convert2table() - Code Metrics - Inspection of "version bumped: internal functions are changed" - ickc/pantable - Measure and Improve Code Quality continuously with Scrutinizer

Completed

Push — master ( 658df1...045d66 )

by Kolen

created 2016-12-05 05:00 UTC

convert2table() C

↳ Parent: Project

Complexity

Conditions

Size

Total Lines

Duplication

Lines	0
Ratio	0 %

Importance

Changes	17
Bugs	0	Features	0

Metric	Value
cc	9
c	17
b	0
f	0
dl	0
loc	55
rs	5.4159

How to fix Long Method

#!/usr/bin/env python3

r"""
Panflute filter to parse table in fenced YAML code blocks.
Currently only CSV table is supported.

7 metadata keys are recognized:

-   caption: the caption of the table. If omitted, no caption will be inserted.
-   alignment: a string of characters among L,R,C,D, case-insensitive,
        corresponds to Left-aligned, Right-aligned,
        Center-aligned, Default-aligned respectively.
    e.g. LCRD for a table with 4 columns
    default: DDD...
-   width: a list of relative width corresponding to the width of each columns.
    default: auto calculate from the length of each line in table cells.
-   table-width: the relative width of the table (e.g. relative to \linewidth).
    default: 1.0
-   header: If it has a header row. default: true
-   markdown: If CSV table cell contains markdown syntax. default: False
-   include: the path to an CSV file.
    If non-empty, override the CSV in the CodeBlock.
    default: None

When the metadata keys is invalid, the default will be used instead.
Note that width and table-width accept fractions as well.

e.g.

```table
---
caption: '*Awesome* **Markdown** Table'
alignment: RC
table-width: 2/3
markdown: True
---
First row,defaulted to be header row,can be disabled
1,cell can contain **markdown**,"It can be aribrary block element:

- following standard markdown syntax
- like this"
2,"Any markdown syntax, e.g.",$$E = mc^2$$
```
"""

import csv
from fractions import Fraction
import io
import panflute


# begin helper functions
def to_bool(to_be_bool, default=True):
    """
    Do nothing if to_be_bool is boolean,
    return `False` if it is "false" or "no" (case-insensitive),
    otherwise return default.
    """
    if not isinstance(to_be_bool, bool):
        try:
            if to_be_bool.lower() in ("false", "no"):
                to_be_bool = False
            elif to_be_bool.lower() in ("true", "yes"):
                to_be_bool = True
            else:
                raise ValueError
        except (ValueError, TypeError, AttributeError):
            to_be_bool = default
            panflute.debug("""pantable: invalid boolean. \
Should be true/false/yes/no, case-insensitive. Default is used.""")
    return to_be_bool


def get_width(options, number_of_columns):
    """
    get width: set to `None` when

    1. not given
    2. not a list
    3. length not equal to the number of columns
    4. negative entries
    """
    if 'width' not in options:
        width = None
    else:
        width = options['width']
        try:
            width = options['width']
            if len(width) != number_of_columns:
                raise ValueError
            width = [float(Fraction(x)) for x in options['width']]
            if not all(i >= 0 for i in width):
                raise ValueError
        except (ValueError, TypeError):
            width = None
            panflute.debug("pantable: invalid width")
    return width


def get_table_width(options):
    """
    `table-width` set to `1.0` if invalid
    """
    if 'table-width' not in options:
        table_width = 1.0
    else:
        try:
            table_width = float(Fraction(options.get('table-width')))
            if table_width <= 0:
                raise ValueError
        except (ValueError, TypeError):
            table_width = 1.0
            panflute.debug("pantable: invalid table-width")
    return table_width
# end helper functions


def auto_width(table_width, number_of_columns, raw_table_list):
    """
    `width` is auto-calculated if not given in YAML
    It also returns None when table is empty.
    """
    # calculate width
    # The +3 match the way pandoc handle width, see jgm/pandoc commit 0dfceda
    width_abs = [3 + max(
        [max(
            [len(line) for line in row[i].split("\n")]
        ) for row in raw_table_list]
    ) for i in range(number_of_columns)]
    try:
        width_tot = sum(width_abs)
        # when all are 3 means all are empty, see comment above
        if width_tot == 3 * number_of_columns:
            raise ValueError
        width = [
            width_abs[i] / width_tot * table_width
            for i in range(number_of_columns)
        ]
    except ValueError:
        panflute.debug("pantable: table is empty")
        width = None
    return width


def parse_alignment(alignment_string, number_of_columns):
    """
    `alignment` string is parsed into pandoc format (AlignDefault, etc.)
    """
    # initialize
    alignment_string = str(alignment_string)
    number_of_alignments = len(alignment_string)
    # truncate and debug if too long
    try:
        if number_of_alignments > number_of_columns:
            raise ValueError
    except ValueError:
        alignment_string = alignment_string[:number_of_columns]
        panflute.debug("pantable: alignment string is too long")
    # parsing
    alignment = [("AlignLeft" if i.lower() == "l"
                  else "AlignCenter" if i.lower() == "c"
                  else "AlignRight" if i.lower() == "r"
                  else "AlignDefault" if i.lower() == "d"
                  else None) for i in alignment_string]
    # debug if invalid; set to default
    try:
        if None in alignment:
            raise ValueError
    except ValueError:
        alignment = [(i if i is not None else "AlignDefault")
                     for i in alignment]
        panflute.debug(
            "pantable: alignment: invalid character found, default is used instead.")
    # fill up with default if too short
    if number_of_columns > number_of_alignments:
        alignment += ["AlignDefault" for __ in range(
            number_of_columns - len(alignment))]
    return alignment


def read_data(include, data):
    """
    read csv and return the table in list
    """
    if include is None:
        with io.StringIO(data) as file:
            raw_table_list = list(csv.reader(file))
    else:
        path2file = str(include)
        try:
            with open(path2file) as file:
                raw_table_list = list(csv.reader(file))
        except FileNotFoundError:
            raw_table_list = None
            panflute.debug("pantable: file not found from the path", path2file)
    return raw_table_list


def regularize_table_list(raw_table_list):
    """
    When the length of rows are uneven, make it as long as the longest row.
    """
    max_number_of_columns = max(
        [len(row) for row in raw_table_list]
    )
    for row in raw_table_list:
        missing_number_of_columns = max_number_of_columns - len(row)
        if missing_number_of_columns > 0:
            row += ['' for __ in range(missing_number_of_columns)]
    return


def parse_table_list(markdown, raw_table_list):
    """
    read table in list and return panflute table format
    """
    if markdown:
        table_body = [panflute.TableRow(*[
            panflute.TableCell(*panflute.convert_text(x))
            for x in row
        ]) for row in raw_table_list]
    else:
        table_body = [panflute.TableRow(*[
            panflute.TableCell(panflute.Plain(panflute.Str(x)))
            for x in row
        ]) for row in raw_table_list]
    return table_body


def convert2table(options, data, **__):
    """
    provided to panflute.yaml_filter to parse its content as pandoc table.
    """
    # prepare table in list from data/include
    raw_table_list = read_data(options.get('include', None), data)
    # delete element if table is empty (by returning [])
    # element unchanged if include is invalid (by returning None)
    try:
        if not raw_table_list or raw_table_list is None:
            raise ValueError
    except ValueError:
        panflute.debug("pantable: table is empty or include is invalid")
        return raw_table_list
    # regularize table: all rows should have same length
    regularize_table_list(raw_table_list)
    # preparation: get no of columns of the table
    number_of_columns = len(raw_table_list[0])

    # Initialize the `options` output from `panflute.yaml_filter`
    # parse width
    width = get_width(options, number_of_columns)
    # auto-width when width is not specified
    if width is None:
        width = auto_width(get_table_width(
            options), number_of_columns, raw_table_list)
    # delete element if table is empty (by returning [])
    # width remains None only when table is empty
    try:
        if width is None:
            raise ValueError
    except ValueError:
        panflute.debug("pantable: table is empty")
        return []
    # parse alignment
    alignment = parse_alignment(options.get(
        'alignment', None), number_of_columns)
    header = to_bool(options.get('header', True), True)
    markdown = to_bool(options.get('markdown', False), False)

    # get caption: parsed as markdown into panflute AST if non-empty.
    caption = panflute.convert_text(str(options['caption']))[
        0].content if 'caption' in options else None
    # parse list to panflute table
    table_body = parse_table_list(markdown, raw_table_list)
    # extract header row
    header_row = table_body.pop(0) if (
        len(table_body) > 1 and header
    ) else None
    return panflute.Table(
        *table_body,
        caption=caption,
        alignment=alignment,
        width=width,
        header=header_row
    )


def main(_=None):
    """
    Fenced code block with class table will be parsed using
    panflute.yaml_filter with the fuction convert2table above.
    """
    return panflute.run_filter(
        panflute.yaml_filter,
        tag='table',
        function=convert2table,
        strict_yaml=True
    )

if __name__ == '__main__':
    main()


1			#!/usr/bin/env python3
2
3			r"""
4			Panflute filter to parse table in fenced YAML code blocks.
5			Currently only CSV table is supported.
6
7			7 metadata keys are recognized:
8
9			- caption: the caption of the table. If omitted, no caption will be inserted.
10			- alignment: a string of characters among L,R,C,D, case-insensitive,
11			corresponds to Left-aligned, Right-aligned,
12			Center-aligned, Default-aligned respectively.
13			e.g. LCRD for a table with 4 columns
14			default: DDD...
15			- width: a list of relative width corresponding to the width of each columns.
16			default: auto calculate from the length of each line in table cells.
17			- table-width: the relative width of the table (e.g. relative to \linewidth).
18			default: 1.0
19			- header: If it has a header row. default: true
20			- markdown: If CSV table cell contains markdown syntax. default: False
21			- include: the path to an CSV file.
22			If non-empty, override the CSV in the CodeBlock.
23			default: None
24
25			When the metadata keys is invalid, the default will be used instead.
26			Note that width and table-width accept fractions as well.
27
28			e.g.
29
30			```table
31			---
32			caption: 'Awesome Markdown Table'
33			alignment: RC
34			table-width: 2/3
35			markdown: True
36			---
37			First row,defaulted to be header row,can be disabled
38			1,cell can contain markdown,"It can be aribrary block element:
39
40			- following standard markdown syntax
41			- like this"
42			2,"Any markdown syntax, e.g.",$$E = mc^2$$
43			```
44			"""
45
46			import csv
47			from fractions import Fraction
48			import io
49			import panflute
50
51
52			# begin helper functions
53			def to_bool(to_be_bool, default=True):
54			"""
55			Do nothing if to_be_bool is boolean,
56			return `False` if it is "false" or "no" (case-insensitive),
57			otherwise return default.
58			"""
59			if not isinstance(to_be_bool, bool):
60			try:
61			if to_be_bool.lower() in ("false", "no"):
62			to_be_bool = False
63			elif to_be_bool.lower() in ("true", "yes"):
64			to_be_bool = True
65			else:
66			raise ValueError
67			except (ValueError, TypeError, AttributeError):
68			to_be_bool = default
69			panflute.debug("""pantable: invalid boolean. \
70			Should be true/false/yes/no, case-insensitive. Default is used.""")
71			return to_be_bool
72
73
74			def get_width(options, number_of_columns):
75			"""
76			get width: set to `None` when
77
78			1. not given
79			2. not a list
80			3. length not equal to the number of columns
81			4. negative entries
82			"""
83			if 'width' not in options:
84			width = None
85			else:
86			width = options['width']
87			try:
88			width = options['width']
89			if len(width) != number_of_columns:
90			raise ValueError
91			width = [float(Fraction(x)) for x in options['width']]
92			if not all(i >= 0 for i in width):
93			raise ValueError
94			except (ValueError, TypeError):
95			width = None
96			panflute.debug("pantable: invalid width")
97			return width
98
99
100			def get_table_width(options):
101			"""
102			`table-width` set to `1.0` if invalid
103			"""
104			if 'table-width' not in options:
105			table_width = 1.0
106			else:
107			try:
108			table_width = float(Fraction(options.get('table-width')))
109			if table_width <= 0:
110			raise ValueError
111			except (ValueError, TypeError):
112			table_width = 1.0
113			panflute.debug("pantable: invalid table-width")
114			return table_width
115			# end helper functions
116
117
118			def auto_width(table_width, number_of_columns, raw_table_list):
119			"""
120			`width` is auto-calculated if not given in YAML
121			It also returns None when table is empty.
122			"""
123			# calculate width
124			# The +3 match the way pandoc handle width, see jgm/pandoc commit 0dfceda
125			width_abs = [3 + max(
126			[max(
127			[len(line) for line in row[i].split("\n")]
128			) for row in raw_table_list]
129			) for i in range(number_of_columns)]
130			try:
131			width_tot = sum(width_abs)
132			# when all are 3 means all are empty, see comment above
133			if width_tot == 3 * number_of_columns:
134			raise ValueError
135			width = [
136			width_abs[i] / width_tot * table_width
137			for i in range(number_of_columns)
138			]
139			except ValueError:
140			panflute.debug("pantable: table is empty")
141			width = None
142			return width
143
144
145			def parse_alignment(alignment_string, number_of_columns):
146			"""
147			`alignment` string is parsed into pandoc format (AlignDefault, etc.)
148			"""
149			# initialize
150			alignment_string = str(alignment_string)
151			number_of_alignments = len(alignment_string)
152			# truncate and debug if too long
153			try:
154			if number_of_alignments > number_of_columns:
155			raise ValueError
156			except ValueError:
157			alignment_string = alignment_string[:number_of_columns]
158			panflute.debug("pantable: alignment string is too long")
159			# parsing
160			alignment = [("AlignLeft" if i.lower() == "l"
161			else "AlignCenter" if i.lower() == "c"
162			else "AlignRight" if i.lower() == "r"
163			else "AlignDefault" if i.lower() == "d"
164			else None) for i in alignment_string]
165			# debug if invalid; set to default
166			try:
167			if None in alignment:
168			raise ValueError
169			except ValueError:
170			alignment = [(i if i is not None else "AlignDefault")
171			for i in alignment]
172			panflute.debug(
173			"pantable: alignment: invalid character found, default is used instead.")
174			# fill up with default if too short
175			if number_of_columns > number_of_alignments:
176			alignment += ["AlignDefault" for __ in range(
177			number_of_columns - len(alignment))]
178			return alignment
179
180
181			def read_data(include, data):
182			"""
183			read csv and return the table in list
184			"""
185			if include is None:
186			with io.StringIO(data) as file:
187			raw_table_list = list(csv.reader(file))
188			else:
189			path2file = str(include)
190			try:
191			with open(path2file) as file:
192			raw_table_list = list(csv.reader(file))
193			except FileNotFoundError:
194			raw_table_list = None
195			panflute.debug("pantable: file not found from the path", path2file)
196			return raw_table_list
197
198
199			def regularize_table_list(raw_table_list):
200			"""
201			When the length of rows are uneven, make it as long as the longest row.
202			"""
203			max_number_of_columns = max(
204			[len(row) for row in raw_table_list]
205			)
206			for row in raw_table_list:
207			missing_number_of_columns = max_number_of_columns - len(row)
208			if missing_number_of_columns > 0:
209			row += ['' for __ in range(missing_number_of_columns)]
210			return
211
212
213			def parse_table_list(markdown, raw_table_list):
214			"""
215			read table in list and return panflute table format
216			"""
217			if markdown:
218			table_body = [panflute.TableRow(*[
219			panflute.TableCell(*panflute.convert_text(x))
220			for x in row
221			]) for row in raw_table_list]
222			else:
223			table_body = [panflute.TableRow(*[
224			panflute.TableCell(panflute.Plain(panflute.Str(x)))
225			for x in row
226			]) for row in raw_table_list]
227			return table_body
228
229
230			def convert2table(options, data, **__):
231			"""
232			provided to panflute.yaml_filter to parse its content as pandoc table.
233			"""
234			# prepare table in list from data/include
235			raw_table_list = read_data(options.get('include', None), data)
236			# delete element if table is empty (by returning [])
237			# element unchanged if include is invalid (by returning None)
238			try:
239			if not raw_table_list or raw_table_list is None:
240			raise ValueError
241			except ValueError:
242			panflute.debug("pantable: table is empty or include is invalid")
243			return raw_table_list
244			# regularize table: all rows should have same length
245			regularize_table_list(raw_table_list)
246			# preparation: get no of columns of the table
247			number_of_columns = len(raw_table_list[0])
248
249			# Initialize the `options` output from `panflute.yaml_filter`
250			# parse width
251			width = get_width(options, number_of_columns)
252			# auto-width when width is not specified
253			if width is None:
254			width = auto_width(get_table_width(
255			options), number_of_columns, raw_table_list)
256			# delete element if table is empty (by returning [])
257			# width remains None only when table is empty
258			try:
259			if width is None:
260			raise ValueError
261			except ValueError:
262			panflute.debug("pantable: table is empty")
263			return []
264			# parse alignment
265			alignment = parse_alignment(options.get(
266			'alignment', None), number_of_columns)
267			header = to_bool(options.get('header', True), True)
268			markdown = to_bool(options.get('markdown', False), False)
269
270			# get caption: parsed as markdown into panflute AST if non-empty.
271			caption = panflute.convert_text(str(options['caption']))[
272			0].content if 'caption' in options else None
273			# parse list to panflute table
274			table_body = parse_table_list(markdown, raw_table_list)
275			# extract header row
276			header_row = table_body.pop(0) if (
277			len(table_body) > 1 and header
278			) else None
279			return panflute.Table(
280			*table_body,
281			caption=caption,
282			alignment=alignment,
283			width=width,
284			header=header_row
285			)
286
287
288			def main(_=None):
289			"""
290			Fenced code block with class table will be parsed using
291			panflute.yaml_filter with the fuction convert2table above.
292			"""
293			return panflute.run_filter(
294			panflute.yaml_filter,
295			tag='table',
296			function=convert2table,
297			strict_yaml=True
298			)
299
300			if __name__ == '__main__':
301			main()
302

ickc / pantable

Push — master ( 658df1...045d66 )

convert2table() C

Complexity

Size

Duplication

Importance

How to fix Long Method

Long Method

Duplication Side-by-Side

Filter issues like