NeoAnd/Docling
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Added handling of code blocks in html with <pre> tag (#302)

Browse Source 
Signed-off-by: Maksym Lysak <mly@zurich.ibm.com>
Co-authored-by: Maksym Lysak <mly@zurich.ibm.com>
This commit is contained in:
Maxim Lysak 2024-11-11 15:00:11 +01:00 committed by GitHub
parent 1239ade275
commit 53bf2d1790
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 12 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
docling/backend/html_backend.py
View File

@ -120,6 +120,8 @@ class HTMLDocumentBackend(DeclarativeDocumentBackend):
self.handle_header(element, idx, doc) self.handle_header(element, idx, doc)
elif element.name in ["p"]: elif element.name in ["p"]:
self.handle_paragraph(element, idx, doc) self.handle_paragraph(element, idx, doc)
elif element.name in ["pre"]:
self.handle_code(element, idx, doc)
elif element.name in ["ul", "ol"]: elif element.name in ["ul", "ol"]:
self.handle_list(element, idx, doc) self.handle_list(element, idx, doc)
elif element.name in ["li"]: elif element.name in ["li"]:
@ -205,6 +207,16 @@ class HTMLDocumentBackend(DeclarativeDocumentBackend):
level=hlevel, level=hlevel,
) )
def handle_code(self, element, idx, doc):
"""Handles monospace code snippets (pre)."""
if element.text is None:
return
text = element.text.strip()
label = DocItemLabel.CODE
if len(text) == 0:
return
doc.add_text(parent=self.parents[self.level], label=label, text=text)
def handle_paragraph(self, element, idx, doc): def handle_paragraph(self, element, idx, doc):
"""Handles paragraph tags (p).""" """Handles paragraph tags (p)."""
if element.text is None: if element.text is None: