Skip to content

GitLab

Commit c9171d1f authored by zhougaofeng's avatar zhougaofeng
Browse files

Update app.py, count_pdfs.py, LICENSE.md, magic-pdf.template.json,... 

Update app.py, count_pdfs.py, LICENSE.md, magic-pdf.template.json, requirements.txt, requirements-docker.txt, requirements-qa.txt, update_version.py, setup.py, magic_pdf/__init__.py, magic_pdf/pdf_parse_by_txt.py, magic_pdf/pdf_parse_by_ocr.py, magic_pdf/user_api.py, magic_pdf/pdf_parse_union_core.py, magic_pdf/__pycache__/pdf_parse_by_ocr.cpython-310.pyc, magic_pdf/__pycache__/__init__.cpython-310.pyc, magic_pdf/__pycache__/pdf_parse_by_txt.cpython-310.pyc, magic_pdf/__pycache__/pdf_parse_union_core.cpython-310.pyc, magic_pdf/__pycache__/user_api.cpython-310.pyc, magic_pdf/dict2md/__init__.py, magic_pdf/dict2md/mkcontent.py, magic_pdf/dict2md/ocr_mkcontent.py, magic_pdf/dict2md/ocr_client.py, magic_pdf/dict2md/ocr_server.py, magic_pdf/dict2md/ocr_server_72.py, magic_pdf/dict2md/tmp.py, magic_pdf/dict2md/__pycache__/__init__.cpython-310.pyc, magic_pdf/dict2md/__pycache__/ocr_client.cpython-310.pyc, magic_pdf/dict2md/__pycache__/ocr_mkcontent.cpython-310.pyc, magic_pdf/filter/__init__.py, magic_pdf/filter/pdf_classify_by_type.py, magic_pdf/filter/pdf_meta_scan.py, magic_pdf/integrations/__init__.py, magic_pdf/integrations/rag/__init__.py, magic_pdf/integrations/rag/api.py, magic_pdf/integrations/rag/type.py, magic_pdf/integrations/rag/utils.py, magic_pdf/layout/__init__.py, magic_pdf/layout/bbox_sort.py, magic_pdf/layout/layout_det_utils.py, magic_pdf/layout/layout_sort.py, magic_pdf/layout/layout_spiler_recog.py, magic_pdf/layout/mcol_sort.py, magic_pdf/libs/__init__.py, magic_pdf/libs/boxbase.py, magic_pdf/libs/calc_span_stats.py, magic_pdf/libs/commons.py, magic_pdf/libs/config_reader.py, magic_pdf/libs/Constants.py, magic_pdf/libs/convert_utils.py, magic_pdf/libs/coordinate_transform.py, magic_pdf/libs/detect_language_from_model.py, magic_pdf/libs/draw_bbox.py, magic_pdf/libs/drop_reason.py, magic_pdf/libs/drop_tag.py, magic_pdf/libs/hash_utils.py, magic_pdf/libs/json_compressor.py, magic_pdf/libs/language.py, magic_pdf/libs/local_math.py, magic_pdf/libs/MakeContentConfig.py, magic_pdf/libs/markdown_utils.py, magic_pdf/libs/nlp_utils.py, magic_pdf/libs/ModelBlockTypeEnum.py, magic_pdf/libs/ocr_content_type.py, magic_pdf/libs/path_utils.py, magic_pdf/libs/pdf_check.py, magic_pdf/libs/pdf_image_tools.py, magic_pdf/libs/safe_filename.py, magic_pdf/libs/textbase.py, magic_pdf/libs/version.py, magic_pdf/libs/vis_utils.py, magic_pdf/model/__init__.py, magic_pdf/model/doc_analyze_by_custom_model.py, magic_pdf/model/magic_model.py, magic_pdf/model/pdf_extract_kit.py, magic_pdf/model/model_list.py, magic_pdf/model/pp_structure_v2.py, magic_pdf/model/ppTableModel.py, magic_pdf/model/pek_sub_modules/__init__.py, magic_pdf/model/pek_sub_modules/post_process.py, magic_pdf/model/pek_sub_modules/self_modify.py, magic_pdf/model/pek_sub_modules/layoutlmv3/__init__.py, magic_pdf/model/pek_sub_modules/layoutlmv3/backbone.py, magic_pdf/model/pek_sub_modules/layoutlmv3/beit.py, magic_pdf/model/pek_sub_modules/layoutlmv3/deit.py, magic_pdf/model/pek_sub_modules/layoutlmv3/model_init.py, magic_pdf/model/pek_sub_modules/layoutlmv3/rcnn_vl.py, magic_pdf/model/pek_sub_modules/layoutlmv3/visualizer.py, magic_pdf/model/pek_sub_modules/layoutlmv3/layoutlmft/__init__.py, magic_pdf/model/pek_sub_modules/layoutlmv3/layoutlmft/models/__init__.py, magic_pdf/model/pek_sub_modules/layoutlmv3/layoutlmft/models/layoutlmv3/__init__.py, magic_pdf/model/pek_sub_modules/layoutlmv3/layoutlmft/models/layoutlmv3/modeling_layoutlmv3.py, magic_pdf/model/pek_sub_modules/layoutlmv3/layoutlmft/models/layoutlmv3/tokenization_layoutlmv3_fast.py, magic_pdf/model/pek_sub_modules/layoutlmv3/layoutlmft/models/layoutlmv3/configuration_layoutlmv3.py, magic_pdf/model/pek_sub_modules/layoutlmv3/layoutlmft/models/layoutlmv3/tokenization_layoutlmv3.py, magic_pdf/model/pek_sub_modules/layoutlmv3/layoutlmft/data/cord.py, magic_pdf/model/pek_sub_modules/layoutlmv3/layoutlmft/data/__init__.py, magic_pdf/model/pek_sub_modules/layoutlmv3/layoutlmft/data/data_collator.py, magic_pdf/model/pek_sub_modules/layoutlmv3/layoutlmft/data/funsd.py, magic_pdf/model/pek_sub_modules/layoutlmv3/layoutlmft/data/image_utils.py, magic_pdf/model/pek_sub_modules/layoutlmv3/layoutlmft/data/xfund.py, magic_pdf/model/pek_sub_modules/structeqtable/__init__.py, magic_pdf/model/pek_sub_modules/structeqtable/StructTableModel.py, magic_pdf/para/__init__.py, magic_pdf/para/block_continuation_processor.py, magic_pdf/para/block_termination_processor.py, magic_pdf/para/commons.py, magic_pdf/para/denoise.py, magic_pdf/para/draw.py, magic_pdf/para/exceptions.py, magic_pdf/para/layout_match_processor.py, magic_pdf/para/para_pipeline.py, magic_pdf/para/para_split.py, magic_pdf/para/para_split_v2.py, magic_pdf/para/raw_processor.py, magic_pdf/para/stats.py, magic_pdf/para/title_processor.py, magic_pdf/parse/__init__.py, magic_pdf/parse/common_parse.py, magic_pdf/parse/excel_parse.py, magic_pdf/parse/pdf_client.py, magic_pdf/pipe/__init__.py, magic_pdf/pipe/AbsPipe.py, magic_pdf/pipe/OCRPipe.py, magic_pdf/pipe/TXTPipe.py, magic_pdf/pipe/UNIPipe.py, magic_pdf/post_proc/__init__.py, magic_pdf/post_proc/pdf_post_filter.py, magic_pdf/post_proc/remove_footnote.py, magic_pdf/post_proc/detect_para.py, magic_pdf/pre_proc/__init__.py, magic_pdf/pre_proc/citationmarker_remove.py, magic_pdf/pre_proc/construct_page_dict.py, magic_pdf/pre_proc/cut_image.py, magic_pdf/pre_proc/detect_equation.py, magic_pdf/pre_proc/detect_footer_header_by_statistics.py, magic_pdf/pre_proc/detect_footer_by_model.py, magic_pdf/pre_proc/detect_footnote.py, magic_pdf/pre_proc/detect_header.py, magic_pdf/pre_proc/detect_images.py, magic_pdf/pre_proc/detect_page_number.py, magic_pdf/pre_proc/detect_tables.py, magic_pdf/pre_proc/equations_replace.py, magic_pdf/pre_proc/fix_image.py, magic_pdf/pre_proc/fix_table.py, magic_pdf/pre_proc/main_text_font.py, magic_pdf/pre_proc/ocr_detect_all_bboxes.py, magic_pdf/pre_proc/ocr_detect_layout.py, magic_pdf/pre_proc/ocr_dict_merge.py, magic_pdf/pre_proc/ocr_span_list_modify.py, magic_pdf/pre_proc/post_layout_split.py, magic_pdf/pre_proc/remove_bbox_overlap.py, magic_pdf/pre_proc/remove_colored_strip_bbox.py, magic_pdf/pre_proc/pdf_pre_filter.py, magic_pdf/pre_proc/remove_footer_header.py, magic_pdf/pre_proc/remove_rotate_bbox.py, magic_pdf/pre_proc/resolve_bbox_conflict.py, magic_pdf/pre_proc/solve_line_alien.py, magic_pdf/pre_proc/statistics.py, magic_pdf/resources/fasttext-langdetect/lid.176.ftz, magic_pdf/resources/model_config/model_configs.yaml, magic_pdf/resources/model_config/layoutlmv3/layoutlmv3_base_inference.yaml, magic_pdf/resources/model_config/UniMERNet/demo.yaml, magic_pdf/rw/__init__.py, magic_pdf/rw/AbsReaderWriter.py, magic_pdf/rw/DiskReaderWriter.py, magic_pdf/rw/S3ReaderWriter.py, magic_pdf/spark/__init__.py, magic_pdf/spark/spark_api.py, magic_pdf/tools/__init__.py, magic_pdf/tools/pdf_client.py, magic_pdf/tools/common.py, magic_pdf/tools/cli_dev.py, magic_pdf/tools/cli.py, magic_pdf/tools/pdf_server.py files
parent 748e3b56
Pipeline #1783 canceled with stages
Hide whitespace changes
Inline Side-by-side
Showing with 2482 additions and 0 deletions
LICENSE.md 0 → 100644
View file @ c9171d1f
GNU AFFERO GENERAL PUBLIC LICENSE
Version 3, 19 November 2007
Copyright (C) 2007 Free Software Foundation, Inc. <https://fsf.org/>
Everyone is permitted to copy and distribute verbatim copies
of this license document, but changing it is not allowed.
Preamble
The GNU Affero General Public License is a free, copyleft license for
software and other kinds of works, specifically designed to ensure
cooperation with the community in the case of network server software.
The licenses for most software and other practical works are designed
to take away your freedom to share and change the works. By contrast,
our General Public Licenses are intended to guarantee your freedom to
share and change all versions of a program--to make sure it remains free
software for all its users.
When we speak of free software, we are referring to freedom, not
price. Our General Public Licenses are designed to make sure that you
have the freedom to distribute copies of free software (and charge for
them if you wish), that you receive source code or can get it if you
want it, that you can change the software or use pieces of it in new
free programs, and that you know you can do these things.
Developers that use our General Public Licenses protect your rights
with two steps: (1) assert copyright on the software, and (2) offer
you this License which gives you legal permission to copy, distribute
and/or modify the software.
A secondary benefit of defending all users' freedom is that
improvements made in alternate versions of the program, if they
receive widespread use, become available for other developers to
incorporate. Many developers of free software are heartened and
encouraged by the resulting cooperation. However, in the case of
software used on network servers, this result may fail to come about.
