def __create_xyzdata(dfs):
chart_data = None
seriesNum = 1
for df in dfs:
colNames = df.columns.tolist()
name = 'Series ' + str(seriesNum)
if hasattr(df, 'name') and df.name != "":
name = df.name
if len(colNames) > 1 and len(colNames) < 4:
if len(colNames) == 2 and chart_data is None:
chart_data = XyChartData()
elif len(colNames) == 3 and chart_data is None:
chart_data = BubbleChartData()
series = chart_data.add_series(name)
for index, row in df.iterrows():
data = []
for colName in colNames:
data.append(row[colName])
if len(colNames) == 2:
series.add_data_point(data[0], data[1])
else:
series.add_data_point(data[0], data[1], data[2])
seriesNum += 1
return chart_data
def replace_xy_data(shape: GraphicFrame, data: XYData):
chart_data = XyChartData()
for name, points in data.series.items():
series_data = chart_data.add_series(name)
for point in points:
series_data.add_data_point(point.x, point.y)
shape.chart.replace_data(chart_data)
def _build_xy_chart_data(csv):
chart_data = XyChartData()
for i in range(1, csv.columns.size):
series = chart_data.add_series(csv.columns[i])
xy_col = csv.iloc[:, [0, i]]
for (_, row) in xy_col.iterrows():
y, x = _nan_to_none(row[0]), _nan_to_none(row[1])
log.debug(u" Adding xy %s,%s" % (x, y))
series.add_data_point(x, y)
return chart_data
def when_I_replace_its_data_with_3_series_of_three_points_each(context):
chart_data = XyChartData()
x = y = 0
for idx in range(3):
series_title = "New Series %d" % (idx + 1)
series = chart_data.add_series(series_title)
for jdx in range(3):
x, y = idx * 3 + jdx, idx * 2 + jdx
series.add_data_point(x, y)
context.chart.replace_data(chart_data)
def create_chart(df):
x = "平均気温"
y = "最高気温"
chart = XyChartData()
c1 = chart.add_series('系列1')
c2 = chart.add_series('系列2')
str_df = df[[x,y]].astype("str")
df_col = list([str_df.columns.tolist()])
df_list = df_col + str_df.values.tolist()
s1.add_data_point()
print(df_list)
def rewrite_fixture(self):
ser_xml, expected_xml = snippet_seq("rewrite-ser")[2:4]
chart_data = XyChartData()
series_data = chart_data.add_series("Series 1")
series_data.add_data_point(1, 2)
series_data.add_data_point(3, 4)
rewriter = _XySeriesXmlRewriter(chart_data)
ser = parse_xml(ser_xml)
return rewriter, ser, series_data, expected_xml
def rewrite_fixture(self):
ser_xml, expected_xml = snippet_seq('rewrite-ser')[2:4]
chart_data = XyChartData()
series_data = chart_data.add_series('Series 1')
series_data.add_data_point(1, 2)
series_data.add_data_point(3, 4)
rewriter = _XySeriesXmlRewriter(chart_data)
ser = parse_xml(ser_xml)
return rewriter, ser, series_data, expected_xml
def yVal_fixture(self, request):
cht_numfmt, ser_numfmt, expected_numfmt = request.param
expected_xml = xml(
'c:yVal/c:numRef/(c:f"Sheet1!$B$2:$B$1",c:numCache/(c:formatCode'
'"%s",c:ptCount{val=0}))' % expected_numfmt)
chart_number_format = () if cht_numfmt is None else (cht_numfmt, )
series_number_format = () if ser_numfmt is None else (ser_numfmt, )
chart_data = XyChartData(*chart_number_format)
series_data = chart_data.add_series(None, *series_number_format)
xml_writer = _XySeriesXmlWriter(series_data)
return xml_writer, expected_xml
def when_I_replace_its_data_with_3_series_of_three_points_each(context):
chart_data = XyChartData()
x = y = 0
for idx in range(3):
series_title = 'New Series %d' % (idx + 1)
series = chart_data.add_series(series_title)
for jdx in range(3):
x, y = idx * 3 + jdx, idx * 2 + jdx
series.add_data_point(x, y)
context.chart.replace_data(chart_data)
def yVal_fixture(self, request):
cht_numfmt, ser_numfmt, expected_numfmt = request.param
expected_xml = xml(
'c:yVal/c:numRef/(c:f"Sheet1!$B$2:$B$1",c:numCache/(c:formatCode'
'"%s",c:ptCount{val=0}))' % expected_numfmt
)
chart_number_format = () if cht_numfmt is None else (cht_numfmt,)
series_number_format = () if ser_numfmt is None else (ser_numfmt,)
chart_data = XyChartData(*chart_number_format)
series_data = chart_data.add_series(None, *series_number_format)
xml_writer = _XySeriesXmlWriter(series_data)
return xml_writer, expected_xml
def _build_xy_chart_data(csv, xy_transpose, number_format):
chart_data = XyChartData()
for i in range(1, csv.columns.size):
# nameに日本語が入ると後続処理中で、python v2.7の場合にUnicodeDecodeErrorが出るため対処。nameは結局pptx内では使われない
series = chart_data.add_series(u"column%s" % i, number_format=_normalize_number_format(number_format))
xy_col = csv.iloc[:, [0, i]]
for (_, row) in xy_col.iterrows():
if xy_transpose:
y, x = _nan_to_none(row[0]), _nan_to_none(row[1])
else:
x, y = _nan_to_none(row[0]), _nan_to_none(row[1])
log.debug(u" Adding xy %s,%s" % (x, y))
series.add_data_point(x, y)
return chart_data
def when_I_add_an_xy_chart_having_2_series_of_3_points(context, xy_type):
chart_type = getattr(XL_CHART_TYPE, xy_type)
data = (('Series 1', ((-0.1, 0.5), (16.2, 0.0), (8.0, 0.2))),
('Series 2', ((12.4, 0.8), (-7.5, -0.5), (-5.1, -0.2))))
chart_data = XyChartData()
for series_data in data:
series_label, points = series_data
series = chart_data.add_series(series_label)
for point in points:
x, y = point
series.add_data_point(x, y)
context.chart = context.slide.shapes.add_chart(chart_type, Inches(1),
Inches(1), Inches(8),
Inches(5), chart_data).chart
def make_xy_chart_data(ser_count, point_count):
"""
Return an |XyChartData| object populated with *ser_count* series each
having *point_count* data points. Values are auto-generated.
"""
points = (
(1.1, 11.1), (2.1, 12.1), (3.1, 13.1),
(1.2, 11.2), (2.2, 12.2), (3.2, 13.2),
)
chart_data = XyChartData()
for i in range(ser_count):
series_label = 'Series %d' % (i+1)
series = chart_data.add_series(series_label)
for j in range(point_count):
point_idx = (i * point_count) + j
x, y = points[point_idx]
series.add_data_point(x, y)
return chart_data
def when_I_add_an_xy_chart_having_2_series_of_3_points(context, xy_type):
chart_type = getattr(XL_CHART_TYPE, xy_type)
data = (
("Series 1", ((-0.1, 0.5), (16.2, 0.0), (8.0, 0.2))),
("Series 2", ((12.4, 0.8), (-7.5, -0.5), (-5.1, -0.2))),
)
chart_data = XyChartData()
for series_data in data:
series_label, points = series_data
series = chart_data.add_series(series_label)
for point in points:
x, y = point
series.add_data_point(x, y)
context.chart = context.slide.shapes.add_chart(
chart_type, Inches(1), Inches(1), Inches(8), Inches(5), chart_data
).chart
def generate_correlation_curve(presentation_context):
x_label = presentation_context["topic"]
y_label = presentation_context["seed"]
if is_too_similar_for_axes(x_label, y_label):
x_label = _CORRELATION_WORD_GENERATOR(y_label)
if is_too_similar_for_axes(x_label, y_label):
x_label = "time"
presentation_context.update({"x_label": x_label, "y_label": y_label})
title = correlation_title_generator(presentation_context)
if not title:
return None
chart_data = XyChartData()
serie = chart_data.add_series("Model")
# Generate some Xs, with chance of exponential differences in size between generated x axes
xs = generate_random_x(0, 2**random.uniform(1, 10),
int(2**random.uniform(3, 8)))
