v1.19.0

• Enable permanently setting almost all command line options via environment variables or a custom .pdfalyzer file, add --env-vars option to display exactly which command line options can be set by which variables
• Add --export-png option to render .png images of output
• Add --echo-command option to save the exact command used along with the output
• Add --no-timestamps option for exported filenames
• Add --suppress-output option
• Highlight some of the interesting object reference keys in the rich tree view
• In rich tree view sort DictionaryObject key/value pairs by key alphabetically, except /Type and /Subtype are at the top
• Display the number of revisions (max generation value) in metadata table
• Display obj types/labels determined when walking the tree instead of newly constructed PdfObjectProperties which may lack nuance in their labeling
• Better labeling of /StructElem objects in a StructTreeRoot hierarchy
• Coerce /Nums number trees into dict like objects for the purposes of assigning addresses
• /Annots and other indeterminate nodes now have /Subtype integrated into their labeling (e.g. /Annots:Link instead of just /Annots)
• Fancier table for PDF metadata that also contains the number of pages, images, and revisions (if possible)
• Test suite now checks results against pre-recorded fixture output

v1.18.1

• Ensure cryptography package is installable as an extra

v1.18.0

• Handle encrypted PDFs via --password option and/or prompting user for the password
• pdfalyze script now returns error code 1 to shell if there's unplaced nodes unless new --allow-missed-nodes option is used
• Upgrade pypdf to 6.6.0 and make use of new Font object
• Send logs to stderr instead of stdout, redirect and reformat pypdf logs, other logging improvements
• Sort Pdfalyzer.font_infos array by node ID
• Placement of formerly orphaned nodes:
  • Force stranded /Pages nodes to be children of /Catalog
  • Better placement of orphaned nodes that are members of an ArrayObject
  • Place special /Linearization nodes under root
  • Force /Xobject nodes with /Subtype of /Form to be children of /AcroForm nodes
  • Remove non_tree_relationships if there's an actual parent/child relationship
  • Insert grandparents in situations where there's nodes that are in any array but also claim a node other than the array is their parent

v1.17.13

• Remove YARA rule invalid_trailer_structure because it's causing YARA to crash with internal error 46: TOO_MANY_RE_FIBERS on some files (opened issue in YARA repo), fixes #15
• Indent the font character maps under the font info panel
• Bump yaralyzer to 1.0.11

v1.17.12

• Set max version of pypdf to below 6.6.0 because of breaking change with _cmap.build_char_map() from

v1.17.11

• Properly escape image OCR text with rich.markup.escape() when printing to page_buffer to avoid exceptions on weird OCR text
• Upgrade pypdf to 6.4.2, bump pymupdf

v1.17.10

• Fix logging bug in create_dir_if_it_does_not_exist()
• Upgrade pypdf to 6.4.0

v1.17.9

• Handle errors in FontInfo extraction more gracefully

v1.17.7

• Bump pypdf to 6.1.3 (fixes #31)
• Bump PyMuPDF to 1.26.5

v1.17.6

• Better handling for errors resulting from bugs in PyPDF
• Properly close file handle when pdfalyzing is complete