The GNU General Public License permits making a modified version and
letting the public access it on a server without ever releasing its
source code to the public.
The GNU Affero General Public License is designed specifically to
ensure that, in such cases, the modified source code becomes available
to the community. It requires the operator of a network server to
provide the source code of the modified version running there to the
users of that server. Therefore, public use of a modified version, on
a publicly accessible server, gives the public access to the source
code of the modified version.
An older license, called the Affero General Public License and
published by Affero, was designed to accomplish similar goals. This is
a different license, not a version of the Affero GPL, but Affero has
released a new version of the Affero GPL which permits relicensing under
this license.
The precise terms and conditions for copying, distribution and
modification follow.
TERMS AND CONDITIONS
0. Definitions.
"This License" refers to version 3 of the GNU Affero General Public License.
"Copyright" also means copyright-like laws that apply to other kinds of
works, such as semiconductor masks.
"The Program" refers to any copyrightable work licensed under this
License. Each licensee is addressed as "you". "Licensees" and
"recipients" may be individuals or organizations.
To "modify" a work means to copy from or adapt all or part of the work
in a fashion requiring copyright permission, other than the making of an
exact copy. The resulting work is called a "modified version" of the
earlier work or a work "based on" the earlier work.
A "covered work" means either the unmodified Program or a work based
on the Program.
To "propagate" a work means to do anything with it that, without
permission, would make you directly or secondarily liable for
infringement under applicable copyright law, except executing it on a
computer or modifying a private copy. Propagation includes copying,
distribution (with or without modification), making available to the
public, and in some countries other activities as well.
To "convey" a work means any kind of propagation that enables other
parties to make or receive copies. Mere interaction with a user through
a computer network, with no transfer of a copy, is not conveying.
An interactive user interface displays "Appropriate Legal Notices"
to the extent that it includes a convenient and prominently visible
feature that (1) displays an appropriate copyright notice, and (2)
tells the user that there is no warranty for the work (except to the
extent that warranties are provided), that licensees may convey the
work under this License, and how to view a copy of this License. If
the interface presents a list of user commands or options, such as a
menu, a prominent item in the list meets this criterion.
1. Source Code.
The "source code" for a work means the preferred form of the work
for making modifications to it. "Object code" means any non-source
form of a work.
A "Standard Interface" means an interface that either is an official
standard defined by a recognized standards body, or, in the case of
interfaces specified for a particular programming language, one that
is widely used among developers working in that language.
The "System Libraries" of an executable work include anything, other
than the work as a whole, that (a) is included in the normal form of
packaging a Major Component, but which is not part of that Major
Component, and (b) serves only to enable use of the work with that
Major Component, or to implement a Standard Interface for which an
implementation is available to the public in source code form. A
"Major Component", in this context, means a major essential component
(kernel, window system, and so on) of the specific operating system
(if any) on which the executable work runs, or a compiler used to
produce the work, or an object code interpreter used to run it.
The "Corresponding Source" for a work in object code form means all
the source code needed to generate, install, and (for an executable
work) run the object code and to modify the work, including scripts to
control those activities. However, it does not include the work's
System Libraries, or general-purpose tools or generally available free
programs which are used unmodified in performing those activities but
which are not part of the work. For example, Corresponding Source
includes interface definition files associated with source files for
the work, and the source code for shared libraries and dynamically
linked subprograms that the work is specifically designed to require,
such as by intimate data communication or control flow between those
subprograms and other parts of the work.
The Corresponding Source need not include anything that users
can regenerate automatically from other parts of the Corresponding
Source.
The Corresponding Source for a work in source code form is that
same work.
2. Basic Permissions.
All rights granted under this License are granted for the term of
copyright on the Program, and are irrevocable provided the stated
conditions are met. This License explicitly affirms your unlimited
permission to run the unmodified Program. The output from running a
covered work is covered by this License only if the output, given its
content, constitutes a covered work. This License acknowledges your
rights of fair use or other equivalent, as provided by copyright law.
You may make, run and propagate covered works that you do not
convey, without conditions so long as your license otherwise remains
in force. You may convey covered works to others for the sole purpose
of having them make modifications exclusively for you, or provide you
with facilities for running those works, provided that you comply with
the terms of this License in conveying all material for which you do
not control copyright. Those thus making or running the covered works
for you must do so exclusively on your behalf, under your direction
and control, on terms that prohibit them from making any copies of
your copyrighted material outside their relationship with you.
Conveying under any other circumstances is permitted solely under
the conditions stated below. Sublicensing is not allowed; section 10
makes it unnecessary.
3. Protecting Users' Legal Rights From Anti-Circumvention Law.
No covered work shall be deemed part of an effective technological
measure under any applicable law fulfilling obligations under article
11 of the WIPO copyright treaty adopted on 20 December 1996, or
similar laws prohibiting or restricting circumvention of such
measures.
When you convey a covered work, you waive any legal power to forbid
circumvention of technological measures to the extent such circumvention
is effected by exercising rights under this License with respect to
the covered work, and you disclaim any intention to limit operation or
modification of the work as a means of enforcing, against the work's
users, your or third parties' legal rights to forbid circumvention of
technological measures.
4. Conveying Verbatim Copies.
You may convey verbatim copies of the Program's source code as you
receive it, in any medium, provided that you conspicuously and
appropriately publish on each copy an appropriate copyright notice;
keep intact all notices stating that this License and any
non-permissive terms added in accord with section 7 apply to the code;
keep intact all notices of the absence of any warranty; and give all
recipients a copy of this License along with the Program.
You may charge any price or no price for each copy that you convey,
and you may offer support or warranty protection for a fee.
5. Conveying Modified Source Versions.
You may convey a work based on the Program, or the modifications to
produce it from the Program, in the form of source code under the
terms of section 4, provided that you also meet all of these conditions:
a) The work must carry prominent notices stating that you modified
it, and giving a relevant date.
b) The work must carry prominent notices stating that it is
released under this License and any conditions added under section
7. This requirement modifies the requirement in section 4 to
"keep intact all notices".
c) You must license the entire work, as a whole, under this
License to anyone who comes into possession of a copy. This
License will therefore apply, along with any applicable section 7
additional terms, to the whole of the work, and all its parts,
regardless of how they are packaged. This License gives no
permission to license the work in any other way, but it does not
invalidate such permission if you have separately received it.
d) If the work has interactive user interfaces, each must display
Appropriate Legal Notices; however, if the Program has interactive
interfaces that do not display Appropriate Legal Notices, your
work need not make them do so.
A compilation of a covered work with other separate and independent
works, which are not by their nature extensions of the covered work,
and which are not combined with it such as to form a larger program,
in or on a volume of a storage or distribution medium, is called an
"aggregate" if the compilation and its resulting copyright are not
used to limit the access or legal rights of the compilation's users
beyond what the individual works permit. Inclusion of a covered work
in an aggregate does not cause this License to apply to the other
parts of the aggregate.
6. Conveying Non-Source Forms.
You may convey a covered work in object code form under the terms
of sections 4 and 5, provided that you also convey the
machine-readable Corresponding Source under the terms of this License,
in one of these ways:
a) Convey the object code in, or embodied in, a physical product
(including a physical distribution medium), accompanied by the
Corresponding Source fixed on a durable physical medium
customarily used for software interchange.
b) Convey the object code in, or embodied in, a physical product
(including a physical distribution medium), accompanied by a
written offer, valid for at least three years and valid for as
long as you offer spare parts or customer support for that product
model, to give anyone who possesses the object code either (1) a
copy of the Corresponding Source for all the software in the
product that is covered by this License, on a durable physical
medium customarily used for software interchange, for a price no
more than your reasonable cost of physically performing this
conveying of source, or (2) access to copy the
Corresponding Source from a network server at no charge.
c) Convey individual copies of the object code with a copy of the
written offer to provide the Corresponding Source. This
alternative is allowed only occasionally and noncommercially, and
only if you received the object code with such an offer, in accord
with subsection 6b.
d) Convey the object code by offering access from a designated
place (gratis or for a charge), and offer equivalent access to the
Corresponding Source in the same way through the same place at no
further charge. You need not require recipients to copy the
Corresponding Source along with the object code. If the place to
copy the object code is a network server, the Corresponding Source
may be on a different server (operated by you or a third party)
that supports equivalent copying facilities, provided you maintain
clear directions next to the object code saying where to find the
Corresponding Source. Regardless of what server hosts the
Corresponding Source, you remain obligated to ensure that it is
available for as long as needed to satisfy these requirements.
e) Convey the object code using peer-to-peer transmission, provided
you inform other peers where the object code and Corresponding
Source of the work are being offered to the general public at no
charge under subsection 6d.
A separable portion of the object code, whose source code is excluded
from the Corresponding Source as a System Library, need not be
included in conveying the object code work.
A "User Product" is either (1) a "consumer product", which means any
tangible personal property which is normally used for personal, family,
or household purposes, or (2) anything designed or sold for incorporation
into a dwelling. In determining whether a product is a consumer product,
doubtful cases shall be resolved in favor of coverage. For a particular
product received by a particular user, "normally used" refers to a
typical or common use of that class of product, regardless of the status
of the particular user or of the way in which the particular user
actually uses, or expects or is expected to use, the product. A product
is a consumer product regardless of whether the product has substantial
commercial, industrial or non-consumer uses, unless such uses represent
the only significant mode of use of the product.
"Installation Information" for a User Product means any methods,
procedures, authorization keys, or other information required to install
and execute modified versions of a covered work in that User Product from
a modified version of its Corresponding Source. The information must
suffice to ensure that the continued functioning of the modified object
code is in no case prevented or interfered with solely because
modification has been made.
If you convey an object code work under this section in, or with, or
specifically for use in, a User Product, and the conveying occurs as
part of a transaction in which the right of possession and use of the
User Product is transferred to the recipient in perpetuity or for a
fixed term (regardless of how the transaction is characterized), the
Corresponding Source conveyed under this section must be accompanied
by the Installation Information. But this requirement does not apply
if neither you nor any third party retains the ability to install
modified object code on the User Product (for example, the work has
been installed in ROM).
The requirement to provide Installation Information does not include a
requirement to continue to provide support service, warranty, or updates
for a work that has been modified or installed by the recipient, or for
the User Product in which it has been modified or installed. Access to a
network may be denied when the modification itself materially and
adversely affects the operation of the network or violates the rules and
protocols for communication across the network.
Corresponding Source conveyed, and Installation Information provided,
in accord with this section must be in a format that is publicly
documented (and with an implementation available to the public in
source code form), and must require no special password or key for
unpacking, reading or copying.
7. Additional Terms.
"Additional permissions" are terms that supplement the terms of this
License by making exceptions from one or more of its conditions.
Additional permissions that are applicable to the entire Program shall
be treated as though they were included in this License, to the extent
that they are valid under applicable law. If additional permissions
apply only to part of the Program, that part may be used separately
under those permissions, but the entire Program remains governed by
this License without regard to the additional permissions.
When you convey a copy of a covered work, you may at your option
remove any additional permissions from that copy, or from any part of
it. (Additional permissions may be written to require their own
removal in certain cases when you modify the work.) You may place
additional permissions on material, added by you to a covered work,
for which you have or can give appropriate copyright permission.
Notwithstanding any other provision of this License, for material you
add to a covered work, you may (if authorized by the copyright holders of
that material) supplement the terms of this License with terms:
a) Disclaiming warranty or limiting liability differently from the
terms of sections 15 and 16 of this License; or
b) Requiring preservation of specified reasonable legal notices or
author attributions in that material or in the Appropriate Legal
Notices displayed by works containing it; or
c) Prohibiting misrepresentation of the origin of that material, or
requiring that modified versions of such material be marked in
reasonable ways as different from the original version; or
d) Limiting the use for publicity purposes of names of licensors or
authors of the material; or
e) Declining to grant rights under trademark law for use of some
trade names, trademarks, or service marks; or
f) Requiring indemnification of licensors and authors of that
material by anyone who conveys the material (or modified versions of
it) with contractual assumptions of liability to the recipient, for
any liability that these contractual assumptions directly impose on
those licensors and authors.
All other non-permissive additional terms are considered "further
restrictions" within the meaning of section 10. If the Program as you
received it, or any part of it, contains a notice stating that it is
governed by this License along with a term that is a further
restriction, you may remove that term. If a license document contains
a further restriction but permits relicensing or conveying under this
License, you may add to a covered work material governed by the terms
of that license document, provided that the further restriction does
not survive such relicensing or conveying.
If you add terms to a covered work in accord with this section, you
must place, in the relevant source files, a statement of the
additional terms that apply to those files, or a notice indicating
where to find the applicable terms.
Additional terms, permissive or non-permissive, may be stated in the
form of a separately written license, or stated as exceptions;
the above requirements apply either way.
8. Termination.
You may not propagate or modify a covered work except as expressly
provided under this License. Any attempt otherwise to propagate or
modify it is void, and will automatically terminate your rights under
this License (including any patent licenses granted under the third
paragraph of section 11).
However, if you cease all violation of this License, then your
license from a particular copyright holder is reinstated (a)
provisionally, unless and until the copyright holder explicitly and
finally terminates your license, and (b) permanently, if the copyright
holder fails to notify you of the violation by some reasonable means
prior to 60 days after the cessation.
Moreover, your license from a particular copyright holder is
reinstated permanently if the copyright holder notifies you of the
violation by some reasonable means, this is the first time you have
received notice of violation of this License (for any work) from that
copyright holder, and you cure the violation prior to 30 days after
your receipt of the notice.
Termination of your rights under this section does not terminate the
licenses of parties who have received copies or rights from you under
this License. If your rights have been terminated and not permanently
reinstated, you do not qualify to receive new licenses for the same
material under section 10.
9. Acceptance Not Required for Having Copies.
You are not required to accept this License in order to receive or
run a copy of the Program. Ancillary propagation of a covered work
occurring solely as a consequence of using peer-to-peer transmission
to receive a copy likewise does not require acceptance. However,
nothing other than this License grants you permission to propagate or
modify any covered work. These actions infringe copyright if you do
not accept this License. Therefore, by modifying or propagating a
covered work, you indicate your acceptance of this License to do so.
10. Automatic Licensing of Downstream Recipients.
Each time you convey a covered work, the recipient automatically
receives a license from the original licensors, to run, modify and
propagate that work, subject to this License. You are not responsible
for enforcing compliance by third parties with this License.
An "entity transaction" is a transaction transferring control of an
organization, or substantially all assets of one, or subdividing an
organization, or merging organizations. If propagation of a covered
work results from an entity transaction, each party to that
transaction who receives a copy of the work also receives whatever
licenses to the work the party's predecessor in interest had or could
give under the previous paragraph, plus a right to possession of the
Corresponding Source of the work from the predecessor in interest, if
the predecessor has it or can get it with reasonable efforts.
You may not impose any further restrictions on the exercise of the
rights granted or affirmed under this License. For example, you may
not impose a license fee, royalty, or other charge for exercise of
rights granted under this License, and you may not initiate litigation
(including a cross-claim or counterclaim in a lawsuit) alleging that
any patent claim is infringed by making, using, selling, offering for
sale, or importing the Program or any portion of it.
11. Patents.
A "contributor" is a copyright holder who authorizes use under this
License of the Program or a work on which the Program is based. The
work thus licensed is called the contributor's "contributor version".
A contributor's "essential patent claims" are all patent claims
owned or controlled by the contributor, whether already acquired or
hereafter acquired, that would be infringed by some manner, permitted
by this License, of making, using, or selling its contributor version,
but do not include claims that would be infringed only as a
consequence of further modification of the contributor version. For
purposes of this definition, "control" includes the right to grant
patent sublicenses in a manner consistent with the requirements of
this License.
Each contributor grants you a non-exclusive, worldwide, royalty-free
patent license under the contributor's essential patent claims, to
make, use, sell, offer for sale, import and otherwise run, modify and
propagate the contents of its contributor version.
In the following three paragraphs, a "patent license" is any express
agreement or commitment, however denominated, not to enforce a patent
(such as an express permission to practice a patent or covenant not to
sue for patent infringement). To "grant" such a patent license to a
party means to make such an agreement or commitment not to enforce a
patent against the party.
If you convey a covered work, knowingly relying on a patent license,
and the Corresponding Source of the work is not available for anyone
to copy, free of charge and under the terms of this License, through a
publicly available network server or other readily accessible means,
then you must either (1) cause the Corresponding Source to be so
available, or (2) arrange to deprive yourself of the benefit of the
patent license for this particular work, or (3) arrange, in a manner
consistent with the requirements of this License, to extend the patent
license to downstream recipients. "Knowingly relying" means you have
actual knowledge that, but for the patent license, your conveying the
covered work in a country, or your recipient's use of the covered work
in a country, would infringe one or more identifiable patents in that
country that you have reason to believe are valid.
If, pursuant to or in connection with a single transaction or
arrangement, you convey, or propagate by procuring conveyance of, a
covered work, and grant a patent license to some of the parties
receiving the covered work authorizing them to use, propagate, modify
or convey a specific copy of the covered work, then the patent license
you grant is automatically extended to all recipients of the covered
work and works based on it.
A patent license is "discriminatory" if it does not include within
the scope of its coverage, prohibits the exercise of, or is
conditioned on the non-exercise of one or more of the rights that are
specifically granted under this License. You may not convey a covered
work if you are a party to an arrangement with a third party that is
in the business of distributing software, under which you make payment
to the third party based on the extent of your activity of conveying
the work, and under which the third party grants, to any of the
parties who would receive the covered work from you, a discriminatory
patent license (a) in connection with copies of the covered work
conveyed by you (or copies made from those copies), or (b) primarily
for and in connection with specific products or compilations that
contain the covered work, unless you entered into that arrangement,
or that patent license was granted, prior to 28 March 2007.
Nothing in this License shall be construed as excluding or limiting
any implied license or other defenses to infringement that may
otherwise be available to you under applicable patent law.
12. No Surrender of Others' Freedom.
If conditions are imposed on you (whether by court order, agreement or
otherwise) that contradict the conditions of this License, they do not
excuse you from the conditions of this License. If you cannot convey a
covered work so as to satisfy simultaneously your obligations under this
License and any other pertinent obligations, then as a consequence you may
not convey it at all. For example, if you agree to terms that obligate you
to collect a royalty for further conveying from those to whom you convey
the Program, the only way you could satisfy both those terms and this
License would be to refrain entirely from conveying the Program.
13. Remote Network Interaction; Use with the GNU General Public License.
Notwithstanding any other provision of this License, if you modify the
Program, your modified version must prominently offer all users
interacting with it remotely through a computer network (if your version
supports such interaction) an opportunity to receive the Corresponding
Source of your version by providing access to the Corresponding Source
from a network server at no charge, through some standard or customary
means of facilitating copying of software. This Corresponding Source
shall include the Corresponding Source for any work covered by version 3
of the GNU General Public License that is incorporated pursuant to the
following paragraph.
Notwithstanding any other provision of this License, you have
permission to link or combine any covered work with a work licensed
under version 3 of the GNU General Public License into a single
combined work, and to convey the resulting work. The terms of this
License will continue to apply to the part which is the covered work,
but the work with which it is combined will remain governed by version
3 of the GNU General Public License.
14. Revised Versions of this License.
The Free Software Foundation may publish revised and/or new versions of
the GNU Affero General Public License from time to time. Such new versions
will be similar in spirit to the present version, but may differ in detail to
address new problems or concerns.
Each version is given a distinguishing version number. If the
Program specifies that a certain numbered version of the GNU Affero General
Public License "or any later version" applies to it, you have the
option of following the terms and conditions either of that numbered
version or of any later version published by the Free Software
Foundation. If the Program does not specify a version number of the
GNU Affero General Public License, you may choose any version ever published
by the Free Software Foundation.
If the Program specifies that a proxy can decide which future
versions of the GNU Affero General Public License can be used, that proxy's
public statement of acceptance of a version permanently authorizes you
to choose that version for the Program.
Later license versions may give you additional or different
permissions. However, no additional obligations are imposed on any
author or copyright holder as a result of your choosing to follow a
later version.
15. Disclaimer of Warranty.
THERE IS NO WARRANTY FOR THE PROGRAM, TO THE EXTENT PERMITTED BY
APPLICABLE LAW. EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT
HOLDERS AND/OR OTHER PARTIES PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY
OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO,
THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
PURPOSE. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE PROGRAM
IS WITH YOU. SHOULD THE PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF
ALL NECESSARY SERVICING, REPAIR OR CORRECTION.
16. Limitation of Liability.
IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING
WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MODIFIES AND/OR CONVEYS
THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY
GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE
USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED TO LOSS OF
DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD
PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER PROGRAMS),
EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF
SUCH DAMAGES.
17. Interpretation of Sections 15 and 16.
If the disclaimer of warranty and limitation of liability provided
above cannot be given local legal effect according to their terms,
reviewing courts shall apply local law that most closely approximates
an absolute waiver of all civil liability in connection with the
Program, unless a warranty or assumption of liability accompanies a
copy of the Program in return for a fee.
END OF TERMS AND CONDITIONS
How to Apply These Terms to Your New Programs
If you develop a new program, and you want it to be of the greatest
possible use to the public, the best way to achieve this is to make it
free software which everyone can redistribute and change under these terms.
To do so, attach the following notices to the program. It is safest
to attach them to the start of each source file to most effectively
state the exclusion of warranty; and each file should have at least
the "copyright" line and a pointer to where the full notice is found.
<one line to give the program's name and a brief idea of what it does.>
Copyright (C) <year> <name of author>
This program is free software: you can redistribute it and/or modify
it under the terms of the GNU Affero General Public License as published
by the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU Affero General Public License for more details.
You should have received a copy of the GNU Affero General Public License
along with this program. If not, see <https://www.gnu.org/licenses/>.
Also add information on how to contact you by electronic and paper mail.
If your software can interact with users remotely through a computer
network, you should also make sure that it provides a way for users to
get its source. For example, if your program is a web application, its
interface could display a "Source" link that leads users to an archive
of the code. There are many ways you could offer source, and different
solutions will be better for different programs; see section 13 for the
specific requirements.
You should also get your employer (if you work as a programmer) or school,
if any, to sign a "copyright disclaimer" for the program, if necessary.
