def _append2latexdoc(doc, content):
if isinstance(content, list):
for item in content:
if item.get('title'):
doc.append(_get_section(**item))
_append2latexdoc(doc, item['content'])
else:
_append2latexdoc(doc, item)
else:
section = _get_last_section(doc)
if content.get('text'):
if isinstance(content['text'], dict):
section.append(open(content['text']['filename']).read())
else:
temp = ''
if isinstance(content['text'],list):
for subtext in content['text']:
temp += subtext
elif isinstance(content['text'],str):
temp = content['text']
else:
raise Exception(f'Did not understand format of text string: \n {content["text"]}')
section.append(temp)
if content.get('latex_code'):
if isinstance(content['latex_code'], dict):
section.append(NoEscape(open(content['latex_code']['filename']).read()))
else:
section.append(NoEscape(content['latex_code']))
if content.get('table'):
for table in content['table']:
section.append(NoEscape('\\begin{table}[H]')) # note require float latex package for H command
if table.get('filename'):
df = pd.read_excel(table['filename'], **table['kwargs'])
elif isinstance(table.get('dataframe'),pd.DataFrame):
df = table.get('dataframe')
section.append(NoEscape(df.to_latex(longtable=True,multicolumn_format='c')))
section.append(NoEscape('\\end{table}'))
if content.get('image'):
for image in content.get('image'):
section.append(NoEscape('\\begin{figure}[H]')) # note require float latex package for H command
Figure.add_image(section, image['filename'])
section.append(NoEscape('\\end{figure}'))
if content.get('subimage'):
figure = Figure(position='H')
for i, subimage in enumerate(content['subimage']):
subfigure = SubFigure(width=NoEscape(
r'{}\linewidth'.format(np.round(1. / subimage.get('nr_horizontal_subimages', 2), 2) - 0.01)))
subfigure.add_image(subimage['filename'])
if subimage.get('caption', False):
subfigure.add_caption(subimage['caption'])
if subimage.get('figure_caption', False) and i == 0:
figure.add_caption(subimage['figure_caption'])
figure.append(subfigure)
if (i + 1) % subimage.get('nr_horizontal_subimages', 2) == 0 and i != 0 or subimage.get(
'nr_horizontal_subimages', 2) == 1:
section.append(figure)
figure = Figure(arguments=NoEscape('\ContinuedFloat'), position='H')
section.append(figure)
if content.get('packages'):
[doc.packages.append(Package(package)) for package in content['packages']]
def de_image(s: lines.Image):
img_path = ImageTool.verify(s.link, config.cacheimg_dir)
if img_path is None:
c = Center()
c.append(NoEscape('Link or path not found: {}'.format(s.link)))
return c
if config.give_rele_path:
img_path = os.path.relpath(img_path, config.output_dir)
img_path = norm_path(img_path)
c = Center()
if isinstance(s.parent, env.Quote):
c.append(
NoEscape(
r"\includegraphics[width=0.8\textwidth]{{{img_path}}}".format(
img_path=img_path)))
else:
fig = Figure(position=config.fig_position)
fig.add_image(img_path, placement='')
if len(s.desc.strip()) > 0:
fig.add_caption(s.desc)
c.append(fig)
return c
def test_graphics():
f = Figure(data=None, position=None)
f.add_image(filename='', width=r'0.8\textwidth', placement=r'\centering')
f.add_caption(caption='')
repr(f)
# Subfigure
s = SubFigure(data=None, position=None, width=r'0.45\linewidth')
s.add_image(filename='', width='r\linewidth',
placement=None)
s.add_caption(caption='')
repr(s)
# Matplotlib
plot = Figure(data=None, position=None)
x = [0, 1, 2, 3, 4, 5, 6]
y = [15, 2, 7, 1, 5, 6, 9]
pyplot.plot(x, y)
plot.add_plot(width=r'0.8\textwidth', placement=r'\centering')
plot.add_caption(caption='I am a caption.')
