Merge branch 'cau/input-format-abstraction' of github.com:DS4SD/docling into cau/input-format-abstraction

docling/backend/mspowerpoint_backend.py

@@ -94,7 +94,7 @@ class MsPowerpointDocumentBackend(DeclarativeDocumentBackend, PaginatedDocumentB
         )
         doc = DoclingDocument(
             description=DescriptionItem(), name="name_without_extension", origin=origin
-        )  # TODO must add origin information
+        )  # must add origin information
         doc = self.walk_linear(self.pptx_obj, doc)
 
         return doc
@@ -183,7 +183,7 @@ class MsPowerpointDocumentBackend(DeclarativeDocumentBackend, PaginatedDocumentB
                         e_is_a_list_item = False
 
                     if e_is_a_list_item:
-                        # TODO: Set marker and enumerated arguments if this is an enumeration element.
+                        # Set marker and enumerated arguments if this is an enumeration element.
                         enum_marker = str(enum_list_item_value) + "."
                         doc.add_list_item(
                             marker=enum_marker,

docling/backend/msword_backend.py

@@ -44,6 +44,7 @@ class MsWordDocumentBackend(DeclarativeDocumentBackend):
             self.parents[i] = None
 
         self.level = 0
+        self.listIter = 0
 
         self.history = {
             "names": [None],
@@ -124,6 +125,7 @@ class MsWordDocumentBackend(DeclarativeDocumentBackend):
         for element in body:
             tag_name = etree.QName(element).localname
 
+            # Check for Inline Images (drawings or blip elements)
             found_drawing = etree.ElementBase.xpath(
                 element, ".//w:drawing", namespaces=self.xml_namespaces
             )
@@ -137,8 +139,7 @@ class MsWordDocumentBackend(DeclarativeDocumentBackend):
                     self.handle_tables(element, docx_obj, doc)
                 except Exception:
                     _log.error("could not parse a table, broken docx table")
-            # Check for Inline Images (drawings or blip elements)
+
-            # elif element.xpath(".//w:drawing", namespaces = self.xml_namespaces) or element.xpath(".//w:pict", namespaces = self.xml_namespaces):
             elif found_drawing or found_pict:
                 self.handle_pictures(element, docx_obj, doc)
             # Check for Text
@@ -166,7 +167,6 @@ class MsWordDocumentBackend(DeclarativeDocumentBackend):
             # Get the numId element and extract the value
             numId_elem = numPr.find("w:numId", namespaces=paragraph._element.nsmap)
             ilvl_elem = numPr.find("w:ilvl", namespaces=paragraph._element.nsmap)
-
             numId = numId_elem.get(self.XML_KEY) if numId_elem is not None else None
             ilvl = ilvl_elem.get(self.XML_KEY) if ilvl_elem is not None else None
 
@@ -215,15 +215,31 @@ class MsWordDocumentBackend(DeclarativeDocumentBackend):
         text = paragraph.text.strip()
         # if len(text)==0 # keep empty paragraphs, they seperate adjacent lists!
 
+        # Common styles for bullet and numbered lists.
+        # "List Bullet", "List Number", "List Paragraph"
+        # TODO: reliably identify wether list is a numbered list or not
+        # is_numbered = "List Bullet" not in paragraph.style.name
+        is_numbered = False
+
         p_style_name, p_level = self.get_label_and_level(paragraph)
         numid, ilevel = self.get_numId_and_ilvl(paragraph)
+        # print("numid: {}, ilevel: {}, text: {}".format(numid, ilevel, text))
 
-        # print("paragraph.text: {} | numid: {} | ilevel: {}".format(paragraph.text, numid, ilevel))
+        if numid == 0:
+            numid = None
 
         # Handle lists
         if numid is not None and ilevel is not None:
             self.add_listitem(
-                element, docx_obj, doc, p_style_name, p_level, numid, ilevel, text
+                element,
+                docx_obj,
+                doc,
+                p_style_name,
+                p_level,
+                numid,
+                ilevel,
+                text,
+                is_numbered,
             )
             self.update_history(p_style_name, p_level, numid, ilevel)
             return
@@ -233,14 +249,12 @@ class MsWordDocumentBackend(DeclarativeDocumentBackend):
                     self.parents[key] = None
             self.level = self.level_at_new_list - 1
             self.level_at_new_list = None
-
         if p_style_name in ["Title"]:
             for key, val in self.parents.items():
                 self.parents[key] = None
             self.parents[0] = doc.add_text(
                 parent=None, label=DocItemLabel.TITLE, text=text
             )
-
         elif "Heading" in p_style_name:
             self.add_header(element, docx_obj, doc, p_style_name, p_level, text)
 