For more information on this, and how to apply and follow the GNU AGPL, see
<https://www.gnu.org/licenses/>.
app.py 0 → 100644
View file @ c9171d1f
# Copyright (c) Opendatalab. All rights reserved.
import base64
import os
import time
import zipfile
from pathlib import Path
import re
from loguru import logger
from magic_pdf.libs.hash_utils import compute_sha256
from magic_pdf.rw.AbsReaderWriter import AbsReaderWriter
from magic_pdf.rw.DiskReaderWriter import DiskReaderWriter
from magic_pdf.tools.common import do_parse, prepare_env
os.system("pip install gradio")
os.system("pip install gradio-pdf")
import gradio as gr
from gradio_pdf import PDF
def read_fn(path):
disk_rw = DiskReaderWriter(os.path.dirname(path))
return disk_rw.read(os.path.basename(path), AbsReaderWriter.MODE_BIN)
def parse_pdf(doc_path, output_dir, end_page_id):
os.makedirs(output_dir, exist_ok=True)
try:
file_name = f"{str(Path(doc_path).stem)}_{time.time()}"
pdf_data = read_fn(doc_path)
parse_method = "auto"
local_image_dir, local_md_dir = prepare_env(output_dir, file_name, parse_method)
do_parse(
output_dir,
file_name,
pdf_data,
[],
parse_method,
False,
end_page_id=end_page_id,
)
return local_md_dir, file_name
except Exception as e:
logger.exception(e)
def compress_directory_to_zip(directory_path, output_zip_path):
"""
压缩指定目录到一个 ZIP 文件。
:param directory_path: 要压缩的目录路径
:param output_zip_path: 输出的 ZIP 文件路径
"""
try:
with zipfile.ZipFile(output_zip_path, 'w', zipfile.ZIP_DEFLATED) as zipf:
# 遍历目录中的所有文件和子目录
for root, dirs, files in os.walk(directory_path):
for file in files:
# 构建完整的文件路径
file_path = os.path.join(root, file)
# 计算相对路径
arcname = os.path.relpath(file_path, directory_path)
# 添加文件到 ZIP 文件
zipf.write(file_path, arcname)
return 0
except Exception as e:
logger.exception(e)
return -1
def image_to_base64(image_path):
with open(image_path, "rb") as image_file:
return base64.b64encode(image_file.read()).decode('utf-8')
def replace_image_with_base64(markdown_text, image_dir_path):
# 匹配Markdown中的图片标签
pattern = r'\!\[(?:[^\]]*)\]\(([^)]+)\)'
# 替换图片链接
def replace(match):
relative_path = match.group(1)
full_path = os.path.join(image_dir_path, relative_path)
base64_image = image_to_base64(full_path)
return f"![{relative_path}](data:image/jpeg;base64,{base64_image})"
# 应用替换
return re.sub(pattern, replace, markdown_text)
def to_markdown(file_path, end_pages):
# 获取识别的md文件以及压缩包文件路径
local_md_dir, file_name = parse_pdf(file_path, './output', end_pages - 1)
archive_zip_path = os.path.join("./output", compute_sha256(local_md_dir) + ".zip")
zip_archive_success = compress_directory_to_zip(local_md_dir, archive_zip_path)
if zip_archive_success == 0:
logger.info("压缩成功")
else:
logger.error("压缩失败")
md_path = os.path.join(local_md_dir, file_name + ".md")
with open(md_path, 'r', encoding='utf-8') as f:
txt_content = f.read()
md_content = replace_image_with_base64(txt_content, local_md_dir)
# 返回转换后的PDF路径
new_pdf_path = os.path.join(local_md_dir, file_name + "_layout.pdf")
return md_content, txt_content, archive_zip_path, new_pdf_path
# def show_pdf(file_path):
# with open(file_path, "rb") as f:
# base64_pdf = base64.b64encode(f.read()).decode('utf-8')
# pdf_display = f'<embed src="data:application/pdf;base64,{base64_pdf}" ' \
# f'width="100%" height="1000" type="application/pdf">'
# return pdf_display
latex_delimiters = [{"left": "$$", "right": "$$", "display": True},
{"left": '$', "right": '$', "display": False}]
def init_model():
from magic_pdf.model.doc_analyze_by_custom_model import ModelSingleton
try:
model_manager = ModelSingleton()
txt_model = model_manager.get_model(False, False)
logger.info(f"txt_model init final")
ocr_model = model_manager.get_model(True, False)
logger.info(f"ocr_model init final")
return 0
except Exception as e:
logger.exception(e)
return -1
model_init = init_model()
logger.info(f"model_init: {model_init}")
if __name__ == "__main__":
with gr.Blocks() as demo:
with gr.Row():
with gr.Column(variant='panel', scale=5):
pdf_show = gr.Markdown()
max_pages = gr.Slider(1, 10, 5, step=1, label="Max convert pages")
with gr.Row() as bu_flow:
change_bu = gr.Button("Convert")
clear_bu = gr.ClearButton([pdf_show], value="Clear")
pdf_show = PDF(label="Please upload pdf", interactive=True, height=800)
with gr.Column(variant='panel', scale=5):
output_file = gr.File(label="convert result", interactive=False)
with gr.Tabs():
with gr.Tab("Markdown rendering"):
md = gr.Markdown(label="Markdown rendering", height=900, show_copy_button=True,
latex_delimiters=latex_delimiters, line_breaks=True)
with gr.Tab("Markdown text"):
md_text = gr.TextArea(lines=45, show_copy_button=True)
change_bu.click(fn=to_markdown, inputs=[pdf_show, max_pages], outputs=[md, md_text, output_file, pdf_show])
clear_bu.add([md, pdf_show, md_text, output_file])
demo.launch()
count_pdfs.py 0 → 100644
View file @ c9171d1f
import os
def find_pdfs_and_save_to_txt(directory, output_file):
pdf_paths = []
# Traverse the directory and find all PDF files
for root, dirs, files in os.walk(directory):
for file in files:
if file.lower().endswith('.pdf'):
pdf_paths.append(os.path.join(root, file))
# Write all PDF paths to the output file
with open(output_file, 'w',encoding='utf-8') as f:
for path in pdf_paths:
f.write(path + '\n')
return len(pdf_paths)
#
# # Use '.' for the current directory, and specify output file
directory = '/home/practice/rag/DCU知识库文档集合'
output_file = 'pdf_paths.txt'
pdf_count = find_pdfs_and_save_to_txt(directory, output_file)
#
print(f"Found {pdf_count} PDF files. Paths are saved in '{output_file}'.")
def remove_duplicate_lines(file_path):
lines_seen = set() # 用于保存已经看到的唯一行
unique_lines = [] # 用于保存不重复的行
# 读取文件，并过滤重复的行
with open(file_path, 'r',encoding='utf-8') as f:
for line in f:
line = line.strip() # 去掉行首尾的空白符
if line not in lines_seen:
unique_lines.append(line)
lines_seen.add(line)
out_path = file_path.split('.txt')[0] + '_uq.txt'
# 将不重复的行写回文件
with open(out_path, 'w',encoding='utf-8') as f:
for line in unique_lines:
f.write(line + '\n')
# 返回不重复行的数量
return len(unique_lines)
# 指定要处理的文件路径
file_path = 'pdf_paths.txt'
unique_line_count = remove_duplicate_lines(file_path)
print(f"Number of unique lines: {unique_line_count}")
magic-pdf.template.json 0 → 100644
View file @ c9171d1f
{
"bucket_info":{
"bucket-name-1":["ak", "sk", "endpoint"],
"bucket-name-2":["ak", "sk", "endpoint"]
},
"models-dir":"/home/practice/model/PDF-Extract-Kit/models",
"device-mode":"cuda",
"table-config": {
"model": "struct_eqtable",
"is_table_recog_enable": True,
"max_time": 400
}
}
magic_pdf/dict2md/mkcontent.py 0 → 100644
View file @ c9171d1f
import math
from loguru import logger
from magic_pdf.libs.boxbase import find_bottom_nearest_text_bbox, find_top_nearest_text_bbox
from magic_pdf.libs.commons import join_path
from magic_pdf.libs.ocr_content_type import ContentType
TYPE_INLINE_EQUATION = ContentType.InlineEquation
TYPE_INTERLINE_EQUATION = ContentType.InterlineEquation
UNI_FORMAT_TEXT_TYPE = ['text', 'h1', 'h2', 'h3', 'h4', 'h5', 'h6']
@DeprecationWarning
def mk_nlp_markdown_1(para_dict: dict):
"""
对排序后的bboxes拼接内容
"""
content_lst = []
for _, page_info in para_dict.items():
para_blocks = page_info.get("para_blocks")
if not para_blocks:
continue
for block in para_blocks:
item = block["paras"]
for _, p in item.items():
para_text = p["para_text"]
is_title = p["is_para_title"]
title_level = p['para_title_level']
md_title_prefix = "#"*title_level
if is_title:
content_lst.append(f"{md_title_prefix} {para_text}")
else:
content_lst.append(para_text)
content_text = "\n\n".join(content_lst)
return content_text
# 找到目标字符串在段落中的索引
def __find_index(paragraph, target):
index = paragraph.find(target)
if index != -1:
return index
else:
return None
def __insert_string(paragraph, target, postion):
new_paragraph = paragraph[:postion] + target + paragraph[postion:]
return new_paragraph
def __insert_after(content, image_content, target):
"""
在content中找到target，将image_content插入到target后面
"""
index = content.find(target)
if index != -1:
content = content[:index+len(target)] + "\n\n" + image_content + "\n\n" + content[index+len(target):]
else:
logger.error(f"Can't find the location of image {image_content} in the markdown file, search target is {target}")
return content
def __insert_before(content, image_content, target):
"""
在content中找到target，将image_content插入到target前面
"""
index = content.find(target)
if index != -1:
content = content[:index] + "\n\n" + image_content + "\n\n" + content[index:]
else:
logger.error(f"Can't find the location of image {image_content} in the markdown file, search target is {target}")
return content
@DeprecationWarning
def mk_mm_markdown_1(para_dict: dict):
"""拼装多模态markdown"""
content_lst = []
for _, page_info in para_dict.items():
page_lst = [] # 一个page内的段落列表
para_blocks = page_info.get("para_blocks")
pymu_raw_blocks = page_info.get("preproc_blocks")
all_page_images = []
all_page_images.extend(page_info.get("images",[]))
all_page_images.extend(page_info.get("image_backup", []) )
all_page_images.extend(page_info.get("tables",[]))
all_page_images.extend(page_info.get("table_backup",[]) )
if not para_blocks or not pymu_raw_blocks: # 只有图片的拼接的场景
for img in all_page_images:
page_lst.append(f"![]({img['image_path']})") # TODO 图片顺序
page_md = "\n\n".join(page_lst)
else:
for block in para_blocks:
item = block["paras"]
for _, p in item.items():
para_text = p["para_text"]
is_title = p["is_para_title"]
title_level = p['para_title_level']
md_title_prefix = "#"*title_level
if is_title:
page_lst.append(f"{md_title_prefix} {para_text}")
else:
page_lst.append(para_text)
"""拼装成一个页面的文本"""
page_md = "\n\n".join(page_lst)
"""插入图片"""
for img in all_page_images:
imgbox = img['bbox']
img_content = f"![]({img['image_path']})"
# 先看在哪个block内
for block in pymu_raw_blocks:
bbox = block['bbox']
if bbox[0]-1 <= imgbox[0] < bbox[2]+1 and bbox[1]-1 <= imgbox[1] < bbox[3]+1:# 确定在block内
for l in block['lines']:
line_box = l['bbox']
if line_box[0]-1 <= imgbox[0] < line_box[2]+1 and line_box[1]-1 <= imgbox[1] < line_box[3]+1: # 在line内的，插入line前面
line_txt = "".join([s['text'] for s in l['spans']])
page_md = __insert_before(page_md, img_content, line_txt)
break
break
else:# 在行与行之间
# 找到图片x0,y0与line的x0,y0最近的line
min_distance = 100000
min_line = None
for l in block['lines']:
line_box = l['bbox']
distance = math.sqrt((line_box[0] - imgbox[0])**2 + (line_box[1] - imgbox[1])**2)
if distance < min_distance:
min_distance = distance
min_line = l
if min_line:
line_txt = "".join([s['text'] for s in min_line['spans']])
img_h = imgbox[3] - imgbox[1]
if min_distance<img_h: # 文字在图片前面
page_md = __insert_after(page_md, img_content, line_txt)
else:
page_md = __insert_before(page_md, img_content, line_txt)
else:
logger.error(f"Can't find the location of image {img['image_path']} in the markdown file #1")
else:# 应当在两个block之间
# 找到上方最近的block，如果上方没有就找大下方最近的block
top_txt_block = find_top_nearest_text_bbox(pymu_raw_blocks, imgbox)
if top_txt_block:
line_txt = "".join([s['text'] for s in top_txt_block['lines'][-1]['spans']])
page_md = __insert_after(page_md, img_content, line_txt)
else:
bottom_txt_block = find_bottom_nearest_text_bbox(pymu_raw_blocks, imgbox)
if bottom_txt_block:
line_txt = "".join([s['text'] for s in bottom_txt_block['lines'][0]['spans']])
page_md = __insert_before(page_md, img_content, line_txt)
else:
logger.error(f"Can't find the location of image {img['image_path']} in the markdown file #2")
content_lst.append(page_md)
"""拼装成全部页面的文本"""
content_text = "\n\n".join(content_lst)
return content_text
def __insert_after_para(text, type, element, content_list):
"""
在content_list中找到text，将image_path作为一个新的node插入到text后面
"""
for i, c in enumerate(content_list):
content_type = c.get("type")
if content_type in UNI_FORMAT_TEXT_TYPE and text in c.get("text", ''):
if type == "image":
content_node = {
"type": "image",
"img_path": element.get("image_path"),
"img_alt": "",
"img_title": "",
"img_caption": "",
}
elif type == "table":
content_node = {
"type": "table",
"img_path": element.get("image_path"),
"table_latex": element.get("text"),
"table_title": "",
"table_caption": "",
"table_quality": element.get("quality"),
}
content_list.insert(i+1, content_node)
break
else:
logger.error(f"Can't find the location of image {element.get('image_path')} in the markdown file, search target is {text}")
def __insert_before_para(text, type, element, content_list):
"""
在content_list中找到text，将image_path作为一个新的node插入到text前面
"""
for i, c in enumerate(content_list):
content_type = c.get("type")
if content_type in UNI_FORMAT_TEXT_TYPE and text in c.get("text", ''):
if type == "image":
content_node = {
"type": "image",
"img_path": element.get("image_path"),
"img_alt": "",
"img_title": "",
"img_caption": "",
}
elif type == "table":
content_node = {
"type": "table",
"img_path": element.get("image_path"),
"table_latex": element.get("text"),
"table_title": "",
"table_caption": "",
"table_quality": element.get("quality"),
}
content_list.insert(i, content_node)
break
else:
logger.error(f"Can't find the location of image {element.get('image_path')} in the markdown file, search target is {text}")
def mk_universal_format(pdf_info_list: list, img_buket_path):
"""
构造统一格式 https://aicarrier.feishu.cn/wiki/FqmMwcH69iIdCWkkyjvcDwNUnTY
"""
content_lst = []
for page_info in pdf_info_list:
page_lst = [] # 一个page内的段落列表
para_blocks = page_info.get("para_blocks")
pymu_raw_blocks = page_info.get("preproc_blocks")
all_page_images = []
all_page_images.extend(page_info.get("images",[]))
all_page_images.extend(page_info.get("image_backup", []) )
# all_page_images.extend(page_info.get("tables",[]))
# all_page_images.extend(page_info.get("table_backup",[]) )
all_page_tables = []
all_page_tables.extend(page_info.get("tables", []))
if not para_blocks or not pymu_raw_blocks: # 只有图片的拼接的场景
for img in all_page_images:
content_node = {
"type": "image",
"img_path": join_path(img_buket_path, img['image_path']),
"img_alt":"",
"img_title":"",
"img_caption":""
}
page_lst.append(content_node) # TODO 图片顺序
for table in all_page_tables:
content_node = {
"type": "table",
"img_path": join_path(img_buket_path, table['image_path']),
"table_latex": table.get("text"),
"table_title": "",
"table_caption": "",
"table_quality": table.get("quality"),
}
page_lst.append(content_node) # TODO 图片顺序
else:
for block in para_blocks:
item = block["paras"]
for _, p in item.items():
font_type = p['para_font_type']# 对于文本来说，要么是普通文本，要么是个行间公式
if font_type == TYPE_INTERLINE_EQUATION:
content_node = {
"type": "equation",
"latex": p["para_text"]
}
page_lst.append(content_node)
else:
para_text = p["para_text"]
is_title = p["is_para_title"]
title_level = p['para_title_level']
if is_title:
content_node = {
"type": f"h{title_level}",
"text": para_text
}
page_lst.append(content_node)
else:
content_node = {
"type": "text",
"text": para_text
}
page_lst.append(content_node)
content_lst.extend(page_lst)
"""插入图片"""
for img in all_page_images:
insert_img_or_table("image", img, pymu_raw_blocks, content_lst)
"""插入表格"""
for table in all_page_tables:
insert_img_or_table("table", table, pymu_raw_blocks, content_lst)
# end for
return content_lst
def insert_img_or_table(type, element, pymu_raw_blocks, content_lst):
element_bbox = element['bbox']
# 先看在哪个block内
for block in pymu_raw_blocks:
bbox = block['bbox']
if bbox[0] - 1 <= element_bbox[0] < bbox[2] + 1 and bbox[1] - 1 <= element_bbox[1] < bbox[
3] + 1: # 确定在这个大的block内，然后进入逐行比较距离
for l in block['lines']:
line_box = l['bbox']
if line_box[0] - 1 <= element_bbox[0] < line_box[2] + 1 and line_box[1] - 1 <= element_bbox[1] < line_box[
3] + 1: # 在line内的，插入line前面
line_txt = "".join([s['text'] for s in l['spans']])
__insert_before_para(line_txt, type, element, content_lst)
break
break
else: # 在行与行之间
# 找到图片x0,y0与line的x0,y0最近的line
min_distance = 100000
min_line = None
for l in block['lines']:
line_box = l['bbox']
distance = math.sqrt((line_box[0] - element_bbox[0]) ** 2 + (line_box[1] - element_bbox[1]) ** 2)
if distance < min_distance:
min_distance = distance
min_line = l
if min_line:
line_txt = "".join([s['text'] for s in min_line['spans']])
img_h = element_bbox[3] - element_bbox[1]
if min_distance < img_h: # 文字在图片前面
__insert_after_para(line_txt, type, element, content_lst)
else:
__insert_before_para(line_txt, type, element, content_lst)
break
else:
logger.error(f"Can't find the location of image {element.get('image_path')} in the markdown file #1")
else: # 应当在两个block之间
# 找到上方最近的block，如果上方没有就找大下方最近的block
top_txt_block = find_top_nearest_text_bbox(pymu_raw_blocks, element_bbox)
if top_txt_block:
line_txt = "".join([s['text'] for s in top_txt_block['lines'][-1]['spans']])
__insert_after_para(line_txt, type, element, content_lst)
else:
bottom_txt_block = find_bottom_nearest_text_bbox(pymu_raw_blocks, element_bbox)
if bottom_txt_block:
line_txt = "".join([s['text'] for s in bottom_txt_block['lines'][0]['spans']])
__insert_before_para(line_txt, type, element, content_lst)
else: # TODO ，图片可能独占一列，这种情况上下是没有图片的
logger.error(f"Can't find the location of image {element.get('image_path')} in the markdown file #2")
def mk_mm_markdown(content_list):
"""
基于同一格式的内容列表，构造markdown，含图片
"""
content_md = []
for c in content_list:
content_type = c.get("type")
if content_type == "text":
content_md.append(c.get("text"))
elif content_type == "equation":
content = c.get("latex")
if content.startswith("$$") and content.endswith("$$"):
content_md.append(content)
else:
content_md.append(f"\n$$\n{c.get('latex')}\n$$\n")
elif content_type in UNI_FORMAT_TEXT_TYPE:
content_md.append(f"{'#'*int(content_type[1])} {c.get('text')}")
elif content_type == "image":
content_md.append(f"![]({c.get('img_path')})")
return "\n\n".join(content_md)
def mk_nlp_markdown(content_list):
"""
基于同一格式的内容列表，构造markdown，不含图片
"""
content_md = []
for c in content_list:
content_type = c.get("type")
if content_type == "text":
content_md.append(c.get("text"))
elif content_type == "equation":
content_md.append(f"$$\n{c.get('latex')}\n$$")
elif content_type == "table":
content_md.append(f"$$$\n{c.get('table_latex')}\n$$$")
elif content_type in UNI_FORMAT_TEXT_TYPE:
content_md.append(f"{'#'*int(content_type[1])} {c.get('text')}")
return "\n\n".join(content_md)
\ No newline at end of file
magic_pdf/dict2md/ocr_client.py 0 → 100644
View file @ c9171d1f
import os
import json
import requests
from loguru import logger
import argparse
import time
from PIL import Image
def parse_args():
parser = argparse.ArgumentParser()
parser.add_argument(
'--url',
default='http://127.0.0.1:6020',
)
parser.add_argument(
'--image_path',
default='/home/wanglch/projects/Qwen2-VL/20240920-163701.png',
)
parser.add_argument(
'--text',
default="描述你在图片中看到的内容",
)
args = parser.parse_args()
return args
def parse_text(text):
lines = text.split("\n")
lines = [line for line in lines if line.strip() != ""] # 去除空行
count = 0
parsed_lines = []
for i, line in enumerate(lines):
if "```" in line:
count += 1
items = line.split("`")
if count % 2 == 1:
# 开始代码块
parsed_lines.append(f'<pre><code class="language-{items[-1]}">')
else:
# 结束代码块
parsed_lines.append(f"</code></pre>")
else:
if i > 0 and count % 2 == 1:
# 转义代码块内的特殊字符
line = line.replace("`", r"\`")
line = line.replace("<", "&lt;")
line = line.replace(">", "&gt;")
line = line.replace(" ", "&nbsp;")
line = line.replace("*", "&ast;")
line = line.replace("_", "&lowbar;")
line = line.replace("-", "&#45;")
line = line.replace(".", "&#46;")
line = line.replace("!", "&#33;")
line = line.replace("(", "&#40;")
line = line.replace(")", "&#41;")
line = line.replace("$", "&#36;")
# 使用空格连接行
if parsed_lines:
parsed_lines[-1] += " " + line
else:
parsed_lines.append(line)
text = "".join(parsed_lines)
return text
def unparse_text(parsed_text):
in_code_block = False
lines = parsed_text.split("\n")
unparsed_lines = []
for line in lines:
if "<pre><code" in line:
in_code_block = True
# 移除开始标签
line = line.split(">", 1)[1]
elif "</code></pre>" in line:
in_code_block = False
# 移除结束标签
line = line.rsplit("<", 1)[0]
# 反转 HTML 实体
line = line.replace("&lt;", "<")
line = line.replace("&gt;", ">")
line = line.replace("&nbsp;", " ")
line = line.replace("&ast;", "*")
line = line.replace("&lowbar;", "_")
line = line.replace("&#45;", "-")
line = line.replace("&#46;", ".")
