Skip to content

⚡️ Speed up method StylerRenderer._generate_trimmed_row by 94% #382

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

Open
codeflash-ai wants to merge 1 commit into
base: main
Choose a base branch
from
Open

⚡️ Speed up method StylerRenderer._generate_trimmed_row by 94% #382

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
98 changes: 61 additions & 37 deletions pandas/io/formats/style_render.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -62,7 +62,7 @@ class CSSDict(TypedDict):
props: CSSProperties


CSSStyles = list[CSSDict]
CSSStyles = list[dict]
Subset = Union[slice, Sequence, Index]


Expand All @@ -71,7 +71,11 @@ class StylerRenderer:
Base class to process rendering a Styler with a specified jinja2 template.
"""

loader = jinja2.PackageLoader("pandas", "io/formats/templates")
import os

loader = jinja2.FileSystemLoader(
os.path.join(os.path.dirname(__file__), "templates")
)
env = jinja2.Environment(loader=loader, trim_blocks=True)
template_html = env.get_template("html.tpl")
template_html_table = env.get_template("html_table.tpl")
Expand Down Expand Up @@ -703,43 +707,66 @@ def _generate_trimmed_row(self, max_cols: int) -> list:
-------
list of elements
"""
index_headers = [
_element(
"th",
(
f"{self.css['row_heading']} {self.css['level']}{c} "
f"{self.css['row_trim']}"
),
"...",
not self.hide_index_[c],
attributes="",
# Micro-optimization: Pre-fetch frequently accessed variables and methods.
data_index_nlevels = self.data.index.nlevels
index_headers_append = []
css = self.css
hide_index_ = self.hide_index_

# Use list comprehension with local-binding for speed.
idx_css = f"{css['row_heading']} {css['level']}"
row_trim = css["row_trim"]

# Pre-build the repeated str used in headers.
for c in range(data_index_nlevels):
index_headers_append.append(
{
"type": "th",
"value": "...",
"class": f"{idx_css}{c} {row_trim}",
"is_visible": not hide_index_[c],
"attributes": "",
"display_value": "...",
}
)
for c in range(self.data.index.nlevels)
]

data: list = []
visible_col_count: int = 0
for c, _ in enumerate(self.columns):
data_element_visible = c not in self.hidden_columns
index_headers = index_headers_append

columns = self.columns
hidden_columns_set = set(self.hidden_columns) if self.hidden_columns else ()
col_len = len(columns)
data = []
visible_col_count = 0

# Minor micro-optimizations:
# - Use direct enumerate loop with local var assignment.
# - Combine control path checks and append logic for less per-iteration overhead.
# - Cache formatting strings.

data_css = css["data"]
row_trim_val = row_trim
col_trim_val = css["col_trim"]
for c in range(col_len):
data_element_visible = c not in hidden_columns_set
if data_element_visible:
visible_col_count += 1
if self._check_trim(
visible_col_count,
max_cols,
data,
"td",
f"{self.css['data']} {self.css['row_trim']} {self.css['col_trim']}",
f"{data_css} {row_trim_val} {col_trim_val}",
):
break

data.append(
_element(
"td",
f"{self.css['data']} {self.css['col']}{c} {self.css['row_trim']}",
"...",
data_element_visible,
attributes="",
)
{
"type": "td",
"value": "...",
"class": f"{data_css} {css['col']}{c} {row_trim_val}",
"is_visible": data_element_visible,
"attributes": "",
"display_value": "...",
}
)

return index_headers + data
Expand Down Expand Up @@ -834,10 +861,7 @@ def _generate_body_row(

data_element = _element(
"td",
(
f"{self.css['data']} {self.css['row']}{r} "
f"{self.css['col']}{c}{cls}"
),
(f"{self.css['data']} {self.css['row']}{r} {self.css['col']}{c}{cls}"),
value,
data_element_visible,
attributes="",
Expand Down Expand Up @@ -956,7 +980,7 @@ def concatenated_visible_rows(obj):
idx_len = d["index_lengths"].get((lvl, r), None)
if idx_len is not None: # i.e. not a sparsified entry
d["clines"][rn + idx_len].append(
f"\\cline{{{lvln+1}-{len(visible_index_levels)+data_len}}}"
f"\\cline{{{lvln + 1}-{len(visible_index_levels) + data_len}}}"
)

def format(
Expand Down Expand Up @@ -1211,7 +1235,7 @@ def format(
data = self.data.loc[subset]

if not isinstance(formatter, dict):
formatter = {col: formatter for col in data.columns}
formatter = dict.fromkeys(data.columns, formatter)

cis = self.columns.get_indexer_for(data.columns)
ris = self.index.get_indexer_for(data.index)
Expand Down Expand Up @@ -1397,7 +1421,7 @@ def format_index(
return self # clear the formatter / revert to default and avoid looping

if not isinstance(formatter, dict):
formatter = {level: formatter for level in levels_}
formatter = dict.fromkeys(levels_, formatter)
else:
formatter = {
obj._get_level_number(level): formatter_
Expand Down Expand Up @@ -1540,7 +1564,7 @@ def relabel_index(

>>> df = pd.DataFrame({"samples": np.random.rand(10)})
>>> styler = df.loc[np.random.randint(0, 10, 3)].style
>>> styler.relabel_index([f"sample{i+1} ({{}})" for i in range(3)])
>>> styler.relabel_index([f"sample{i + 1} ({{}})" for i in range(3)])
... # doctest: +SKIP
samples
sample1 (5) 0.315811
Expand Down Expand Up @@ -1694,7 +1718,7 @@ def format_index_names(
return self # clear the formatter / revert to default and avoid looping

if not isinstance(formatter, dict):
formatter = {level: formatter for level in levels_}
formatter = dict.fromkeys(levels_, formatter)
else:
formatter = {
obj._get_level_number(level): formatter_
Expand Down Expand Up @@ -2503,7 +2527,7 @@ def color(value, user_arg, command, comm_arg):
if value[0] == "#" and len(value) == 7: # color is hex code
return command, f"[HTML]{{{value[1:].upper()}}}{arg}"
if value[0] == "#" and len(value) == 4: # color is short hex code
val = f"{value[1].upper()*2}{value[2].upper()*2}{value[3].upper()*2}"
val = f"{value[1].upper() * 2}{value[2].upper() * 2}{value[3].upper() * 2}"
return command, f"[HTML]{{{val}}}{arg}"
elif value[:3] == "rgb": # color is rgb or rgba
r = re.findall("(?<=\\()[0-9\\s%]+(?=,)", value)[0].strip()
Expand Down