repr(plot)
def test_graphics():
f = Figure(data=None, position=None)
f.add_image(filename='', width=r'0.8\textwidth', placement=r'\centering')
f.add_caption(caption='')
repr(f)
# Subfigure
s = SubFigure(data=None, position=None, width=r'0.45\linewidth')
s.add_image(filename='', width='r\linewidth', placement=None)
s.add_caption(caption='')
repr(s)
# Matplotlib
plot = Figure(data=None, position=None)
x = [0, 1, 2, 3, 4, 5, 6]
y = [15, 2, 7, 1, 5, 6, 9]
pyplot.plot(x, y)
plot.add_plot(width=r'0.8\textwidth', placement=r'\centering')
plot.add_caption(caption='I am a caption.')
repr(plot)
# StandAloneGraphic
stand_alone_graphic = StandAloneGraphic(
filename='', image_options=r"width=0.8\textwidth")
repr(stand_alone_graphic)
def test_graphics():
f = Figure(data=None, position=None)
f.add_image(filename="", width=r"0.8\textwidth", placement=r"\centering")
f.add_caption(caption="")
# Subfigure
s = SubFigure(data=None, position=None, width=r"0.45\linewidth", seperate_paragraph=False)
s.add_image(filename="", width="r\linewidth", placement=None)
s.add_caption(caption="")
# Matplotlib
plot = MatplotlibFigure(data=None, position=None)
x = [0, 1, 2, 3, 4, 5, 6]
y = [15, 2, 7, 1, 5, 6, 9]
pyplot.plot(x, y)
plot.add_plot(width=r"0.8\textwidth", placement=r"\centering")
plot.add_caption(caption="I am a caption.")
# Quantities
Quantity(quantity=1 * pq.kg)
Quantity(quantity=1 * pq.kg, format_cb=lambda x: str(int(x)))
def test_graphics():
f = Figure(data=None, position=None)
f.add_image(filename='', width=r'0.8\textwidth', placement=r'\centering')
f.add_caption(caption='')
# Subfigure
s = SubFigure(data=None,
position=None,
width=r'0.45\linewidth',
seperate_paragraph=False)
s.add_image(filename='', width='r\linewidth', placement=None)
s.add_caption(caption='')
# Matplotlib
plot = MatplotlibFigure(data=None, position=None)
x = [0, 1, 2, 3, 4, 5, 6]
y = [15, 2, 7, 1, 5, 6, 9]
pyplot.plot(x, y)
plot.add_plot(width=r'0.8\textwidth', placement=r'\centering')
plot.add_caption(caption='I am a caption.')
# Quantities
Quantity(quantity=1 * pq.kg)
Quantity(quantity=1 * pq.kg, format_cb=lambda x: str(int(x)))
def latex(self):
width = "auto"
height = "auto"
self._parse_attr_latex()
if 'width' in self.attrs:
width = self.attrs['width']
if 'height' in self.attrs:
width = self.attrs['height']
_figure = Figure(position = 'h')
_figure.add_image(self.uri, width = NoEscape(width), placement= NoEscape('\centering'))
_figure.add_caption(self.caption)
return _figure
def add_figure(self,file=None,caption=None,width='240px'):
""" 添加图形
:param str file: 图形文件路径
:param str caption: 图形标题
:param str width: 图形宽度
:return: 无返回值
"""
graph = Figure(position='h!')
graph.add_image(file, width=width)
if caption is not None:
graph.add_caption(caption)
self.doc.append(graph)
def add_figure(self, file=None, caption=None, width='240px'):
""" 添加图形
:param str file: 图形文件路径
:param str caption: 图形标题
:param str width: 图形宽度
:return: 无返回值
"""
graph = Figure(position='h!')
graph.add_image(file, width=width)
if caption is not None:
graph.add_caption(caption)
self.doc.append(graph)
def test_graphics():
f = Figure(data=None, position=None)
f.add_image(filename="", width=r"0.8\textwidth", placement=r"\centering")
f.add_caption(caption="")
# Subfigure
s = SubFigure(data=None, position=None, width=r"0.45\linewidth", separate_paragraph=False)
s.add_image(filename="", width="r\linewidth", placement=None)
s.add_caption(caption="")
# Matplotlib
plot = Figure(data=None, position=None)
x = [0, 1, 2, 3, 4, 5, 6]
y = [15, 2, 7, 1, 5, 6, 9]
pyplot.plot(x, y)
plot.add_plot(width=r"0.8\textwidth", placement=r"\centering")
plot.add_caption(caption="I am a caption.")
def write_word_cloud_single_character(
self,
doc,
dta_holder,
tpath,
number_of_wc,
head_of_file_name="wordcloud_for_single_character",
wc_context_selection="MFW",
words_in_word_cloud=12):
doc.packages.append(Package('float'))
if wc_context_selection == "PMI":
section = Section(
"Word Cloud for single characters (method: pointwise mutual information",
numbering=True)
section.append(
"These word clouds were constructed based on pointwise mutual information (PMI). PMI is a measure of how strongly each term is associated with each character mention. "
)
if wc_context_selection == "MFW":
section = Section(
"Word Cloud for single characters (method: most frequent contexts words",
numbering=True)
section.append(
"These word clouds were constructed based on most frequent words. They show the most frequent words that appear around character mention. "
)
network_parameters = dta_holder["network_parameters"]
weighted_degrees = network_parameters[5]
weighted_degrees_sorted = sorted(weighted_degrees,
key=operator.itemgetter(1),
reverse=True)
if wc_context_selection == "PMI":
character_names_with_highest_degree = list()
for weighted_degree in weighted_degrees_sorted[0:number_of_wc]:
character_names_with_highest_degree += [weighted_degree[0]]
for index, context_dic in enumerate(
dta_holder["single_character_context"]):
if dta_holder["single_character_context"][index][
"character_names"] in character_names_with_highest_degree:
if len(context_dic["tf_sorted_list"]
[0:words_in_word_cloud]) == 0:
text_string = "<<empty_word_cloud>>"
if len(context_dic["tf_sorted_list"]
[0:words_in_word_cloud]) > 0:
#print(context_dic)
text_string = str()
# for word_freq_tuple in relation["tf_sorted_list"][0:10]:
for word_freq_tuple in context_dic["tf_sorted_list"][
0:words_in_word_cloud]:
if word_freq_tuple[1] < 10:
for i in range(word_freq_tuple[1]):
text_string += "%s " % word_freq_tuple[0]
if word_freq_tuple[1] >= 10:
for i in range(9):
# print(i)
text_string += "%s " % word_freq_tuple[0]
# print(text_string)
word_cloud.generate_wordcloud_simple(
text=text_string,
ending_number=index,
temppath=tpath,
file_name_head=head_of_file_name)
#wc = word_cloud.generate_wordcloud_simple(text=str(text_string))
wordcloud_pic = Figure(position="H")
wordcloud_pic.add_image(
os.path.join(
tpath,
"wordcloud_for_single_character%s.png" % index),
width='240px') # , placement="center")
wordcloud_pic.add_caption(
'word cloud of "%s"' %
(dta_holder["single_character_context"][index]
["character_names"]))
section.append(wordcloud_pic)
doc.append(section)
if wc_context_selection == "MFW":
character_names_with_highest_degree = list()
for weighted_degree in weighted_degrees_sorted[0:number_of_wc]:
character_names_with_highest_degree += [weighted_degree[0]]
for index in range(0, len(dta_holder["single_character_context"])):
if dta_holder["single_character_context"][index][
"character_names"] in character_names_with_highest_degree:
if len(dta_holder["single_character_context"][index]
["tf_sorted_list"][0:words_in_word_cloud]) == 0:
text_string = "<<empty_word_cloud>>"
if len(dta_holder["single_character_context"][index]
["tf_sorted_list"][0:words_in_word_cloud]) > 0:
#print(dta_holder["single_character_context"][index]["tf_sorted_list"][0:12])
text_string = str()
# for word_freq_tuple in relation["tf_sorted_list"][0:10]:
for word_freq_tuple in dta_holder[
"single_character_context"][index][
"tf_sorted_list"][0:words_in_word_cloud]:
#print(word_freq_tuple)
# for i in range(word_freq_tuple[1]):
# text_string += "%s " % word_freq_tuple[0]
if word_freq_tuple[1] < 10:
for i in range(word_freq_tuple[1]):
text_string += "%s " % word_freq_tuple[0]
if word_freq_tuple[1] >= 10:
for i in range(9):
#print(i)
text_string += "%s " % word_freq_tuple[0]
#print(text_string)
word_cloud.generate_wordcloud_simple(
text=text_string,
ending_number=index,
temppath=tpath,
file_name_head=head_of_file_name)
# wc = word_cloud.generate_wordcloud_simple(text=str(text_string))
#######
wordcloud_pic = Figure(position="H")
wordcloud_pic.add_image(
os.path.join(
tpath,
"wordcloud_for_single_character%s.png" % index),
width='240px') # , placement="center")
wordcloud_pic.add_caption(
'word cloud of "%s"' %
(dta_holder["single_character_context"][index]
["character_names"]))
# subs.append(wordcloud_pic)
# section.append(subs)
section.append(wordcloud_pic)
doc.append(section)
def visualize_zeta(self,
doc,
zeta_edge_pair_results,
path,
top_n_results=10):
# def visualize_zeta(self, edge_pair_result, top_n_results=10):
section = Section("Zeta Scores for Pairs with highest edge weights")
section.append(
"Zeta score is a stylometry measure that measures preferred and avoided terms in the context of character pairs."