line = line.replace("&#33;", "!")
line = line.replace("&#40;", "(")
line = line.replace("&#41;", ")")
line = line.replace("&#36;", "$")
# 如果在代码块内，还原反斜杠转义
if in_code_block:
line = line.replace(r"\`", "`")
unparsed_lines.append(line)
# 合并所有行
unparsed_text = "\n".join(unparsed_lines)
return unparsed_text
def compress_image(image_path, max_size=(1024, 1024)):
img = Image.open(image_path)
width, height = img.size
aspect_ratio = width / height
if width > max_size[0] or height > max_size[1]:
if width > height:
new_width = max_size[0]
new_height = int(new_width / aspect_ratio)
else:
new_height = max_size[1]
new_width = int(new_height * aspect_ratio)
img = img.resize((new_width, new_height), Image.LANCZOS)
img.save(image_path, optimize=True, quality=80)
class PredictClient:
def __init__(self, api_url):
self.api_url = api_url
def predict(self, image_path: str, text: str):
payload = {
"image_path": image_path,
"text": text
}
headers = {'Content-Type': 'application/json'}
response = requests.post(f"{self.api_url}/predict", json=payload, headers=headers)
if response.status_code == 200:
result = response.json()
return result.get('Generated Text', '')
else:
raise Exception(f"Predict API request failed with status code {response.status_code}")
def main():
args = parse_args()
client = PredictClient(args.url)
try:
start_time = time.time() # 记录开始时间
# 压缩图片
#compress_image(args.image_path)
generated_text = client.predict(args.image_path, parse_text(args.text))
end_time = time.time() # 记录结束时间
elapsed_time = end_time - start_time # 计算运行时间
if generated_text:
clean_text = unparse_text(generated_text) # 解析生成的文本
logger.info(f"Image Path: {args.image_path}")
logger.info(f"Generated Text: {clean_text}")
logger.info(f"耗时为: {elapsed_time}秒") # 打印运行时间
else:
logger.warning("Received empty generated text.")
except requests.exceptions.RequestException as e:
logger.error(f"Error while making request to predict service: {e}")
except Exception as e:
logger.error(f"Unexpected error occurred: {e}")
if __name__ == "__main__":
main()
magic_pdf/dict2md/ocr_mkcontent.py 0 → 100644
View file @ c9171d1f
import re
import time
import wordninja
from loguru import logger
from magic_pdf.libs.commons import join_path
from magic_pdf.libs.language import detect_lang
from magic_pdf.libs.MakeContentConfig import DropMode, MakeMode
from magic_pdf.libs.markdown_utils import ocr_escape_special_markdown_char
from magic_pdf.libs.ocr_content_type import BlockType, ContentType
# import pypandoc
from magic_pdf.dict2md.ocr_client import PredictClient,compress_image
def __is_hyphen_at_line_end(line):
"""
Check if a line ends with one or more letters followed by a hyphen.
Args:
line (str): The line of text to check.
Returns:
bool: True if the line ends with one or more letters followed by a hyphen, False otherwise.
"""
# Use regex to check if the line ends with one or more letters followed by a hyphen
return bool(re.search(r'[A-Za-z]+-\s*$', line))
def split_long_words(text):
segments = text.split(' ')
for i in range(len(segments)):
words = re.findall(r'\w+|[^\w]', segments[i], re.UNICODE)
for j in range(len(words)):
if len(words[j]) > 10:
words[j] = ' '.join(wordninja.split(words[j]))
segments[i] = ''.join(words)
return ' '.join(segments)
def ocr_mk_mm_markdown_with_para(pdf_info_list: list, img_buket_path):
markdown = []
for page_info in pdf_info_list:
paras_of_layout = page_info.get('para_blocks')
page_markdown = ocr_mk_markdown_with_para_core_v2(
paras_of_layout, 'mm', img_buket_path)
markdown.extend(page_markdown)
return '\n\n'.join(markdown)
def ocr_mk_nlp_markdown_with_para(pdf_info_dict: list):
markdown = []
for page_info in pdf_info_dict:
paras_of_layout = page_info.get('para_blocks')
page_markdown = ocr_mk_markdown_with_para_core_v2(
paras_of_layout, 'nlp')
markdown.extend(page_markdown)
return '\n\n'.join(markdown)
def ocr_mk_mm_markdown_with_para_and_pagination(pdf_info_dict: list,
img_buket_path):
markdown_with_para_and_pagination = []
page_no = 0
for page_info in pdf_info_dict:
paras_of_layout = page_info.get('para_blocks')
if not paras_of_layout:
continue
page_markdown = ocr_mk_markdown_with_para_core_v2(
paras_of_layout, 'mm', img_buket_path)
markdown_with_para_and_pagination.append({
'page_no':
page_no,
'md_content':
'\n\n'.join(page_markdown)
})
page_no += 1
return markdown_with_para_and_pagination
def ocr_mk_markdown_with_para_core(paras_of_layout, mode, img_buket_path=''):
page_markdown = []
for paras in paras_of_layout:
for para in paras:
para_text = ''
for line in para:
for span in line['spans']:
span_type = span.get('type')
content = ''
language = ''
if span_type == ContentType.Text:
content = span['content']
language = detect_lang(content)
if (language == 'en'): # 只对英文长词进行分词处理，中文分词会丢失文本
content = ocr_escape_special_markdown_char(
split_long_words(content))
else:
content = ocr_escape_special_markdown_char(content)
elif span_type == ContentType.InlineEquation:
content = f"${span['content']}$"
#content = pypandoc.convert_text(content, to='plain', format='latex')
elif span_type == ContentType.InterlineEquation:
content = f"\n$$\n{span['content']}\n$$\n"
#content = pypandoc.convert_text(content, to='plain', format='latex')
elif span_type in [ContentType.Image, ContentType.Table]:
if mode == 'mm':
content = f"\n![]({join_path(img_buket_path, span['image_path'])})\n"
elif mode == 'nlp':
pass
if content != '':
if language == 'en': # 英文语境下 content间需要空格分隔
para_text += content + ' '
else: # 中文语境下，content间不需要空格分隔
para_text += content
if para_text.strip() == '':
continue
else:
page_markdown.append(para_text.strip() + ' ')
return page_markdown
def ocr_mk_markdown_with_para_core_v2(paras_of_layout,
mode,
img_buket_path=''):
page_markdown = []
url = 'http://127.0.0.1:6020'
client = PredictClient(url)
for para_block in paras_of_layout:
para_text = ''
para_type = para_block['type']
if para_type == BlockType.Text:
para_text = str(merge_para_with_text(para_block)).strip()
# 处理标题
elif para_type == BlockType.Title:
para_text = f'{merge_para_with_text(para_block)}'
elif para_type == BlockType.InterlineEquation:
para_text = merge_para_with_text(para_block)
elif para_type == BlockType.Image:
if mode == 'nlp':
continue
elif mode == 'mm':
for block in para_block['blocks']: # 1st.拼image_body
if block['type'] == BlockType.ImageBody:
for line in block['lines']:
for span in line['spans']:
if span['type'] == ContentType.Image:
para_text += f"\n----------------这是ocr图片内容({join_path(img_buket_path, span['image_path'])})------------------ \n"
for block in para_block['blocks']: # 2nd.拼image_caption
if block['type'] == BlockType.ImageCaption:
para_text += merge_para_with_text(block)
for block in para_block['blocks']: # 2nd.拼image_caption
if block['type'] == BlockType.ImageFootnote:
para_text += merge_para_with_text(block)
# 表格类型
elif para_type == BlockType.Table:
if mode == 'nlp':
continue
elif mode == 'mm':
for block in para_block['blocks']: # 1st.拼table_caption
if block['type'] == BlockType.TableCaption:
para_text += merge_para_with_text(block)
for block in para_block['blocks']: # 2nd.拼table_body
if block['type'] == BlockType.TableBody:
for line in block['lines']:
for span in line['spans']:
if span['type'] == ContentType.Table:
# if processed by table model
if span.get('latex', ''):
para_text += f"\n\n$\n {span['latex']}\n$\n\n"
elif span.get('html', ''):
para_text += f"\n\n{span['html']}\n\n"
else:
# 处理图片
# para_text += f"----------------这是ocr表格内容({join_path(img_buket_path, span['image_path'])})------------------- \n"
text = '解析图片内容，直接返回一段带有逻辑性的中文书面语描述，要求表达精准，不脱离图片中的实际内容，不要带换行,文中所有的名词不要用指代词'
start = time.time()
image_path = join_path(img_buket_path, span['image_path'])
compress_image(image_path)
generated_text = client.predict(image_path, text)
end = time.time()
logger.info(f'qwen解析{image_path}表格的内容为：{generated_text},耗时为：{end-start}')
para_text += generated_text
for block in para_block['blocks']: # 3rd.拼table_footnote
if block['type'] == BlockType.TableFootnote:
para_text += merge_para_with_text(block)
if para_text.strip() == '':
continue
else:
page_markdown.append(para_text.strip() + ' ')
return page_markdown
def merge_para_with_text(para_block):
def detect_language(text):
en_pattern = r'[a-zA-Z]+'
en_matches = re.findall(en_pattern, text)
en_length = sum(len(match) for match in en_matches)
if len(text) > 0:
if en_length / len(text) >= 0.5:
return 'en'
else:
return 'unknown'
else:
return 'empty'
para_text = ''
for line in para_block['lines']:
line_text = ''
line_lang = ''
for span in line['spans']:
span_type = span['type']
if span_type == ContentType.Text:
line_text += span['content'].strip()
if line_text != '':
line_lang = detect_lang(line_text)
for span in line['spans']:
span_type = span['type']
content = ''
if span_type == ContentType.Text:
content = span['content']
# language = detect_lang(content)
language = detect_language(content)
if language == 'en': # 只对英文长词进行分词处理，中文分词会丢失文本
content = ocr_escape_special_markdown_char(
split_long_words(content))
else:
content = ocr_escape_special_markdown_char(content)
elif span_type == ContentType.InlineEquation:
content = f" ${span['content']}$ "
#content = pypandoc.convert_text(content, to='plain', format='latex')
elif span_type == ContentType.InterlineEquation:
content = f"\n$$\n{span['content']}\n$$\n"
#content = pypandoc.convert_text(content, to='plain', format='latex')
if content != '':
langs = ['zh', 'ja', 'ko']
if line_lang in langs: # 遇到一些一个字一个span的文档，这种单字语言判断不准，需要用整行文本判断
para_text += content # 中文/日语/韩文语境下，content间不需要空格分隔
elif line_lang == 'en':
# 如果是前一行带有-连字符，那么末尾不应该加空格
if __is_hyphen_at_line_end(content):
para_text += content[:-1]
else:
para_text += content + ' '
else:
para_text += content + ' ' # 西方文本语境下 content间需要空格分隔
return para_text
def para_to_standard_format(para, img_buket_path):
para_content = {}
if len(para) == 1:
para_content = line_to_standard_format(para[0], img_buket_path)
elif len(para) > 1:
para_text = ''
inline_equation_num = 0
for line in para:
for span in line['spans']:
language = ''
span_type = span.get('type')
content = ''
if span_type == ContentType.Text:
content = span['content']
language = detect_lang(content)
if language == 'en': # 只对英文长词进行分词处理，中文分词会丢失文本
content = ocr_escape_special_markdown_char(
split_long_words(content))
else:
content = ocr_escape_special_markdown_char(content)
elif span_type == ContentType.InlineEquation:
content = f"${span['content']}$"
#content = pypandoc.convert_text(content, to='plain', format='latex')
inline_equation_num += 1
if language == 'en': # 英文语境下 content间需要空格分隔
para_text += content + ' '
else: # 中文语境下，content间不需要空格分隔
para_text += content
para_content = {
'type': 'text',
'text': para_text,
'inline_equation_num': inline_equation_num,
}
return para_content
def para_to_standard_format_v2(para_block, img_buket_path, page_idx):
para_type = para_block['type']
if para_type == BlockType.Text:
para_content = {
'type': 'text',
'text': merge_para_with_text(para_block),
'page_idx': page_idx,
}
elif para_type == BlockType.Title:
para_content = {
'type': 'text',
'text': merge_para_with_text(para_block),
'text_level': 1,
'page_idx': page_idx,
}
elif para_type == BlockType.InterlineEquation:
para_content = {
'type': 'equation',
'text': merge_para_with_text(para_block),
'text_format': 'latex',