@@ -312,8 +326,20 @@ class MsWordDocumentBackend(DeclarativeDocumentBackend):
         return
 
     def add_listitem(
-        self, element, docx_obj, doc, p_style_name, p_level, numid, ilevel, text: str
+        self,
+        element,
+        docx_obj,
+        doc,
+        p_style_name,
+        p_level,
+        numid,
+        ilevel,
+        text: str,
+        is_numbered=False,
     ):
+        # is_numbered = is_numbered
+        enum_marker = ""
+
         level = self.get_level()
         if self.prev_numid() is None:  # Open new list
             self.level_at_new_list = level  # type: ignore
@@ -323,7 +349,16 @@ class MsWordDocumentBackend(DeclarativeDocumentBackend):
             )
 
             # TODO: Set marker and enumerated arguments if this is an enumeration element.
-            doc.add_list_item(parent=self.parents[level], text=text)
+            self.listIter += 1
+            if is_numbered:
+                enum_marker = str(self.listIter) + "."
+                is_numbered = True
+            doc.add_list_item(
+                marker=enum_marker,
+                enumerated=is_numbered,
+                parent=self.parents[level],
+                text=text,
+            )
 
         elif (
             self.prev_numid() == numid and self.prev_indent() < ilevel
@@ -334,12 +369,26 @@ class MsWordDocumentBackend(DeclarativeDocumentBackend):
             ):
                 # TODO: determine if this is an unordered list or an ordered list.
                 #  Set GroupLabel.ORDERED_LIST when it fits.
+                self.listIter = 0
+                if is_numbered:
+                    self.parents[i] = doc.add_group(
+                        label=GroupLabel.ORDERED_LIST,
+                        name="list",
+                        parent=self.parents[i - 1],
+                    )
+                else:
                     self.parents[i] = doc.add_group(
                         label=GroupLabel.LIST, name="list", parent=self.parents[i - 1]
                     )
 
             # TODO: Set marker and enumerated arguments if this is an enumeration element.
+            self.listIter += 1
+            if is_numbered:
+                enum_marker = str(self.listIter) + "."
+                is_numbered = True
             doc.add_list_item(
+                marker=enum_marker,
+                enumerated=is_numbered,
                 parent=self.parents[self.level_at_new_list + ilevel],
                 text=text,
             )
@@ -350,14 +399,30 @@ class MsWordDocumentBackend(DeclarativeDocumentBackend):
                     self.parents[k] = None
 
             # TODO: Set marker and enumerated arguments if this is an enumeration element.
+            self.listIter += 1
+            if is_numbered:
+                enum_marker = str(self.listIter) + "."
+                is_numbered = True
             doc.add_list_item(
+                marker=enum_marker,
+                enumerated=is_numbered,
                 parent=self.parents[self.level_at_new_list + ilevel],
                 text=text,
             )
+            self.listIter = 0
 
         elif self.prev_numid() == numid or self.prev_indent() == ilevel:
             # TODO: Set marker and enumerated arguments if this is an enumeration element.
-            doc.add_list_item(parent=self.parents[level - 1], text=text)
+            self.listIter += 1
+            if is_numbered:
+                enum_marker = str(self.listIter) + "."
+                is_numbered = True
+            doc.add_list_item(
+                marker=enum_marker,
+                enumerated=is_numbered,
+                parent=self.parents[level - 1],
+                text=text,
+            )
         return
 
     def handle_tables(self, element, docx_obj, doc):