# Generate y
data_points = generate_y(xs, create_interesting_curve_function())
max_x = max(xs)
data_points = add_gaussian_noise_to_multidim_points(
1.5 * random.uniform(0, max_x / 10), data_points)
# Remove negatives
data_points = [(abs(datapoint[0]), abs(datapoint[1]))
for datapoint in data_points]
add_data_to_series(serie, data_points)
return (
title,
XL_CHART_TYPE.XY_SCATTER,
chart_data,
create_set_scatter_properties(x_label, y_label),
)
def populate_fixture(self, workbook_, worksheet_):
chart_data = XyChartData()
series_1 = chart_data.add_series("Series 1")
for pt in ((1, 1.1), (2, 2.2)):
series_1.add_data_point(*pt)
series_2 = chart_data.add_series("Series 2")
for pt in ((3, 3.3), (4, 4.4)):
series_2.add_data_point(*pt)
workbook_writer = XyWorkbookWriter(chart_data)
expected_calls = [
call.write_column(1, 0, [1, 2], ANY),
call.write(0, 1, "Series 1"),
call.write_column(1, 1, [1.1, 2.2], ANY),
call.write_column(5, 0, [3, 4], ANY),
call.write(4, 1, "Series 2"),
call.write_column(5, 1, [3.3, 4.4], ANY),
]
return workbook_writer, workbook_, worksheet_, expected_calls
def populate_fixture(self, workbook_, worksheet_):
chart_data = XyChartData()
series_1 = chart_data.add_series('Series 1')
for pt in ((1, 1.1), (2, 2.2)):
series_1.add_data_point(*pt)
series_2 = chart_data.add_series('Series 2')
for pt in ((3, 3.3), (4, 4.4)):
series_2.add_data_point(*pt)
workbook_writer = XyWorkbookWriter(chart_data)
expected_calls = [
call.write_column(1, 0, [1, 2], ANY),
call.write(0, 1, 'Series 1'),
call.write_column(1, 1, [1.1, 2.2], ANY),
call.write_column(5, 0, [3, 4], ANY),
call.write(4, 1, 'Series 2'),
call.write_column(5, 1, [3.3, 4.4], ANY)
]
return workbook_writer, workbook_, worksheet_, expected_calls
def make_xy_chart_data(ser_count, point_count):
"""
Return an |XyChartData| object populated with *ser_count* series each
having *point_count* data points. Values are auto-generated.
"""
points = (
(1.1, 11.1),
(2.1, 12.1),
(3.1, 13.1),
(1.2, 11.2),
(2.2, 12.2),
(3.2, 13.2),
)
chart_data = XyChartData()
for i in range(ser_count):
series_label = "Series %d" % (i + 1)
series = chart_data.add_series(series_label)
for j in range(point_count):
point_idx = (i * point_count) + j
x, y = points[point_idx]
series.add_data_point(x, y)
return chart_data
def df_to_chartdata(df, datatype, number_format=None):
'''
根据给定的图表数据类型生成相应的数据
Chartdata:一般的数据
XyChartData: 散点图数据
BubbleChartData:气泡图数据
'''
if isinstance(df, pd.Series):
df = pd.DataFrame(df)
df.fillna(0, inplace=True)
datatype = datatype.lower()
if datatype == 'chartdata':
chart_data = ChartData()
chart_data.categories = ['%s' % (c) for c in list(df.index)]
for col_name in df.columns:
chart_data.add_series('%s' % (col_name), list(df[col_name]),
number_format)
return chart_data
if datatype == 'xychartdata':
chart_data = XyChartData()
if not isinstance(df, list):
df = [df]
for d in df:
series_name = '%s' % (d.columns[0]) + ' vs ' + '%s' % (
d.columns[1])
series_ = chart_data.add_series(series_name)
for i in range(len(d)):
series_.add_data_point(d.iloc[i, 0], d.iloc[i, 1])
return chart_data
if datatype == 'bubblechartdata':
chart_data = BubbleChartData()
if not isinstance(df, list):
df = [df]
for d in df:
series_name = '%s' % (d.columns[0]) + ' vs ' + '%s' % (
d.columns[1])
series_ = chart_data.add_series(series_name)
for i in range(len(d)):
series_.add_data_point(d.iloc[i, 0], d.iloc[i, 1], d.iloc[i,
2])
return chart_data
def add_scatter_chart(self, slide, x, y, cx, cy, *series, unit='cm'):
'''
Add scatter chart into given slide.
