fix(docx): parse integrals as n-ary objects without chr element (#2712)

Signed-off-by: Cesar Berrospi Ramis <ceb@zurich.ibm.com>

docling/backend/docx/latex/latex_dict.py

@@ -65,6 +65,11 @@ CHR_BO = {
     "\u2210": "\\coprod",
     "\u2211": "\\sum",
     "\u222b": "\\int",
+    "\u222c": "\\iint",
+    "\u222d": "\\iiint",
+    "\u222e": "\\oint",
+    "\u222f": "\\oiint",
+    "\u2230": "\\oiiint",
     "\u22c0": "\\bigwedge",
     "\u22c1": "\\bigvee",
     "\u22c2": "\\bigcap",

docling/backend/docx/latex/omml.py

@@ -381,7 +381,8 @@ class oMath2Latex(Tag2Method):
         bo = ""
         for stag, t, e in self.process_children_list(elm):
             if stag == "naryPr":
-                bo = get_val(t.chr, store=CHR_BO)
+                # if <m:naryPr> contains no <m:chr>, the n-ary represents an integral
+                bo = get_val(t.chr, default="\\int", store=CHR_BO)
             else:
                 res.append(t)
         return bo + BLANK.join(res)

tests/data/groundtruth/docling_v2/equations.docx.itxt

@@ -37,4 +37,17 @@ item-0 at level 0: unspecified: group _root_
   item-36 at level 1: formula: e^{x}=1+\frac{x}{1!}+\frac{x^{2} ... xtellipsis } , - \infty  < x <  \infty
   item-37 at level 1: text: 
   item-38 at level 1: text: And that is an equation by itself. Cheers!
-  item-39 at level 1: text: 
+  item-39 at level 1: text: 
+  item-40 at level 1: text: Large operators and integrals ar ... sented with n-ary objects in OMML XML:
+  item-41 at level 1: text: 
+  item-42 at level 1: formula: \sum_{0}^{2}x
+  item-43 at level 1: formula: \bigcup_{n=1}^{m}\left(X_{n} \cap Y_{n}\right)
+  item-44 at level 1: formula: \prod_{k=1}^{n}A_{k}
+  item-45 at level 1: formula: \bigwedge_{}^{}x
+  item-46 at level 1: formula: \int_{}^{}(2x+1)dx
+  item-47 at level 1: formula: \iint_{0}^{1}xdx
+  item-48 at level 1: formula: \iiint_{}^{}ydy
+  item-49 at level 1: formula: \oint_{}^{}\frac{dy}{dx}
+  item-50 at level 1: formula: \oiint_{0}^{2 \pi }idt
+  item-51 at level 1: formula: \oiiint_{C}^{}\frac{1}{z}dz
+  item-52 at level 1: text: 

