContents) {
+        generatedContents = generatedContents.map((content) => {
+            const { title, ...rest } = content;
+            return rest;
+        });
+    }
     return createNodeContents({
         nodeType,
         count,
@@ -81,7 +95,7 @@ function toFakeContentOutputTag(fieldNameOrType) {
             tagName: `${camelToKebabCase(key)}-md`,
             key,
             format: "markdown",
-            tagDescription: "Write document content here (150-200 words) in markdown format",
+            tagDescription: "Write document content here (150-200 words) in markdown format. It MUST NOT include the title.",
         };
     }
     if (key === "introduction") {

package/gen-backend/dist/generator/fake-content-generator.ts/invoke-generate-fake-content.js

@@ -4,9 +4,10 @@ import { readPromptFile } from "../lib/create-prompt.js";
 import { debugBatchLlmOutputs } from "../lib/debug-utils.js";
 import { parseLlmResponseAsList } from "../lib/parse-llm-response.js";
 const singlePromptTpl = PromptTemplate.fromTemplate(await readPromptFile("generate-fake-content-single.md"));
-const multiplePromptTpl = PromptTemplate.fromTemplate(await readPromptFile("generate-fake-content-multiple.md"));
+const multipleDocumentsPromptTpl = PromptTemplate.fromTemplate(await readPromptFile("generate-fake-content-multiple-documents.md"));
+const multiplePartsPromptTpl = PromptTemplate.fromTemplate(await readPromptFile("generate-fake-content-multiple-parts.md"));
 export async function invokeGenerateFakeContent(ctx, input, outputTags, options) {
-    const { language } = input;
+    const { language, typeKind } = input;
     const single = input.count === 1;
     const debugName = `fake-content-${options.debugName}${single ? "" : `-${input.count}`}`;
     const tagAndDescriptions = outputTags
@@ -26,6 +27,9 @@ export async function invokeGenerateFakeContent(ctx, input, outputTags, options)
             siteTheme: input.siteTheme,
             language: languageLabelIn(language, "en"),
         };
+        if (typeKind === "part") {
+            llmInput.documentDescription = input.documentDescription;
+        }
         if (!single) {
             llmInput.count = `${end - startIndex}`;
         }
@@ -35,7 +39,11 @@ export async function invokeGenerateFakeContent(ctx, input, outputTags, options)
     const debug = await debugBatchLlmOutputs(ctx, debugName, ctx.cheapModelName, batchInputs);
     let contents = debug.storedContents;
     if (!contents) {
-        const llmMessages = await (single ? singlePromptTpl : multiplePromptTpl)
+        const llmMessages = await (single
+            ? singlePromptTpl
+            : typeKind === "document"
+                ? multipleDocumentsPromptTpl
+                : multiplePartsPromptTpl)
             .pipe(ctx.cheapModel)
             .batch(batchInputs);
         contents = await debug.getMessageContents(llmMessages);

package/gen-backend/dist/generator/llm-queries/invoke-update-site-schema.js

@@ -97,10 +97,20 @@ async function invokeUpdateSiteSchemaStep2(ctx, input) {
         ...input.generatedSchema,
     };
     if (parsed.siteSchema) {
-        result.siteSchema = parsed.siteSchema;
+        result.siteSchema = fixSiteSchema(parsed.siteSchema);
     }
     if (parsed.l10n) {
         result.l10n = parsed.l10n;
     }
     return result;
 }
+function fixSiteSchema(siteSchema) {
+    for (const nodeType of siteSchema.nodeTypes ?? []) {
+        for (const field of nodeType.fields ?? []) {
+            if (typeof field !== "string" && field.dataType === "quillDelta") {
+                field.plugin = "@paroicms/quill-editor-plugin";
+            }
+        }
+    }
+    return siteSchema;
+}

package/gen-backend/prompts/{generate-fake-content-multiple.md → generate-fake-content-multiple-documents.md}

@@ -1,4 +1,4 @@
-Generate the text content of **{count}** different web pages.
+Generate the text content of **{count}** different **web pages**.
 
 All the produced texts must be in **{language}**.
 
@@ -14,9 +14,9 @@ Make sure web pages are distinct from each other by:
 
 Guidelines:
 
+- NEVER repeat the same title among several sub-contents.
 - Never invite the user to write a comment. There is no comment system.
-- Do not repeat the title in markdown content.
 