)
for index, edge_pair_result in enumerate(zeta_edge_pair_results):
#features().visualize_zeta(zeta_results[index], name_for_figure="zeta_pair_%s" %index, path=dirpath)
#subsection = Subsection("Zeta Scores for Pairs with highest edge weights", numbering=False)
subsection = Subsection("Edge Pair: %s -- %s" %
(edge_pair_result["name_target"],
edge_pair_result["name_comparison"]),
numbering=False)
#Target (Character A)
# the following index [::-1] inverts the list for the figure
objects = [
el[0] for el in edge_pair_result["zeta_scores_target_sorted"]
[0:top_n_results]
][::-1]
y_pos = np.arange(len(objects))
performance = [
el[1] for el in edge_pair_result["zeta_scores_target_sorted"]
[0:top_n_results]
][::-1]
plt.barh(y_pos, performance, align='center', alpha=0.5)
plt.yticks(y_pos, objects)
plt.xlabel('Zeta Score')
plt.title('%s-context' % edge_pair_result["name_target"])
# plt.show()
# plt.savefig("zeta.pdf", bbox_inches='tight')
#print(path)
plt.savefig("%s/zeta_pair_%s_a_target.pdf" % (path, index),
bbox_inches='tight')
target_pic = Figure(position="H")
target_pic.add_image(os.path.join(
path, "zeta_pair_%s_a_target.pdf" % index),
width='240px')
target_pic.add_caption(
"Prefered terms in context of %s (compared to %s)" %
(edge_pair_result["name_target"],
edge_pair_result["name_comparison"]))
#wordcloud_pic.add_caption('word cloud of "%s -- %s"' % (
subsection.append(target_pic)
################ Comparison (Character B)
objects = [
el[0]
for el in edge_pair_result["zeta_scores_comparison_sorted"]
[0:top_n_results]
][::-1]
y_pos = np.arange(len(objects))
performance = [
el[1]
for el in edge_pair_result["zeta_scores_comparison_sorted"]
[0:top_n_results]
][::-1]
plt.barh(y_pos, performance, align='center', alpha=0.5)
plt.yticks(y_pos, objects)
plt.xlabel('Zeta Score')
plt.title('%s-context' % edge_pair_result["name_comparison"])
# plt.show()
# plt.savefig("zeta.pdf", bbox_inches='tight')
plt.savefig("%s/zeta_pair_%s_b_comparison.pdf" % (path, index),
bbox_inches='tight')
comparison_pic = Figure(position="H")
comparison_pic.add_image(os.path.join(
path, "zeta_pair_%s_b_comparison.pdf" % index),
width='240px')
comparison_pic.add_caption(
"Prefered terms in context of %s (compared to %s)" %
(edge_pair_result["name_comparison"],
edge_pair_result["name_target"]))
subsection.append(comparison_pic)
#subsection.append(subsubsection)
section.append(subsection)
doc.append(section)
def write_word_cloud(self,
doc,
dta_holder,
tpath,
number_of_wc,
head_of_file_name="wordcloud",
wc_context_selection="MFW",
words_in_word_cloud=12):
# import float package for -> Figure(position="H")
doc.packages.append(Package('float'))
#section = Section("Word Cloud for character pairs", numbering=True)
if wc_context_selection == "PMI":
section = Section(
"Word Cloud for character pairs (method: pointwise mutual information",
numbering=True)
section.append(
"These word clouds were constructed based on pointwise mutual information (PMI). PMI is a measure of how strongly each term is associated with the character pair. "
)
if wc_context_selection == "MFW":
section = Section(
"Word Cloud for character pairs (method: most frequent contexts words",
numbering=True)
section.append(
"These word clouds were constructed based on most frequent words. They show the most frequent words that appear in the context of the character pair. "
)
network_parameters = dta_holder["network_parameters"]
edge_weights = network_parameters[6]
edge_weights_sorted = sorted(edge_weights,
key=operator.itemgetter(2),