'page_idx': page_idx,
}
elif para_type == BlockType.Image:
para_content = {'type': 'image', 'page_idx': page_idx}
for block in para_block['blocks']:
if block['type'] == BlockType.ImageBody:
para_content['img_path'] = join_path(
img_buket_path,
block['lines'][0]['spans'][0]['image_path'])
if block['type'] == BlockType.ImageCaption:
para_content['img_caption'] = merge_para_with_text(block)
if block['type'] == BlockType.ImageFootnote:
para_content['img_footnote'] = merge_para_with_text(block)
elif para_type == BlockType.Table:
para_content = {'type': 'table', 'page_idx': page_idx}
for block in para_block['blocks']:
if block['type'] == BlockType.TableBody:
if block["lines"][0]["spans"][0].get('latex', ''):
para_content['table_body'] = f"\n\n$\n {block['lines'][0]['spans'][0]['latex']}\n$\n\n"
elif block["lines"][0]["spans"][0].get('html', ''):
para_content['table_body'] = f"\n\n{block['lines'][0]['spans'][0]['html']}\n\n"
para_content['img_path'] = join_path(img_buket_path, block["lines"][0]["spans"][0]['image_path'])
if block['type'] == BlockType.TableCaption:
para_content['table_caption'] = merge_para_with_text(block)
if block['type'] == BlockType.TableFootnote:
para_content['table_footnote'] = merge_para_with_text(block)
return para_content
def make_standard_format_with_para(pdf_info_dict: list, img_buket_path: str):
content_list = []
for page_info in pdf_info_dict:
paras_of_layout = page_info.get('para_blocks')
if not paras_of_layout:
continue
for para_block in paras_of_layout:
para_content = para_to_standard_format_v2(para_block,
img_buket_path)
content_list.append(para_content)
return content_list
def line_to_standard_format(line, img_buket_path):
line_text = ''
inline_equation_num = 0
for span in line['spans']:
if not span.get('content'):
if not span.get('image_path'):
continue
else:
if span['type'] == ContentType.Image:
content = {
'type': 'image',
'img_path': join_path(img_buket_path,
span['image_path']),
}
return content
elif span['type'] == ContentType.Table:
content = {
'type': 'table',
'img_path': join_path(img_buket_path,
span['image_path']),
}
return content
else:
if span['type'] == ContentType.InterlineEquation:
interline_equation = span['content']
content = {
'type': 'equation',
'latex': f'$$\n{interline_equation}\n$$'
}
return content
elif span['type'] == ContentType.InlineEquation:
inline_equation = span['content']
line_text += f'${inline_equation}$'
inline_equation_num += 1
elif span['type'] == ContentType.Text:
text_content = ocr_escape_special_markdown_char(
span['content']) # 转义特殊符号
line_text += text_content
content = {
'type': 'text',
'text': line_text,
'inline_equation_num': inline_equation_num,
}
return content
def ocr_mk_mm_standard_format(pdf_info_dict: list):
"""content_list type string
image/text/table/equation(行间的单独拿出来，行内的和text合并) latex string
latex文本字段。 text string 纯文本格式的文本数据。 md string
markdown格式的文本数据。 img_path string s3://full/path/to/img.jpg."""
content_list = []
for page_info in pdf_info_dict:
blocks = page_info.get('preproc_blocks')
if not blocks:
continue
for block in blocks:
for line in block['lines']:
content = line_to_standard_format(line)
content_list.append(content)
return content_list
def union_make(pdf_info_dict: list,
make_mode: str,
drop_mode: str,
img_buket_path: str = ''):
output_content = []
for page_info in pdf_info_dict:
if page_info.get('need_drop', False):
drop_reason = page_info.get('drop_reason')
if drop_mode == DropMode.NONE:
pass
elif drop_mode == DropMode.WHOLE_PDF:
raise Exception((f'drop_mode is {DropMode.WHOLE_PDF} ,'
f'drop_reason is {drop_reason}'))
elif drop_mode == DropMode.SINGLE_PAGE:
logger.warning((f'drop_mode is {DropMode.SINGLE_PAGE} ,'
f'drop_reason is {drop_reason}'))
continue
else:
raise Exception('drop_mode can not be null')
paras_of_layout = page_info.get('para_blocks')
#logger.info(f'paras_of_layout:\n{paras_of_layout}')
page_idx = page_info.get('page_idx')
if not paras_of_layout:
continue
if make_mode == MakeMode.MM_MD:
page_markdown = ocr_mk_markdown_with_para_core_v2(
paras_of_layout, 'mm', img_buket_path)
output_content.extend(page_markdown)
elif make_mode == MakeMode.NLP_MD:
page_markdown = ocr_mk_markdown_with_para_core_v2(
paras_of_layout, 'nlp')
output_content.extend(page_markdown)
elif make_mode == MakeMode.STANDARD_FORMAT:
for para_block in paras_of_layout:
para_content = para_to_standard_format_v2(
para_block, img_buket_path, page_idx)
output_content.append(para_content)
if make_mode in [MakeMode.MM_MD, MakeMode.NLP_MD]:
return '\n\n'.join(output_content)
elif make_mode == MakeMode.STANDARD_FORMAT:
return output_content
magic_pdf/dict2md/ocr_server.py 0 → 100644
View file @ c9171d1f
# Copyright (c) Alibaba Cloud.
#
# This source code is licensed under the license found in the
# LICENSE file in the root directory of this source tree
import copy
import re
import gc
import torch
from argparse import ArgumentParser
from threading import Thread
from qwen_vl_utils import process_vision_info
from transformers import AutoProcessor, Qwen2VLForConditionalGeneration, TextIteratorStreamer
from fastapi import FastAPI
from pydantic import BaseModel
from typing import Optional
app = FastAPI()
DEFAULT_CKPT_PATH = '/home/practice/model/Qwen2-VL-7B-Instruct'
REVISION = 'v1.0.4'
BOX_TAG_PATTERN = r"<box>([\s\S]*?)</box>"
PUNCTUATION = "！？。＂＃＄％＆＇（）＊＋，－／：；＜＝＞＠［＼］＾＿｀｛｜｝～｟｠｢｣､、〃》「」『』【】〔〕〖〗〘〙〚〛〜〝〞〟〰〾〿–—‘’‛“”„‟…‧﹏."
def _get_args():
parser = ArgumentParser()
parser.add_argument('-c', '--checkpoint_path', type=str, default=DEFAULT_CKPT_PATH,
help='Checkpoint name or path, default to %(default)r')
parser.add_argument('--cpu_only', action='store_true', help='Run demo with CPU only')
parser.add_argument('--flash_attn2', action='store_true', default=False,
help='Enable flash_attention_2 when loading the model.')
parser.add_argument('--share', action='store_true', default=False,
help='Create a publicly shareable link for the interface.')
parser.add_argument('--inbrowser', action='store_true', default=False,
help='Automatically launch the interface in a new tab on the default browser.')
parser.add_argument('--server_port', type=int, default=6020, help='Demo server port.')
parser.add_argument('--server_name', type=str, default='127.0.0.1', help='Demo server name.')
parser.add_argument('--dcu_id', type=str, default='0', help='Specify the GPU ID to load the model onto.')
args = parser.parse_args()
return args
def _load_model_processor(args):
if args.cpu_only:
device_map = 'cpu'
else:
if args.dcu_id is not None:
device_map = {'': f'cuda:{args.dcu_id}'}
print('使用DCU推理:', f'cuda:{args.dcu_id}')
else:
device_map = 'auto'
if args.flash_attn2:
model = Qwen2VLForConditionalGeneration.from_pretrained(
args.checkpoint_path,
torch_dtype=torch.float16,
attn_implementation='flash_attention_2',
device_map=device_map
)
else:
model = Qwen2VLForConditionalGeneration.from_pretrained(
args.checkpoint_path,
torch_dtype=torch.float16,
device_map=device_map
)
processor = AutoProcessor.from_pretrained(args.checkpoint_path)
return model, processor
def _parse_text(text):
lines = text.split("\n")
lines = [line for line in lines if line.strip() != ""] # 去除空行
count = 0
parsed_lines = []
for i, line in enumerate(lines):
if "```" in line:
count += 1
items = line.split("`")
if count % 2 == 1:
# 开始代码块
parsed_lines.append(f'<pre><code class="language-{items[-1]}">')
else:
# 结束代码块
parsed_lines.append(f"</code></pre>")
else:
if i > 0 and count % 2 == 1:
# 转义代码块内的特殊字符
line = line.replace("`", r"\`")
line = line.replace("<", "&lt;")
line = line.replace(">", "&gt;")
line = line.replace(" ", "&nbsp;")
line = line.replace("*", "&ast;")
line = line.replace("_", "&lowbar;")
line = line.replace("-", "&#45;")
line = line.replace(".", "&#46;")
line = line.replace("!", "&#33;")
line = line.replace("(", "&#40;")
line = line.replace(")", "&#41;")
line = line.replace("$", "&#36;")
# 使用空格连接行
if parsed_lines:
parsed_lines[-1] += " " + line
else:
parsed_lines.append(line)
text = "".join(parsed_lines)
return text
def _remove_image_special(text):
text = text.replace('<ref>', '').replace('</ref>', '')
return re.sub(r'<box>.*?(</box>|$)', '', text)
def _is_video_file(filename):
video_extensions = ['.mp4', '.avi', '.mkv', '.mov', '.wmv', '.flv', '.webm', '.mpeg']
return any(filename.lower().endswith(ext) for ext in video_extensions)
def _transform_messages(original_messages):
transformed_messages = []
for message in original_messages:
new_content = []
for item in message['content']:
if 'image' in item:
new_item = {'type': 'image', 'image': item['image']}
elif 'text' in item:
new_item = {'type': 'text', 'text': item['text']}
elif 'video' in item:
new_item = {'type': 'video', 'video': item['video']}
else:
continue
new_content.append(new_item)
new_message = {'role': message['role'], 'content': new_content}
transformed_messages.append(new_message)
return transformed_messages
def _gc():
gc.collect()
if torch.cuda.is_available():
torch.cuda.empty_cache()
def call_local_model(model, processor, messages):
messages = _transform_messages(messages)
text = processor.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
image_inputs, video_inputs = process_vision_info(messages)
inputs = processor(text=[text], images=image_inputs, videos=video_inputs, padding=True, return_tensors='pt')
inputs = inputs.to(model.device)
tokenizer = processor.tokenizer
streamer = TextIteratorStreamer(tokenizer, timeout=20.0, skip_prompt=True, skip_special_tokens=True)
gen_kwargs = {'max_new_tokens': 512, 'streamer': streamer, **inputs}
thread = Thread(target=model.generate, kwargs=gen_kwargs)
thread.start()
generated_text = ''
for new_text in streamer:
generated_text += new_text
yield _parse_text(generated_text)
def create_predict_fn(model, processor):
def predict(_chatbot, task_history):
chat_query = _chatbot[-1][0]
query = task_history[-1][0]
if len(chat_query) == 0:
_chatbot.pop()
task_history.pop()
return _chatbot
print('User: ' + _parse_text(query))
history_cp = copy.deepcopy(task_history)
full_response = ''
messages = []
content = []
for q, a in history_cp:
if isinstance(q, (tuple, list)):
if _is_video_file(q[0]):
content.append({'video': f'file://{q[0]}'})
else:
content.append({'image': f'file://{q[0]}'})
else:
content.append({'text': q})
messages.append({'role': 'user', 'content': content})
messages.append({'role': 'assistant', 'content': [{'text': a}]})
content = []
messages.pop()
for response in call_local_model(model, processor, messages):
_chatbot[-1] = (_parse_text(chat_query), _remove_image_special(_parse_text(response)))
yield _chatbot
full_response = _parse_text(response)
task_history[-1] = (query, full_response)
print('Qwen-VL-Chat: ' + _parse_text(full_response))
yield _chatbot
return predict
# 启用加载模型
args = _get_args()
model, processor = _load_model_processor(args)
class Item(BaseModel):
image_path: str
text: str
@app.post("/predict")
async def predict(item: Item):
messages = [
{
'role': 'user',
'content': [
{'image': item.image_path},
{'text': item.text}
]
}
]
generated_text = ''
for response in call_local_model(model, processor, messages):
generated_text = _parse_text(response)
_gc()
return {"Generated Text": generated_text}
if __name__ == "__main__":
import uvicorn
args = _get_args()
uvicorn.run(app, host=args.server_name, port=args.server_port)
