File size: 18,623 Bytes
ae312d0
94d93d2
ae312d0
94d93d2
9c0dccd
07115d9
ae312d0
3e68ccf
 
e065b20
3e68ccf
c0b5a2b
3e68ccf
c6643c7
e065b20
 
 
 
 
 
 
 
 
 
3e68ccf
e55d16a
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
3e68ccf
9c0dccd
aa4f694
3e68ccf
 
c6643c7
 
 
9c0dccd
c6643c7
 
3e68ccf
 
 
 
 
 
 
c0b5a2b
 
 
94d93d2
c0b5a2b
3e68ccf
08aca02
3e68ccf
9c0dccd
 
 
 
3e68ccf
 
08aca02
 
3e68ccf
9c0dccd
e690364
ae312d0
 
c0b5a2b
008172e
e065b20
3e68ccf
 
 
 
 
 
e690364
 
 
 
3e68ccf
c6643c7
3e68ccf
008172e
3e68ccf
f04f0b2
 
008172e
 
 
f04f0b2
3e68ccf
f04f0b2
 
008172e
f04f0b2
 
008172e
f04f0b2
 
 
008172e
 
 
 
 
 
3e68ccf
 
 
 
 
 
 
94d93d2
3e68ccf
c6643c7
 
3e68ccf
07115d9
 
08aca02
 
07115d9
9c0dccd
 
 
07115d9
 
 
 
 
 
 
 
 
 
 
 
 
e065b20
008172e
e065b20
008172e
 
e065b20
 
 
 
008172e
e065b20
008172e
 
e065b20
 
008172e
 
 
e065b20
 
 
 
 
 
 
 
 
 
 
 
008172e
 
 
 
 
 
 
 
 
 
 
 
 
 
e065b20
 
f04f0b2
 
008172e
 
 
f04f0b2
 
 
 
 
 
008172e
 
f04f0b2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
008172e
f04f0b2
 
 
 
 
 
 
 
 
 
 
008172e
 
 
 
 
 
 
f04f0b2
 
 
 
 
 
 
 
008172e
 
 
 
 
 
 
 
 
 
 
 
 
 
f04f0b2
 
 
 
 
 
e065b20
008172e
e065b20
 
 
 
 
 
 
008172e
e065b20
 
 
f04f0b2
e065b20
 
 
 
 
2c68fcd
 
 
e065b20
 
 
 
 
2c68fcd
 
 
 
 
 
 
 
 
 
 
 
 
 
e065b20
2c68fcd
 
4396061
 
 
 
2c68fcd
e065b20
008172e
 
e065b20
 
 
 
 
 
 
 
 
 
 
 
 
 
59502a0
e065b20
 
 
 
 
4396061
 
 
 
 
 
 
 
 
 
 
 
 
 
 
e065b20
 
 
 
 
 
 
59502a0
e065b20
 
 
 
 
 
008172e
e065b20
 
 
 
 
 
 
008172e
e065b20
 
 
 
 
 
 
 
 
 
008172e
e065b20
 
 
 
 
 
 
 
 
008172e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
3e68ccf
e065b20
07115d9
 
 
 
 
 
 
 
 
 
e065b20
 
07115d9
 
 
 
 
 
 
 
 
 
 
 
 
e065b20
 
07115d9
 
 
 
 
 
 
 
 
 
 
 
 
 
e065b20
 
 
 
 
 
 
 
 
 
 
 
 
07115d9
f04f0b2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
07115d9
 
 
 
 
 
 
e065b20
 
07115d9
 
 
 
 
 
 
 
e065b20
 
07115d9
 
 
 
 
 
e065b20
 
07115d9
 
 
94d93d2
 
 
 
3e68ccf
e065b20
94d93d2
e065b20
3e68ccf
e065b20
1189403
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
import logging
import pathlib
import re
import tempfile

from typing import List, Tuple

import json5
import pptx
from pptx.enum.shapes import MSO_AUTO_SHAPE_TYPE

from global_config import GlobalConfig


# English Metric Unit (used by PowerPoint) to inches
EMU_TO_INCH_SCALING_FACTOR = 1.0 / 914400
INCHES_1_5 = pptx.util.Inches(1.5)
INCHES_1 = pptx.util.Inches(1)
INCHES_0_5 = pptx.util.Inches(0.5)
INCHES_0_4 = pptx.util.Inches(0.4)
INCHES_0_3 = pptx.util.Inches(0.3)

