-
Notifications
You must be signed in to change notification settings - Fork 14.8k
/
pdf.py
948 lines (789 loc) Β· 32.2 KB
/
pdf.py
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
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
import json
import logging
import os
import re
import tempfile
import time
from abc import ABC
from io import StringIO
from pathlib import Path
from typing import (
TYPE_CHECKING,
Any,
Dict,
Iterator,
List,
Mapping,
Optional,
Sequence,
Union,
)
from urllib.parse import urlparse
import requests
from langchain_core.documents import Document
from langchain_core.utils import get_from_dict_or_env
from langchain_community.document_loaders.base import BaseLoader
from langchain_community.document_loaders.blob_loaders import Blob
from langchain_community.document_loaders.dedoc import DedocBaseLoader
from langchain_community.document_loaders.parsers.pdf import (
AmazonTextractPDFParser,
DocumentIntelligenceParser,
PDFMinerParser,
PDFPlumberParser,
PyMuPDFParser,
PyPDFium2Parser,
PyPDFParser,
)
from langchain_community.document_loaders.unstructured import UnstructuredFileLoader
if TYPE_CHECKING:
from textractor.data.text_linearization_config import TextLinearizationConfig
logger = logging.getLogger(__file__)
class UnstructuredPDFLoader(UnstructuredFileLoader):
"""Load `PDF` files using `Unstructured`.
You can run the loader in one of two modes: "single" and "elements".
If you use "single" mode, the document will be returned as a single
langchain Document object. If you use "elements" mode, the unstructured
library will split the document into elements such as Title and NarrativeText.
You can pass in additional unstructured kwargs after mode to apply
different unstructured settings.
Examples
--------
from langchain_community.document_loaders import UnstructuredPDFLoader
loader = UnstructuredPDFLoader(
"example.pdf", mode="elements", strategy="fast",
)
docs = loader.load()
References
----------
https://unstructured-io.github.io/unstructured/bricks.html#partition-pdf
"""
def _get_elements(self) -> List:
from unstructured.partition.pdf import partition_pdf
return partition_pdf(filename=self.file_path, **self.unstructured_kwargs)
class BasePDFLoader(BaseLoader, ABC):
"""Base Loader class for `PDF` files.
If the file is a web path, it will download it to a temporary file, use it, then
clean up the temporary file after completion.
"""
def __init__(self, file_path: Union[str, Path], *, headers: Optional[Dict] = None):
"""Initialize with a file path.
Args:
file_path: Either a local, S3 or web path to a PDF file.
headers: Headers to use for GET request to download a file from a web path.
"""
self.file_path = str(file_path)
self.web_path = None
self.headers = headers
if "~" in self.file_path:
self.file_path = os.path.expanduser(self.file_path)
# If the file is a web path or S3, download it to a temporary file, and use that
if not os.path.isfile(self.file_path) and self._is_valid_url(self.file_path):
self.temp_dir = tempfile.TemporaryDirectory()
_, suffix = os.path.splitext(self.file_path)
if self._is_s3_presigned_url(self.file_path):
suffix = urlparse(self.file_path).path.split("/")[-1]
temp_pdf = os.path.join(self.temp_dir.name, f"tmp{suffix}")
self.web_path = self.file_path
if not self._is_s3_url(self.file_path):
r = requests.get(self.file_path, headers=self.headers)
if r.status_code != 200:
raise ValueError(
"Check the url of your file; returned status code %s"
% r.status_code
)
with open(temp_pdf, mode="wb") as f:
f.write(r.content)
self.file_path = str(temp_pdf)
elif not os.path.isfile(self.file_path):
raise ValueError("File path %s is not a valid file or url" % self.file_path)
def __del__(self) -> None:
if hasattr(self, "temp_dir"):
self.temp_dir.cleanup()
@staticmethod
def _is_valid_url(url: str) -> bool:
"""Check if the url is valid."""