-For each web page, use the following exact format:
+Each web page is composed by several sub-contents. Use the following exact format:
 
 {tagAndDescriptions}

package/gen-backend/prompts/generate-fake-content-multiple-parts.md

@@ -0,0 +1,22 @@
+Generate the text content of **{count}** different **sections** of a web page.
+
+All the produced texts must be in **{language}**.
+
+The sections in question are of type: **{typeLabel}** ({typeDescription}).
+
+For the context, the website's theme is: "{siteTheme}", and the web page is about "{documentDescription}".
+
+Make sure sections are distinct from each other by:
+
+- Cover different aspects in line with the type of sections;
+- Using varied writing styles;
+- Having different lengths within the specified range.
+
+Guidelines:
+
+- NEVER repeat a title among several sub-contents.
+- Never invite the user to write a comment. There is no comment system.
+
+Each section is composed by several sub-contents. Use the following exact format:
+
+{tagAndDescriptions}

package/gen-backend/prompts/new-site-1-analysis.md

@@ -1,4 +1,4 @@
-You will be given a description of a website's structure. Your task is to generate three things:
+You will be given a description of a website's structure. Your task is to **analyze** the website description, then you will generate three things:
 
 1. A YAML for website properties.
 2. A Markdown bulleted list that represents the hierarchy of node types (routing documents, regular documents and parts) within the site.
@@ -55,10 +55,10 @@ Guidelines for creating the hierarchical bullet list:
   - Write `"list of"`, and then the **type name** within backquotes.
   - Append `(parts)`.
   - Add a coma, then write: `"list name:"` and append an identifier for the list name.
-- Each distinct node type must have a unique name through the whole tree structure.
-- Reusing the same node type several times (for recursivity etc.):
-  - When the same node type is reused as a child in several places, then write its children only on the first occurence.
+- Avoid recursivity. But if you can't avoid reusing a node type, then write its children only on the first occurence.
 - On this task, do not focus on fields. If the website description provide fields then you will report them in the unused information. In particular, a collection of medias (a carousel, a slider, an image gallery, a slideshow, etc.) is a type of field.
+- Don't try to provide a comment mechanism when the user requests a blog. Blog comments do not belong to the hierarchy of node types.
+- NEVER ever reuse the same type name for 2 distinct node type. Pay particular attention to words spelled identically in the singular and plural. Each distinct node type must have a unique name through the whole tree structure.
 
 Here's an example of a correct output:
 
@@ -82,27 +82,18 @@ Here is another incorrect example:
 
 <incorrect_example>
 * `home` (routing document)
-  * `animals`
-    * `pages` (regular document)
-      * list of `page` (regular documents)
-  * `vegetals`
-    * `pages` (regular document)
-      * list of `page` (regular documents)
+  * `species` (routing document)
+    * list of `species` (regular documents)
 </incorrect_example>
 
-This incorrect example fails because the children of `pages` node should not be written the second time we use this node type. Here is how to fix that:
+This incorrect example fails because the type name `species` is used for 2 different types. You can distinguish them by suffixing both the collection and the item types:
 
 <correct_example>
 * `home` (routing document)
-  * `animals`
-    * `pages` (regular document)
-      * list of `page` (regular documents)
-  * `vegetals`
-    * `pages` (regular document)
+  * `speciesColl` (routing document)
+    * list of `speciesItem` (regular documents)
 </correct_example>
 
-Notice: the `pages` node will have the same child types for both usage. But we describe them only the first time it appears on the tree.
-
 Here's an example of correct output using parts, and with the default contact and search pages:
 
 <correct_example>
@@ -125,6 +116,7 @@ Guidelines for creating the dictionnary YAML:
 
 - Use a YAML syntax: pay attention to the indentation.
 - Keys: each key is the name of a node type or the name of a list of parts.
+- If the same type is reused several times, then write only one entry in this dictionnary.
 - Values contains the properties. For a node type, provide the following properties:
   - confidence: Your confidence level for the accuracy of this node type (0.0-1.0).
   - kind: Can be `routingDocument`, `regularDocument`, `part`.

package/gen-backend/prompts/update-site-schema-1-write-details.md

@@ -1,4 +1,4 @@
-You are tasked with modifying a JSON object of a _site schema_, based on a given TypeScript type definition and an update message, and its localized labels. Follow these steps carefully:
+You are tasked to plan how modifying a JSON object of a _site schema_, based on a given TypeScript type definition and an update message, and its localized labels. Follow these steps carefully:
 
 # 1. Review the TypeScript type definition of the JSON structure: