1
|
|
|
# SPDX-License-Identifier: LGPL-3.0-only |
2
|
|
|
|
3
|
|
|
"""Functions to export documents and items.""" |
4
|
|
|
|
5
|
|
|
import datetime |
6
|
|
|
import os |
7
|
|
|
from collections import defaultdict |
8
|
|
|
from typing import Any, Dict |
9
|
|
|
|
10
|
|
|
import openpyxl |
11
|
|
|
import yaml |
12
|
|
|
|
13
|
|
|
from doorstop import common, settings |
14
|
|
|
from doorstop.common import DoorstopError |
15
|
|
|
from doorstop.core.types import iter_documents, iter_items |
16
|
|
|
|
17
|
|
|
LIST_SEP = "\n" # string separating list values when joined in a string |
18
|
|
|
|
19
|
|
|
XLSX_MAX_WIDTH = 65.0 # maximum width for a column |
20
|
|
|
XLSX_FILTER_PADDING = 3.5 # column padding to account for filter button |
21
|
|
|
|
22
|
|
|
log = common.logger(__name__) |
23
|
|
|
|
24
|
|
|
|
25
|
|
|
def export(obj, path, ext=None, **kwargs): |
26
|
|
|
"""Export an object to a given format. |
27
|
|
|
|
28
|
|
|
The function can be called in two ways: |
29
|
|
|
|
30
|
|
|
1. document or item-like object + output file path |
31
|
|
|
2. tree-like object + output directory path |
32
|
|
|
|
33
|
|
|
:param obj: (1) Item, list of Items, Document or (2) Tree |
34
|
|
|
:param path: (1) output file path or (2) output directory path |
35
|
|
|
:param ext: file extension to override output extension |
36
|
|
|
|
37
|
|
|
:raises: :class:`doorstop.common.DoorstopError` for unknown file formats |
38
|
|
|
|
39
|
|
|
:return: output location if files created, else None |
40
|
|
|
|
41
|
|
|
""" |
42
|
|
|
# Determine the output format |
43
|
|
|
ext = ext or os.path.splitext(path)[-1] or ".csv" |
44
|
|
|
check(ext) |
45
|
|
|
|
46
|
|
|
# Export documents |
47
|
|
|
count = 0 |
48
|
|
|
for obj2, path2 in iter_documents(obj, path, ext): |
49
|
|
|
count += 1 |
50
|
|
|
|
51
|
|
|
# Export content to the specified path |
52
|
|
|
common.create_dirname(path2) |
53
|
|
|
log.info("exporting to {}...".format(path2)) |
54
|
|
|
if ext in FORMAT_LINES: |
55
|
|
|
lines = export_lines(obj2, ext, **kwargs) |
56
|
|
|
common.write_lines(lines, path2, end=settings.WRITE_LINESEPERATOR) |
57
|
|
|
else: |
58
|
|
|
export_file(obj2, path2, ext, **kwargs) |
59
|
|
|
|
60
|
|
|
# Return the exported path |
61
|
|
|
if count: |
62
|
|
|
msg = "exported to {} file{}".format(count, "s" if count > 1 else "") |
63
|
|
|
log.info(msg) |
64
|
|
|
return path |
65
|
|
|
else: |
66
|
|
|
log.warning("nothing to export") |
67
|
|
|
return None |
68
|
|
|
|
69
|
|
|
|
70
|
|
|
def export_lines(obj, ext=".yml", **kwargs): |
71
|
|
|
"""Yield lines for an export in the specified format. |
72
|
|
|
|
73
|
|
|
:param obj: Item, list of Items, or Document to export |
74
|
|
|
:param ext: file extension to specify the output format |
75
|
|
|
|
76
|
|
|
:raises: :class:`doorstop.common.DoorstopError` for unknown file formats |
77
|
|
|
|
78
|
|
|
:return: lines generator |
79
|
|
|
|
80
|
|
|
""" |
81
|
|
|
gen = check(ext, get_lines_gen=True) |
82
|
|
|
log.debug("yielding {} as lines of {}...".format(obj, ext)) |
83
|
|
|
yield from gen(obj, **kwargs) |
84
|
|
|
|
85
|
|
|
|
86
|
|
|
def export_file(obj, path, ext=None, **kwargs): |
87
|
|
|
"""Create a file object for an export in the specified format. |
88
|
|
|
|
89
|
|
|
:param obj: Item, list of Items, or Document to export |
90
|
|
|
:param path: output file location with desired extension |
91
|
|
|
