annotate doc/misc/semantic.texi @ 106534:cad98e90ecc5

* verilog-mode.el (verilog-vmm-begin-re, verilog-vmm-end-re, verilog-vmm-statement-re, verilog-ovm-statement-re, verilog-defun-level-not-generate-re, verilog-calculate-indent, verilog-leap-to-head, verilog-backward-token): Fix indenting VMM macros. Reported by Jonathan Ashbrook. * verilog-mode.el (verilog-auto-lineup, verilog-nameable-item-re): Cleanup user-visible spelling and documentation errors. One reported by Gary Delp. (verilog-submit-bug-report): Mention bug tracking and CC co-author. (verilog-read-decls): Fix AUTOWIRE with types declared in a package, bug195. Reported by Pierre-David Pfister.

\input texinfo
@setfilename ../../info/semantic
@set TITLE  Semantic Manual
@set AUTHOR Eric M. Ludlam and David Ponce
@settitle @value{TITLE}

@c *************************************************************************
@c @ Header
@c *************************************************************************

@c Merge all indexes into a single index for now.
@c We can always separate them later into two or more as needed.
@syncodeindex vr cp
@syncodeindex fn cp
@syncodeindex ky cp
@syncodeindex pg cp
@syncodeindex tp cp

@c @footnotestyle separate
@c @paragraphindent 2
@c @@smallbook
@c %**end of header

@copying
This manual documents the Semantic library and utilities.

Copyright @copyright{} 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2007,
2009 Free Software Foundation, Inc.

@quotation
Permission is granted to copy, distribute and/or modify this document
under the terms of the GNU Free Documentation License, Version 1.3 or
any later version published by the Free Software Foundation; with no
Invariant Sections, with the Front-Cover texts being ``A GNU Manual,''
and with the Back-Cover Texts as in (a) below.  A copy of the license
is included in the section entitled ``GNU Free Documentation License.''

(a) The FSF's Back-Cover Text is: ``You have the freedom to copy and
modify this GNU manual.  Buying copies from the FSF supports it in
developing GNU and promoting software freedom.''
@end quotation
@end copying

@ifinfo
@format
START-INFO-DIR-ENTRY
* Semantic: (semantic).      Source code parser library and utilities.
END-INFO-DIR-ENTRY
@end format
@end ifinfo

@titlepage
@center @titlefont{Semantic}
@sp 4
@center by @value{AUTHOR}
@end titlepage
@page

@macro semantic{}
@i{Semantic}
@end macro

@macro keyword{kw}
@anchor{\kw\}
@b{\kw\}
@end macro

@macro obsolete{old,new}
@sp 1
@strong{Compatibility}:
@code{\new\} introduced in @semantic{} version 2.0 supercedes
@code{\old\} which is now obsolete.
@end macro

@c *************************************************************************
@c @ Document
@c *************************************************************************
@contents

@node top
@top @value{TITLE}

@semantic{} is a suite of Emacs libraries and utilities for parsing
source code.  At its core is a lexical analyzer and two parser
generators (@code{bovinator} and @code{wisent}) written in Emacs Lisp.
@semantic{} provides a variety of tools for making use of the parser
output, including user commands for code navigation and completion, as
well as enhancements for imenu, speedbar, whichfunc, eldoc,
hippie-expand, and several other parts of Emacs.

To send bug reports, or participate in discussions about semantic,
use the mailing list cedet-semantic@@sourceforge.net via the URL:
@url{http://lists.sourceforge.net/lists/listinfo/cedet-semantic}

@ifnottex
@insertcopying
@end ifnottex

@menu
* Introduction::
* Using Semantic::
* Semantic Internals::
* Glossary::
* GNU Free Documentation License::
* Index::
@end menu

@node Introduction
@chapter Introduction

This chapter gives an overview of @semantic{} and its goals.

Ordinarily, Emacs uses regular expressions (and syntax tables) to
analyze source code for purposes such as syntax highlighting.  This
approach, though simple and efficient, has its limitations: roughly
speaking, it only ``guesses'' the meaning of each piece of source code
in the context of the programming language, instead of rigorously
``understanding'' it.

@semantic{} provides a new infrastructure to analyze source code using
@dfn{parsers} instead of regular expressions.  It contains two
built-in parser generators (an @acronym{LL} generator named
@code{Bovine} and an @acronym{LALR} generator named @code{Wisent},
both written in Emacs Lisp), and parsers for several common
programming languages.  It can also make use of @dfn{external
parsers}---programs such as GNU Global and GNU IDUtils.

@semantic{} provides a uniform, language-independent @acronym{API} for
accessing the parser output.  This output can be used by other Emacs
Lisp programs to implement ``syntax-aware'' behavior.  @semantic{}
itself includes several such utilities, including user-level Emacs
commands for navigating, searching, and completing source code.

The following diagram illustrates the structure of the @semantic{}
package:

@table @strong
@item Please Note:
The words in all-capital are those that @semantic{} itself provides.
Others are current or future languages or applications that are not
distributed along with @semantic{}.
@end table

@example
                                                             Applications
                                                                 and
                                                              Utilities
                                                                -------
                                                               /       \
               +---------------+    +--------+    +--------+
         C --->| C      PARSER |--->|        |    |        |
               +---------------+    |        |    |        |
               +---------------+    | COMMON |    | COMMON |<--- SPEEDBAR
      Java --->| JAVA   PARSER |--->| PARSE  |    |        |
               +---------------+    | TREE   |    | PARSE  |<--- SEMANTICDB
               +---------------+    | FORMAT |    | API    |<--- ecb
    Scheme --->| SCHEME PARSER |--->|        |    |        |
               +---------------+    |        |    |        |
               +---------------+    |        |    |        |
   Texinfo --->| TEXI.  PARSER |--->|        |    |        |
               +---------------+    |        |    |        |

                    ...                ...           ...         ...

               +---------------+    |        |    |        |<--- app. 1
   Lang. A --->| A      Parser |--->|        |    |        |
               +---------------+    |        |    |        |<--- app. 2
               +---------------+    |        |    |        |
   Lang. B --->| B      Parser |--->|        |    |        |<--- app. 3
               +---------------+    |        |    |        |

                     ...        ...     ...          ...       ...

               +---------------+    |        |    |        |
   Lang. Y --->| Y      Parser |--->|        |    |        |<--- app. ?
               +---------------+    |        |    |        |
               +---------------+    |        |    |        |<--- app. ?
   Lang. Z --->| Z      Parser |--->|        |    |        |
               +---------------+    +--------+    +--------+
@end example

@menu
* Semantic Components::
@end menu

@node Semantic Components
@section Semantic Components

In this section, we provide a more detailed description of the major
components of @semantic{}, and how they interact with one another.

The first step in parsing a source code file is to break it up into
its fundamental components.  This step is called lexical analysis:

@example
        syntax table, keywords list, and options
                         |
                         |
                         v
    input file  ---->  Lexer   ----> token stream
@end example

@noindent
The output of the lexical analyzer is a list of tokens that make up
the file.  The next step is the actual parsing, shown below:

@example
                    parser tables
                         |
                         v
    token stream --->  Parser  ----> parse tree
@end example

@noindent
The end result, the parse tree, is @semantic{}'s internal
representation of the language grammar.  @semantic{} provides an
@acronym{API} for Emacs Lisp programs to access the parse tree.

Parsing large files can take several seconds or more.  By default,
@semantic{} automatically caches parse trees by saving them in your
@file{.emacs.d} directory.  When you revisit a previously-parsed file,
the parse tree is automatically reloaded from this cache, to save
time.  @xref{SemanticDB}.

@node Using Semantic
@chapter Using Semantic

@include sem-user.texi

@node Semantic Internals
@chapter Semantic Internals

This chapter provides an overview of the internals of @semantic{}.
This information would not be needed by neither application developers
nor grammar developers.

It would be useful mostly for the hackers who would like to learn
more about how @semantic{} works.

@menu
* Parser code ::	  Code used for the parsers
* Tag handling ::	  Code used for manipulating tags
* Semanticdb Internals :: Code used in the semantic database
* Analyzer Internals ::   Code used in the code analyzer
* Tools ::                Code used in user tools
* Tests ::                Code used for testing
@end menu

@node Parser code
@section Parser code

@semantic{} parsing code is spread across a range of files.

@table @file
@item semantic.el
The core infrastructure sets up buffers for parsing, and has all the
core parsing routines.  Most parsing routines are overloadable, so the
actual implementation may be somewhere else.

@item semantic-edit.el
Incremental reparse based on user edits.

@item semantic-grammar.el
@itemx semantic-grammar.wy
Parser for the different grammar languages, and a major mode for
editing grammars in Emacs.

@item semantic-lex.el
Infrastructure for implementing lexical analyzers.  Provides macros
for creating individual analyzers for specific features, and a way to
combine them together.

@item semantic-lex-spp.el
Infrastructure for a lexical symbolic preprocessor.  This was written
to implement the C preprocessor, but could be used for other lexical
preprocessors.

@item bovine/bovine-grammar.el
@itemx bovine/bovine-grammar-macros.el
@itemx bovine/semantic-bovine.el
The ``bovine'' grammar.  This is the first grammar mode written for
@semantic{} and is useful for simple creating simple parsers.

@item wisent/wisent.el
@itemx wisent/bison-wisent.el
@itemx wisent/semantic-wisent.el
@itemx wisent/semantic-debug-grammar.el
A port of bison to Emacs.  This infrastructure lets you create LALR
based parsers for @semantic{}.