STEP_BY_STEP_PROCESS_MARKER = '>> '

PATTERN = re.compile(r"^slide[ ]+\d+:", re.IGNORECASE)
SAMPLE_JSON_FOR_PPTX = '''
{
    "title": "Understanding AI",
    "slides": [
        {
            "heading": "Introduction",
            "bullet_points": [
                "Brief overview of AI",
                [
                    "Importance of understanding AI"
                ]
            ]
        }
    ]
}
'''

logger = logging.getLogger(__name__)


def remove_slide_number_from_heading(header: str) -> str:
    """
    Remove the slide number from a given slide header.

    :param header: The header of a slide.
    """

    if PATTERN.match(header):
        idx = header.find(':')
        header = header[idx + 1:]

    return header


def generate_powerpoint_presentation(
        structured_data: str,
        slides_template: str,
        output_file_path: pathlib.Path
) -> List:
    """
    Create and save a PowerPoint presentation file containing the content in JSON format.

    :param structured_data: The presentation contents as "JSON" (may contain trailing commas).
    :param slides_template: The PPTX template to use.
    :param output_file_path: The path of the PPTX file to save as.
    :return A list of presentation title and slides headers.
    """

    # The structured "JSON" might contain trailing commas, so using json5
    parsed_data = json5.loads(structured_data)

    logger.debug(
        '*** Using PPTX template: %s',
        GlobalConfig.PPTX_TEMPLATE_FILES[slides_template]['file']
    )
    presentation = pptx.Presentation(GlobalConfig.PPTX_TEMPLATE_FILES[slides_template]['file'])
    slide_width_inch, slide_height_inch = _get_slide_width_height_inches(presentation)

    # The title slide
    title_slide_layout = presentation.slide_layouts[0]
    slide = presentation.slides.add_slide(title_slide_layout)
    title = slide.shapes.title
    subtitle = slide.placeholders[1]
    title.text = parsed_data['title']
    logger.info(
        'PPT title: %s | #slides: %d',
        title.text, len(parsed_data['slides'])
    )
    subtitle.text = 'by Myself and SlideDeck AI :)'
    all_headers = [title.text, ]

    # Add content in a loop
    for a_slide in parsed_data['slides']:
        is_processing_done = _handle_double_col_layout(
            presentation=presentation,
            slide_json=a_slide,
            slide_width_inch=slide_width_inch,
            slide_height_inch=slide_height_inch
        )

        if not is_processing_done:
            is_processing_done = _handle_step_by_step_process(
                presentation=presentation,
                slide_json=a_slide,
                slide_width_inch=slide_width_inch,
                slide_height_inch=slide_height_inch
            )

        if not is_processing_done:
            _handle_default_display(
                presentation=presentation,
                slide_json=a_slide,
                slide_width_inch=slide_width_inch,
                slide_height_inch=slide_height_inch
            )

    # The thank-you slide
    last_slide_layout = presentation.slide_layouts[0]
    slide = presentation.slides.add_slide(last_slide_layout)
    title = slide.shapes.title
    title.text = 'Thank you!'

    presentation.save(output_file_path)

    return all_headers


def get_flat_list_of_contents(items: list, level: int) -> List[Tuple]:
    """
    Flatten a (hierarchical) list of bullet points to a single list containing each item and
    its level.

    :param items: A bullet point (string or list).
    :param level: The current level of hierarchy.
    :return: A list of (bullet item text, hierarchical level) tuples.
    """

    flat_list = []

    for item in items:
        if isinstance(item, str):
            flat_list.append((item, level))
        elif isinstance(item, list):
            flat_list = flat_list + get_flat_list_of_contents(item, level + 1)

    return flat_list


def _handle_default_display(
        presentation: pptx.Presentation,
        slide_json: dict,
        slide_width_inch: float,
        slide_height_inch: float
):
    """
    Display a list of text in a slide.

