add html placeholder

Signed-off-by: Michele Dolfi <[email protected]>

Author: dolfim-ibm

docling_core/types/doc/document.py

@@ -1393,6 +1393,17 @@ class DoclingDocument(BaseModel):
     math annotation {
     display: none;
     }
+    .formula-not-decoded {
+    background: repeating-linear-gradient(
+    45deg, /* Angle of the stripes */
+    LightGray, /* First color */
+    LightGray 10px, /* Length of the first color */
+    White 10px, /* Second color */
+    White 20px /* Length of the second color */
+    );
+    margin: 0;
+    text-align: center;
+    }
     </style>
     </head>"""
 
@@ -2503,6 +2514,10 @@ def _prepare_tag_content(
 
                 if text != "":
                     html_texts.append(text)
+                else:
+                    html_texts.append(
+                        '<div class="formula-not-decoded">Formula not decoded</div>'
+                    )
 
             elif isinstance(item, ListItem):
 

test/data/doc/2206.01062.yaml.html

@@ -56,6 +56,17 @@
     math annotation {
     display: none;
     }
+    .formula-not-decoded {
+    background: repeating-linear-gradient(
+    45deg, /* Angle of the stripes */
+    LightGray, /* First color */
+    LightGray 10px, /* Length of the first color */
+    White 10px, /* Second color */
+    White 20px /* Length of the second color */
+    );
+    margin: 0;
+    text-align: center;
+    }
     </style>
     </head>
 <h2>DocLayNet: A Large Human-Annotated Dataset for Document-Layout Analysis</h2>

test/data/doc/bad_doc.yaml.html

@@ -56,6 +56,17 @@
     math annotation {
     display: none;
     }
+    .formula-not-decoded {
+    background: repeating-linear-gradient(
+    45deg, /* Angle of the stripes */
+    LightGray, /* First color */
+    LightGray 10px, /* Length of the first color */
+    White 10px, /* Second color */
+    White 20px /* Length of the second color */
+    );
+    margin: 0;
+    text-align: center;
+    }
     </style>
     </head>
 <h1>This is the title</h1>

test/data/doc/constructed_doc.embedded.html.gt

@@ -56,6 +56,17 @@
     math annotation {
     display: none;
     }
+    .formula-not-decoded {
+    background: repeating-linear-gradient(
+    45deg, /* Angle of the stripes */
+    LightGray, /* First color */
+    LightGray 10px, /* Length of the first color */
+    White 10px, /* Second color */
+    White 20px /* Length of the second color */
+    );
+    margin: 0;
+    text-align: center;
+    }
     </style>
     </head>
 <h1>Title of the Document</h1>

test/data/doc/constructed_doc.placeholder.html.gt

@@ -56,6 +56,17 @@
     math annotation {
     display: none;
     }
+    .formula-not-decoded {
+    background: repeating-linear-gradient(
+    45deg, /* Angle of the stripes */
+    LightGray, /* First color */
+    LightGray 10px, /* Length of the first color */
+    White 10px, /* Second color */
+    White 20px /* Length of the second color */
+    );
+    margin: 0;
+    text-align: center;
+    }
     </style>
     </head>
 <h1>Title of the Document</h1>

test/data/doc/constructed_doc.referenced.html.gt

@@ -56,6 +56,17 @@
     math annotation {
     display: none;
     }
+    .formula-not-decoded {
+    background: repeating-linear-gradient(
+    45deg, /* Angle of the stripes */
+    LightGray, /* First color */
+    LightGray 10px, /* Length of the first color */
+    White 10px, /* Second color */
+    White 20px /* Length of the second color */
+    );
+    margin: 0;
+    text-align: center;
+    }
     </style>
     </head>
 <h1>Title of the Document</h1>

test/data/doc/constructed_document.yaml.html

@@ -56,6 +56,17 @@
     math annotation {
     display: none;
     }
+    .formula-not-decoded {
+    background: repeating-linear-gradient(
+    45deg, /* Angle of the stripes */
+    LightGray, /* First color */
+    LightGray 10px, /* Length of the first color */
+    White 10px, /* Second color */
+    White 20px /* Length of the second color */
+    );
+    margin: 0;
+    text-align: center;
+    }
     </style>
     </head>
 <h1>Title of the Document</h1>

test/data/doc/dummy_doc.yaml.html

@@ -56,6 +56,17 @@
     math annotation {
     display: none;
     }
+    .formula-not-decoded {
+    background: repeating-linear-gradient(
+    45deg, /* Angle of the stripes */
+    LightGray, /* First color */
+    LightGray 10px, /* Length of the first color */
+    White 10px, /* Second color */
+    White 20px /* Length of the second color */
+    );
+    margin: 0;
+    text-align: center;
+    }
     </style>
     </head>
 <h1>DocLayNet: A Large Human-Annotated Dataset for Document-Layout Analysis</h1>