(*:maxLineLen=78:*)
theory Presentation
imports Base
begin
chapter ‹Presenting theories \label{ch:present}›
text ‹
Isabelle provides several ways to present the outcome of formal
developments, including WWW-based browsable libraries or actual printable
documents. Presentation is centered around the concept of 🪙‹sessions›
(\chref{ch:session}). The global session structure is that of a tree, with
Isabelle Pure at its root, further object-logics derived (e.g.\ HOLCF from
HOL, and HOL from Pure), and application sessions further on in the
hierarchy.
The command-line tools @{tool_ref mkroot} and @{tool_ref build} provide the
primary means for managing Isabelle sessions, including options for
presentation: ``🍋‹document=pdf›'' generates PDF output from the theory
session, and ``🍋‹document_output=dir›'' emits a copy of the document sources
with the PDF into the given directory (relative to the session directory).
Alternatively, @{tool_ref document} may be used to turn the generated
{\LaTeX} sources of a session (exports from its session build database) into PDF.
›
section ‹Generating HTML browser information \label{sec:info}›
text ‹
As a side-effect of building sessions, Isabelle is able to generate theory
browsing information, including HTML documents that show the theory sources
and the relationship with its ancestors and descendants. Besides the HTML
file that is generated for every theory, Isabelle stores links to all
theories of a session in an index file. As a second hierarchy, groups of
sessions are organized as 🪙‹chapters›,
implicit tree structure of the session build hierarchy is 🪙‹not› relevant
for the presentation.
🪙
To generate theory browsing information for an existing session, just invoke
@{tool build} with suitable options:
@{verbatim [display] ‹isabelle build -o browser_info -v -c FOL›}
The presentation output will appear in a sub-directory
🍋‹$ISABELLE_BROWSER_INFO›, according to the chapter and session name.
Many Isabelle sessions (such as 🪙‹HOL-Library› in
🍋‹~~/src/HOL/Library›) also provide theory documents in PDF. These are
prepared automatically as well if enabled like this:
@{verbatim [display] ‹isabelle build -o browser_info -o document -v -c HOL-Library›}
Enabling both browser info and document preparation simultaneously causes an
appropriate ``document'' link to be included in the HTML index. Documents
may be generated independently of browser information as well, see
\secref{sec:tool-document} for further details.
🪙
The theory browsing information is stored in the directory determined by the
@{setting_ref ISABELLE_BROWSER_INFO} setting, with sub-directory structure
according to the chapter and session name. In order to present Isabelle
applications on the web, the corresponding subdirectory from @{setting
ISABELLE_BROWSER_INFO} can be put on a WWW server.
›
section ‹Creating session root directories \label{sec:tool-mkroot}›
text ‹
The @{tool_def mkroot} tool configures a given directory as session root,
with some 🍋‹ROOT› f
@{verbatim [display]
‹Usage: isabelle mkroot [OPTIONS] [DIRECTORY]
Options are:
-A LATEX provide author in LaTeX notation (default: user name)
-I init Mercurial repository and add generated files
-T LATEX provide title in LaTeX notation (default: session name)
-n NAME alternative session name (default: directory base name)
-q quiet mode: less verbosity
Create session root directory (default: current directory).
›}
The results are placed
in the given directory
‹dir›, which refers
to the
current directory
by default. The @{tool mkroot} tool
is conservative
in the
sense that it does not overwrite existing files or directories. Earlier
attempts
to generate a session root need
to be deleted manually.
The generated session template will be accompanied
by a formal document,
with ‹DIRECTORY›\<^verbatim>
‹/document/root.tex› as its {
\LaTeX} entry point (see
also
\chref{ch:present}).
Options
🍋‹-T› and 🍋‹-A› specify the document title
and author explicitly,
using {
\LaTeX} source
notation.
Option
🍋‹-I› initializes a Mercurial repository
in the target directory,
and
adds all generated files (without
commit).
Option
🍋‹-n› specifies an alternative session name;
otherwise the base name
of the given directory
is used.
Option
🍋‹-q› reduces verbosity.
🪙
The implicit Isabelle settings variable @{setting ISABELLE_LOGIC} specifies
the parent session.
›
subsubsection
‹Examples›
text ‹
Produce session 🍋‹Test› w
@{verbatim [display] ‹isabelle mkroot -q Test && isabelle build -D Test›}
🪙
Upgrade the current directory into a session ROOT with document preparation,
and build it:
@{verbatim [display] ‹isabelle mkroot -q && isabelle build -D .›}
›
section ‹Preparing Isabelle session documents \label{sec:tool-document}›
text ‹
The @{tool_def document} tool prepares logic session documents. Its usage
is:
@{verbatim [display]
‹Usage: isabelle document [OPTIONS] SESSION
Options are:
-O DIR output directory for LaTeX sources and resulting PDF
-P DIR output directory for resulting PDF
-S DIR output directory for LaTeX sources
-V verbose latex
-d DIR include session directory
-o OPTION override Isabelle system OPTION (via NAME=VAL or NAME)
-v verbose build
Prepare the theory document of a session.›}
Generated {
\LaTeX} sources are taken
from the session build database:
@{tool_ref build}
is invoked beforehand
to ensure that it
is up-to-date.
Further files are generated on the spot, notably essential Isabelle style
files,
and 🍋‹session.tex› to input all
theory sources
from the session
(excluding
imports from other sessions).
🪙 Options
🍋‹-d›,
🍋‹-o›,
🍋‹-v› have the same meaning as
for @{tool
build}.
🪙 Option
🍋‹-V› prints full
output of {
\LaTeX} tools.
🪙 Option
🍋‹-O›~
‹dir› specifies the
output directory
for generated {
\LaTeX}
sources
and the result PDF
file. Options
🍋‹-P› and 🍋‹-S› only refer
to the
PDF
and sources, respectively.
For example,
for output directory ``
🍋‹output›''
and the default document
variant ``
🍋‹document›'', the generated document sources are placed into the
subdirectory
🍋‹output/document/› and the resulting PDF into
🍋‹output/document.pdf›.
🪙 Isabelle
is usually smart enough
to create the PDF
from the given
🍋‹root.tex› and optional
🍋‹root.bib› (bibliography)
and 🍋‹root.idx› (index)
using standard {
\LaTeX} tools. Actual command-lines are given
by settings
@{setting_ref ISABELLE_LUALATEX} (or @{setting_ref ISABELLE_PDFLATEX}),
@{setting_ref ISABELLE_BIBTEX}, @{setting_ref ISABELLE_MAKEINDEX}: these
variables are used without quoting
in shell scripts,
and thus may contain
additional options.
The system option @{system_option_def
"document_build"} specifies an
alternative build engine, e.g. within the session
🍋‹ROOT› file as
``
🍋‹options [document_build = pdflatex]›''. The following standard engines
are available:
🪙 🍋‹lualatex› (default)
uses the shell command
🍋‹$ISABELLE_LUALATEX› on
the main
🍋‹root.tex› file,
with further runs of
🍋‹$ISABELLE_BIBTEX› and
🍋‹$ISABELLE_MAKEINDEX› as required.
🪙 🍋‹pdflatex› uses 🍋‹$ISABELLE_PDFLATEX› instead of
🍋‹$ISABELLE_LUALATEX›,
and the other tools as above.
🪙 🍋‹build› invokes an executable script of the same name
in a private
directory containing all
\isakeyword{document
\_files}
and other generated
document sources. The script
is invoked as ``
🍋‹./build pdf›~
‹name›''
for
the document variant name; it needs
to produce a corresponding
‹name›\<^verbatim>
‹.pdf› file by arbitrary means on its own.
Further engines can be defined
by add-on components
in Isabelle/Scala
(
\secref{sec:scala-build}), providing a service
class derived
from
🍋‹isabelle.Document_Build.Engine›.
›
subsubsection
‹Examples›
text ‹
Produce the document from session 🍋‹FOL› w
the current directory (subdirectory 🍋‹document› and file 🍋‹document.pdf)›:
@{verbatim [display] ‹isabelle document -v -V -O. FOL›}
›
section ‹Full-text search for formal theory content›
text ‹
The session information of a regular @{tool_ref build} can also be used to
generate a search index for full-text search over formal theory content. To
that end, the 🍋‹Find_Facts› m
a full-text search engine, that can run embedded in a JVM process. Solr is
bundled as a separate Isabelle component, and its run-time dependencies
(as specified in @{setting SOLR_JARS}) need to be added separately to the
classpath of a regular Isabelle/Scala process.
🪙
A search index can be created using the @{tool_def find_facts_index} tool,
which has options similar to the regular @{tool_ref build}. User data such
as search indexes is stored in @{setting FIND_FACTS_HOME_USER}. The name of
the search index can be specified via system option
@{system_option find_facts_database_name}. A finished search index can be
packed for later use as a regular Isabelle component using the
@{tool_def find_facts_index_build} tool, with a 🍋‹.db› file and
🍋‹etc/settings› to augment @{setting FIND_FACTS_INDEXES} for use by @{tool
find_facts_server}.
🪙
The user interface of the search is available as web application that
can be started with the @{tool_def find_facts_server} tool. Its usage is:
@{verbatim [display]
‹Usage: isabelle find_facts_server [OPTIONS]
Options are:
-d devel mode
-o OPTION override Isabelle system OPTION (via NAME=VAL or NAME)
-p PORT explicit web server port
-v verbose server
Run server for Find_Facts.
›}
This Isabelle/Scala HTTP server provides the both the front-end
(implemented
in the Elm
🪙‹🪙‹https://elm-lang.org›\› language)
and REST
endpoints
for search queries
with JSON data.
Options
🍋‹-o› and 🍋‹-v› have the same meaning as
in @{tool build}.
Option
🍋‹-d› re-compiles the front-end
in 🍋‹$FIND_FACTS_HOME_USER/web›
on page reload (when sources are changed).
Option
🍋‹-p› specifies an explicit TCP port
for the server socket (assigned
by the operating system per default).
For public-facing servers, a common
scheme
is 🍋‹-p 8080› that
is access-restricted via firewall rules,
with a
reverse proxy
🪙‹E.g. via Caddy 🪙‹https://caddyserver.com/docs›\› in system
space (that
also handles SSL) on ports 80
and 443.
›
end