Skip to content

Add style name caching to improve performance #598

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
JayHuLBL merged 1 commit into from
Aug 25, 2026
Merged

Add style name caching to improve performance #598

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
63 changes: 49 additions & 14 deletions server/scripts/sequence-doc/src/mogrifier.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -25,12 +25,44 @@
Selections = Dict[str, List]

elements_to_delete = []
_run_style_name_cache = {}
_paragraph_style_name_cache = {}

def initialize_remove_list():
''' Resets global remove list to an empty array
''' Resets global remove list and style name caches for a fresh doc
'''
global elements_to_delete
elements_to_delete = []
_run_style_name_cache.clear()
_paragraph_style_name_cache.clear()

def get_run_style_name(run):
''' Cached lookup of a run's character style name.
python-docx re-walks the whole style tree on every call to
Styles.default()/get_by_id() with no internal caching, so calling
run.style.name for every run in the document (most of which fall
back to the default style) made style resolution the dominant cost.
The style id -> name mapping can't change while we process a doc,
so cache it here keyed by (part, style id).
'''
key = (run.part, run.element.style)
try:
return _run_style_name_cache[key]
except KeyError:
name = run.style.name
_run_style_name_cache[key] = name
return name

def get_paragraph_style_name(paragraph):
''' Cached lookup of a paragraph's style name, see get_run_style_name
'''
key = (paragraph.part, paragraph._p.style)
try:
return _paragraph_style_name_cache[key]
except KeyError:
name = paragraph.style.name
_paragraph_style_name_cache[key] = name
return name

def remove_node(node):
''' Controls access to a global list for nodes to be deleted
Expand All @@ -50,7 +82,7 @@ def remove_element(element):
def get_heading_level(paragraph):
''' Gets heading level of paragraph
'''
style = paragraph.style.name
style = get_paragraph_style_name(paragraph)
match = re.match(r'Heading (\d+)', style)

if not match:
Expand Down Expand Up @@ -83,7 +115,7 @@ def remove_info_box(paragraph, run_op_lookup: Dict):
def remove_section(paragraph: Paragraph, run_op_lookup: Dict):
''' Removes a section of text and its siblings
'''
style = paragraph.style.name
style = get_paragraph_style_name(paragraph)
level = get_heading_level(paragraph)

if style in ('Info. box', 'InfoboxList', 'InfoTableTitle'):
Expand Down Expand Up @@ -144,25 +176,26 @@ def create_control_structures(doc):
## Sections
for para in doc.paragraphs:
for run in para.runs:
run_style_name = get_run_style_name(run)
if (
(current and run.style.name != ANNOTATION_STYLE)
(current and run_style_name != ANNOTATION_STYLE)
or
(current and para is not current['paragraph'])
):
current['text'] = current['text'].strip()
if current['text']:
control_structures.append(current)
current = None
if not current and run.style.name == ANNOTATION_STYLE:

if not current and run_style_name == ANNOTATION_STYLE:
current = {
'paragraph': para,
'text': '',
'runs': [],
}
run_op_lookup[run.element] = current

if current and run.style.name == ANNOTATION_STYLE:
if current and run_style_name == ANNOTATION_STYLE:
run_op_lookup[run.element] = current
current['runs'].append(run)
current['text'] += run.text
Expand All @@ -173,8 +206,9 @@ def create_control_structures(doc):
for cell in row.cells:
for para in cell.paragraphs:
for run in para.runs:
run_style_name = get_run_style_name(run)
if (
(current and run.style.name != ANNOTATION_STYLE)
(current and run_style_name != ANNOTATION_STYLE)
or
(current and para is not current['paragraph'])
):
Expand All @@ -183,7 +217,7 @@ def create_control_structures(doc):
control_structures.append(current)
current = None

if not current and run.style.name == ANNOTATION_STYLE:
if not current and run_style_name == ANNOTATION_STYLE:
current = {
'paragraph': para,
'text': '',
Expand All @@ -195,7 +229,7 @@ def create_control_structures(doc):
}
run.op = current

if current and run.style.name == ANNOTATION_STYLE:
if current and run_style_name == ANNOTATION_STYLE:
run.op = current
current['runs'].append(run)
current['text'] += run.text
Expand All @@ -220,9 +254,10 @@ def remove_info_and_instr_boxes(doc, selections: Selections):
instr_box_style = 'Instr. box'

for para in doc.paragraphs:
if utils.reduce_to_boolean(selections['DEL_INFO_BOX']) and para.style.name in info_box_styles:
para_style_name = get_paragraph_style_name(para)
if utils.reduce_to_boolean(selections['DEL_INFO_BOX']) and para_style_name in info_box_styles:
remove_node(para)
if para.style.name == instr_box_style:
if para_style_name == instr_box_style:
remove_node(para)

def evaluate_annotation(op, name_map, selections: Selections):
Expand Down Expand Up @@ -509,15 +544,15 @@ def remove_toggles(doc):
'''
for para in doc.paragraphs:
for run in para.runs:
if run.style.name == ANNOTATION_STYLE:
if get_run_style_name(run) == ANNOTATION_STYLE:
remove_node(run)

for table in doc.tables:
for row in table.rows:
for cell in row.cells:
for para in cell.paragraphs:
for run in para.runs:
if run.style.name == ANNOTATION_STYLE:
if get_run_style_name(run) == ANNOTATION_STYLE:
remove_node(run)

def mogrify_doc(doc: Document, name_map: Dict, selections: Selections) -> Document:
Expand Down