:param ext: file extension to override output path's extension |
92
|
|
|
|
93
|
|
|
:raises: :class:`doorstop.common.DoorstopError` for unknown file formats |
94
|
|
|
|
95
|
|
|
:return: path to created file |
96
|
|
|
|
97
|
|
|
""" |
98
|
|
|
ext = ext or os.path.splitext(path)[-1] |
99
|
|
|
func = check(ext, get_file_func=True) |
100
|
|
|
log.debug("converting %s to file format %s...", obj, ext) |
101
|
|
|
try: |
102
|
|
|
return func(obj, path, **kwargs) |
103
|
|
|
except IOError: |
104
|
|
|
msg = "unable to write to: {}".format(path) |
105
|
|
|
raise common.DoorstopFileError(msg) from None |
106
|
|
|
|
107
|
|
|
|
108
|
|
|
def _lines_yaml(obj, **_): |
109
|
|
|
"""Yield lines for a YAML export. |
110
|
|
|
|
111
|
|
|
:param obj: Item, list of Items, or Document to export |
112
|
|
|
|
113
|
|
|
:return: iterator of lines of text |
114
|
|
|
|
115
|
|
|
""" |
116
|
|
|
for item in iter_items(obj): |
117
|
|
|
data = {str(item.uid): item.data} |
118
|
|
|
text = yaml.dump(data, default_flow_style=False, allow_unicode=True) |
119
|
|
|
yield text |
120
|
|
|
|
121
|
|
|
|
122
|
|
|
def _tabulate(obj, sep=LIST_SEP, auto=False): |
123
|
|
|
"""Yield lines of header/data for tabular export. |
124
|
|
|
|
125
|
|
|
:param obj: Item, list of Items, or Document to export |
126
|
|
|
:param sep: string separating list values when joined in a string |
127
|
|
|
:param auto: include placeholders for new items on import |
128
|
|
|
|
129
|
|
|
:return: iterator of rows of data |
130
|
|
|
|
131
|
|
|
""" |
132
|
|
|
|
133
|
|
|
header = ["level", "text", "ref", "links"] |
134
|
|
|
|
135
|
|
|
# 'at_least_one_ref' detects if at least one of the items still have a deprecated 'ref' field. |
136
|
|
|
# If there is none, 'ref' header is excluded from the headers and is not exported. |
137
|
|
|
at_least_one_ref = False |
138
|
|
|
for item in iter_items(obj): |
139
|
|
|
data = item.data |
140
|
|
|
|
141
|
|
|
for value in sorted(data.keys()): |
142
|
|
|
if value not in header: |
143
|
|
|
header.append(value) |
144
|
|
|
|
145
|
|
|
ref_value = data.get("ref") |
146
|
|
|
if ref_value: |
147
|
|
|
at_least_one_ref = True |
148
|
|
|
|
149
|
|
|
try: |
150
|
|
|
reference_index = header.index("references") |
151
|
|
|
|
152
|
|
|
# Inserting 'references' header after the 'ref' header. |
153
|
|
|
header.insert(3, header.pop(reference_index)) |
154
|
|
|
|
155
|
|
|
if not at_least_one_ref: |
156
|
|
|
header.remove("ref") |
157
|
|
|
except ValueError: |
158
|
|
|
pass |
159
|
|
|
|
160
|
|
|
yield ["uid"] + header |
161
|
|
|
|
162
|
|
|
for item in iter_items(obj): |
163
|
|
|
data = item.data |
164
|
|
|
|
165
|
|
|
# Yield row |
166
|
|
|
row = [item.uid] |
167
|
|
|
for key in header: |
168
|
|
|
value = data.get(key) |
169
|
|
|
if key == "level": |
170
|
|
|
# some levels are floats for YAML presentation |
171
|
|
|
value = str(value) |
172
|
|
|
elif key == "links": |
173
|
|
|
# separate identifiers with a delimiter |
174
|
|
|
value = sep.join(uid.string for uid in item.links) |
175
|
|
|
elif key == "references": |
176
|
|
|
if value is None: |
177
|
|
|
value = "" |
178
|
|
|
else: |
179
|
|
|
ref_strings = [] |
180
|
|
|
for ref_item in value: |
181
|
|
|
ref_type = ref_item["type"] |
182
|
|
|
ref_path = ref_item["path"] |
183
|
|
|
|
184
|
|
|
ref_string = "type:{},path:{}".format(ref_type, ref_path) |
185
|
|
|
|
186
|
|
|
if "keyword" in ref_item: |
187
|
|
|
keyword = ref_item["keyword"] |
188
|
|
|
ref_string += ",keyword:{}".format(keyword) |
189
|
|
|
|
190
|
|
|
ref_strings.append(ref_string) |
191
|
|
|
value = "\n".join(ref_string for ref_string in ref_strings) |
192
|
|
|
elif isinstance(value, str) and key not in ("reviewed",): |
193
|
|
|
# remove sentence boundaries and line wrapping |
194
|
|
|
value = item.get(key) |
195
|
|
|
elif value is None: |
196
|
|
|
value = "" |
197
|
|
|
row.append(value) |
198
|
|
|
yield row |
199
|
|
|
|
200
|
|
|
# Yield placeholders for new items |
201
|
|
|
if auto: |
202
|
|
|
for _ in range(settings.PLACEHOLDER_COUNT): |
203
|
|
|
yield [settings.PLACEHOLDER] |
204
|
|
|
|
205
|
|
|
|
206
|
|
|
def _file_csv(obj, path, delimiter=",", auto=False): |
207
|
|
|
"""Create a CSV file at the given path. |
208
|
|
|
|
209
|
|
|
:param obj: Item, list of Items, or Document to export |
210
|
|
|
:param path: location to export CSV file |
211
|
|
|
:param delimiter: character to delimit fields |
212
|
|
|
:param auto: include placeholders for new items on import |
213
|
|
|
|
214
|
|
|
:return: path of created file |
215
|
|
|
|
216
|
|
|
""" |
217
|
|
|
return common.write_csv( |
218
|
|
|
_tabulate(obj, auto=auto), |
219
|
|
|
path, |
220
|
|
|
delimiter=delimiter, |
221
|
|
|
newline="", |
222
|
|
|
encoding="utf-8", |
223
|
|
|
) |
224
|
|
|
|
225
|
|
|
|
226
|
|
|
def _file_tsv(obj, path, auto=False): |
227
|
|
|
"""Create a TSV file at the given path. |
228
|
|
|
|
229
|
|
|
:param obj: Item, list of Items, or Document to export |
230
|
|
|
:param path: location to export TSV file |
231
|
|
|
:param auto: include placeholders for new items on import |
232
|
|
|
|
233
|
|
|
:return: path of created file |
234
|
|
|
|
235
|
|
|
""" |
236
|
|
|
return _file_csv(obj, path, delimiter="\t", auto=auto) |
237
|
|
|
|
238
|
|
|
|
239
|
|
|
def _file_xlsx(obj, path, auto=False): |
240
|
|
|
"""Create an XLSX file at the given path. |
241
|
|
|
|
242
|
|
|
:param obj: Item, list of Items, or Document to export |
243
|
|
|
:param path: location to export XLSX file |
244
|
|
|
:param auto: include placeholders for new items on import |
245
|
|
|
|
246
|
|
|
:return: path of created file |
247
|
|
|
|
248
|
|
|
""" |
249
|
|
|
workbook = _get_xlsx(obj, auto) |
250
|
|
|
workbook.save(path) |
251
|
|
|
|
252
|
|
|
return path |
253
|
|
|
|
254
|
|
|
|
255
|
|
|
def _get_xlsx(obj, auto): |
256
|
|
|
"""Create an XLSX workbook object. |
257
|
|
|
|
258
|
|
|
:param obj: Item, list of Items, or Document to export |
259
|
|
|
:param auto: include placeholders for new items on import |
260
|
|
|
|
261
|
|
|
:return: new workbook |
262
|
|
|
|
263
|
|
|
""" |
264
|
|
|
col_widths: Dict[Any, float] = defaultdict(float) |
265
|
|
|
col = "A" |
266
|
|
|
|
267
|
|
|
# Create a new workbook |
268
|
|
|
workbook = openpyxl.Workbook() |
269
|
|
|
worksheet = workbook.active |
270
|
|
|
|
271
|
|
|
# Populate cells |
272
|
|
|
for row, data in enumerate(_tabulate(obj, auto=auto), start=1): |
273
|
|
|
for col_idx, value in enumerate(data, start=1): |
274
|
|
|
cell = worksheet.cell(column=col_idx, row=row) |
275
|
|
|
|
276
|
|
|
# wrap text in every cell |
277
|
|
|
alignment = openpyxl.styles.Alignment( |
278
|
|
|
vertical="top", horizontal="left", wrap_text=True |
279
|
|
|
) |
280
|
|
|
cell.alignment = alignment |
281
|
|
|
# and bold header rows |
282
|
|
|
if row == 1: |
283
|
|
|
cell.font = openpyxl.styles.Font(bold=True) |
