Merge pull request #7189 from quarto-dev/bugfix/7187

`html-table-processing` options in document, project and attributes

Author: cscheid

news/changelog-1.4.md

@@ -37,6 +37,7 @@
 - ([#6714](https://github.com/quarto-dev/quarto-cli/issues/6714)): Display title block for HTML when other (non-title/author/subtitle) metadata is present.
 - ([#6833](https://github.com/quarto-dev/quarto-cli/issues/6833)): Handle partially-specified aspect ratio, width, and height attributes in `video` shortcode.
 - ([#7137](https://github.com/quarto-dev/quarto-cli/discussions/7137)): Automatically set `rel="noopener"` when setting a target on external links
+- ([#7187](https://github.com/quarto-dev/quarto-cli/issues/7187)): Add `html-table-processing: none` to document- and project-level metadata to disable HTML table processing. Add `{html-table-processing="none"}` to a fenced div to disable HTML table processing for the elements in that div.
 
 ## Appendix
 

src/command/render/filters.ts

@@ -23,6 +23,7 @@ import {
   kFormatIdentifier,
   kHeaderIncludes,
   kHtmlMathMethod,
+  kHtmlTableProcessing,
   kIncludeAfter,
   kIncludeAfterBody,
   kIncludeBefore,
@@ -553,6 +554,10 @@ async function quartoFilterParams(
   if (foldCode) {
     params[kCodeFold] = foldCode;
   }
+  const htmlTableProcessing = format.render[kHtmlTableProcessing];
+  if (htmlTableProcessing) {
+    params[kHtmlTableProcessing] = htmlTableProcessing;
+  }
   const tblColwidths = format.render[kTblColwidths];
   if (tblColwidths !== undefined) {
     params[kTblColwidths] = tblColwidths;

src/config/constants.ts

@@ -130,6 +130,8 @@ export const kQuartoRequired = "quarto-required";
 export const kPreviewMode = "preview-mode";
 export const kPreviewModeRaw = "raw";
 
+export const kHtmlTableProcessing = "html-table-processing";
+
 export const kIdentifierDefaultsKeys = [
   kTargetFormat,
   kDisplayName,
@@ -214,6 +216,7 @@ export const kRenderDefaultsKeys = [
   kNotebookViewStyle,
   kNotebookPreserveCells,
   kClearCellOptions,
+  kHtmlTableProcessing,
 ];
 
 // language fields

src/config/types.ts

@@ -87,6 +87,7 @@ import {
   kGladtex,
   kHighlightStyle,
   kHtmlMathMethod,
+  kHtmlTableProcessing,
   kInclude,
   kIncludeAfterBody,
   kIncludeBeforeBody,
@@ -471,6 +472,7 @@ export interface FormatRender {
   [kNotebookPreserveCells]?: boolean;
   [kClearCellOptions]?: boolean;
   [kIpynbProduceSourceNotebook]?: boolean;
+  [kHtmlTableProcessing]?: "none";
 }
 
 export interface FormatExecute {

src/resources/editor/tools/vs-code.mjs

@@ -16200,6 +16200,15 @@ var require_yaml_intelligence_resources = __commonJS({
             short: "Specify the default dpi (dots per inch) value for conversion from pixels to inch/\ncentimeters and vice versa.\n",
             long: "Specify the default dpi (dots per inch) value for conversion from pixels to inch/\ncentimeters and vice versa. (Technically, the correct term would be ppi: pixels per\ninch.) The default is `96`. When images contain information about dpi internally, the\nencoded value is used instead of the default specified by this option.\n"
           }
+        },
+        {
+          name: "html-table-processing",
+          schema: {
+            enum: [
+              "none"
+            ]
+          },
+          description: "If `none`, do not process tables in HTML input."
         }
       ],
       "schema/document-reveal-content.yml": [
@@ -21019,7 +21028,7 @@ var require_yaml_intelligence_resources = __commonJS({
         "Keep the markdown file generated by executing code",
         "Keep the notebook file generated from executing code.",
         "Filters to pre-process ipynb files before rendering to markdown",
-        "Speicfy which nodes should be run interactively (displaying output\nfrom expressions)",
+        "Specify which nodes should be run interactively (displaying output\nfrom expressions)",
         "Keep the intermediate typst file used during render.",
         "Keep the intermediate tex file used during render.",
         {
@@ -21908,7 +21917,8 @@ var require_yaml_intelligence_resources = __commonJS({
         },
         "Disambiguating year suffix in author-date styles (e.g. \u201Ca\u201D in \u201CDoe,\n1999a\u201D).",
         "Manuscript configuration",
-        "internal-schema-hack"
+        "internal-schema-hack",
+        "If <code>none</code>, do not process tables in HTML input."
       ],
       "schema/external-schemas.yml": [
         {
@@ -22132,12 +22142,12 @@ var require_yaml_intelligence_resources = __commonJS({
         mermaid: "%%"
       },
       "handlers/mermaid/schema.yml": {
-        _internalId: 168151,
+        _internalId: 168154,
         type: "object",
         description: "be an object",
         properties: {
           "mermaid-format": {
-            _internalId: 168143,
+            _internalId: 168146,
             type: "enum",
             enum: [
               "png",
@@ -22153,7 +22163,7 @@ var require_yaml_intelligence_resources = __commonJS({
             exhaustiveCompletions: true
           },
           theme: {
-            _internalId: 168150,
+            _internalId: 168153,
             type: "anyOf",
             anyOf: [
               {