parsed = urlparse(url)
return bool(parsed.netloc) and bool(parsed.scheme)
@staticmethod
def _is_s3_url(url: str) -> bool:
"""check if the url is S3"""
try:
result = urlparse(url)
if result.scheme == "s3" and result.netloc:
return True
return False
except ValueError:
return False
@staticmethod
def _is_s3_presigned_url(url: str) -> bool:
"""Check if the url is a presigned S3 url."""
try:
result = urlparse(url)
return bool(re.search(r"\.s3\.amazonaws\.com$", result.netloc))
except ValueError:
return False
@property
def source(self) -> str:
return self.web_path if self.web_path is not None else self.file_path
class OnlinePDFLoader(BasePDFLoader):
"""Load online `PDF`."""
def load(self) -> List[Document]:
"""Load documents."""
loader = UnstructuredPDFLoader(str(self.file_path))
return loader.load()
class PyPDFLoader(BasePDFLoader):
"""
PyPDFLoader document loader integration
Setup:
Install ``langchain-community``.
.. code-block:: bash
pip install -U langchain-community
Instantiate:
.. code-block:: python
from langchain_community.document_loaders import PyPDFLoader
loader = PyPDFLoader(
file_path = "./example_data/layout-parser-paper.pdf",
password = "my-pasword",
extract_images = True,
# headers = None
# extraction_mode = "plain",
# extraction_kwargs = None,
)
Lazy load:
.. code-block:: python
docs = []
docs_lazy = loader.lazy_load()
# async variant:
# docs_lazy = await loader.alazy_load()
for doc in docs_lazy:
docs.append(doc)
print(docs[0].page_content[:100])
print(docs[0].metadata)
.. code-block:: python
LayoutParser : A Uniο¬ed Toolkit for Deep
Learning Based Document Image Analysis
Zejiang Shen1( ), R
{'source': './example_data/layout-parser-paper.pdf', 'page': 0}
Async load:
.. code-block:: python
docs = await loader.aload()
print(docs[0].page_content[:100])
print(docs[0].metadata)
.. code-block:: python
LayoutParser : A Uniο¬ed Toolkit for Deep
Learning Based Document Image Analysis
Zejiang Shen1( ), R
{'source': './example_data/layout-parser-paper.pdf', 'page': 0}
""" # noqa: E501
def __init__(
self,
file_path: str,
password: Optional[Union[str, bytes]] = None,
headers: Optional[Dict] = None,
extract_images: bool = False,
*,
extraction_mode: str = "plain",
extraction_kwargs: Optional[Dict] = None,
) -> None:
"""Initialize with a file path."""
try:
import pypdf # noqa:F401
except ImportError:
raise ImportError(
"pypdf package not found, please install it with `pip install pypdf`"
)
super().__init__(file_path, headers=headers)
self.parser = PyPDFParser(
password=password,
extract_images=extract_images,
extraction_mode=extraction_mode,
extraction_kwargs=extraction_kwargs,
)
def lazy_load(
self,
) -> Iterator[Document]:
"""Lazy load given path as pages."""
if self.web_path:
blob = Blob.from_data(open(self.file_path, "rb").read(), path=self.web_path) # type: ignore[attr-defined]
else:
blob = Blob.from_path(self.file_path) # type: ignore[attr-defined]
yield from self.parser.parse(blob)
class PyPDFium2Loader(BasePDFLoader):
"""Load `PDF` using `pypdfium2` and chunks at character level."""
def __init__(
self,
file_path: str,
*,
headers: Optional[Dict] = None,
extract_images: bool = False,
):
"""Initialize with a file path."""
super().__init__(file_path, headers=headers)
self.parser = PyPDFium2Parser(extract_images=extract_images)
def lazy_load(
self,
) -> Iterator[Document]:
"""Lazy load given path as pages."""
if self.web_path:
blob = Blob.from_data(open(self.file_path, "rb").read(), path=self.web_path) # type: ignore[attr-defined]
else:
blob = Blob.from_path(self.file_path) # type: ignore[attr-defined]
yield from self.parser.parse(blob)
class PyPDFDirectoryLoader(BaseLoader):
"""Load a directory with `PDF` files using `pypdf` and chunks at character level.