Args:
slide: Slide Given slide
x: x aixs location of chart
y: float y aixs location of chart
cx: float Total width of chart
cy: float Total height of chart
series: list List of data series, one of each has 2 element: 'series name', 'point list (x, y)'
eg: *series = *[('Series 1', [(1.3, 3.6), (2.3, 1.5)]),
('Series 2', [(-0.5, 1.6), (0.4, 2.9)])]
unit: str Must 'cm' or 'inches'
Return:
scatter_chart: Scatter chart
'''
x, y, cx, cy = self._convert_nums(x, y, cx, cy, unit=unit)
chart_data = self._add_series_datapoints(XyChartData(), *series)
graphic_frame = slide.shapes.add_chart(XL_CHART_TYPE.XY_SCATTER, x, y,
cx, cy, chart_data)
return graphic_frame.chart
def given_a_XyChartData_object_with_number_format(context, strval):
params = {}
if strval != 'None':
params['number_format'] = int(strval)
context.chart_data = XyChartData(**params)
def add_series_fixture(self, request, XySeriesData_, series_data_):
chart_data = XyChartData()
label = 'Series Label'
return chart_data, label, XySeriesData_, series_data_
def it_is_a__BaseChartData_object(self):
assert isinstance(XyChartData(), _BaseChartData)
def add_chart_slide(slide,
df,
chart_type,
title_size=None,
text_size=None,
ind=None,
cols=None,
data_labels=None,
name=None,
legend_pos=None,
font=None,
datalabel_size=None,
hide_zeros=None):
"""
Add a chart to a slide using a DataFrame
TODO : create a function to create the ChartData from DF
TODO : create a function to insert chart in placeholder or frame
TODO : create a function to apply style to created chart
"""
if datalabel_size is None:
datalabel_size = text_size
if chart_type == CHART.XY_SCATTER:
data_labels = False
# Select dataframe slice
if ind is None:
ind = (None, None)
if cols is None:
cols = (None, None)
df = df.iloc[ind[0]:ind[1], cols[0]:cols[1]]
if chart_type == CHART.XY_SCATTER:
# Create ChartData
chart_data = XyChartData()
df = df.fillna(0)
for ind, row in df.iterrows():
s = chart_data.add_series(ind)
s.add_data_point(row[0], row[1])
else:
# Create ChartData
chart_data = ChartData()
chart_data.categories = list(df.columns)
# In order to not print 0 data_labels
if hide_zeros:
df = df.replace(0, np.nan)
df = df.fillna('')
for ind, row in df.iterrows():
chart_data.add_series(ind, row)
# Insert chart in slide or placeholder
try:
shp = slide.insert_chart(chart_type, chart_data)
except AttributeError:
x, y, cx, cy = Inches(2), Inches(2), Inches(6), Inches(4.5)
shp = slide.shapes.add_chart(chart_type, x, y, cx, cy, chart_data)
# Define chart style
if name:
title = shp.chart.chart_title.text_frame
title.text = name
for p in title.paragraphs:
if font:
p.font.name = font
if title_size:
p.font.size = Pt(title_size)
# Legend
shp.chart.has_legend = True
shp.chart.legend.include_in_layout = False
if legend_pos:
shp.chart.legend.position = legend_pos
# Text sizes
if text_size is not None:
text_size = Pt(text_size)
# Lower indexes
try:
shp.chart.value_axis.tick_labels.font.size = text_size
except ValueError:
()
for plot in shp.chart.plots:
# Legend
plot.chart.legend.font.size = text_size
# Indexes
if chart_type not in [CHART.DOUGHNUT, CHART.PIE]:
plot.chart.category_axis.tick_labels.font.size = text_size
try:
plot.has_data_labels = data_labels
if plot.has_data_labels:
# Numbers in the plot (on the bars)
plot.data_labels.font.size = Pt(datalabel_size)
except AttributeError:
logging.info('No datalabels available for {} chart'.format(
chart_type._member_name))
return shp
from pptx import Presentation
from pptx.chart.data import XyChartData, BubbleChartData
from pptx.enum.chart import XL_CHART_TYPE
from pptx.util import Inches
# create presentation with 1 slide ------
prs = Presentation()
slide = prs.slides.add_slide(prs.slide_layouts[5])
# define chart data ---------------------
chart_data = XyChartData()
series_1 = chart_data.add_series('Model 1')
series_1.add_data_point(0.7, 2.7)
series_1.add_data_point(1.8, 3.2)
series_1.add_data_point(2.6, 0.8)
series_2 = chart_data.add_series('Model 2')
series_2.add_data_point(1.3, 3.7)
series_2.add_data_point(2.7, 2.3)
series_2.add_data_point(1.6, 1.8)
# add chart to slide --------------------
x, y, cx, cy = Inches(2), Inches(2), Inches(6), Inches(4.5)
chart = slide.shapes.add_chart(XL_CHART_TYPE.XY_SCATTER, x, y, cx, cy,
chart_data).chart
prs.save('BubbleChart.pptx')