poetry.lock

@@ -885,7 +885,7 @@ files = []
 develop = false
 
 [package.dependencies]
-docling-core = {git = "https://github.com/DS4SD/docling-core.git", rev = "6fee533a101ca08f62e88826218c33e0aab2f417"}
+docling-core = {git = "https://github.com/DS4SD/docling-core.git", rev = "7c104d61aa5d003dd8d9711c37e23ce04799f4c9"}
 docutils = "!=0.21"
 matplotlib = "^3.7.1"
 networkx = "^3.1"
@@ -909,8 +909,8 @@ toolkit = ["deepsearch-toolkit (>=0.31.0)"]
 [package.source]
 type = "git"
 url = "https://github.com/DS4SD/deepsearch-glm.git"
-reference = "c13a6cdda25206911d63a5a28e990217ad823068"
+reference = "c185c4f985ccd29a470a1cddd3bec43880b739ee"
-resolved_reference = "c13a6cdda25206911d63a5a28e990217ad823068"
+resolved_reference = "c185c4f985ccd29a470a1cddd3bec43880b739ee"
 
 [[package]]
 name = "dill"
@@ -952,14 +952,15 @@ json-schema-for-humans = "^1.0.0"
 jsonref = "^1.1.0"
 jsonschema = "^4.16.0"
 pandas = "^2.1.4"
+pillow = "^10.3.0"
 pydantic = "^2.6.0"
 tabulate = "^0.9.0"
 
 [package.source]
 type = "git"
 url = "https://github.com/DS4SD/docling-core.git"
-reference = "6fee533a101ca08f62e88826218c33e0aab2f417"
+reference = "7c104d61aa5d003dd8d9711c37e23ce04799f4c9"
-resolved_reference = "6fee533a101ca08f62e88826218c33e0aab2f417"
+resolved_reference = "7c104d61aa5d003dd8d9711c37e23ce04799f4c9"
 
 [[package]]
 name = "docling-ibm-models"
@@ -3441,9 +3442,9 @@ files = [
 [package.dependencies]
 numpy = [
     {version = ">=1.26.0", markers = "python_version >= \"3.12\""},
-    {version = ">=1.23.5", markers = "python_version >= \"3.11\" and python_version < \"3.12\""},
     {version = ">=1.21.4", markers = "python_version >= \"3.10\" and platform_system == \"Darwin\" and python_version < \"3.11\""},
     {version = ">=1.21.2", markers = "platform_system != \"Darwin\" and python_version >= \"3.10\" and python_version < \"3.11\""},
+    {version = ">=1.23.5", markers = "python_version >= \"3.11\" and python_version < \"3.12\""},
 ]
 
 [[package]]
@@ -3577,8 +3578,8 @@ files = [
 [package.dependencies]
 numpy = [
     {version = ">=1.26.0", markers = "python_version >= \"3.12\""},
-    {version = ">=1.23.2", markers = "python_version == \"3.11\""},
     {version = ">=1.22.4", markers = "python_version < \"3.11\""},
+    {version = ">=1.23.2", markers = "python_version == \"3.11\""},
 ]
 python-dateutil = ">=2.8.2"
 pytz = ">=2020.1"
@@ -7114,4 +7115,4 @@ tesserocr = ["tesserocr"]
 [metadata]
 lock-version = "2.0"
 python-versions = "^3.10"
-content-hash = "46f6c1eb76034223f7d65760f6ebe0989ba9e8aff46fcdbce82c147030fcb8be"
+content-hash = "14143d6cc79f4c2c8a4d021711198697e91ca01ecf290dd270b483984461c3d1"

pyproject.toml

@@ -37,9 +37,9 @@ torchvision = [
 ######################
 python = "^3.10"
 pydantic = "^2.0.0"
-docling-core = {git = "https://github.com/DS4SD/docling-core.git", rev = "6fee533a101ca08f62e88826218c33e0aab2f417"}
+docling-core = {git = "https://github.com/DS4SD/docling-core.git", rev = "7c104d61aa5d003dd8d9711c37e23ce04799f4c9"}
 docling-ibm-models = {git = "https://github.com/DS4SD/docling-ibm-models.git", rev = "1d2e2a2e6eb152c237f1383cdba20cf85db80b97"}
-deepsearch-glm = {git = "https://github.com/DS4SD/deepsearch-glm.git", rev = "c13a6cdda25206911d63a5a28e990217ad823068"}
+deepsearch-glm = {git = "https://github.com/DS4SD/deepsearch-glm.git", rev = "c185c4f985ccd29a470a1cddd3bec43880b739ee"}
 docling-parse = "^1.5.1"
 
 filetype = "^1.2.0"