Loader also stores page numbers in metadata.
"""
def __init__(
self,
path: Union[str, Path],
glob: str = "**/[!.]*.pdf",
silent_errors: bool = False,
load_hidden: bool = False,
recursive: bool = False,
extract_images: bool = False,
):
self.path = path
self.glob = glob
self.load_hidden = load_hidden
self.recursive = recursive
self.silent_errors = silent_errors
self.extract_images = extract_images
@staticmethod
def _is_visible(path: Path) -> bool:
return not any(part.startswith(".") for part in path.parts)
def load(self) -> List[Document]:
p = Path(self.path)
docs = []
items = p.rglob(self.glob) if self.recursive else p.glob(self.glob)
for i in items:
if i.is_file():
if self._is_visible(i.relative_to(p)) or self.load_hidden:
try:
loader = PyPDFLoader(str(i), extract_images=self.extract_images)
sub_docs = loader.load()
for doc in sub_docs:
doc.metadata["source"] = str(i)
docs.extend(sub_docs)
except Exception as e:
if self.silent_errors:
logger.warning(e)
else:
raise e
return docs
class PDFMinerLoader(BasePDFLoader):
"""Load `PDF` files using `PDFMiner`."""
def __init__(
self,
file_path: str,
*,
headers: Optional[Dict] = None,
extract_images: bool = False,
concatenate_pages: bool = True,
) -> None:
"""Initialize with file path.
Args:
extract_images: Whether to extract images from PDF.
concatenate_pages: If True, concatenate all PDF pages into one a single
document. Otherwise, return one document per page.
"""
try:
from pdfminer.high_level import extract_text # noqa:F401
except ImportError:
raise ImportError(
"`pdfminer` package not found, please install it with "
"`pip install pdfminer.six`"
)
super().__init__(file_path, headers=headers)
self.parser = PDFMinerParser(
extract_images=extract_images, concatenate_pages=concatenate_pages
)
def lazy_load(
self,
) -> Iterator[Document]:
"""Lazily load documents."""
if self.web_path:
blob = Blob.from_data(open(self.file_path, "rb").read(), path=self.web_path) # type: ignore[attr-defined]
else:
blob = Blob.from_path(self.file_path) # type: ignore[attr-defined]
yield from self.parser.parse(blob)
class PDFMinerPDFasHTMLLoader(BasePDFLoader):
"""Load `PDF` files as HTML content using `PDFMiner`."""
def __init__(self, file_path: str, *, headers: Optional[Dict] = None):
"""Initialize with a file path."""
try:
from pdfminer.high_level import extract_text_to_fp # noqa:F401
except ImportError:
raise ImportError(
"`pdfminer` package not found, please install it with "
"`pip install pdfminer.six`"
)
super().__init__(file_path, headers=headers)
def lazy_load(self) -> Iterator[Document]:
"""Load file."""
from pdfminer.high_level import extract_text_to_fp
from pdfminer.layout import LAParams
from pdfminer.utils import open_filename
output_string = StringIO()
with open_filename(self.file_path, "rb") as fp:
extract_text_to_fp(
fp,
output_string,
codec="",
laparams=LAParams(),
output_type="html",
)
metadata = {
"source": self.file_path if self.web_path is None else self.web_path
}
yield Document(page_content=output_string.getvalue(), metadata=metadata)
class PyMuPDFLoader(BasePDFLoader):
"""Load `PDF` files using `PyMuPDF`."""
def __init__(
self,
file_path: str,
*,
headers: Optional[Dict] = None,
extract_images: bool = False,
**kwargs: Any,
) -> None:
"""Initialize with a file path."""
try:
import fitz # noqa:F401
except ImportError:
raise ImportError(
"`PyMuPDF` package not found, please install it with "
"`pip install pymupdf`"
)
super().__init__(file_path, headers=headers)
self.extract_images = extract_images
self.text_kwargs = kwargs
def _lazy_load(self, **kwargs: Any) -> Iterator[Document]:
if kwargs:
logger.warning(
f"Received runtime arguments {kwargs}. Passing runtime args to `load`"
f" is deprecated. Please pass arguments during initialization instead."