src/resources/editor/tools/yaml/web-worker.js

@@ -9172,6 +9172,15 @@
         "short": "Specify the default dpi (dots per inch) value for conversion from pixels to inch/\ncentimeters and vice versa.\n",
         "long": "Specify the default dpi (dots per inch) value for conversion from pixels to inch/\ncentimeters and vice versa. (Technically, the correct term would be ppi: pixels per\ninch.) The default is `96`. When images contain information about dpi internally, the\nencoded value is used instead of the default specified by this option.\n"
       }
+    },
+    {
+      "name": "html-table-processing",
+      "schema": {
+        "enum": [
+          "none"
+        ]
+      },
+      "description": "If `none`, do not process tables in HTML input."
     }
   ],
   "schema/document-reveal-content.yml": [
@@ -13991,7 +14000,7 @@
     "Keep the markdown file generated by executing code",
     "Keep the notebook file generated from executing code.",
     "Filters to pre-process ipynb files before rendering to markdown",
-    "Speicfy which nodes should be run interactively (displaying output\nfrom expressions)",
+    "Specify which nodes should be run interactively (displaying output\nfrom expressions)",
     "Keep the intermediate typst file used during render.",
     "Keep the intermediate tex file used during render.",
     {
@@ -14880,7 +14889,8 @@
     },
     "Disambiguating year suffix in author-date styles (e.g. “a” in “Doe,\n1999a”).",
     "Manuscript configuration",
-    "internal-schema-hack"
+    "internal-schema-hack",
+    "If <code>none</code>, do not process tables in HTML input."
   ],
   "schema/external-schemas.yml": [
     {
@@ -15104,12 +15114,12 @@
     "mermaid": "%%"
   },
   "handlers/mermaid/schema.yml": {
-    "_internalId": 168151,
+    "_internalId": 168154,
     "type": "object",
     "description": "be an object",
     "properties": {
       "mermaid-format": {
-        "_internalId": 168143,
+        "_internalId": 168146,
         "type": "enum",
         "enum": [
           "png",
@@ -15125,7 +15135,7 @@
         "exhaustiveCompletions": true
       },
       "theme": {
-        "_internalId": 168150,
+        "_internalId": 168153,
         "type": "anyOf",
         "anyOf": [
           {

src/resources/editor/tools/yaml/yaml-intelligence-resources.json

@@ -65,6 +65,7 @@ local kIncludeInHeader = "include-in-header"
 
 local kCopyright = "copyright"
 local kLicense = "license"
+local kHtmlTableProcessing = "html-table-processing"
 
 -- for a given language, the comment character(s)
 local kLangCommentChars = {
@@ -187,5 +188,6 @@ return {
   kCopyright = kCopyright,
   kLicense = kLicense,
 
-  kLangCommentChars = kLangCommentChars
+  kLangCommentChars = kLangCommentChars,
+  kHtmlTableProcessing = kHtmlTableProcessing
 }

src/resources/filters/modules/constants.lua

@@ -4,9 +4,12 @@
 function quarto_ast_pipeline()
   return {
     { name = "normalize-table-merge-raw-html", filter = table_merge_raw_html() },
-    
+
+    -- this can't be combined because it's top-down processing.
+    -- unfortunate.
+    { name = "normalize-html-table-processing", filter = parse_html_tables() },
+
     { name = "normalize-combined-1", filter = combineFilters({
-        parse_html_tables(),
         parse_extended_nodes(),
         code_filename(),
         normalize_fixup_data_uri_image_extension(),

src/resources/filters/normalize/astpipeline.lua

@@ -21,7 +21,16 @@ end
 
 function parse_html_tables()
   local filter
+  if param(constants.kHtmlTableProcessing) == "none" then
+    return {}
+  end
   filter = {
+    traverse = "topdown",
+    Div = function(div)
+      if div.attributes[constants.kHtmlTableProcessing] == "none" then
+        return div.contents, false
+      end
+    end,
     RawBlock = function(el)
       if _quarto.format.isRawHtml(el) then
         -- if we have a raw html table in a format that doesn't handle raw_html