tests/data/groundtruth/docling_v2/equations.docx.json

@@ -1,10 +1,10 @@
 {
   "schema_name": "DoclingDocument",
-  "version": "1.7.0",
+  "version": "1.8.0",
   "name": "equations",
   "origin": {
     "mimetype": "application/vnd.openxmlformats-officedocument.wordprocessingml.document",
-    "binary_hash": 11121138535595486899,
+    "binary_hash": 8638432756089077257,
     "filename": "equations.docx"
   },
   "furniture": {
@@ -106,6 +106,45 @@
       },
       {
         "$ref": "#/texts/35"
+      },
+      {
+        "$ref": "#/texts/36"
+      },
+      {
+        "$ref": "#/texts/37"
+      },
+      {
+        "$ref": "#/texts/38"
+      },
+      {
+        "$ref": "#/texts/39"
+      },
+      {
+        "$ref": "#/texts/40"
+      },
+      {
+        "$ref": "#/texts/41"
+      },
+      {
+        "$ref": "#/texts/42"
+      },
+      {
+        "$ref": "#/texts/43"
+      },
+      {
+        "$ref": "#/texts/44"
+      },
+      {
+        "$ref": "#/texts/45"
+      },
+      {
+        "$ref": "#/texts/46"
+      },
+      {
+        "$ref": "#/texts/47"
+      },
+      {
+        "$ref": "#/texts/48"
       }
     ],
     "content_layer": "body",
@@ -655,6 +694,169 @@
       "prov": [],
       "orig": "",
       "text": ""
+    },
+    {
+      "self_ref": "#/texts/36",
+      "parent": {
+        "$ref": "#/body"
+      },
+      "children": [],
+      "content_layer": "body",
+      "label": "text",
+      "prov": [],
+      "orig": "Large operators and integrals are represented with n-ary objects in OMML XML:",
+      "text": "Large operators and integrals are represented with n-ary objects in OMML XML:",
+      "formatting": {
+        "bold": false,
+        "italic": false,
+        "underline": false,
+        "strikethrough": false,
+        "script": "baseline"
+      }
+    },
+    {
+      "self_ref": "#/texts/37",
+      "parent": {
+        "$ref": "#/body"
+      },
+      "children": [],
+      "content_layer": "body",
+      "label": "text",
+      "prov": [],
+      "orig": "",
+      "text": ""
+    },
+    {
+      "self_ref": "#/texts/38",
+      "parent": {
+        "$ref": "#/body"
+      },
+      "children": [],
+      "content_layer": "body",
+      "label": "formula",
+      "prov": [],
+      "orig": "\\sum_{0}^{2}x",
+      "text": "\\sum_{0}^{2}x"
+    },
+    {
+      "self_ref": "#/texts/39",
+      "parent": {
+        "$ref": "#/body"
+      },
+      "children": [],
+      "content_layer": "body",
+      "label": "formula",
+      "prov": [],
+      "orig": "\\bigcup_{n=1}^{m}\\left(X_{n} \\cap Y_{n}\\right)",
+      "text": "\\bigcup_{n=1}^{m}\\left(X_{n} \\cap Y_{n}\\right)"
+    },
+    {
+      "self_ref": "#/texts/40",
+      "parent": {
+        "$ref": "#/body"
+      },
+      "children": [],
+      "content_layer": "body",
+      "label": "formula",
+      "prov": [],
+      "orig": "\\prod_{k=1}^{n}A_{k}",
+      "text": "\\prod_{k=1}^{n}A_{k}"
+    },
+    {
+      "self_ref": "#/texts/41",
+      "parent": {
+        "$ref": "#/body"
+      },
+      "children": [],
+      "content_layer": "body",
+      "label": "formula",
+      "prov": [],
+      "orig": "\\bigwedge_{}^{}x",
+      "text": "\\bigwedge_{}^{}x"
+    },
+    {
+      "self_ref": "#/texts/42",
+      "parent": {
+        "$ref": "#/body"
+      },
+      "children": [],
+      "content_layer": "body",
+      "label": "formula",
+      "prov": [],
+      "orig": "\\int_{}^{}(2x+1)dx",
+      "text": "\\int_{}^{}(2x+1)dx"
+    },
+    {
+      "self_ref": "#/texts/43",
+      "parent": {
+        "$ref": "#/body"
+      },
+      "children": [],
+      "content_layer": "body",
+      "label": "formula",
+      "prov": [],
+      "orig": "\\iint_{0}^{1}xdx",
+      "text": "\\iint_{0}^{1}xdx"
+    },
+    {
+      "self_ref": "#/texts/44",
+      "parent": {
+        "$ref": "#/body"
+      },
+      "children": [],
+      "content_layer": "body",
+      "label": "formula",
+      "prov": [],
+      "orig": "\\iiint_{}^{}ydy",
+      "text": "\\iiint_{}^{}ydy"
+    },
+    {
+      "self_ref": "#/texts/45",
+      "parent": {
+        "$ref": "#/body"
+      },
+      "children": [],
+      "content_layer": "body",
+      "label": "formula",
+      "prov": [],
+      "orig": "\\oint_{}^{}\\frac{dy}{dx}",
+      "text": "\\oint_{}^{}\\frac{dy}{dx}"
+    },
+    {
+      "self_ref": "#/texts/46",
+      "parent": {
+        "$ref": "#/body"
+      },
+      "children": [],
+      "content_layer": "body",
+      "label": "formula",
+      "prov": [],
+      "orig": "\\oiint_{0}^{2 \\pi }idt",
+      "text": "\\oiint_{0}^{2 \\pi }idt"
+    },
+    {
+      "self_ref": "#/texts/47",
+      "parent": {
+        "$ref": "#/body"
+      },
+      "children": [],
+      "content_layer": "body",
+      "label": "formula",
+      "prov": [],
+      "orig": "\\oiiint_{C}^{}\\frac{1}{z}dz",
+      "text": "\\oiiint_{C}^{}\\frac{1}{z}dz"
+    },
+    {
+      "self_ref": "#/texts/48",
+      "parent": {
+        "$ref": "#/body"
+      },
+      "children": [],
+      "content_layer": "body",
+      "label": "text",
+      "prov": [],
+      "orig": "",
+      "text": ""
     }
   ],
   "pictures": [],

tests/data/groundtruth/docling_v2/equations.docx.md

@@ -26,4 +26,26 @@ This is a word document and this is an inline equation:  $A= \pi r^{2}$ . If ins
 
 $$e^{x}=1+\frac{x}{1!}+\frac{x^{2}}{2!}+\frac{x^{3}}{3!}+ \text{ \textellipsis } , - \infty  < x <  \infty$$
 
-And that is an equation by itself. Cheers!
+And that is an equation by itself. Cheers!
+
+Large operators and integrals are represented with n-ary objects in OMML XML:
+
+$$\sum_{0}^{2}x$$
+
+$$\bigcup_{n=1}^{m}\left(X_{n} \cap Y_{n}\right)$$
+
+$$\prod_{k=1}^{n}A_{k}$$
+
+$$\bigwedge_{}^{}x$$
+
+$$\int_{}^{}(2x+1)dx$$
+
+$$\iint_{0}^{1}xdx$$
+
+$$\iiint_{}^{}ydy$$
+
+$$\oint_{}^{}\frac{dy}{dx}$$
+
+$$\oiint_{0}^{2 \pi }idt$$
+
+$$\oiiint_{C}^{}\frac{1}{z}dz$$