)
text_kwargs = {**self.text_kwargs, **kwargs}
parser = PyMuPDFParser(
text_kwargs=text_kwargs, extract_images=self.extract_images
)
if self.web_path:
blob = Blob.from_data(open(self.file_path, "rb").read(), path=self.web_path) # type: ignore[attr-defined]
else:
blob = Blob.from_path(self.file_path) # type: ignore[attr-defined]
yield from parser.lazy_parse(blob)
def load(self, **kwargs: Any) -> List[Document]:
return list(self._lazy_load(**kwargs))
def lazy_load(self) -> Iterator[Document]:
yield from self._lazy_load()
# MathpixPDFLoader implementation taken largely from Daniel Gross's:
# https://gist.github.com/danielgross/3ab4104e14faccc12b49200843adab21
class MathpixPDFLoader(BasePDFLoader):
"""Load `PDF` files using `Mathpix` service."""
def __init__(
self,
file_path: str,
processed_file_format: str = "md",
max_wait_time_seconds: int = 500,
should_clean_pdf: bool = False,
extra_request_data: Optional[Dict[str, Any]] = None,
**kwargs: Any,
) -> None:
"""Initialize with a file path.
Args:
file_path: a file for loading.
processed_file_format: a format of the processed file. Default is "md".
max_wait_time_seconds: a maximum time to wait for the response from
the server. Default is 500.
should_clean_pdf: a flag to clean the PDF file. Default is False.
extra_request_data: Additional request data.
**kwargs: additional keyword arguments.
"""
self.mathpix_api_key = get_from_dict_or_env(
kwargs, "mathpix_api_key", "MATHPIX_API_KEY"
)
self.mathpix_api_id = get_from_dict_or_env(
kwargs, "mathpix_api_id", "MATHPIX_API_ID"
)
# The base class isn't expecting these and doesn't collect **kwargs
kwargs.pop("mathpix_api_key", None)
kwargs.pop("mathpix_api_id", None)
super().__init__(file_path, **kwargs)
self.processed_file_format = processed_file_format
self.extra_request_data = (
extra_request_data if extra_request_data is not None else {}
)
self.max_wait_time_seconds = max_wait_time_seconds
self.should_clean_pdf = should_clean_pdf
@property
def _mathpix_headers(self) -> Dict[str, str]:
return {"app_id": self.mathpix_api_id, "app_key": self.mathpix_api_key}
@property
def url(self) -> str:
return "https://api.mathpix.com/v3/pdf"
@property
def data(self) -> dict:
options = {
"conversion_formats": {self.processed_file_format: True},
**self.extra_request_data,
}
return {"options_json": json.dumps(options)}
def send_pdf(self) -> str:
with open(self.file_path, "rb") as f:
files = {"file": f}
response = requests.post(
self.url, headers=self._mathpix_headers, files=files, data=self.data
)
response_data = response.json()
if "error" in response_data:
raise ValueError(f"Mathpix request failed: {response_data['error']}")
if "pdf_id" in response_data:
pdf_id = response_data["pdf_id"]
return pdf_id
else:
raise ValueError("Unable to send PDF to Mathpix.")
def wait_for_processing(self, pdf_id: str) -> None:
"""Wait for processing to complete.
Args:
pdf_id: a PDF id.
Returns: None
"""
url = self.url + "/" + pdf_id
for _ in range(0, self.max_wait_time_seconds, 5):
response = requests.get(url, headers=self._mathpix_headers)
response_data = response.json()
# This indicates an error with the request (e.g. auth problems)
error = response_data.get("error", None)
error_info = response_data.get("error_info", None)
if error is not None:
error_msg = f"Unable to retrieve PDF from Mathpix: {error}"
if error_info is not None:
error_msg += f" ({error_info['id']})"
raise ValueError(error_msg)
status = response_data.get("status", None)
if status == "completed":
return
elif status == "error":
# This indicates an error with the PDF processing
raise ValueError("Unable to retrieve PDF from Mathpix")
else:
print(f"Status: {status}, waiting for processing to complete") # noqa: T201
time.sleep(5)
raise TimeoutError
def get_processed_pdf(self, pdf_id: str) -> str:
self.wait_for_processing(pdf_id)
url = f"{self.url}/{pdf_id}.{self.processed_file_format}"
response = requests.get(url, headers=self._mathpix_headers)
return response.content.decode("utf-8")
def clean_pdf(self, contents: str) -> str:
"""Clean the PDF file.
Args:
contents: a PDF file contents.
Returns:
"""
contents = "\n".join(
[line for line in contents.split("\n") if not line.startswith("![]")]
)
# replace \section{Title} with # Title
contents = contents.replace("\\section{", "# ").replace("}", "")
# replace the "\" slash that Mathpix adds to escape $, %, (, etc.
contents = (
contents.replace(r"\$", "$")
.replace(r"\%", "%")
.replace(r"\(", "(")
.replace(r"\)", ")")
)
return contents
def load(self) -> List[Document]:
pdf_id = self.send_pdf()
contents = self.get_processed_pdf(pdf_id)
if self.should_clean_pdf:
contents = self.clean_pdf(contents)
metadata = {"source": self.source, "file_path": self.source, "pdf_id": pdf_id}
return [Document(page_content=contents, metadata=metadata)]
class PDFPlumberLoader(BasePDFLoader):
"""Load `PDF` files using `pdfplumber`."""
def __init__(
self,
file_path: str,
text_kwargs: Optional[Mapping[str, Any]] = None,
dedupe: bool = False,
headers: Optional[Dict] = None,
extract_images: bool = False,
) -> None:
"""Initialize with a file path."""
try:
import pdfplumber # noqa:F401
except ImportError:
raise ImportError(
"pdfplumber package not found, please install it with "
"`pip install pdfplumber`"
)
super().__init__(file_path, headers=headers)
self.text_kwargs = text_kwargs or {}
self.dedupe = dedupe
self.extract_images = extract_images
def load(self) -> List[Document]:
"""Load file."""
parser = PDFPlumberParser(
text_kwargs=self.text_kwargs,
dedupe=self.dedupe,
extract_images=self.extract_images,
)
if self.web_path:
blob = Blob.from_data(open(self.file_path, "rb").read(), path=self.web_path) # type: ignore[attr-defined]
else:
blob = Blob.from_path(self.file_path) # type: ignore[attr-defined]
return parser.parse(blob)
class AmazonTextractPDFLoader(BasePDFLoader):
"""Load `PDF` files from a local file system, HTTP or S3.
To authenticate, the AWS client uses the following methods to
automatically load credentials:
https://boto3.amazonaws.com/v1/documentation/api/latest/guide/credentials.html
If a specific credential profile should be used, you must pass
the name of the profile from the ~/.aws/credentials file that is to be used.
Make sure the credentials / roles used have the required policies to
access the Amazon Textract service.
Example:
.. code-block:: python
from langchain_community.document_loaders import AmazonTextractPDFLoader
loader = AmazonTextractPDFLoader(
file_path="s3://pdfs/myfile.pdf"
)
document = loader.load()
"""
def __init__(
self,
file_path: str,
textract_features: Optional[Sequence[str]] = None,
client: Optional[Any] = None,
credentials_profile_name: Optional[str] = None,
region_name: Optional[str] = None,
endpoint_url: Optional[str] = None,
headers: Optional[Dict] = None,
*,
linearization_config: Optional["TextLinearizationConfig"] = None,
) -> None:
"""Initialize the loader.
Args:
file_path: A file, url or s3 path for input file
textract_features: Features to be used for extraction, each feature
should be passed as a str that conforms to the enum
`Textract_Features`, see `amazon-textract-caller` pkg
client: boto3 textract client (Optional)
credentials_profile_name: AWS profile name, if not default (Optional)
region_name: AWS region, eg us-east-1 (Optional)
endpoint_url: endpoint url for the textract service (Optional)
linearization_config: Config to be used for linearization of the output
should be an instance of TextLinearizationConfig from
the `textractor` pkg
"""
super().__init__(file_path, headers=headers)
try:
import textractcaller as tc
except ImportError:
raise ImportError(
"Could not import amazon-textract-caller python package. "
"Please install it with `pip install amazon-textract-caller`."
)
if textract_features:
features = [tc.Textract_Features[x] for x in textract_features]
else:
features = []
if credentials_profile_name or region_name or endpoint_url:
try:
import boto3
if credentials_profile_name is not None:
session = boto3.Session(profile_name=credentials_profile_name)
else:
# use default credentials
session = boto3.Session()
client_params = {}
if region_name:
client_params["region_name"] = region_name
if endpoint_url:
client_params["endpoint_url"] = endpoint_url
client = session.client("textract", **client_params)
except ImportError:
raise ImportError(
"Could not import boto3 python package. "
"Please install it with `pip install boto3`."
)
except Exception as e:
raise ValueError(
"Could not load credentials to authenticate with AWS client. "
"Please check that credentials in the specified "
f"profile name are valid. {e}"
) from e
self.parser = AmazonTextractPDFParser(
textract_features=features,
client=client,
linearization_config=linearization_config,
)
def load(self) -> List[Document]:
"""Load given path as pages."""
return list(self.lazy_load())
def lazy_load(
self,
) -> Iterator[Document]:
"""Lazy load documents"""
# the self.file_path is local, but the blob has to include
# the S3 location if the file originated from S3 for multi-page documents
# raises ValueError when multi-page and not on S3"""
if self.web_path and self._is_s3_url(self.web_path):
blob = Blob(path=self.web_path) # type: ignore[call-arg] # type: ignore[misc]
else:
blob = Blob.from_path(self.file_path) # type: ignore[attr-defined]
if AmazonTextractPDFLoader._get_number_of_pages(blob) > 1:
raise ValueError(
f"the file {blob.path} is a multi-page document, \
but not stored on S3. \
Textract requires multi-page documents to be on S3."
)
yield from self.parser.parse(blob)
@staticmethod
def _get_number_of_pages(blob: Blob) -> int: # type: ignore[valid-type]
try:
import pypdf
from PIL import Image, ImageSequence
except ImportError:
raise ImportError(
"Could not import pypdf or Pilloe python package. "
"Please install it with `pip install pypdf Pillow`."
)
if blob.mimetype == "application/pdf": # type: ignore[attr-defined]
with blob.as_bytes_io() as input_pdf_file: # type: ignore[attr-defined]
pdf_reader = pypdf.PdfReader(input_pdf_file)
return len(pdf_reader.pages)
elif blob.mimetype == "image/tiff": # type: ignore[attr-defined]
num_pages = 0
img = Image.open(blob.as_bytes()) # type: ignore[attr-defined]
for _, _ in enumerate(ImageSequence.Iterator(img)):
num_pages += 1
return num_pages
elif blob.mimetype in ["image/png", "image/jpeg"]: # type: ignore[attr-defined]
return 1
else:
raise ValueError(f"unsupported mime type: {blob.mimetype}") # type: ignore[attr-defined]
class DedocPDFLoader(DedocBaseLoader):
"""
DedocPDFLoader document loader integration to load PDF files using `dedoc`.
The file loader can automatically detect the correctness of a textual layer in the
PDF document.
Note that `__init__` method supports parameters that differ from ones of
DedocBaseLoader.
Setup:
Install ``dedoc`` package.
.. code-block:: bash
pip install -U dedoc
Instantiate:
.. code-block:: python
from langchain_community.document_loaders import DedocPDFLoader
loader = DedocPDFLoader(
file_path="example.pdf",
# split=...,
# with_tables=...,
# pdf_with_text_layer=...,
# pages=...,
# ...
)
Load:
.. code-block:: python
docs = loader.load()
print(docs[0].page_content[:100])
print(docs[0].metadata)
.. code-block:: python
Some text
{
'file_name': 'example.pdf',
'file_type': 'application/pdf',
# ...
}
Lazy load:
.. code-block:: python
docs = []
docs_lazy = loader.lazy_load()
for doc in docs_lazy:
docs.append(doc)
print(docs[0].page_content[:100])
print(docs[0].metadata)
.. code-block:: python
Some text
{
'file_name': 'example.pdf',
'file_type': 'application/pdf',
# ...
}
Parameters used for document parsing via `dedoc`
(https://dedoc.readthedocs.io/en/latest/parameters/pdf_handling.html):
with_attachments: enable attached files extraction
recursion_deep_attachments: recursion level for attached files extraction,
works only when with_attachments==True
pdf_with_text_layer: type of handler for parsing, available options
["true", "false", "tabby", "auto", "auto_tabby" (default)]
language: language of the document for PDF without a textual layer,
available options ["eng", "rus", "rus+eng" (default)], the list of
languages can be extended, please see
https://dedoc.readthedocs.io/en/latest/tutorials/add_new_language.html
pages: page slice to define the reading range for parsing
is_one_column_document: detect number of columns for PDF without a textual
layer, available options ["true", "false", "auto" (default)]
document_orientation: fix document orientation (90, 180, 270 degrees) for PDF
without a textual layer, available options ["auto" (default), "no_change"]
need_header_footer_analysis: remove headers and footers from the output result
need_binarization: clean pages background (binarize) for PDF without a textual
layer
need_pdf_table_analysis: parse tables for PDF without a textual layer
"""
def _make_config(self) -> dict:
from dedoc.utils.langchain import make_manager_pdf_config
return make_manager_pdf_config(
file_path=self.file_path,
parsing_params=self.parsing_parameters,
split=self.split,
)
class DocumentIntelligenceLoader(BasePDFLoader):
"""Load a PDF with Azure Document Intelligence"""
def __init__(
self,
file_path: str,
client: Any,
model: str = "prebuilt-document",
headers: Optional[Dict] = None,
) -> None:
"""
Initialize the object for file processing with Azure Document Intelligence
(formerly Form Recognizer).
This constructor initializes a DocumentIntelligenceParser object to be used
for parsing files using the Azure Document Intelligence API. The load method
generates a Document node including metadata (source blob and page number)
for each page.
Parameters:
-----------
file_path : str
The path to the file that needs to be parsed.
client: Any
A DocumentAnalysisClient to perform the analysis of the blob
model : str
The model name or ID to be used for form recognition in Azure.
Examples:
---------
>>> obj = DocumentIntelligenceLoader(
... file_path="path/to/file",
... client=client,
... model="prebuilt-document"
... )
"""
self.parser = DocumentIntelligenceParser(client=client, model=model)
super().__init__(file_path, headers=headers)
def load(self) -> List[Document]:
"""Load given path as pages."""
return list(self.lazy_load())
def lazy_load(
self,
) -> Iterator[Document]:
"""Lazy load given path as pages."""
blob = Blob.from_path(self.file_path) # type: ignore[attr-defined]
yield from self.parser.parse(blob)
# Legacy: only for backwards compatibility. Use PyPDFLoader instead
PagedPDFSplitter = PyPDFLoader