284
|
|
|
|
285
|
|
|
# convert incompatible Excel types: |
286
|
|
|
# http://pythonhosted.org/openpyxl/api.html#openpyxl.cell.Cell.value |
287
|
|
|
if isinstance(value, (int, float, datetime.datetime)): |
288
|
|
|
cell.value = value |
289
|
|
|
else: |
290
|
|
|
cell.value = str(value) |
291
|
|
|
|
292
|
|
|
# track cell width |
293
|
|
|
col_widths[col_idx] = max(col_widths[col_idx], _width(str(value))) |
294
|
|
|
|
295
|
|
|
# Add filter up to the last column |
296
|
|
|
col_letter = openpyxl.utils.get_column_letter(len(col_widths)) |
297
|
|
|
worksheet.auto_filter.ref = "A1:%s1" % col_letter |
298
|
|
|
|
299
|
|
|
# Set column width based on column contents |
300
|
|
|
for col in col_widths: |
301
|
|
|
if col_widths[col] > XLSX_MAX_WIDTH: |
302
|
|
|
width = XLSX_MAX_WIDTH |
303
|
|
|
else: |
304
|
|
|
width = col_widths[col] + XLSX_FILTER_PADDING |
305
|
|
|
col_letter = openpyxl.utils.get_column_letter(col) |
306
|
|
|
worksheet.column_dimensions[col_letter].width = width |
307
|
|
|
|
308
|
|
|
# Freeze top row |
309
|
|
|
worksheet.freeze_panes = worksheet.cell(row=2, column=1) |
310
|
|
|
|
311
|
|
|
return workbook |
312
|
|
|
|
313
|
|
|
|
314
|
|
|
def _width(text): |
315
|
|
|
"""Get the maximum length in a multiline string.""" |
316
|
|
|
if text: |
317
|
|
|
return max(len(line) for line in text.splitlines()) |
318
|
|
|
else: |
319
|
|
|
return 0 |
320
|
|
|
|
321
|
|
|
|
322
|
|
|
# Mapping from file extension to lines generator |
323
|
|
|
FORMAT_LINES = {".yml": _lines_yaml} |
324
|
|
|
# Mapping from file extension to file generator |
325
|
|
|
FORMAT_FILE = {".csv": _file_csv, ".tsv": _file_tsv, ".xlsx": _file_xlsx} |
326
|
|
|
# Union of format dictionaries |
327
|
|
|
FORMAT = dict(list(FORMAT_LINES.items()) + list(FORMAT_FILE.items())) # type: ignore |
328
|
|
|
|
329
|
|
|
|
330
|
|
|
def check(ext, get_lines_gen=False, get_file_func=False): |
331
|
|
|
"""Confirm an extension is supported for export. |
332
|
|
|
|
333
|
|
|
:param get_lines_func: return a lines generator if available |
334
|
|
|
:param get_file_func: return a file creator if available |
335
|
|
|
|
336
|
|
|
:raises: :class:`doorstop.common.DoorstopError` for unknown formats |
337
|
|
|
|
338
|
|
|
:return: function requested if available |
339
|
|
|
|
340
|
|
|
""" |
341
|
|
|
exts = ", ".join(ext for ext in FORMAT) |
342
|
|
|
lines_exts = ", ".join(ext for ext in FORMAT_LINES) |
343
|
|
|
file_exts = ", ".join(ext for ext in FORMAT_FILE) |
344
|
|
|
fmt = "unknown {{}} format: {} (options: {{}})".format(ext or None) |
345
|
|
|
|
346
|
|
|
if get_lines_gen: |
347
|
|
|
try: |
348
|
|
|
gen = FORMAT_LINES[ext] |
349
|
|
|
except KeyError: |
350
|
|
|
exc = DoorstopError(fmt.format("lines export", lines_exts)) |
351
|
|
|
raise exc from None |
352
|
|
|
else: |
353
|
|
|
log.debug("found lines generator for: {}".format(ext)) |
354
|
|
|
return gen |
355
|
|
|
|
356
|
|
|
if get_file_func: |
357
|
|
|
try: |
358
|
|
|
func = FORMAT_FILE[ext] |
359
|
|
|
except KeyError: |
360
|
|
|
exc = DoorstopError(fmt.format("file export", file_exts)) |
361
|
|
|
raise exc from None |
362
|
|
|
else: |
363
|
|
|
log.debug("found file creator for: {}".format(ext)) |
364
|
|
|
return func |
365
|
|
|
|
366
|
|
|
if ext not in FORMAT: |
367
|
|
|
exc = DoorstopError(fmt.format("export", exts)) |
368
|
|
|
raise exc |
369
|
|
|
|
370
|
|
|
return None |
371
|
|
|
|