reverse=True)
### here we iterate over all context_terms in "character_relations_context"
### it is done with the index since a numbering for the word cloud file is necessary
# for relation in dta_holder["character_relations_context"]:
if wc_context_selection == "PMI":
#print(dta_holder["character_relations_context"])
PMI_all_pairs = features().PMI(
context_words=dta_holder["character_relations_context"])
#print(PMI_all_pairs[0:3])
# for i in PMI_all_pairs[0:3]:
# print(i)
for edge_pair_list in edge_weights_sorted[0:number_of_wc]:
for index, character_context_dic in enumerate(PMI_all_pairs):
#print(index)
if character_context_dic["character_names"][
0] == edge_pair_list[0] and character_context_dic[
"character_names"][1] == edge_pair_list[1]:
#print(index, character_context_dic["character_names"], character_context_dic["PMI"])
if len(character_context_dic["PMI"]
[0:words_in_word_cloud]) == 0:
text_string = "<<empty_word_cloud>>"
if len(character_context_dic["PMI"]
[0:words_in_word_cloud]) > 0:
text_string = str()
# for word_freq_tuple in relation["tf_sorted_list"][0:10]:
#print(index,character_context_dic["character_names"],character_context_dic["PMI"][0:12])
for word_freq_list in character_context_dic["PMI"][
0:words_in_word_cloud]:
if round(word_freq_list[1]) < 10:
if round(word_freq_list[1]) > 0:
for i in range(round(
word_freq_list[1])):
text_string += "%s " % word_freq_list[
0]
if round(word_freq_list[1]) <= 0:
text_string += "%s " % word_freq_list[0]
if round(word_freq_list[1]) >= 10:
for i in range(9):
text_string += "%s " % word_freq_list[0]
#print(text_string)
word_cloud.generate_wordcloud_simple(
text=text_string,
ending_number=index,
temppath=tpath,
file_name_head=head_of_file_name)
# wc = word_cloud.generate_wordcloud_simple(text=str(text_string))
#######
wordcloud_pic = Figure(position="H")
wordcloud_pic.add_image(
os.path.join(tpath, "wordcloud%s.png" % index),
width='240px') # , placement="center")
# wordcloud_pic.add_image(wc, width='240px')#, placement="center")
# wordcloud_pic.add_caption('word cloud of "%s -- %s"' % (relation["character_names"][0], relation["character_names"][1]))
wordcloud_pic.add_caption(
'word cloud of "%s -- %s"' %
(dta_holder["character_relations_context"][index]
["character_names"][0],
dta_holder["character_relations_context"][index]
["character_names"][1]))
# subs.append(wordcloud_pic)
# section.append(subs)
section.append(wordcloud_pic)
doc.append(section)
if wc_context_selection == "MFW":
#BETTER WITH ENUMERATE (CHANGE, IF HAVE SOME TIME)
# for edge_pair_list in edge_weights_sorted[0:number_of_wc]:
# for index, character_context_dic in enumerate(dta_holder["character_relations_context"]):
# #print(index)
# if character_context_dic["character_names"][0] == edge_pair_list[0] and character_context_dic["character_names"][1] == edge_pair_list[1]:
# print(index, character_context_dic["character_names"], character_context_dic["tf_sorted_list"])
#
for edge_pair_list in edge_weights_sorted[0:number_of_wc]:
for index in range(
len(dta_holder["character_relations_context"])):
if dta_holder["character_relations_context"][index][
"character_names"][0] == edge_pair_list[
0] and dta_holder[
"character_relations_context"][index][
"character_names"][
1] == edge_pair_list[1]:
#print(dta_holder["character_relations_context"][index]["tf_sorted_list"][0:12])
if len(dta_holder["character_relations_context"][index]
["tf_sorted_list"][0:words_in_word_cloud]) == 0:
text_string = "<<empty_word_cloud>>"