magic_pdf/dict2md/ocr_server_72.py 0 → 100644
View file @ c9171d1f
# Copyright (c) Alibaba Cloud.
#
# This source code is licensed under the license found in the
# LICENSE file in the root directory of this source tree
import copy
import re
import gc
import torch
from argparse import ArgumentParser
from threading import Thread
from qwen_vl_utils import process_vision_info
from transformers import AutoProcessor, Qwen2VLForConditionalGeneration, TextIteratorStreamer
import os
from fastapi import FastAPI
from pydantic import BaseModel
from typing import Optional
app = FastAPI()
DEFAULT_CKPT_PATH = '/home/practice/model/Qwen2-VL-72B-Instruct'
REVISION = 'v1.0.4'
BOX_TAG_PATTERN = r"<box>([\s\S]*?)</box>"
PUNCTUATION = "！？。＂＃＄％＆＇（）＊＋，－／：；＜＝＞＠［＼］＾＿｀｛｜｝～｟｠｢｣､、〃》「」『』【】〔〕〖〗〘〙〚〛〜〝〞〟〰〾〿–—‘’‛“”„‟…‧﹏."
def get_args():
parser = ArgumentParser()
parser.add_argument('-c', '--checkpoint_path', type=str, default=DEFAULT_CKPT_PATH,
help='Checkpoint name or path, default to %(default)r')
parser.add_argument('--cpu_only', action='store_true', help='Run demo with CPU only')
parser.add_argument('--flash_attn2', action='store_true', default=False,
help='Enable flash_attention_2 when loading the model.')
parser.add_argument('--share', action='store_true', default=False,
help='Create a publicly shareable link for the interface.')
parser.add_argument('--inbrowser', action='store_true', default=False,
help='Automatically launch the interface in a new tab on the default browser.')
parser.add_argument('--server_port', type=int, default=6020, help='Demo server port.')
parser.add_argument('--server_name', type=str, default='127.0.0.1', help='Demo server name.')
parser.add_argument('--dcu_id', type=str, default=None, help='Specify the GPU ID to load the model onto.')
args = parser.parse_args()
return args
def load_model_processor(args):
if args.cpu_only:
device_map = 'cpu'
else:
os.environ['CUDA_VISIBLE_DEVICES'] = args.dcu_id
device_map = 'auto'
print(f"Visible CUDA devices: {os.environ.get('CUDA_VISIBLE_DEVICES')}")
if args.flash_attn2:
model = Qwen2VLForConditionalGeneration.from_pretrained(
args.checkpoint_path,
torch_dtype=torch.float16,
attn_implementation='flash_attention_2',
device_map=device_map
)
else:
model = Qwen2VLForConditionalGeneration.from_pretrained(
args.checkpoint_path,
torch_dtype=torch.float16,
device_map=device_map
)
processor = AutoProcessor.from_pretrained(args.checkpoint_path)
return model, processor
def parse_text(text):
lines = text.split("\n")
lines = [line for line in lines if line.strip() != ""] # 去除空行
count = 0
parsed_lines = []
for i, line in enumerate(lines):
if "```" in line:
count += 1
items = line.split("`")
if count % 2 == 1:
# 开始代码块
parsed_lines.append(f'<pre><code class="language-{items[-1]}">')
else:
# 结束代码块
parsed_lines.append(f"</code></pre>")
else:
if i > 0 and count % 2 == 1:
# 转义代码块内的特殊字符
line = line.replace("`", r"\`")
line = line.replace("<", "&lt;")
line = line.replace(">", "&gt;")
line = line.replace(" ", "&nbsp;")
line = line.replace("*", "&ast;")
line = line.replace("_", "&lowbar;")
line = line.replace("-", "&#45;")
line = line.replace(".", "&#46;")
line = line.replace("!", "&#33;")
line = line.replace("(", "&#40;")
line = line.replace(")", "&#41;")
line = line.replace("$", "&#36;")
# 使用空格连接行
if parsed_lines:
parsed_lines[-1] += " " + line
else:
parsed_lines.append(line)
text = "".join(parsed_lines)
return text
def unparse_text(parsed_text):
in_code_block = False
lines = parsed_text.split("\n")
unparsed_lines = []
for line in lines:
if "<pre><code" in line:
in_code_block = True
# 移除开始标签
line = line.split(">", 1)[1]
elif "</code></pre>" in line:
in_code_block = False
# 移除结束标签
line = line.rsplit("<", 1)[0]
# 反转 HTML 实体
line = line.replace("&lt;", "<")
line = line.replace("&gt;", ">")
line = line.replace("&nbsp;", " ")
line = line.replace("&ast;", "*")
line = line.replace("&lowbar;", "_")
line = line.replace("&#45;", "-")
line = line.replace("&#46;", ".")
line = line.replace("&#33;", "!")
line = line.replace("&#40;", "(")
line = line.replace("&#41;", ")")
line = line.replace("&#36;", "$")
# 如果在代码块内，还原反斜杠转义
if in_code_block:
line = line.replace(r"\`", "`")
unparsed_lines.append(line)
# 合并所有行
unparsed_text = "\n".join(unparsed_lines)
return unparsed_text
def remove_image_special(text):
text = text.replace('<ref>', '').replace('</ref>', '')
return re.sub(r'<box>.*?(</box>|$)', '', text)
def is_video_file(filename):
video_extensions = ['.mp4', '.avi', '.mkv', '.mov', '.wmv', '.flv', '.webm', '.mpeg']
return any(filename.lower().endswith(ext) for ext in video_extensions)
def transform_messages(original_messages):
transformed_messages = []
for message in original_messages:
new_content = []
for item in message['content']:
if 'image' in item:
new_item = {'type': 'image', 'image': item['image']}
elif 'text' in item:
new_item = {'type': 'text', 'text': item['text']}
elif 'video' in item:
new_item = {'type': 'video', 'video': item['video']}
else:
continue
new_content.append(new_item)
new_message = {'role': message['role'], 'content': new_content}
transformed_messages.append(new_message)
return transformed_messages
def _gc():
gc.collect()
if torch.cuda.is_available():
torch.cuda.empty_cache()
def call_local_model(model, processor, messages):
messages = transform_messages(messages)
text = processor.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
image_inputs, video_inputs = process_vision_info(messages)
inputs = processor(text=[text], images=image_inputs, videos=video_inputs, padding=True, return_tensors='pt')
inputs = inputs.to(model.device)
tokenizer = processor.tokenizer
streamer = TextIteratorStreamer(tokenizer, timeout=2000.0, skip_prompt=True, skip_special_tokens=True)
gen_kwargs = {'max_new_tokens': 512, 'streamer': streamer, **inputs}
thread = Thread(target=model.generate, kwargs=gen_kwargs)
thread.start()
generated_text = ''
for new_text in streamer:
generated_text += new_text
yield parse_text(generated_text)
def create_predict_fn(model, processor):
def predict(_chatbot, task_history):
chat_query = _chatbot[-1][0]
query = task_history[-1][0]
if len(chat_query) == 0:
_chatbot.pop()
task_history.pop()
return _chatbot
print('User: ' + parse_text(query))
history_cp = copy.deepcopy(task_history)
full_response = ''
messages = []
content = []
for q, a in history_cp:
if isinstance(q, (tuple, list)):
if is_video_file(q[0]):
content.append({'video': f'file://{q[0]}'})
else:
content.append({'image': f'file://{q[0]}'})
else:
content.append({'text': q})
messages.append({'role': 'user', 'content': content})
messages.append({'role': 'assistant', 'content': [{'text': a}]})
content = []
messages.pop()
for response in call_local_model(model, processor, messages):
_chatbot[-1] = (parse_text(chat_query), remove_image_special(parse_text(response)))
yield _chatbot
full_response = parse_text(response)
task_history[-1] = (query, full_response)
print('Qwen-VL-Chat: ' + unparse_text(full_response))
yield _chatbot
return predict
# 启用加载模型
args = get_args()
model, processor = load_model_processor(args)
class Item(BaseModel):
image_path: str
text: str
@app.post("/predict")
async def predict(item: Item):
messages = [
{
'role': 'user',
'content': [
{'image': item.image_path},
{'text': item.text}
]
}
]
generated_text = ''
for response in call_local_model(model, processor, messages):
generated_text = unparse_text(response)
_gc()
return {"Generated Text": generated_text}
if __name__ == "__main__":
import uvicorn
args = get_args()
uvicorn.run(app, host=args.server_name, port=args.server_port)
magic_pdf/dict2md/tmp.py 0 → 100644
View file @ c9171d1f
# -*- coding: utf-8 -*-
import time
import os
import ocr_client
text = '识别图片的内容，如果是发票就执行以下操作识别图中的文字信息，并以json格式返回，如果不是发票返回False'
image_dir = '/home/practice/dcu_Mineru/images'
false_count = 0
file_count = 0
true_count = 0
for root, dirs, files in os.walk(image_dir):
for file in files:
# 构建完整的文件路径
file_path = os.path.join(root, file)
# 打印文件名
image_path = file_path
start = time.time()
ocr_client.compress_image(image_path)
generated_text = ocr_client.PredictClient('http://127.0.0.1:6020').predict(image_path, text)
if generated_text == 'False' or generated_text == 'false':
false_count += 1
else:
print(f'{image_path}提取到的内容是：\n{generated_text}')
true_count += 1
file_count +=1
print(f'总共{file_count}个文件，其中qwen返回{false_count}个False，返回{true_count}个json')
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment