Compare commits
| Author | SHA1 | Date | |
|---|---|---|---|
|
|
17b25c7d9b | ||
|
|
b7fab64801 | ||
|
|
609bcda76a | ||
|
|
ef230847ac | ||
|
|
d1d1b23f2f | ||
|
|
abfbe98897 | ||
|
|
e7701ce2ac | ||
|
|
c239f79b2c |
No files matched your search
+3
-2
@@ -16,7 +16,7 @@ a interpreter of the pdf-standard commands
|
||||
|
||||
### Example
|
||||
```py
|
||||
from pdfmajor.interpreter import PageInterpreter
|
||||
from pdfmajor.interpreter import PDFInterpreter
|
||||
|
||||
for page in PDFInterpreter("/path/to/pdf.pdf"):
|
||||
print("page start", page.page_num)
|
||||
@@ -58,7 +58,8 @@ All layout items extend the `LTItem` class. There are two kinds of layout items:
|
||||
All of these classes extend the LTContainer class.
|
||||
|
||||
- LTXObject: a layout item containing other additional layout items
|
||||
- LTCharBlock: a layout item containing LTChars
|
||||
- LTCharBlock: a layout item containing LTChars, this corresponds to whenever a `TJ` or `Tj` operators is issued within a text object.
|
||||
- LTTextBlock: a layout item containing LTCharBlocks, note that this directly corresponds to the `BT` and `ET` operators pair in the pdf standard
|
||||
|
||||
#### Layout Components
|
||||
All of these classes extend the LTComponent class.
|
||||
|
||||
+1
-1
@@ -19,7 +19,7 @@ See documentation [here](https://asosnovsky.github.io/pdfmajor/).
|
||||
|
||||
## Basic Example
|
||||
```py
|
||||
from pdfmajor.interpreter import PageInterpreter
|
||||
from pdfmajor.interpreter import PDFInterpreter
|
||||
|
||||
for page in PDFInterpreter("/path/to/pdf.pdf"):
|
||||
print("page start", page.page_num)
|
||||
|
||||
+71
-65
@@ -62,14 +62,14 @@
|
||||
|
||||
<ul>
|
||||
|
||||
<li><a class="toctree-l3" href="#parser">parser</a></li>
|
||||
|
||||
<li><a class="toctree-l3" href="#interpreter">interpreter</a></li>
|
||||
|
||||
<li><a class="toctree-l3" href="#converters">converters</a></li>
|
||||
|
||||
<li><a class="toctree-l3" href="#imagewriter">imagewriter</a></li>
|
||||
|
||||
<li><a class="toctree-l3" href="#layouts">layouts</a></li>
|
||||
|
||||
</ul>
|
||||
|
||||
|
||||
@@ -111,37 +111,82 @@
|
||||
<h1 id="api">API</h1>
|
||||
<p>The library is constructed from 4 base modules:</p>
|
||||
<ul>
|
||||
<li><strong>interpreter</strong>: contains low-level classes for extracting fundamental data-structures from the documents</li>
|
||||
<li><strong>converters</strong>: contains high-level classes for conversion of the fundamental pdf structures to other formats</li>
|
||||
<li><strong>parser</strong>: contains low-level classes for extracting fundamental data-structures from the documents</li>
|
||||
<li><strong>interpreter</strong>: a interpreter of the pdf-standard commands</li>
|
||||
<li><strong>converters</strong>: contains high-level functions for conversion of the fundamental pdf structures to other formats</li>
|
||||
<li><strong>imagewriter</strong>: contains a simple implementation for converting PDF Image Streams to png/bmp/img formats</li>
|
||||
<li><strong>layouts</strong>: contains an abstraction layer for the data-structures found in the pdf</li>
|
||||
</ul>
|
||||
<h2 id="parser">parser</h2>
|
||||
<p>WIP</p>
|
||||
<h2 id="interpreter">interpreter</h2>
|
||||
<pre><code>WIP
|
||||
</code></pre>
|
||||
<h2 id="converters">converters</h2>
|
||||
<p>Contains high-level classes for conversion of the fundamental pdf structures to other formats. This library includes 4 high-level conversion classes:</p>
|
||||
<ul>
|
||||
<li>HTMLConverter</li>
|
||||
<li>JSONConverter</li>
|
||||
<li>XMLConverter</li>
|
||||
<li>TextConverter</li>
|
||||
</ul>
|
||||
<p>These classes all extend the <a href="#converters.PDFConverter">PDFConverter</a>. To use them simply call the static method <a href="#converterspdfconverterparse_fileinput_file-output_file-image_folder_path-codec-maxpages-password-caching-check_extractable-pagenos">parse_file</a>.</p>
|
||||
<p>a interpreter of the pdf-standard commands</p>
|
||||
<h3 id="example">Example</h3>
|
||||
<pre><code class="py">from pdfmajor.converters import HTMLConverter
|
||||
<pre><code class="py">from pdfmajor.interpreter import PageInterpreter
|
||||
|
||||
input_file = open('path/to/input/file.pdf', 'rb')
|
||||
output_file = open('path/to/output/file.html', 'wb')
|
||||
HTMLConverter.parse_file(
|
||||
input_file,
|
||||
output_file
|
||||
)
|
||||
input_file.close()
|
||||
output_file.close()
|
||||
for page in PDFInterpreter("/path/to/pdf.pdf"):
|
||||
print("page start", page.page_num)
|
||||
for item in page:
|
||||
print(" >", item)
|
||||
print("page end", page.page_num)
|
||||
</code></pre>
|
||||
|
||||
<h3 id="convertersconvert_fileinput_file-output_file-image_folder_path-codec-maxpages-password-caching-check_extractable-pagenos-out_type">converters.convert_file(input_file, output_file, image_folder_path, codec, maxpages, password, caching, check_extractable, pagenos, out_type)</h3>
|
||||
<h3 id="interpreterpdfinterpreter">interpreter.PDFInterpreter</h3>
|
||||
<p>This generator-function yields individual pages which contain their respected items.</p>
|
||||
<h4 id="arguments">Arguments</h4>
|
||||
<ul>
|
||||
<li><code>input_file_path</code>: <a href="#">str</a></li>
|
||||
<li><code>preload</code>: <a href="#">bool</a> defaults to False</li>
|
||||
<li><code>maxpages</code>: <a href="#">int</a> defaults to 0 </li>
|
||||
<li><code>password</code>: <a href="#">str</a> defaults to None </li>
|
||||
<li><code>caching</code>: <a href="#">bool</a> defaults to True </li>
|
||||
<li><code>check_extractable</code>: <a href="#">bool</a> defaults to True</li>
|
||||
<li><code>pagenos</code>: <a href="#">List[int]</a> defaults to None</li>
|
||||
<li><code>debug_level</code>: <a href="#https://docs.python.org/3/library/logging.html#levels">logging.levels</a> defaults logging.WARNING</li>
|
||||
</ul>
|
||||
<h4 id="yield-value">Yield Value</h4>
|
||||
<p>This function returns a generator that yields <a href="#interpreterpageinterpreter">PDFInterpreter</a>.</p>
|
||||
<h3 id="interpreterpageinterpreter">interpreter.PageInterpreter</h3>
|
||||
<p>This generator-function-class yields individual <a href="#layout-items">layout items</a>.</p>
|
||||
<h3 id="layout-items">Layout Items</h3>
|
||||
<p>All layout items extend the <code>LTItem</code> class. There are two kinds of layout items:</p>
|
||||
<ul>
|
||||
<li>LTComponent: extends the base <code>LTItem</code> class, this class will have additional values such as boundary boxes, height and width</li>
|
||||
<li>LTContainer: extends the <code>LTComponent</code> class, this class is used to contain elements of the pdf that would have child elements. Iterating on this element will output its child elements.</li>
|
||||
</ul>
|
||||
<h4 id="layout-containers">Layout Containers</h4>
|
||||
<p>All of these classes extend the LTContainer class.</p>
|
||||
<ul>
|
||||
<li>LTXObject: a layout item containing other additional layout items</li>
|
||||
<li>LTCharBlock: a layout item containing LTChars, this corresponds to whenever a <code>TJ</code> or <code>Tj</code> operators is issued within a text object.</li>
|
||||
<li>LTTextBlock: a layout item containing LTCharBlocks, note that this directly corresponds to the <code>BT</code> and <code>ET</code> operators pair in the pdf standard</li>
|
||||
</ul>
|
||||
<h4 id="layout-components">Layout Components</h4>
|
||||
<p>All of these classes extend the LTComponent class.</p>
|
||||
<ul>
|
||||
<li>LTChar: an individual character</li>
|
||||
<li>LTCurves: represents a collection of svg-paths (available under <code>self.paths</code>)</li>
|
||||
<li>LTImage: a component containing information regarding an image</li>
|
||||
</ul>
|
||||
<h2 id="converters">converters</h2>
|
||||
<p>Contains high-level functions for conversion of the fundamental pdf structures to other formats. This library includes 4 high-level conversion cases:</p>
|
||||
<ul>
|
||||
<li>HTML</li>
|
||||
<li>JSON</li>
|
||||
<li>XML</li>
|
||||
<li>Text</li>
|
||||
</ul>
|
||||
<p>These formats are all generated using the <a href="#PDFInterpreter">PageInterpreter</a>. To use them simply call the static method <a href="#converterspdfconverter">parse_file</a>.</p>
|
||||
<h3 id="example_1">Example</h3>
|
||||
<pre><code class="py">from pdfmajor.converters import convert_file
|
||||
|
||||
convert_file(
|
||||
"path/to/input/file.pdf",
|
||||
"path/to/output/file.html",
|
||||
out_type="html"
|
||||
)
|
||||
</code></pre>
|
||||
|
||||
<h3 id="convertersconvert_file">converters.convert_file</h3>
|
||||
<p>A high-level abstraction for the conversion classes.</p>
|
||||
<ul>
|
||||
<li><code>input_file</code>: <a href="https://docs.python.org/3/library/io.html#io.TextIOWrapper">TextIOWrapper</a> </li>
|
||||
@@ -155,46 +200,7 @@ output_file.close()
|
||||
<li><code>pagenos</code>: <a href="#">List[int]</a> defaults to None</li>
|
||||
<li><code>out_type</code>: <a href="#">str</a> defaults to 'html'</li>
|
||||
</ul>
|
||||
<h3 id="example-convertersconvert_file">Example (converters.convert_file)</h3>
|
||||
<pre><code class="py">from pdfmajor.converters import convert_file
|
||||
|
||||
input_file = open('path/to/input/file.pdf', 'rb')
|
||||
output_file = open('path/to/output/file.html', 'wb')
|
||||
convert_file(
|
||||
input_file,
|
||||
output_file
|
||||
)
|
||||
input_file.close()
|
||||
output_file.close()
|
||||
</code></pre>
|
||||
|
||||
<h3 id="converterspdfconverter">converters.PDFConverter</h3>
|
||||
<p>Extends <code>layouts.PDFLayoutAnalyzer</code>.</p>
|
||||
<h4 id="converterspdfconverter9595init9595rsrcmgr-outfp-imagewriter-codec-pageno">converters.PDFConverter.__init__(rsrcmgr, outfp, imagewriter, codec, pageno)</h4>
|
||||
<ul>
|
||||
<li><code>rsrcmgr</code>: <a href="#interpreter.PDFResourceManager">PDFResourceManager</a></li>
|
||||
<li><code>outfp</code>: <a href="https://docs.python.org/3/library/io.html#io.TextIOWrapper">TextIOWrapper</a></li>
|
||||
<li><code>imagewriter</code>: <a href="#imagewriter.ImageWriter">ImageWriter</a></li>
|
||||
<li><code>codec</code>: <a href="#">str</a> defaults to 'utf-8'</li>
|
||||
<li><code>pageno</code>: <a href="#">int</a> defaults to 1</li>
|
||||
</ul>
|
||||
<p>initialization function for the class</p>
|
||||
<h4 id="converterspdfconverterparse_fileinput_file-output_file-image_folder_path-codec-maxpages-password-caching-check_extractable-pagenos">converters.PDFConverter.parse_file(input_file, output_file, image_folder_path, codec, maxpages, password, caching, check_extractable, pagenos)</h4>
|
||||
<p>returns the output file <a href="https://docs.python.org/3/library/io.html#io.TextIOWrapper">TextIOWrapper</a>.</p>
|
||||
<ul>
|
||||
<li><code>input_file</code>: <a href="https://docs.python.org/3/library/io.html#io.TextIOWrapper">TextIOWrapper</a> </li>
|
||||
<li><code>output_file</code>: <a href="https://docs.python.org/3/library/io.html#io.TextIOWrapper">TextIOWrapper</a> </li>
|
||||
<li><code>image_folder_path</code>: <a href="#">str</a> defaults to None</li>
|
||||
<li><code>codec</code>: <a href="#">str</a> defaults to 'utf-8'</li>
|
||||
<li><code>maxpages</code>: <a href="#">int</a> defaults to 0 </li>
|
||||
<li><code>password</code>: <a href="#">str</a> defaults to None </li>
|
||||
<li><code>caching</code>: <a href="#">bool</a> defaults to True </li>
|
||||
<li><code>check_extractable</code>: <a href="#">bool</a> defaults to True</li>
|
||||
<li><code>pagenos</code>: <a href="#">List[int]</a> defaults to None</li>
|
||||
</ul>
|
||||
<h2 id="imagewriter">imagewriter</h2>
|
||||
<p>WIP </p>
|
||||
<h2 id="layouts">layouts</h2>
|
||||
<p>WIP </p>
|
||||
|
||||
</div>
|
||||
|
||||
+20
-7
@@ -59,6 +59,8 @@
|
||||
|
||||
<li><a class="toctree-l3" href="#documentation">Documentation</a></li>
|
||||
|
||||
<li><a class="toctree-l3" href="#basic-example">Basic Example</a></li>
|
||||
|
||||
<li><a class="toctree-l3" href="#why-another-fork">Why Another Fork?</a></li>
|
||||
|
||||
<li><a class="toctree-l3" href="#features">Features</a></li>
|
||||
@@ -122,23 +124,34 @@
|
||||
</tr>
|
||||
</table>
|
||||
|
||||
<p>PDF-Major is a fork of <code>PDFMiner.six</code>. It is meant to be a more light-weight implementation that makes fewer assumptions on the data. Additionally, this version intends to be more up to date with recent versions of python and make liberal use of the typing capabilities of python for improved future maintainability.</p>
|
||||
<p>PDF-Major is a fork of <code>PDFMiner.six</code>. It is meant to be a more light-weight implementation that makes fewer assumptions on the data. The ultimate goal of this project is to provide a simpler, faster and more functional library that both casual and low-level developers could build upon.</p>
|
||||
<h2 id="documentation">Documentation</h2>
|
||||
<p>See documentation <a href="https://asosnovsky.github.io/pdfmajor/">here</a>.</p>
|
||||
<h2 id="basic-example">Basic Example</h2>
|
||||
<pre><code class="py">from pdfmajor.interpreter import PageInterpreter
|
||||
|
||||
for page in PDFInterpreter("/path/to/pdf.pdf"):
|
||||
print("page start", page.page_num)
|
||||
for item in page:
|
||||
print(" >", item)
|
||||
print("page end", page.page_num)
|
||||
|
||||
</code></pre>
|
||||
|
||||
<h2 id="why-another-fork">Why Another Fork?</h2>
|
||||
<p><code>PDFMiner</code> was designed to run in python 2.7>, <code>PDFMiner.six</code> was written to bring cross-version support for python 2 and 3. This versions intends to make use of the latest improvement in the python language while remaining fairly simple to develop with. </p>
|
||||
<p><code>PDFMiner</code> was designed to run in python 2.7>, <code>PDFMiner.six</code> was written to bring cross-version support for python 2 and 3. However, after attempting to build and extend upon <code>PDFMiner.six</code> I found it rather difficult to extend and identify additional details regarding the items encoded in the pdf. This library will attempt to expose as much information as possible to end-users, without having them solely rely on to-xml or to-json exports.</p>
|
||||
<h2 id="features">Features</h2>
|
||||
<ul>
|
||||
<li>Written entirely in Python.</li>
|
||||
<li>Functional extraction method based on generators</li>
|
||||
<li>Parse, analyze, and convert PDF documents.</li>
|
||||
<li>PDF-1.7 specification support. (well, almost)</li>
|
||||
<li>Font-Color extraction</li>
|
||||
<li>Shape fill and stroke color extraction</li>
|
||||
<li>CJK languages and vertical writing scripts support.</li>
|
||||
<li>Various font types (Type1, TrueType, Type3, and CID) support.</li>
|
||||
<li>Basic encryption (RC4) support.</li>
|
||||
<li>Font-Color extraction</li>
|
||||
<li>Shape fill and stroke color extraction</li>
|
||||
</ul>
|
||||
<p><strong>Note</strong>: We took out the layout-analysis process in this version (there is no more LTTextHorizontal or LTTextVertical). While the mathematics behind the grouping process was sound, the coupling of the layout-analysis process with the parsing and interpretation process produced unfriendly-code. This feature could be brought back by extending the <code>pdfmajor.layouts.PDFLayoutAnalyzer</code> class, but at the current time is not supported.</p>
|
||||
<p><strong>Note</strong>: We took out the layout-analysis process in this version (there is no more LTTextHorizontal or LTTextVertical). While the mathematics behind the grouping process was sound, the coupling of the layout-analysis process with the parsing and interpretation process produced unfriendly-code. This feature could be brought back by running your own implementation of it on the <code>pdfmajor.interpreter.PageInterpreter</code> class, but at the current time is not supported.</p>
|
||||
<h2 id="how-to-install">How to Install</h2>
|
||||
<h3 id="source">Source</h3>
|
||||
<ul>
|
||||
@@ -227,5 +240,5 @@ SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.</p>
|
||||
|
||||
<!--
|
||||
MkDocs version : 1.0.4
|
||||
Build Date UTC : 2018-11-04 21:27:38
|
||||
Build Date UTC : 2018-12-24 06:52:28
|
||||
-->
|
||||
@@ -1 +1 @@
|
||||
{"config":{"lang":["en"],"prebuild_index":false,"separator":"[\\s\\-]+"},"docs":[{"location":"","text":"PDFMajor Latest Release PDF-Major is a fork of PDFMiner.six . It is meant to be a more light-weight implementation that makes fewer assumptions on the data. Additionally, this version intends to be more up to date with recent versions of python and make liberal use of the typing capabilities of python for improved future maintainability. Documentation See documentation here . Why Another Fork? PDFMiner was designed to run in python 2.7>, PDFMiner.six was written to bring cross-version support for python 2 and 3. This versions intends to make use of the latest improvement in the python language while remaining fairly simple to develop with. Features Written entirely in Python. Parse, analyze, and convert PDF documents. PDF-1.7 specification support. (well, almost) CJK languages and vertical writing scripts support. Various font types (Type1, TrueType, Type3, and CID) support. Basic encryption (RC4) support. Font-Color extraction Shape fill and stroke color extraction Note : We took out the layout-analysis process in this version (there is no more LTTextHorizontal or LTTextVertical). While the mathematics behind the grouping process was sound, the coupling of the layout-analysis process with the parsing and interpretation process produced unfriendly-code. This feature could be brought back by extending the pdfmajor.layouts.PDFLayoutAnalyzer class, but at the current time is not supported. How to Install Source Install Python 3.6.4 or newer. clone this repo git clone https://github.com/asosnovsky/pdfmajor * install repo python setup.py install Pypi Install Python 3.6.4 or newer. install repo pip install pdfmajor Terms and Conditions (This is so-called MIT/X License) Copyright (c) 2018-2019 Ariel Sosnovsky Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the \"Software\"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED \"AS IS\", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.","title":"Home"},{"location":"#pdfmajor","text":"Latest Release PDF-Major is a fork of PDFMiner.six . It is meant to be a more light-weight implementation that makes fewer assumptions on the data. Additionally, this version intends to be more up to date with recent versions of python and make liberal use of the typing capabilities of python for improved future maintainability.","title":"PDFMajor"},{"location":"#documentation","text":"See documentation here .","title":"Documentation"},{"location":"#why-another-fork","text":"PDFMiner was designed to run in python 2.7>, PDFMiner.six was written to bring cross-version support for python 2 and 3. This versions intends to make use of the latest improvement in the python language while remaining fairly simple to develop with.","title":"Why Another Fork?"},{"location":"#features","text":"Written entirely in Python. Parse, analyze, and convert PDF documents. PDF-1.7 specification support. (well, almost) CJK languages and vertical writing scripts support. Various font types (Type1, TrueType, Type3, and CID) support. Basic encryption (RC4) support. Font-Color extraction Shape fill and stroke color extraction Note : We took out the layout-analysis process in this version (there is no more LTTextHorizontal or LTTextVertical). While the mathematics behind the grouping process was sound, the coupling of the layout-analysis process with the parsing and interpretation process produced unfriendly-code. This feature could be brought back by extending the pdfmajor.layouts.PDFLayoutAnalyzer class, but at the current time is not supported.","title":"Features"},{"location":"#how-to-install","text":"","title":"How to Install"},{"location":"#source","text":"Install Python 3.6.4 or newer. clone this repo git clone https://github.com/asosnovsky/pdfmajor * install repo python setup.py install","title":"Source"},{"location":"#pypi","text":"Install Python 3.6.4 or newer. install repo pip install pdfmajor","title":"Pypi"},{"location":"#terms-and-conditions","text":"(This is so-called MIT/X License) Copyright (c) 2018-2019 Ariel Sosnovsky Permission is hereby granted, free of chargeLine truncated
|
||||
{"config":{"lang":["en"],"prebuild_index":false,"separator":"[\\s\\-]+"},"docs":[{"location":"","text":"PDFMajor Latest Release PDF-Major is a fork of PDFMiner.six . It is meant to be a more light-weight implementation that makes fewer assumptions on the data. The ultimate goal of this project is to provide a simpler, faster and more functional library that both casual and low-level developers could build upon. Documentation See documentation here . Basic Example from pdfmajor.interpreter import PageInterpreter for page in PDFInterpreter(\"/path/to/pdf.pdf\"): print(\"page start\", page.page_num) for item in page: print(\" >\", item) print(\"page end\", page.page_num) Why Another Fork? PDFMiner was designed to run in python 2.7>, PDFMiner.six was written to bring cross-version support for python 2 and 3. However, after attempting to build and extend upon PDFMiner.six I found it rather difficult to extend and identify additional details regarding the items encoded in the pdf. This library will attempt to expose as much information as possible to end-users, without having them solely rely on to-xml or to-json exports. Features Functional extraction method based on generators Parse, analyze, and convert PDF documents. PDF-1.7 specification support. (well, almost) Font-Color extraction Shape fill and stroke color extraction CJK languages and vertical writing scripts support. Various font types (Type1, TrueType, Type3, and CID) support. Basic encryption (RC4) support. Note : We took out the layout-analysis process in this version (there is no more LTTextHorizontal or LTTextVertical). While the mathematics behind the grouping process was sound, the coupling of the layout-analysis process with the parsing and interpretation process produced unfriendly-code. This feature could be brought back by running your own implementation of it on the pdfmajor.interpreter.PageInterpreter class, but at the current time is not supported. How to Install Source Install Python 3.6.4 or newer. clone this repo git clone https://github.com/asosnovsky/pdfmajor * install repo python setup.py install Pypi Install Python 3.6.4 or newer. install repo pip install pdfmajor Terms and Conditions (This is so-called MIT/X License) Copyright (c) 2018-2019 Ariel Sosnovsky Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the \"Software\"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED \"AS IS\", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.","title":"Home"},{"location":"#pdfmajor","text":"Latest Release PDF-Major is a fork of PDFMiner.six . It is meant to be a more light-weight implementation that makes fewer assumptions on the data. The ultimate goal of this project is to provide a simpler, faster and more functional library that both casual and low-level developers could build upon.","title":"PDFMajor"},{"location":"#documentation","text":"See documentation here .","title":"Documentation"},{"location":"#basic-example","text":"from pdfmajor.interpreter import PageInterpreter for page in PDFInterpreter(\"/path/to/pdf.pdf\"): print(\"page start\", page.page_num) for item in page: print(\" >\", item) print(\"page end\", page.page_num)","title":"Basic Example"},{"location":"#why-another-fork","text":"PDFMiner was designed to run in python 2.7>, PDFMiner.six was written to bring cross-version support for python 2 and 3. However, after attempting to build and extend upon PDFMiner.six I found it rather difficult to extend and identify additional details regarding the items encoded in the pdf. This library will attempt to expose as much information as possible to end-users, without having them solely rely on to-xml or to-json exports.","title":"Why Another Fork?"},{"location":"#features","text":"Functional extraction method based on generators Parse, analyze, and convert PDF documents. PDF-1.7 specification support. (well, almost) Font-Color extraction Shape fill and stroke color extraction CJK languages and vertical writing scripts support. Various font types (Type1, TrueType, Type3, and CID) support. Basic encryption (RC4) support. Note : We took out the layout-analysis process in this version (there is no more LTTLine truncated
|
||||
+2
-2
@@ -2,12 +2,12 @@
|
||||
<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9">
|
||||
<url>
|
||||
<loc>None</loc>
|
||||
<lastmod>2018-11-04</lastmod>
|
||||
<lastmod>2018-12-24</lastmod>
|
||||
<changefreq>daily</changefreq>
|
||||
</url>
|
||||
<url>
|
||||
<loc>None</loc>
|
||||
<lastmod>2018-11-04</lastmod>
|
||||
<lastmod>2018-12-24</lastmod>
|
||||
<changefreq>daily</changefreq>
|
||||
</url>
|
||||
</urlset>
|
||||
Binary file not shown.
@@ -0,0 +1,60 @@
|
||||
|
||||
import struct
|
||||
from io import BytesIO
|
||||
|
||||
from ..parser.PDFStream import LITERALS_DCT_DECODE, PDFStream
|
||||
|
||||
def align32(x):
|
||||
return ((x+3)//4)*4
|
||||
|
||||
## BMPWriter
|
||||
##
|
||||
class BMPWriter(object):
|
||||
|
||||
@classmethod
|
||||
def write_all(cls, fp, stream: PDFStream, bits: int, width: float, height: float, step: float):
|
||||
bmp = cls(fp, 1, width, height)
|
||||
data = stream.get_data()
|
||||
i = 0
|
||||
for y in range(height):
|
||||
bmp.write_line(y, data[i:i+step])
|
||||
i += step
|
||||
|
||||
def __init__(self, fp, bits, width, height):
|
||||
self.fp = fp
|
||||
self.bits = bits
|
||||
self.width = width
|
||||
self.height = height
|
||||
if bits == 1:
|
||||
ncols = 2
|
||||
elif bits == 8:
|
||||
ncols = 256
|
||||
elif bits == 24:
|
||||
ncols = 0
|
||||
else:
|
||||
raise ValueError(bits)
|
||||
self.linesize = align32((self.width*self.bits+7)//8)
|
||||
self.datasize = self.linesize * self.height
|
||||
headersize = 14+40+ncols*4
|
||||
info = struct.pack('<IiiHHIIIIII', 40, self.width, self.height, 1, self.bits, 0, self.datasize, 0, 0, ncols, 0)
|
||||
assert len(info) == 40, str(len(info))
|
||||
header = struct.pack('<ccIHHI', b'B', b'M', headersize+self.datasize, 0, 0, headersize)
|
||||
assert len(header) == 14, str(len(header))
|
||||
self.fp.write(header)
|
||||
self.fp.write(info)
|
||||
if ncols == 2:
|
||||
# B&W color table
|
||||
for i in (0, 255):
|
||||
self.fp.write(struct.pack('BBBx', i, i, i))
|
||||
elif ncols == 256:
|
||||
# grayscale color table
|
||||
for i in range(256):
|
||||
self.fp.write(struct.pack('BBBx', i, i, i))
|
||||
self.pos0 = self.fp.tell()
|
||||
self.pos1 = self.pos0 + self.datasize
|
||||
return
|
||||
|
||||
def write_line(self, y, data):
|
||||
self.fp.seek(self.pos1 - (y+1)*self.linesize)
|
||||
self.fp.write(data)
|
||||
return
|
||||
@@ -97,7 +97,7 @@ OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
|
||||
SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
|
||||
|
||||
"""
|
||||
__version__ = (1, 3, 0)
|
||||
__version__ = (1, 3, 2b)
|
||||
|
||||
if __name__ == '__main__':
|
||||
print(
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
from typing import List
|
||||
from .writers.html import HTMLMaker
|
||||
from ..interpreter import PDFInterpreter, PageInterpreter, logging
|
||||
from ..interpreter import LTImage, LTCharBlock, LTChar, LTCurve, LTXObject
|
||||
from ..interpreter import LTImage, LTTextBlock, LTCharBlock, LTChar, LTCurve, LTXObject
|
||||
from ..interpreter.commands import LTItem
|
||||
from ..interpreter.commands.state import CurvePath, PDFColor
|
||||
|
||||
@@ -76,8 +76,12 @@ def render_page(html: HTMLMaker, ltpage: PageInterpreter):
|
||||
'max-height': item.height,
|
||||
}):
|
||||
html.write(item.get_text())
|
||||
elif isinstance(item, LTTextBlock):
|
||||
for char in item:
|
||||
render(char)
|
||||
elif isinstance(item, LTCharBlock):
|
||||
with html.elm('span', { 'class': 'char-block' }, {
|
||||
with html.elm('span', {'class': 'char-block'}):
|
||||
with html.elm('p', { 'class': 'text-block' }, {
|
||||
'position': 'absolute',
|
||||
'left': f'{item.x0}',
|
||||
'bottom': f'{item.y0}',
|
||||
@@ -98,8 +102,8 @@ def render_page(html: HTMLMaker, ltpage: PageInterpreter):
|
||||
'justify-content': 'space-between',
|
||||
# 'padding-left': item.font.leading
|
||||
}):
|
||||
for char in item:
|
||||
render(char)
|
||||
for char in item:
|
||||
render(char)
|
||||
|
||||
with html.elm('div', { "class": 'page', "id": f"page-{ltpage.page_num}" }, {
|
||||
'width': f'{ltpage.width}px',
|
||||
|
||||
@@ -2,7 +2,7 @@ from typing import List
|
||||
|
||||
from .writers.json import JSONMaker, JSONMakerObject, JSONMakerArray
|
||||
from ..interpreter import PDFInterpreter, PageInterpreter, logging
|
||||
from ..interpreter import LTImage, LTCharBlock, LTChar, LTCurve, LTXObject
|
||||
from ..interpreter import LTImage, LTTextBlock, LTCharBlock, LTCurve, LTXObject
|
||||
from ..interpreter.commands import LTItem
|
||||
from ..interpreter.commands import LTRect, LTLine
|
||||
from ..interpreter.commands.state import CurvePath, PDFColor
|
||||
@@ -76,11 +76,21 @@ def render(json: JSONMakerObject, item: LTItem):
|
||||
json.number("x1", item.x1)
|
||||
json.number("y0", item.y0)
|
||||
json.number("y1", item.y1)
|
||||
elif isinstance(item, LTCharBlock):
|
||||
place_char_block(json, item)
|
||||
elif isinstance(item, LTTextBlock):
|
||||
place_text_block(json, item)
|
||||
|
||||
def place_text_block(json: JSONMakerObject, text_block: LTTextBlock):
|
||||
json.string("type", 'text-block')
|
||||
json.number("x0", text_block.x0)
|
||||
json.number("x1", text_block.x1)
|
||||
json.number("y0", text_block.y0)
|
||||
json.number("y1", text_block.y1)
|
||||
with json.array("char_blocks") as cb_arr:
|
||||
for cb in text_block:
|
||||
with cb_arr.object() as cb_obj:
|
||||
place_char_block(cb_obj, cb)
|
||||
|
||||
def place_char_block(json: JSONMakerObject, char_block: LTCharBlock):
|
||||
json.string("type", 'char-block')
|
||||
json.number("size", char_block.size)
|
||||
json.number("x0", char_block.x0)
|
||||
json.number("x1", char_block.x1)
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
from typing import List
|
||||
|
||||
from ..interpreter import PDFInterpreter, PageInterpreter, logging
|
||||
from ..interpreter import LTCharBlock, LTXObject
|
||||
from ..interpreter import LTTextBlock, LTXObject
|
||||
|
||||
|
||||
def convert_to_text(
|
||||
@@ -28,9 +28,10 @@ def convert_to_text(
|
||||
with open(output_file_path, 'wb') as outfp:
|
||||
def process_container(container: LTXObject):
|
||||
for item in container:
|
||||
if isinstance(item, LTCharBlock):
|
||||
for char in item:
|
||||
outfp.write(char.get_text().encode(codec))
|
||||
if isinstance(item, LTTextBlock):
|
||||
for text in item:
|
||||
for char in text:
|
||||
outfp.write(char.get_text().encode(codec))
|
||||
elif isinstance(item, LTXObject):
|
||||
process_container(item)
|
||||
for page in intepreter:
|
||||
|
||||
@@ -4,7 +4,7 @@ from xml.sax import saxutils
|
||||
|
||||
from .writers.xml import XMLMaker
|
||||
from ..interpreter import PDFInterpreter, PageInterpreter, logging
|
||||
from ..interpreter import LTImage, LTCharBlock, LTChar, LTCurve, LTXObject
|
||||
from ..interpreter import LTImage, LTTextBlock, LTCharBlock, LTChar, LTCurve, LTXObject
|
||||
from ..interpreter.commands import LTItem
|
||||
from ..interpreter.commands import LTRect, LTLine
|
||||
from ..interpreter.commands.state import CurvePath, PDFColor
|
||||
@@ -86,8 +86,8 @@ def render_page(xml: XMLMaker, ltpage: PageInterpreter):
|
||||
"y0": item.y0,
|
||||
"y1": item.y1,
|
||||
})
|
||||
elif isinstance(item, LTCharBlock):
|
||||
place_char_block(xml, item)
|
||||
elif isinstance(item, LTTextBlock):
|
||||
place_text_block(xml, item)
|
||||
with xml.elm('page', {
|
||||
"id": ltpage.page_num,
|
||||
"width": ltpage.width,
|
||||
@@ -121,6 +121,17 @@ def place_char_block(xml: XMLMaker, char_block: LTCharBlock):
|
||||
for char in char_block:
|
||||
place_char(xml, char)
|
||||
|
||||
def place_text_block(xml: XMLMaker, text_block: LTCharBlock):
|
||||
attr = {
|
||||
"x0": text_block.x0,
|
||||
"x1": text_block.x1,
|
||||
"y0": text_block.y0,
|
||||
"y1": text_block.y1,
|
||||
}
|
||||
with xml.elm("text-block", attr):
|
||||
for char_block in text_block:
|
||||
place_char_block(xml, char_block)
|
||||
|
||||
def place_curve(xml: XMLMaker, item: LTCurve):
|
||||
attr = {
|
||||
"x0": item.x0,
|
||||
|
||||
@@ -1,127 +0,0 @@
|
||||
|
||||
import struct
|
||||
import os
|
||||
import os.path
|
||||
from io import BytesIO
|
||||
from ..interpreter.PDFStream import LITERALS_DCT_DECODE
|
||||
from ..interpreter.PDFColorSpace import LITERAL_DEVICE_GRAY
|
||||
from ..interpreter.PDFColorSpace import LITERAL_DEVICE_RGB
|
||||
from ..interpreter.PDFColorSpace import LITERAL_DEVICE_CMYK
|
||||
from ..layouts import LTImage
|
||||
|
||||
def align32(x):
|
||||
return ((x+3)//4)*4
|
||||
|
||||
## BMPWriter
|
||||
##
|
||||
class BMPWriter(object):
|
||||
|
||||
def __init__(self, fp, bits, width, height):
|
||||
self.fp = fp
|
||||
self.bits = bits
|
||||
self.width = width
|
||||
self.height = height
|
||||
if bits == 1:
|
||||
ncols = 2
|
||||
elif bits == 8:
|
||||
ncols = 256
|
||||
elif bits == 24:
|
||||
ncols = 0
|
||||
else:
|
||||
raise ValueError(bits)
|
||||
self.linesize = align32((self.width*self.bits+7)//8)
|
||||
self.datasize = self.linesize * self.height
|
||||
headersize = 14+40+ncols*4
|
||||
info = struct.pack('<IiiHHIIIIII', 40, self.width, self.height, 1, self.bits, 0, self.datasize, 0, 0, ncols, 0)
|
||||
assert len(info) == 40, str(len(info))
|
||||
header = struct.pack('<ccIHHI', b'B', b'M', headersize+self.datasize, 0, 0, headersize)
|
||||
assert len(header) == 14, str(len(header))
|
||||
self.fp.write(header)
|
||||
self.fp.write(info)
|
||||
if ncols == 2:
|
||||
# B&W color table
|
||||
for i in (0, 255):
|
||||
self.fp.write(struct.pack('BBBx', i, i, i))
|
||||
elif ncols == 256:
|
||||
# grayscale color table
|
||||
for i in range(256):
|
||||
self.fp.write(struct.pack('BBBx', i, i, i))
|
||||
self.pos0 = self.fp.tell()
|
||||
self.pos1 = self.pos0 + self.datasize
|
||||
return
|
||||
|
||||
def write_line(self, y, data):
|
||||
self.fp.seek(self.pos1 - (y+1)*self.linesize)
|
||||
self.fp.write(data)
|
||||
return
|
||||
|
||||
|
||||
## ImageWriter
|
||||
##
|
||||
class ImageWriter(object):
|
||||
|
||||
def __init__(self, outdir: str = None):
|
||||
if outdir is None:
|
||||
self.inactive = True
|
||||
else:
|
||||
self.inactive = False
|
||||
self.outdir = outdir
|
||||
if not os.path.exists(self.outdir):
|
||||
os.makedirs(self.outdir)
|
||||
|
||||
def export_image(self, image: LTImage) -> str:
|
||||
if self.inactive:
|
||||
return "N/A"
|
||||
|
||||
stream = image.stream
|
||||
filters = stream.get_filters()
|
||||
(width, height) = image.srcsize
|
||||
if len(filters) == 1 and filters[0][0] in LITERALS_DCT_DECODE:
|
||||
ext = '.jpg'
|
||||
elif (image.bits == 1 or
|
||||
image.bits == 8 and image.colorspace in (LITERAL_DEVICE_RGB, LITERAL_DEVICE_GRAY)):
|
||||
ext = '.%dx%d.bmp' % (width, height)
|
||||
else:
|
||||
ext = '.%d.%dx%d.img' % (image.bits, width, height)
|
||||
name = image.name+ext
|
||||
path = os.path.join(self.outdir, name)
|
||||
fp=open(path, 'wb')
|
||||
if ext == '.jpg':
|
||||
raw_data = stream.get_rawdata()
|
||||
if LITERAL_DEVICE_CMYK in image.colorspace:
|
||||
from PIL import Image
|
||||
from PIL import ImageChops
|
||||
ifp = BytesIO(raw_data)
|
||||
i = Image.open(ifp)
|
||||
i = ImageChops.invert(i)
|
||||
i = i.convert('RGB')
|
||||
i.save(fp, 'JPEG')
|
||||
else:
|
||||
fp.write(raw_data)
|
||||
elif image.bits == 1:
|
||||
bmp = BMPWriter(fp, 1, width, height)
|
||||
data = stream.get_data()
|
||||
i = 0
|
||||
width = (width+7)//8
|
||||
for y in range(height):
|
||||
bmp.write_line(y, data[i:i+width])
|
||||
i += width
|
||||
elif image.bits == 8 and image.colorspace is LITERAL_DEVICE_RGB:
|
||||
bmp = BMPWriter(fp, 24, width, height)
|
||||
data = stream.get_data()
|
||||
i = 0
|
||||
width = width*3
|
||||
for y in range(height):
|
||||
bmp.write_line(y, data[i:i+width])
|
||||
i += width
|
||||
elif image.bits == 8 and image.colorspace is LITERAL_DEVICE_GRAY:
|
||||
bmp = BMPWriter(fp, 8, width, height)
|
||||
data = stream.get_data()
|
||||
i = 0
|
||||
for y in range(height):
|
||||
bmp.write_line(y, data[i:i+width])
|
||||
i += width
|
||||
else:
|
||||
fp.write(stream.get_data())
|
||||
fp.close()
|
||||
return name
|
||||
@@ -1,13 +1,8 @@
|
||||
from pdfmajor.parser.PDFPage import PDFPage
|
||||
from pdfmajor.parser.PDFStream import list_value
|
||||
from pdfmajor.parser.PDFContentParser import PDFContentParser
|
||||
from pdfmajor.parser.PSStackParser import PSKeyword
|
||||
from pdfmajor.parser.PSStackParser import keyword_name
|
||||
|
||||
from .commands import PDFStateStack, PDFCommands
|
||||
from .commands import LTXObject
|
||||
from .utils import init_resources
|
||||
from .XObjectInterpreter import XObjectInterpreter
|
||||
from .commands import PDFStateStack
|
||||
from .commands import process_command_stream, prep_state
|
||||
|
||||
class PageInterpreter:
|
||||
class PDFInterpreterError(Exception): pass
|
||||
@@ -31,52 +26,21 @@ class PageInterpreter:
|
||||
self.width = x1-x0
|
||||
|
||||
# Init State
|
||||
self.state: PDFStateStack = PDFStateStack()
|
||||
self.state.t_matrix = ctm
|
||||
self.state.resources = page.resources
|
||||
# set some global states.
|
||||
self.state.graphics.ncolspace = self.state.graphics.scolspace = None
|
||||
if self.state.colorspace_map:
|
||||
col_space = next(iter(self.state.colorspace_map.values()))
|
||||
self.state.graphics.ncolspace = self.state.graphics.scolspace = col_space
|
||||
# Init Resources
|
||||
init_resources(self.state, self.font_cache)
|
||||
self.state: PDFStateStack = prep_state(
|
||||
PDFStateStack(),
|
||||
ctm=ctm,
|
||||
resources=page.resources,
|
||||
font_cache=self.font_cache
|
||||
)
|
||||
|
||||
|
||||
def __iter__(self):
|
||||
parser = PDFContentParser(list_value(self.page.contents))
|
||||
history = []
|
||||
for obj in parser:
|
||||
history.append(obj)
|
||||
if isinstance(obj, PSKeyword):
|
||||
name = keyword_name(obj)
|
||||
method = name.replace('*', '_a').replace('"', '_w').replace("'", '_q')
|
||||
if method in PDFCommands.commands.keys():
|
||||
func = PDFCommands.commands.get(method)
|
||||
nargs = func.__code__.co_argcount-1
|
||||
if nargs:
|
||||
args = self.state.pop(nargs)
|
||||
if len(args) == nargs:
|
||||
func(self.state, *args)
|
||||
else:
|
||||
raise self.PDFInterpreterError(f"Invalid Number of Args provided {method}")
|
||||
else:
|
||||
# log.debug('exec: %s', name)
|
||||
func(self.state)
|
||||
else:
|
||||
raise self.PDFInterpreterError('Unknown operator: %r' % name)
|
||||
else:
|
||||
self.state.argstack.append(obj)
|
||||
|
||||
for complete_item in self.state.complete_layout_items:
|
||||
if isinstance(complete_item, LTXObject):
|
||||
interpeter = XObjectInterpreter(
|
||||
streams=[complete_item.stream],
|
||||
resources=complete_item.resources,
|
||||
ctm=complete_item.t_matrix
|
||||
)
|
||||
for item in interpeter:
|
||||
complete_item.add(item)
|
||||
yield complete_item
|
||||
self.state.complete_layout_items = []
|
||||
for item in process_command_stream(
|
||||
streams=list_value(self.page.contents),
|
||||
font_cache=self.font_cache,
|
||||
state=self.state
|
||||
):
|
||||
yield item
|
||||
|
||||
def __repr__(self) -> str:
|
||||
return f"<Page:{self.page_num}/>"
|
||||
return f"<Page:{self.page_num} width={self.width} height={self.height}/>"
|
||||
@@ -1,70 +0,0 @@
|
||||
from pdfmajor.utils import MATRIX_IDENTITY
|
||||
from pdfmajor.parser.PSStackParser import literal_name
|
||||
from pdfmajor.parser.PDFParser import PDFStream
|
||||
from pdfmajor.parser.PSStackParser import PSEOF
|
||||
from pdfmajor.parser.PSStackParser import PSKeyword
|
||||
from pdfmajor.parser.PSStackParser import keyword_name
|
||||
from pdfmajor.parser.PSStackParser import PSStackParser
|
||||
from pdfmajor.parser.constants import LITERAL_PDF, LITERAL_TEXT, LITERAL_FONT
|
||||
|
||||
from pdfmajor.parser.PDFStream import resolve1, list_value, dict_value
|
||||
from pdfmajor.parser.PDFStream.PDFObjRef import PDFObjRef
|
||||
from pdfmajor.parser.PDFContentParser import PDFContentParser
|
||||
|
||||
from .commands.state import PDFStateStack, PDFColorSpace, PREDEFINED_COLORSPACE
|
||||
from .commands import PDFCommands, LTXObject
|
||||
from .utils import init_resources
|
||||
|
||||
class XObjectInterpreter:
|
||||
|
||||
class PDFInterpreterError(Exception): pass
|
||||
|
||||
def __init__(self, streams, resources, ctm):
|
||||
self.streams = streams
|
||||
self.font_cache = {}
|
||||
|
||||
self.state = PDFStateStack()
|
||||
self.state.t_matrix = ctm
|
||||
self.state.resources = resources
|
||||
|
||||
# set some global states.
|
||||
self.state.graphics.ncolspace = self.state.graphics.scolspace = None
|
||||
if self.state.colorspace_map:
|
||||
col_space = next(iter(self.state.colorspace_map.values()))
|
||||
self.state.graphics.ncolspace = self.state.graphics.scolspace = col_space
|
||||
init_resources(self.state, self.font_cache)
|
||||
|
||||
def __iter__(self):
|
||||
parser = PDFContentParser(self.streams)
|
||||
for obj in parser:
|
||||
if isinstance(obj, PSKeyword):
|
||||
name = keyword_name(obj)
|
||||
method = name.replace('*', '_a').replace('"', '_w').replace("'", '_q')
|
||||
if method in PDFCommands.commands.keys():
|
||||
func = PDFCommands.commands.get(method)
|
||||
nargs = func.__code__.co_argcount-1
|
||||
if nargs:
|
||||
args = self.state.pop(nargs)
|
||||
if len(args) == nargs:
|
||||
func(self.state, *args)
|
||||
else:
|
||||
raise self.PDFInterpreterError(f"Invalid Number of Args provided {method}")
|
||||
else:
|
||||
# log.debug('exec: %s', name)
|
||||
func(self.state)
|
||||
else:
|
||||
raise self.PDFInterpreterError('Unknown operator: %r' % name)
|
||||
else:
|
||||
self.state.argstack.append(obj)
|
||||
|
||||
for complete_item in self.state.complete_layout_items:
|
||||
if isinstance(complete_item, LTXObject):
|
||||
interpeter = self.__class__(
|
||||
streams=[complete_item.stream],
|
||||
resources=complete_item.resources,
|
||||
ctm=complete_item.t_matrix
|
||||
)
|
||||
for item in interpeter:
|
||||
complete_item.add(item)
|
||||
yield complete_item
|
||||
self.state.complete_layout_items = []
|
||||
@@ -6,7 +6,7 @@ from ..parser.PDFPage import PDFPage
|
||||
from ..utils import set_log_level, get_logger, logging
|
||||
from .commands.state import PDFStateStack, PDFColorSpace, PREDEFINED_COLORSPACE
|
||||
from .commands import PDFCommands
|
||||
from .commands import LTCharBlock, LTChar
|
||||
from .commands import LTTextBlock, LTCharBlock, LTChar
|
||||
from .commands import LTCurve, LTLine, LTHorizontalLine, LTVerticalLine, LTRect
|
||||
from .commands import LTImage
|
||||
from .commands import LTXObject
|
||||
|
||||
@@ -1,9 +1,75 @@
|
||||
from typing import List
|
||||
|
||||
from pdfmajor.parser.PDFContentParser import PDFContentParser, PDFStream
|
||||
from pdfmajor.parser.PSStackParser import PSKeyword, keyword_name
|
||||
from pdfmajor.utils import get_logger
|
||||
|
||||
from .commands import PDFCommands
|
||||
from .state import PDFStateStack, PDFGraphicState, PDFTextState
|
||||
from .state.PDFGraphicState import PDFColor, PDFColorSpace, PREDEFINED_COLORSPACE
|
||||
from .state import PDFFont, get_font
|
||||
from .state import LTCharBlock, LTChar
|
||||
from .state import LTTextBlock, LTCharBlock, LTChar
|
||||
from .state import LTCurve, LTLine, LTHorizontalLine, LTVerticalLine, LTRect
|
||||
from .state import LTImage
|
||||
from .state import LTXObject
|
||||
from .state import LTItem, LTComponent, LTContainer
|
||||
from .state import LTItem, LTComponent, LTContainer
|
||||
from .utils import init_resources
|
||||
|
||||
log = get_logger('process_command_stream')
|
||||
|
||||
class CommandProcessorError(Exception): pass
|
||||
|
||||
def process_command_stream(streams: List[PDFStream], font_cache: dict = None, state: PDFStateStack = None):
|
||||
if state is None:
|
||||
state = PDFStateStack()
|
||||
if font_cache is None:
|
||||
font_cache = {}
|
||||
parser = PDFContentParser(streams)
|
||||
for obj in parser:
|
||||
if isinstance(obj, PSKeyword):
|
||||
name = keyword_name(obj)
|
||||
method = name.replace('*', '_a').replace('"', '_w').replace("'", '_q')
|
||||
if method in PDFCommands.commands.keys():
|
||||
func = PDFCommands.commands.get(method)
|
||||
nargs = func.__code__.co_argcount-1
|
||||
if nargs:
|
||||
args = state.pop(nargs)
|
||||
if len(args) == nargs:
|
||||
func(state, *args)
|
||||
else:
|
||||
raise CommandProcessorError(f"Invalid Number of Args provided {method}")
|
||||
else:
|
||||
func(state)
|
||||
else:
|
||||
log.warning('Unknown operator: %r (i.e. %r)' % (name, method))
|
||||
else:
|
||||
state.argstack.append(obj)
|
||||
|
||||
for complete_item in state.complete_layout_items:
|
||||
if isinstance(complete_item, LTXObject):
|
||||
xobj_state = prep_state(
|
||||
state=PDFGraphicState(),
|
||||
ctm=complete_item.t_matrix,
|
||||
resources=complete_item.resources,
|
||||
font_cache=font_cache
|
||||
)
|
||||
for item in process_command_stream(
|
||||
[complete_item.stream],
|
||||
font_cache=font_cache,
|
||||
state=xobj_state
|
||||
):
|
||||
complete_item.add(item)
|
||||
yield complete_item
|
||||
state.complete_layout_items = []
|
||||
|
||||
def prep_state(state: PDFStateStack, ctm: tuple, resources: dict, font_cache: dict) -> PDFStateStack:
|
||||
state.t_matrix = ctm
|
||||
state.resources = resources
|
||||
|
||||
# set some global states.
|
||||
state.graphics.ncolspace = state.graphics.scolspace = None
|
||||
if state.colorspace_map:
|
||||
col_space = next(iter(state.colorspace_map.values()))
|
||||
state.graphics.ncolspace = state.graphics.scolspace = col_space
|
||||
init_resources(state, font_cache)
|
||||
return state
|
||||
@@ -1,33 +1,51 @@
|
||||
from pdfmajor.utils import mult_matrix, MATRIX_IDENTITY, apply_matrix_pt
|
||||
from pdfmajor.utils import mult_matrix, MATRIX_IDENTITY, apply_matrix_pt, get_logger
|
||||
from pdfmajor.parser.PSStackParser import literal_name
|
||||
from pdfmajor.parser.PDFStream import PDFStream, list_value, dict_value
|
||||
from pdfmajor.parser.constants import LITERAL_FORM, LITERAL_IMAGE
|
||||
|
||||
from .PDFCommands import PDFCommands
|
||||
from .state import PDFStateStack, PDFGraphicState
|
||||
from .state import PDFStateStack, PDFGraphicState, LTTextBlock
|
||||
from .state.Curves import CurveMethod, CurvePath, CurvePoint
|
||||
# from .state.PDFItem import PDFImage, PDFShape, PDFText, PDFXObject
|
||||
from .state import make_char_block, make_curve, make_image, make_xobject
|
||||
|
||||
log = get_logger('commands')
|
||||
|
||||
# No Support
|
||||
@PDFCommands.add('W','W_a', 'sh', 'ET', 'BX', 'EX' ,'BI', 'ID', 'gs')
|
||||
def do_no_support(stack: PDFStateStack) -> PDFStateStack:
|
||||
return stack
|
||||
# @PDFCommands.add('W','W_a', 'sh', 'ET', 'BX', 'EX' ,'BI', 'ID', 'gs')
|
||||
# def do_no_support(stack: PDFStateStack) -> PDFStateStack:
|
||||
# # log.warning("Unsupported command")
|
||||
# return stack
|
||||
|
||||
# marked content operators
|
||||
@PDFCommands.add('MP', 'DP')
|
||||
def do_MP(stack: PDFStateStack) -> PDFStateStack:
|
||||
@PDFCommands.add('MP')
|
||||
def do_MP(stack: PDFStateStack, tag) -> PDFStateStack:
|
||||
log.warning(f"Unsupported command - do_tag {[tag]}")
|
||||
# self.device.do_tag(tag)
|
||||
return stack
|
||||
|
||||
@PDFCommands.add('BMC', 'BDC')
|
||||
def do_BMC(stack: PDFStateStack) -> PDFStateStack:
|
||||
@PDFCommands.add('DP')
|
||||
def do_DP(stack: PDFStateStack, tag, props=None) -> PDFStateStack:
|
||||
log.warning(f"Unsupported command - do_tag {[tag.name, type(tag), props]}")
|
||||
# self.device.do_tag(tag)
|
||||
return stack
|
||||
|
||||
@PDFCommands.add('BMC')
|
||||
def do_BMC(stack: PDFStateStack, tag) -> PDFStateStack:
|
||||
log.warning(f"Unsupported command - begin_tag {[tag.name, type(tag)]}")
|
||||
# self.device.begin_tag(tag)
|
||||
return stack
|
||||
|
||||
@PDFCommands.add('BDC')
|
||||
def do_BDC(stack: PDFStateStack, tag, props=None) -> PDFStateStack:
|
||||
log.warning(f"Unsupported command - begin_tag {[tag.name, type(tag), props]}")
|
||||
# self.device.begin_tag(tag)
|
||||
return stack
|
||||
|
||||
@PDFCommands.add('EMC')
|
||||
def do_EMC(stack: PDFStateStack) -> PDFStateStack:
|
||||
# self.device.end_tag(tag)
|
||||
log.warning("Unsupported command - end_tag")
|
||||
return stack
|
||||
|
||||
@PDFCommands.add('q')
|
||||
@@ -332,6 +350,16 @@ def do_scn(stack: PDFStateStack) -> PDFStateStack:
|
||||
def do_BT(stack: PDFStateStack) -> PDFStateStack:
|
||||
stack.text.matrix = MATRIX_IDENTITY
|
||||
stack.text.linematrix = [0, 0]
|
||||
if stack.current_textblock is None:
|
||||
stack.current_textblock = LTTextBlock()
|
||||
else:
|
||||
raise PDFCommands.InvalidOperation("Last TextBlock did not clear")
|
||||
return stack
|
||||
|
||||
@PDFCommands.add('ET')
|
||||
def do_ET(stack: PDFStateStack) -> PDFStateStack:
|
||||
stack.complete_layout_items.append(stack.current_textblock)
|
||||
stack.current_textblock = None
|
||||
return stack
|
||||
|
||||
# setcharspace
|
||||
@@ -417,12 +445,14 @@ def do_T_a(stack: PDFStateStack) -> PDFStateStack:
|
||||
def do_TJ(stack: PDFStateStack, seq: bytearray) -> PDFStateStack:
|
||||
if stack.text.font is None:
|
||||
raise PDFCommands.InvalidOperation("No Font Specified")
|
||||
stack.complete_layout_items.append(make_char_block(
|
||||
if stack.current_textblock is None:
|
||||
raise PDFCommands.InvalidOperation("No TextBlock initilized")
|
||||
stack.current_textblock.add_char_block(
|
||||
seq,
|
||||
stack.t_matrix,
|
||||
stack.text.copy(),
|
||||
stack.graphics.ncolor.copy()
|
||||
))
|
||||
)
|
||||
return stack
|
||||
|
||||
# show
|
||||
|
||||
@@ -10,7 +10,7 @@ from .PDFTextState import PDFTextState, PDFFont, get_font
|
||||
from .Curves import CurvePath
|
||||
|
||||
from .layout import LTItem, LTComponent, LTContainer
|
||||
from .layout import LTCharBlock, LTChar
|
||||
from .layout import LTTextBlock, LTCharBlock, LTChar
|
||||
from .layout import LTCurve, LTLine, LTHorizontalLine, LTVerticalLine, LTRect
|
||||
from .layout import LTXObject
|
||||
from .layout import LTImage
|
||||
@@ -30,6 +30,7 @@ class PDFStateStack:
|
||||
self.fontmap = {}
|
||||
self.xobjmap = {}
|
||||
self.resources = {}
|
||||
self.current_textblock: LTTextBlock = None
|
||||
|
||||
def pop(self, n: int) -> bytearray:
|
||||
if n == 0:
|
||||
|
||||
@@ -1,7 +1,19 @@
|
||||
import os
|
||||
from io import BytesIO
|
||||
|
||||
import PIL as pillow
|
||||
|
||||
from pdfmajor.BMPWriter import BMPWriter
|
||||
from pdfmajor.utils import Bbox
|
||||
from pdfmajor.parser.PDFStream import PDFStream
|
||||
from pdfmajor.parser.PDFStream import PDFStream, LITERALS_DCT_DECODE
|
||||
from pdfmajor.parser.constants import LIT
|
||||
|
||||
from ._base import LTComponent
|
||||
|
||||
LITERAL_DEVICE_GRAY = LIT('DeviceGray')
|
||||
LITERAL_DEVICE_RGB = LIT('DeviceRGB')
|
||||
LITERAL_DEVICE_CMYK = LIT('DeviceCMYK')
|
||||
|
||||
## LTImage
|
||||
##
|
||||
class LTImage(LTComponent):
|
||||
@@ -22,4 +34,46 @@ class LTImage(LTComponent):
|
||||
def __repr__(self):
|
||||
return ('<%s(%s) %s %r>' %
|
||||
(self.__class__.__name__, self.name,
|
||||
str(self.bbox), self.srcsize))
|
||||
str(self.bbox), self.srcsize))
|
||||
|
||||
def save_image(self, outdir: str, file_name: str = None):
|
||||
stream = self.stream
|
||||
filters = stream.get_filters()
|
||||
width = self.bbox.width
|
||||
height = self.bbox.height
|
||||
bits = self.bits
|
||||
|
||||
if len(filters) == 1 and filters[0][0] in LITERALS_DCT_DECODE:
|
||||
ext = '.jpg'
|
||||
elif (bits == 1 or
|
||||
bits == 8 and self.colorspace in (LITERAL_DEVICE_RGB, LITERAL_DEVICE_GRAY)):
|
||||
ext = '.w%dh%d.bmp' % (width, height)
|
||||
else:
|
||||
ext = '.b%dw%dh%d.img' % (bits, width, height)
|
||||
|
||||
if file_name is None:
|
||||
file_name = self.name + ext
|
||||
else:
|
||||
file_name += ext
|
||||
path = os.path.join(outdir, file_name)
|
||||
|
||||
with open(path, 'wb') as fp:
|
||||
if ext == '.jpg':
|
||||
raw_data = stream.get_rawdata()
|
||||
if LITERAL_DEVICE_CMYK in self.colorspace:
|
||||
ifp = BytesIO(raw_data)
|
||||
i = pillow.Image.open(ifp)
|
||||
i = pillow.ImageChops.invert(i)
|
||||
i = i.convert('RGB')
|
||||
i.save(fp, 'JPEG')
|
||||
else:
|
||||
fp.write(raw_data)
|
||||
elif bits == 1:
|
||||
BMPWriter.write_all(fp, stream, 1, width, height, (width+7)//8)
|
||||
elif bits == 8 and self.colorspace is LITERAL_DEVICE_RGB:
|
||||
BMPWriter.write_all(fp, stream, 24, width, height, width*3)
|
||||
elif bits == 8 and self.colorspace is LITERAL_DEVICE_GRAY:
|
||||
BMPWriter.write_all(fp, stream, 8, width, height, width)
|
||||
else:
|
||||
fp.write(stream.get_data())
|
||||
return path
|
||||
@@ -0,0 +1,40 @@
|
||||
from typing import List
|
||||
|
||||
from pdfmajor.utils import apply_matrix_norm, apply_matrix_pt, matrix2str, Bbox, Point
|
||||
|
||||
from ..PDFGraphicState import PDFColor
|
||||
from ..PDFTextState import PDFFont
|
||||
from ..PDFTextState import PDFTextState
|
||||
|
||||
from ._base import LTContainer
|
||||
from .LTCharBlock import LTCharBlock
|
||||
from .utils.textdecoder import decode_text_seq
|
||||
|
||||
|
||||
## LTTextBlock
|
||||
##
|
||||
class LTTextBlock(LTContainer):
|
||||
def __init__(self, x0=None, y0=None, x1=None, y1=None):
|
||||
LTContainer.__init__(self,
|
||||
Bbox(x0, y0, x1, y1),
|
||||
[]
|
||||
)
|
||||
|
||||
def add_char_block(self, seq: bytearray, ctm: tuple, textstate: PDFTextState, color: PDFColor):
|
||||
(char_meta_datas, textstate, color) = decode_text_seq(seq, ctm, textstate, color)
|
||||
char_block = LTCharBlock(
|
||||
chars=char_meta_datas,
|
||||
color=color,
|
||||
textstate=textstate
|
||||
)
|
||||
if None in [self.bbox.x0, self.bbox.x1, self.bbox.y0, self.bbox.y1]:
|
||||
self.bbox.x0 = char_block.x0
|
||||
self.bbox.x1 = char_block.x1
|
||||
self.bbox.y0 = char_block.y0
|
||||
self.bbox.y1 = char_block.y1
|
||||
else:
|
||||
self.bbox.x0 = min([self.bbox.x0, char_block.x0])
|
||||
self.bbox.x1 = max([self.bbox.x1, char_block.x1])
|
||||
self.bbox.y0 = min([self.bbox.y0, char_block.y0])
|
||||
self.bbox.y1 = max([self.bbox.y1, char_block.y1])
|
||||
self.add(char_block)
|
||||
@@ -6,6 +6,7 @@ from ..PDFGraphicState import PDFGraphicState, PDFColor
|
||||
from ..PDFTextState import PDFTextState
|
||||
|
||||
from ._base import LTItem, LTComponent, LTContainer
|
||||
from .LTTextBlock import LTTextBlock
|
||||
from .LTCharBlock import LTCharBlock, LTChar
|
||||
from .LTCurves import LTCurve, LTLine, LTHorizontalLine, LTRect, LTVerticalLine
|
||||
from .LTImage import LTImage
|
||||
|
||||
@@ -4,8 +4,8 @@ from pdfmajor.parser.PDFStream.PDFObjRef import PDFObjRef
|
||||
from pdfmajor.parser.PSStackParser import literal_name
|
||||
from pdfmajor.parser.constants import LITERAL_PDF, LITERAL_TEXT
|
||||
|
||||
from .commands import PDFColorSpace, PREDEFINED_COLORSPACE
|
||||
from .commands import PDFStateStack, get_font
|
||||
from .state import PDFColorSpace, PREDEFINED_COLORSPACE
|
||||
from .state import PDFStateStack, get_font
|
||||
|
||||
def get_colorspace(spec):
|
||||
if isinstance(spec, list):
|
||||
@@ -1,35 +0,0 @@
|
||||
import collections
|
||||
from .PSStackParser import LIT
|
||||
|
||||
## PDFColorSpace
|
||||
##
|
||||
LITERAL_DEVICE_GRAY = LIT('DeviceGray')
|
||||
LITERAL_DEVICE_RGB = LIT('DeviceRGB')
|
||||
LITERAL_DEVICE_CMYK = LIT('DeviceCMYK')
|
||||
|
||||
|
||||
class PDFColorSpace(object):
|
||||
|
||||
def __init__(self, name: str, ncomponents: int):
|
||||
self.name: str = name
|
||||
self.ncomponents: int = ncomponents
|
||||
return
|
||||
|
||||
def __repr__(self):
|
||||
return '<PDFColorSpace: %s, ncomponents=%d>' % (self.name, self.ncomponents)
|
||||
|
||||
|
||||
PREDEFINED_COLORSPACE = collections.OrderedDict()
|
||||
|
||||
for (name, n) in [
|
||||
('DeviceGray', 1), # default value first
|
||||
('CalRGB', 3),
|
||||
('CalGray', 1),
|
||||
('Lab', 3),
|
||||
('DeviceRGB', 3),
|
||||
('DeviceCMYK', 4),
|
||||
('Separation', 1),
|
||||
('Indexed', 1),
|
||||
('Pattern', 1),
|
||||
]:
|
||||
PREDEFINED_COLORSPACE[name]=PDFColorSpace(name, n)
|
||||
@@ -10,7 +10,7 @@ import sys
|
||||
with open('README.md') as readme_file:
|
||||
SETUP = {
|
||||
"name": "pdfmajor",
|
||||
"version": "1.3.0",
|
||||
"version": "1.3.2b",
|
||||
"packages": find_packages(include=['pdfmajor*']),
|
||||
"install_requires": [
|
||||
'pycryptodome',
|
||||
@@ -18,6 +18,7 @@ with open('README.md') as readme_file:
|
||||
],
|
||||
"description": 'PDF parser',
|
||||
"long_description": readme_file.read(),
|
||||
'long_description_content_type': 'text/markdown',
|
||||
"license": 'MIT/X',
|
||||
"author": 'Ariel Sosnovsky + Yusuke Shinyama + Philippe Guglielmetti',
|
||||
"author_email": 'ariel@sosnovsky.ca',
|
||||
|
||||
Binary file not shown.
Binary file not shown.
@@ -0,0 +1,33 @@
|
||||
import os
|
||||
|
||||
from pdfmajor.interpreter import LTImage, logging
|
||||
from pdfmajor.interpreter import PageInterpreter, PDFInterpreter
|
||||
from pdfmajor.interpreter.commands import LTItem
|
||||
from pdfmajor.interpreter.commands import LTCharBlock, LTChar
|
||||
|
||||
CUR_PATH = os.path.dirname(os.path.dirname(__file__))
|
||||
FILES = [
|
||||
"jpg",
|
||||
"looseless"
|
||||
]
|
||||
OUTPUT_FOLDER = os.path.join(
|
||||
CUR_PATH,
|
||||
'samples/output/images'
|
||||
)
|
||||
|
||||
if not os.path.exists(OUTPUT_FOLDER):
|
||||
os.makedirs(OUTPUT_FOLDER)
|
||||
|
||||
for file_name in FILES:
|
||||
for page in PDFInterpreter(os.path.join(
|
||||
CUR_PATH,
|
||||
f"samples/pdf/{file_name}.pdf"
|
||||
), debug_level=logging.ERROR):
|
||||
assert isinstance(page, PageInterpreter)
|
||||
for item_idx, item in enumerate(page):
|
||||
assert isinstance(item, LTItem)
|
||||
if isinstance(item, LTImage):
|
||||
print(item.save_image(
|
||||
OUTPUT_FOLDER,
|
||||
file_name=f'{file_name}_p{page.page_num}i{item_idx}'
|
||||
))
|
||||
Reference in new issue
Block a user