    :param presentation: The presentation object.
    :param slide_json: The content of the slide as JSON data.
    :param slide_width_inch: The width of the slide in inches.
    :param slide_height_inch: The height of the slide in inches.
    """

    bullet_slide_layout = presentation.slide_layouts[1]
    slide = presentation.slides.add_slide(bullet_slide_layout)

    shapes = slide.shapes
    title_shape = shapes.title
    body_shape = shapes.placeholders[1]
    title_shape.text = remove_slide_number_from_heading(slide_json['heading'])
    text_frame = body_shape.text_frame

    # The bullet_points may contain a nested hierarchy of JSON arrays
    # In some scenarios, it may contain objects (dictionaries) because the LLM generated so
    #  ^ The second scenario is not covered

    flat_items_list = get_flat_list_of_contents(slide_json['bullet_points'], level=0)

    for idx, an_item in enumerate(flat_items_list):
        if idx == 0:
            text_frame.text = an_item[0].removeprefix(STEP_BY_STEP_PROCESS_MARKER)
        else:
            paragraph = text_frame.add_paragraph()
            paragraph.text = an_item[0].removeprefix(STEP_BY_STEP_PROCESS_MARKER)
            paragraph.level = an_item[1]

    _handle_key_message(
        the_slide=slide,
        slide_json=slide_json,
        slide_height_inch=slide_height_inch,
        slide_width_inch=slide_width_inch
    )


def _handle_double_col_layout(
        presentation: pptx.Presentation(),
        slide_json: dict,
        slide_width_inch: float,
        slide_height_inch: float
) -> bool:
    """
    Add a slide with a double column layout for comparison.

    :param presentation: The presentation object.
    :param slide_json: The content of the slide as JSON data.
    :param slide_width_inch: The width of the slide in inches.
    :param slide_height_inch: The height of the slide in inches.
    :return: True if double col layout has been added; False otherwise.
    """

    if 'bullet_points' in slide_json and slide_json['bullet_points']:
        double_col_content = slide_json['bullet_points']

        if double_col_content and (
                len(double_col_content) == 2
        ) and isinstance(double_col_content[0], dict) and isinstance(double_col_content[1], dict):
            slide = presentation.slide_layouts[4]
            slide = presentation.slides.add_slide(slide)

            shapes = slide.shapes
            title_placeholder = shapes.title
            title_placeholder.text = remove_slide_number_from_heading(slide_json['heading'])

            left_heading, right_heading = shapes.placeholders[1], shapes.placeholders[3]
            left_col, right_col = shapes.placeholders[2], shapes.placeholders[4]
            left_col_frame, right_col_frame = left_col.text_frame, right_col.text_frame

            if 'heading' in double_col_content[0]:
                left_heading.text = double_col_content[0]['heading']
            if 'bullet_points' in double_col_content[0]:
                flat_items_list = get_flat_list_of_contents(
                    double_col_content[0]['bullet_points'], level=0
                )

                for idx, an_item in enumerate(flat_items_list):
                    if idx == 0:
                        left_col_frame.text = an_item[0].removeprefix(STEP_BY_STEP_PROCESS_MARKER)
                    else:
                        paragraph = left_col_frame.add_paragraph()
                        paragraph.text = an_item[0].removeprefix(STEP_BY_STEP_PROCESS_MARKER)
                        paragraph.level = an_item[1]

            if 'heading' in double_col_content[1]:
                right_heading.text = double_col_content[1]['heading']
            if 'bullet_points' in double_col_content[1]:
                flat_items_list = get_flat_list_of_contents(
                    double_col_content[1]['bullet_points'], level=0
                )

                for idx, an_item in enumerate(flat_items_list):
                    if idx == 0:
                        right_col_frame.text = an_item[0].removeprefix(STEP_BY_STEP_PROCESS_MARKER)
                    else:
                        paragraph = right_col_frame.add_paragraph()
                        paragraph.text = an_item[0].removeprefix(STEP_BY_STEP_PROCESS_MARKER)
                        paragraph.level = an_item[1]

            _handle_key_message(
                the_slide=slide,
                slide_json=slide_json,
                slide_height_inch=slide_height_inch,
                slide_width_inch=slide_width_inch
            )

            return True

    return False


def _handle_step_by_step_process(
        presentation: pptx.Presentation,
        slide_json: dict,
        slide_width_inch: float,
        slide_height_inch: float
) -> bool:
    """
    Add shapes to display a step-by-step process in the slide, if available.

    :param presentation: The presentation object.
    :param slide_json: The content of the slide as JSON data.
    :param slide_width_inch: The width of the slide in inches.
    :param slide_height_inch: The height of the slide in inches.
    :return True if this slide has a step-by-step process depiction added; False otherwise.
    """

    if 'bullet_points' in slide_json and slide_json['bullet_points']:
        steps = slide_json['bullet_points']

        no_marker_count = 0.0
        n_steps = len(steps)

        # Ensure that it is a single list of strings without any sub-list
        for step in steps:
            if not isinstance(step, str):
                return False

            # In some cases, one or two steps may not begin with >>, e.g.:
            # {
            #     "heading": "Step-by-Step Process: Creating a Legacy",
            #     "bullet_points": [
            #         "Identify your unique talents and passions",
            #         ">> Develop your skills and knowledge",
            #         ">> Create meaningful work",
            #         ">> Share your work with the world",
            #         ">> Continuously learn and adapt"
            #     ],
            #     "key_message": ""
            # },
            #
            # Use a threshold, e.g., at most 20%
            if not step.startswith(STEP_BY_STEP_PROCESS_MARKER):
                no_marker_count += 1

        slide_header = slide_json['heading'].lower()
        if (no_marker_count / n_steps > 0.25) and not (
                ('step-by-step' in slide_header) or ('step by step' in slide_header)
        ):
            return False

        bullet_slide_layout = presentation.slide_layouts[1]
        slide = presentation.slides.add_slide(bullet_slide_layout)
        shapes = slide.shapes
        shapes.title.text = remove_slide_number_from_heading(slide_json['heading'])

        if 3 <= n_steps <= 4:
            # Horizontal display
            height = INCHES_1_5
            width = pptx.util.Inches(slide_width_inch / n_steps - 0.01)
            top = pptx.util.Inches(slide_height_inch / 2)
            left = pptx.util.Inches((slide_width_inch - width.inches * n_steps) / 2 + 0.05)

            for step in steps:
                shape = shapes.add_shape(MSO_AUTO_SHAPE_TYPE.CHEVRON, left, top, width, height)
                shape.text = step.removeprefix(STEP_BY_STEP_PROCESS_MARKER)
                left += width - INCHES_0_4
        elif 4 < n_steps <= 6:
            # Vertical display
            height = pptx.util.Inches(0.65)
            top = pptx.util.Inches(slide_height_inch / 4)
            left = INCHES_1  # slide_width_inch - width.inches)

            # Find the close to median width, based on the length of each text, to be set
            # for the shapes
            width = pptx.util.Inches(slide_width_inch * 2 / 3)
            lengths = [len(step) for step in steps]
            font_size_20pt = pptx.util.Pt(20)
            widths = sorted(
                [
                    min(
                        pptx.util.Inches(font_size_20pt.inches * a_len),
                        width
                    ) for a_len in lengths
                ]
            )
            width = widths[len(widths) // 2]

            for step in steps:
                shape = shapes.add_shape(MSO_AUTO_SHAPE_TYPE.PENTAGON, left, top, width, height)
                shape.text = step.removeprefix(STEP_BY_STEP_PROCESS_MARKER)
                top += height + INCHES_0_3
                left += INCHES_0_5
        else:
            # Two steps -- probably not a process
            # More than 5--6 steps -- would likely cause a visual clutter
            return False

    return True


def _handle_key_message(
        the_slide: pptx.slide.Slide,
        slide_json: dict,
        slide_width_inch: float,
        slide_height_inch: float
):
    """
    Add a shape to display the key message in the slide, if available.

