This is a style that is derivative of the article class to be used for legal
proceedings in the United States District Court, District of Oregon. It likely
follows most of the local rules for other federal district courts. Some control
sequences must be defined before loading the style, and here that is
accomplished by the line \input{test-params} to input the macros contained in
test-params.tex before \usepackage{usdc-or}.
This package automates the case caption, case header (=caption + counsel information, a macro should be provided to typeset the counsel information), signature, and certificate of service. It can be used with the legbib package, also by this author, to automate legal citations, including those to exhibits and to counsel (an @counsel entry is defined in a .bib used to typeset the counsel contact information by Biblatex). The LaTeX article class and other standard (tex-live) packages (hyperref, fancyhdr) are used to automate footers, page numbering, table of contents, section and paragraph references (including links), and backlinks.
A dot file showing an example document format is at doc/dom.dot. The document object model here assumes that all text is placed in a paragraph or subparagraph. But any depth of section (section, subsection, and subsubsection) can have a paragraph, and possibly then a subparagraph. There must be at least a section for any paragraph.
The references to sections are arabic numbers, the paragraphs alphabetical, and
the subparagraphs roman, with delimitation by dots between sections and opening
and closing parantheses around paragraphs and subparagraphs. Therefore the
general reference is, in an approximate regex,
\d+(\.\d+)?(\.\d+)?\([a-z]\)(\([romannumeral]\))? (see
https://stackoverflow.com/questions/267399 for regex for roman numerals). The
named destinations are given the same text as what is typeset to allow for easy
cross-referencing, with a prefix as to what (section, subsection,
subsubsection, paragraph, and subparagraph) is being referenced separated by a dot.
Redaction is often required in legal proceedings where there is protected information, examples being private information of individuals and commercially sensitive information of businesses. Generally redaction is done a posteriori using a pdf editor that removes the text data and puts a black stripe in its place. There is an advantage in a priori redaction, where protected information in the text is never typeset. This poses some technical difficulties for a program which does as much complicated computing as TeX does if one wants to do what might be called conservative a priori redaction, where there is no difference in the positions of the unredacted text between redacted and unredacted documents. The advantage of conservative a priori redaction is that jurists can then make equivalent references to sealed and unsealed documents in terms of positional locators (page and line numbers). But even if you replace all characters with boxes of the equivalent width, ligature and hyphenation will lead to different outputs of the unredacted text. The only way to accomplish conservative redaction, to the author's knowledge, is to use LuaLaTeX to have TeX construct the boxes for labeled characters, then remove the characters from the constructed boxes. The different a priori redaction techniques are reviewed at https://tex.stackexchange.com/questions/596971, including a possible virtual font solution. Here, nonconservative a priori redaction is supported. The command \litredact replaces characters with equivalent width white space (\litredact), and unless used very often or on large segments of text should be approximately conservative. There is also the highly nonconservative a priori redaction command \redact, that indicates how much redacted text there was by a character count or replaces the text to be redacted with a description of what was redacted.
Note that perfect redaction can be a security compromise for non-monospaced fonts, because an adversary could determine based on the dimensions of the unredacted text what sequence(s) of characters could have been used.
- Add cross-referencing (in \externalcitenameddest) to named destinations, which are the pdf equivalent of HTML fragments. All pdf viewers tested (Okular, Evince, PDFStudio) don't support links to named destinations, however.
- Add conservative a priori redaction (LuaLaTeX).
- Add redaction for section headings (protected commands).
- Add width-control for page marks to allow section headings without overlap (protected commands).