if len(dta_holder["character_relations_context"][index]
["tf_sorted_list"][0:words_in_word_cloud]) > 0:
text_string = str()
# for word_freq_tuple in relation["tf_sorted_list"][0:10]:
for word_freq_tuple in dta_holder[
"character_relations_context"][index][
"tf_sorted_list"][
0:words_in_word_cloud]:
if word_freq_tuple[1] < 10:
for i in range(word_freq_tuple[1]):
text_string += "%s " % word_freq_tuple[
0]
if word_freq_tuple[1] >= 10:
for i in range(9):
# print(i)
text_string += "%s " % word_freq_tuple[
0]
#print(text_string)
word_cloud.generate_wordcloud_simple(
text=text_string,
ending_number=index,
temppath=tpath,
file_name_head=head_of_file_name)
# wc = word_cloud.generate_wordcloud_simple(text=str(text_string))
#######
wordcloud_pic = Figure(position="H")
wordcloud_pic.add_image(
os.path.join(tpath, "wordcloud%s.png" % index),
width='240px') # , placement="center")
# wordcloud_pic.add_image(wc, width='240px')#, placement="center")
# wordcloud_pic.add_caption('word cloud of "%s -- %s"' % (relation["character_names"][0], relation["character_names"][1]))
wordcloud_pic.add_caption(
'word cloud of "%s -- %s"' %
(dta_holder["character_relations_context"][index]
["character_names"][0],
dta_holder["character_relations_context"][index]
["character_names"][1]))
# subs.append(wordcloud_pic)
# section.append(subs)
section.append(wordcloud_pic)
doc.append(section)
# MultiColumn/MultiRow.
t.add_row((MultiColumn(size=2, align='|c|', data='MultiColumn'),))
t.add_row((MultiRow(size=2, width='*', data='MultiRow'),))
# Command
c = Command(command='documentclass', arguments=None, options=None,
packages=None)
# Figure
f = Figure(data=None, position=None)
f.add_image(filename='', width=r'0.8\textwidth', placement=r'\centering')
f.add_caption(caption='')
# Subfigure
s = SubFigure(data=None, position=None,
width=r'0.45\linewidth', seperate_paragraph=False)
s.add_image(filename='', width='r\linewidth',
placement=None)
s.add_caption(caption='')
# Plt
plot = Plt(data=None, position=None)
x = [0, 1, 2, 3, 4, 5, 6]
y = [15, 2, 7, 1, 5, 6, 9]
t.add_multicolumn(size=2, align='|c|', content='Multicol', cells=None,
escape=False)
t.add_multirow(size=3, align='*', content='Multirow', hlines=True, cells=None,
escape=False)
# Command
c = Command('documentclass', arguments=None, options=None, packages=None)
# Figure
f = Figure(data=None, position=None)
f.add_image(filename='', width=r'0.8\textwidth', placement=r'\centering')
f.add_caption('')
# Numpy
v = VectorName(name='')
M = np.matrix([[2, 3, 4],
[0, 0, 1],
[0, 0, 2]])
m = Matrix(matrix=M, name='', mtype='p', alignment=None)
# Package
p = Package(name='', base='usepackage', options=None)
# PGFPlots
tikz = TikZ(data=None)
# MultiColumn/MultiRow.
t.add_row((MultiColumn(size=2, align='|c|', data='MultiColumn'), ))
t.add_row((MultiRow(size=2, width='*', data='MultiRow'), ))
# Command
c = Command(command='documentclass',
arguments=None,
options=None,
packages=None)
# Figure
f = Figure(data=None, position=None)
f.add_image(filename='', width=r'0.8\textwidth', placement=r'\centering')
f.add_caption(caption='')
# Subfigure
s = SubFigure(data=None,
position=None,
width=r'0.45\linewidth',
seperate_paragraph=False)
s.add_image(filename='', width='r\linewidth', placement=None)
s.add_caption(caption='')
# Plt
plot = Plt(data=None, position=None)
x = [0, 1, 2, 3, 4, 5, 6]
def plot_show(caption, width=r'0.5\textwidth', *args, **kwargs):
fig = Figure(position='htbp')
fig.add_plot(width=NoEscapeStr(width), *args, **kwargs)
fig.add_caption(NoEscapeStr(caption))
return fig