    :param the_slide: The slide to be processed.
    :param slide_json: The content of the slide as JSON data.
    :param slide_width_inch: The width of the slide in inches.
    :param slide_height_inch: The height of the slide in inches.
    """

    if 'key_message' in slide_json and slide_json['key_message']:
        height = pptx.util.Inches(1.6)
        width = pptx.util.Inches(slide_width_inch / 2.3)
        top = pptx.util.Inches(slide_height_inch - height.inches - 0.1)
        left = pptx.util.Inches((slide_width_inch - width.inches) / 2)
        shape = the_slide.shapes.add_shape(
            MSO_AUTO_SHAPE_TYPE.ROUNDED_RECTANGLE,
            left=left,
            top=top,
            width=width,
            height=height
        )
        shape.text = slide_json['key_message']


def _get_slide_width_height_inches(presentation: pptx.Presentation) -> Tuple[float, float]:
    """
    Get the dimensions of a slide in inches.

    :param presentation: The presentation object.
    :return: The width and the height.
    """

    slide_width_inch = EMU_TO_INCH_SCALING_FACTOR * presentation.slide_width
    slide_height_inch = EMU_TO_INCH_SCALING_FACTOR * presentation.slide_height
    # logger.debug('Slide width: %f, height: %f', slide_width_inch, slide_height_inch)

    return slide_width_inch, slide_height_inch


if __name__ == '__main__':
    _JSON_DATA = '''
    {
    "title": "Understanding AI",
    "slides": [
        {
            "heading": "Introduction",
            "bullet_points": [
                "Brief overview of AI",
                [
                    "Importance of understanding AI"
                ]
            ],
            "key_message": ""
        },
        {
            "heading": "What is AI?",
            "bullet_points": [
                "Definition of AI",
                [
                    "Types of AI",
                    [
                        "Narrow or weak AI",
                        "General or strong AI"
                    ]
                ],
                "Differences between AI and machine learning"
            ],
            "key_message": ""
        },
        {
            "heading": "How AI Works",
            "bullet_points": [
                "Overview of AI algorithms",
                [
                    "Types of AI algorithms",
                    [
                        "Rule-based systems",
                        "Decision tree systems",
                        "Neural networks"
                    ]
                ],
                "How AI processes data"
            ],
            "key_message": ""
        },
        {
            "heading": "Building AI Models",
            "bullet_points": [
                ">> Collect data",
                ">> Select model or architecture to use",
                ">> Set appropriate parameters",
                ">> Train model with data",
                ">> Run inference",
            ],
            "key_message": ""
        },
        {
            "heading": "Pros and Cons: Deep Learning vs. Classical Machine Learning",
            "bullet_points": [
                {
                    "heading": "Classical Machine Learning",
                    "bullet_points": [
                        "Interpretability: Easy to understand the model",
                        "Faster Training: Quicker to train models",
                        "Scalability: Can handle large datasets"
                    ]
                },
                {
                    "heading": "Deep Learning",
                    "bullet_points": [
                        "Handling Complex Data: Can learn from raw data",
                        "Feature Extraction: Automatically learns features",
                        "Improved Accuracy: Achieves higher accuracy"
                    ]
                }
            ],
            "key_message": ""
        },
        {
            "heading": "Pros of AI",
            "bullet_points": [
                "Increased efficiency and productivity",
                "Improved accuracy and precision",
                "Enhanced decision-making capabilities",
                "Personalized experiences"
            ],
            "key_message": "AI can be used for many different purposes"
        },
        {
            "heading": "Cons of AI",
            "bullet_points": [
                "Job displacement and loss of employment",
                "Bias and discrimination",
                "Privacy and security concerns",
                "Dependence on technology"
            ],
            "key_message": ""
        },
        {
            "heading": "Future Prospects of AI",
            "bullet_points": [
                "Advancements in fields such as healthcare and finance",
                "Increased use"
            ],
            "key_message": ""
        }
    ]
}'''

    temp = tempfile.NamedTemporaryFile(delete=False, suffix='.pptx')
    path = pathlib.Path(temp.name)

    generate_powerpoint_presentation(
        json5.loads(_JSON_DATA),
        output_file_path=path,
        slides_template='Basic'
    )
    print(f'File path: {path}')

    temp.close()