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
172import pikepdf
from pdfminer.high_level import extract_text
import pandas as pd
import requests
from PyPDF2 import PdfReader
from io import BytesIO
import logging
import gc
from tqdm import tqdm
logging.basicConfig(level=logging.INFO, format='%(message)s')
# Initialize output CSV with headers
output_headers = [
'Link Type', 'Location Type', 'Title', 'Link', 'URL',
'PDF Size (bytes)', 'Page Count', 'Text-based', 'Has Title',
'Language Set', 'Tagged', 'Notes'
]
pd.DataFrame(columns=output_headers).to_csv('output.csv', index=False)
# Load input CSV
df = pd.read_csv('input.csv')
logging.info("Starting PDF accessibility analysis...")
results_batch = []
BATCH_SIZE = 100
for i, url in enumerate(tqdm(df['Link'], desc="Processing PDFs", unit="file")):
logging.info(f"\nProcessing: {url}")
link_type = str(df.iloc[i]['Link Type']).strip().lower()
if link_type == 'box':
row = df.iloc[i].to_dict()
row.update({
'PDF Size (bytes)': None,
'Page Count': None,
'Text-based': None,
'Has Title': None,
'Language Set': None,
'Tagged': None,
'Notes': 'Skipped: Box link'
})
filtered_row = {key: row.get(key, None) for key in output_headers}
results_batch.append(filtered_row)
if len(results_batch) >= BATCH_SIZE:
pd.DataFrame(results_batch).to_csv('output.csv', mode='a', header=False, index=False)
results_batch = []
gc.collect()
logging.info("→ Skipped: Box link")
continue
if not url.lower().endswith('.pdf'):
row = df.iloc[i].to_dict()
row.update({
'PDF Size (bytes)': None,
'Page Count': None,
'Text-based': None,
'Has Title': None,
'Language Set': None,
'Tagged': None,
'Notes': 'Skipped: Not a PDF link'
})
filtered_row = {key: row.get(key, None) for key in output_headers}
results_batch.append(filtered_row)
if len(results_batch) >= BATCH_SIZE:
pd.DataFrame(results_batch).to_csv('output.csv', mode='a', header=False, index=False)
results_batch = []
gc.collect()
continue
try:
response = requests.get(url, timeout=15)
response.raise_for_status()
if 'application/pdf' not in response.headers.get('Content-Type', ''):
raise ValueError("Not a PDF based on Content-Type")
except Exception as e:
logging.warning(f"→ Failed to download PDF: {e}")
row = df.iloc[i].to_dict()
row.update({
'PDF Size (bytes)': None,
'Page Count': None,
'Text-based': None,
'Has Title': None,
'Language Set': None,
'Tagged': None,
'Notes': f"Download failed: {e}"
})
filtered_row = {key: row.get(key, None) for key in output_headers}
results_batch.append(filtered_row)
if len(results_batch) >= BATCH_SIZE:
pd.DataFrame(results_batch).to_csv('output.csv', mode='a', header=False, index=False)
results_batch = []
gc.collect()
continue
# Default values
size = None
pages = None
is_text_based = None
has_title = None
lang = None
is_tagged = None
notes = []
# Size
size = len(response.content)
# Page Count
try:
reader = PdfReader(BytesIO(response.content))
pages = len(reader.pages
)
except Exception as e:
if "invalid float value" in str(e).lower():
logging.warning("→ PDF parsing issue: invalid float in color setting (non-fatal).")
else:
logging.warning(f"→ Failed to read page count: {e}")
notes.append("Failed to read page count")
pages = None
# Text-based check
try:
text = extract_text(BytesIO(response.content))
is_text_based = bool(text.strip())
except Exception as e:
logging.warning(f"→ Failed to extract text: {e}")
notes.append("Failed to extract text")
# Metadata
try:
with pikepdf.open(BytesIO(response.content)) as pdf:
docinfo = pdf.docinfo
has_title = bool(docinfo.get('/Title'))
root = getattr(pdf, "root", None)
if root and '/Lang' in root:
lang = root.get('/Lang', 'Not Set')
else:
lang = 'Unknown'
notes.append("Missing /Lang in outline root")
mark_info = root.get('/MarkInfo') if root else None
is_tagged = mark_info.get('/Marked') if mark_info and '/Marked' in mark_info else False
if not mark_info:
notes.append("Missing /MarkInfo in outline root")
except Exception as e:
logging.warning(f"→ Failed to extract metadata: {e}")
notes.append("Failed to extract metadata")
has_title = None
lang = 'Unknown'
is_tagged = None
row = df.iloc[i].to_dict()
row.update({
'PDF Size (bytes)': size,
'Page Count': pages,
'Text-based': is_text_based,
'Has Title': has_title,
'Language Set': lang,
'Tagged': is_tagged,
'Notes': "; ".join(notes)
})
# Filter row to only include output_headers keys in correct order
filtered_row = {key: row.get(key, None) for key in output_headers}
results_batch.append(filtered_row)
if len(results_batch) >= BATCH_SIZE:
pd.DataFrame(results_batch).to_csv('output.csv', mode='a', header=False, index=False)
results_batch = []
gc.collect()
if results_batch:
pd.DataFrame(results_batch).to_csv('output.csv', mode='a', header=False, index=False)
logging.info("\nAnalysis complete. Results saved to 'output.csv'.")