mirrors/docling
1
0
Fork 0
mirror of https://github.com/DS4SD/docling.git synced 2025-07-27 12:34:22 +00:00
Code Issues Packages Projects Releases Wiki Activity

Remove prints and backend flag

Browse Source 
Signed-off-by: Rafael Teixeira de Lima <Rafael.td.lima@gmail.com>
This commit is contained in:
Rafael Teixeira de Lima 2025-01-28 13:39:54 +01:00
parent 31e30a2cb7
commit b3b7c387ca
2 changed files with 4 additions and 20 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
docling/backend/docx_latex/omml.py
View File

@ -187,7 +187,7 @@ class oMath2Latex(Tag2Method):
self._latex = self.process_children(element) self._latex = self.process_children(element)
def __str__(self): def __str__(self):
return self.latex return self.latex.replace(" ", " ")
def __unicode__(self): def __unicode__(self):
return self.__str__(self) return self.__str__(self)
@ -231,20 +231,13 @@ class oMath2Latex(Tag2Method):
pr = c_dict["dPr"] pr = c_dict["dPr"]
null = D_DEFAULT.get("null") null = D_DEFAULT.get("null")
print(pr.text)
s_val = get_val(pr.begChr, default=D_DEFAULT.get("left"), store=T) s_val = get_val(pr.begChr, default=D_DEFAULT.get("left"), store=T)
print(pr.begChr, D_DEFAULT.get("left"), s_val)
e_val = get_val(pr.endChr, default=D_DEFAULT.get("right"), store=T) e_val = get_val(pr.endChr, default=D_DEFAULT.get("right"), store=T)
print(pr.endChr, D_DEFAULT.get("right"), s_val)
delim = pr.text + D.format( delim = pr.text + D.format(
left=null if not s_val else escape_latex(s_val), left=null if not s_val else escape_latex(s_val),
text=c_dict["e"], text=c_dict["e"],
right=null if not e_val else escape_latex(e_val), right=null if not e_val else escape_latex(e_val),
) )
print(delim)
print()
return delim return delim
def do_spre(self, elm): def do_spre(self, elm):
@ -355,7 +348,7 @@ class oMath2Latex(Tag2Method):
""" """
rows = [] rows = []
for stag, t, e in self.process_children_list(elm): for stag, t, e in self.process_children_list(elm):
if stag is "mPr": if stag == "mPr":
pass pass
elif stag == "mr": elif stag == "mr":
rows.append(t) rows.append(t)

13
docling/backend/msword_backend.py
View File

@ -27,12 +27,7 @@ _log = logging.getLogger(__name__)
class MsWordDocumentBackend(DeclarativeDocumentBackend): class MsWordDocumentBackend(DeclarativeDocumentBackend):
def __init__( def __init__(self, in_doc: "InputDocument", path_or_stream: Union[BytesIO, Path]):
self,
in_doc: "InputDocument",
path_or_stream: Union[BytesIO, Path],
get_latex=False,
):
super().__init__(in_doc, path_or_stream) super().__init__(in_doc, path_or_stream)
self.XML_KEY = ( self.XML_KEY = (
"{http://schemas.openxmlformats.org/wordprocessingml/2006/main}val" "{http://schemas.openxmlformats.org/wordprocessingml/2006/main}val"
@ -54,9 +49,6 @@ class MsWordDocumentBackend(DeclarativeDocumentBackend):
self.level = 0 self.level = 0
self.listIter = 0 self.listIter = 0
# Transform MSWord equations to latex
self.get_latex = get_latex
self.history = { self.history = {
"names": [None], "names": [None],
"levels": [None], "levels": [None],
@ -248,8 +240,7 @@ class MsWordDocumentBackend(DeclarativeDocumentBackend):
paragraph = docx.text.paragraph.Paragraph(element, docx_obj) paragraph = docx.text.paragraph.Paragraph(element, docx_obj)
text = paragraph.text text = paragraph.text
if self.get_latex: text = self.handle_equations_in_text(element=element, text=text)
text = self.handle_equations_in_text(element=element, text=text)
if text is None: if text is None:
return return