Skip to content

Latest commit

 

History

History
142 lines (87 loc) · 7.56 KB

Readme.md

File metadata and controls

142 lines (87 loc) · 7.56 KB
Raw

pdfalto

Build Status

pdfalto is a command line executable for parsing PDF files and producing structured XML representations of the PDF content in ALTO format.

pdfalto is a fork of pdf2xml, developed at XRCE, with modifications for robustness, addition of features and output enhanced format in ALTO (including in particular space information, useful for instance for further machine learning processing). It is based on the Xpdf library.

The latest (non-)stable version is 0.2.

Requirements

  • clang 3.6 or gcc 4.9
  • cmake 3.12.0

Usage

General usage is as follow:

 pdfalto [options] <PDF-file> [<xml-file>]
  -f <int>               : first page to convert
  -l <int>               : last page to convert
  -verbose               : display pdf attributes
  -noText                : do not extract textual objects
  -noImage               : do not extract Images (Bitmap and Vectorial)
  -noImageInline         : do not include images inline in the stream
  -outline               : create an outline file xml (i.e. a table of content) as additional file
  -annotation            : create an annotations file xml as additional file
  -cutPages              : cut all pages in separately files
  -blocks                : add blocks informations whithin the structure
  -readingOrder          : blocks follow the reading order
  -fullFontName          : fonts names are not normalized
  -nsURI <string>        : add the specified namespace URI
  -opw <string>          : owner password (for encrypted files)
  -upw <string>          : user password (for encrypted files)
  -filesLimit <int>      : limit of asset files be extracted to the value specified
  -q                     : don't print any messages or errors
  -v                     : print version info
  -h                     : print usage information
  -help                  : print usage information
  --help                 : print usage information
  -?                     : print usage information
  --saveconf <string>    : save all command line parameters in the specified XML <file>
  -ocr                   : recognises all characters that are missing from unicode, or replace with placeholder to flag it.

In addition to the ALTO file describing the PDF content, the following files are generated:

  • _metadata.xml file containing a pdf file metadata (generate metadata information in a separate XML file as ALTO schema does not support that).

  • _annot.xml file containing a description of the annotations in the PDF (e.g. GOTO, external http links, ...) obtained with -annotation option

  • _outline.xml file containing a possible PDF-embedded table of content (aka outline) obtained with -outline option

  • .xml_data/ subdirectory containing the vectorial (.vec) and bitmap images (.png) embedded in the PDF, this is generated by default when the option -noImage is not present

Dependencies

All dependencies are provided as static libraries corresponding to each operating system.

Dependencies can be recompiled by running this script

See compiling dependencies procedures for further details.

Known issues (issue 41) might occur whille building, in this case you'll need to compile the dependencies before build pdflato.

Build

  • NOTE for windows : it's recommended to use Cygwin and install standard libraries (either for cland or gcc)

git clone https://github.com/kermitt2/pdfalto.git && cd pdfalto

  • Xpdf 4.00 is shipped as git submodule, to download it:

git submodule update --init --recursive

  • Build pdfalto:

cmake ./

make

The executable pdfalto is generated in the root directory. Additionally, this will create a static library for xpdf-4.00 at the following path xpdf-4.00/build/xpdf/lib/libxpdf.a and all the libraries and their respective subdirectory.

Future work

  • map special characters in secondary fonts to their expected unicode

  • try OCR for unsolved character unicode value based on their associated glyph in embedded font

  • try OCR for unsolved character unicode value in context based on their occurences in the document

  • try to optimize speed and memory

  • ..see the issue tracker

Changes

  • support unicode composition of characters

  • generalize reading order to all blocks (now it is limited to the blocks of the first page)

  • use subscript/superscript text font style attribut.

  • use SVG as a format for vectorial images.

  • propagate unsolved character unicode value (free unicode range for embedded fonts) as encoded special character in ALTO (so-called "placeholder" approach)

  • generate metadata information in a separate XML file (as ALTO schema does not support that)

  • use the latest version of xpdf, version 4.00.

  • add cmake

  • ALTO output is replacing custom Xerox XML format

  • encode URI (using xmlURIEscape from libxml2) for the @href attribute content to avoid blocking XML wellformedness issues. From our experiments, this problem happens in average for 2-3 scholar PDF out of one thousand.

  • output coordinates attributes for the BLOCK elements when the -block option is selected,

  • add a parameter -readingOrder which re-order the blocks following the reading order when the -block option is selected. By default in pdf2xml, the elements follow the PDF content stream (the so-called raw order). In pdf2txt from xpdf, several text flow orders are available including the raw order and the reading order. Note that, with this modification and this new option, only the blocks are re-ordered.

    From our experiments, the raw order can diverge quite significantly from the order of elements according to the visual/reading layout in 2-4% of scholar PDF (e.g. title element is introduced at the end of the page element, while visually present at the top of the page), and minor changes can be present in up to 100% of PDF for some scientific publishers (e.g. headnote introduced at the end of the page content). This additional mode can be thus quite useful for information/structure extraction applications exploiting pdf2xml output.

  • use the latest version of xpdf, version 3.04.

Known Limitations

See the issue tracker and future work :

  1. Text like containing block element characters (https://unicode.org/charts/PDF/U2B00.pdf) might be placeholders for unknown characters unicodes instead of what you expect when you are extracting text. This is because the actual characters are glyphs that are embedded in the PDF document. The only way to access the text is to use OCR. This may be a future enhancement.

Contributors

xpdf is developed by Glyph & Cog, LLC (1996-2017) and distributed under GPL2 or GPL3 license.

pdf2xml is orignally written by Hervé Déjean, Sophie Andrieu, Jean-Yves Vion-Dury and Emmanuel Giguet (XRCE) under GPL2 license.

pdfalto has been modified and forked by Patrice Lopez ([email protected]) and Achraf Azhar ([email protected]).

The windows version has been built originally by @pboumenot and ported on windows 7 for 64 bit, then for windows (native and cygwin) by @lfoppiano and @flydutch.

License

As the original pdf2xml, pdfalto is distributed under GPL2 license.