@item semantic-ast.el
Manage Abstract Syntax Trees for parsers.

@item semantic-debug.el
Infrastructure for debugging grammars.

@item semantic-util.el
Various utilities for manipulating tags, such as describing the tag
under point, adding labels, and the all important
@code{semantic-something-to-tag-table}.

@end table

@node Tag handling
@section Tag handling

A tag represents an individual item found in a buffer, such as a
function or variable.  Tag handling is handled in several source
files.

@table @file
@item semantic-tag.el
Basic tag creation, queries, cloning, binding, and unbinding.

@item semantic-tag-write.el
Write a tag or tag list to a stream.  These routines are used by
@file{semanticdb-file.el} when saving a list of tags.

@item semantic-tag-file.el
Files associated with tags.  Goto-tag, file for include, and file for
a prototype.

@item semantic-tag-ls.el
Language dependant features of a tag, such as parent calculation, slot
protection, and other states like abstract, virtual, static, and leaf.

@item semantic-dep.el
Include file handling.  Contains the include path concepts, and
routines for looking up file names in the include path.

@item semantic-format.el
Convert a tag into a nicely formatted and colored string.  Use
@code{semantic-test-all-format-tag-functions} to test different output
options.

@item semantic-find.el
Find tags matching different conditions in a tag table.
These routines are used by @file{semanticdb-find.el} once the database
has been converted into a simpler tag table.

@item semantic-sort.el
Sorting lists of tags in different ways.  Includes sorting a plain
list of tags forward or backward.  Includes binning tags based on
attributes (bucketize), and tag adoption for multiple references to
the same thing.

@item semantic-doc.el
Capture documentation comments from near a tag.

@end table

@node Semanticdb Internals
@section Semanticdb Internals

@acronym{Semanticdb} complexity is certainly an issue.  It is a rather
hairy problem to try and solve.

@table @file
@item semanticdb.el
Defines a @dfn{database} and a @dfn{table} base class.  You can
instantiate these classes, and use them, but they are not persistent.

This file also provides support for @code{semanticdb-minor-mode},
which automatically associates files with tables in databases so that
tags are @emph{saved} while a buffer is not in memory.

The database and tables both also provide applicate cache information,
and cache flushing system.  The semanticdb search routines use caches
to save datastructures that are complex to calculate.

Lastly, it provides the concept of @dfn{project root}.  It is a system
by which a file can be associated with the root of a project, so if
you have a tree of directories and source files, it can find the root,
and allow a tag-search to span all available databases in that
directory hierarchy.

@item semanticdb-file.el
Provides a subclass of the basic table so that it can be saved to
disk.  Implements all the code needed to unbind/rebind tags to a
buffer and writing them to a file.

@item semanticdb-el.el
Implements a special kind of @dfn{system} database that uses Emacs
internals to perform queries.

@item semanticdb-ebrowse.el
Implements a system database that uses Ebrowse to parse files into a
table that can be queried for tag names.  Successful tag hits during a
find causes @semantic{} to pick up and parse the reference files to
get the full details.

@item semanticdb-find.el
Infrastructure for searching groups @semantic{} databases, and dealing
with the search results format.

@item semanticdb-ref.el
Tracks crossreferences.   Cross references are needed when buffer is
reparsed, and must alert other tables that any dependant caches may
need to be flushed.  References are in the form of include files.

@end table

@node Analyzer Internals
@section Analyzer Internals

The @semantic{} analyzer is a complex engine which has been broken
down across several modules.  When the @semantic{} analyzer fails,
start with @code{semantic-analyze-debug-assist}, then dive into some
of these files.

@table @file
@item semantic-analyze.el
The core analyzer for defining the @dfn{current context}.  The
current context is an object that contains references to aspects of
the local context including the current prefix, and a tag list
defining what the prefix means.

@item semantic-analyze-complete.el
Provides @code{semantic-analyze-possible-completions}.

@item semantic-analyze-debug.el
The analyzer debugger.  Useful when attempting to get everything
configured.

@item semantic-analyze-fcn.el
Various support functions needed by the analyzer.

@item semantic-ctxt.el
Local context parser.  Contains overloadable functions used to move
around through different scopes, get local variables, and collect the
current prefix used when doing completion.

@item semantic-scope.el
Calculate @dfn{scope} for a location in a buffer.  The scope includes
local variables, and tag lists in scope for various reasons, such as
C++ using statements.

@item semanticdb-typecache.el
The typecache is part of @code{semanticdb}, but is used primarilly by
the analyzer to look up datatypes and complex names.  The typecache is
bound across source files and builds a master lookup table for data
type names.

@item semantic-ia.el
Interactive Analyzer functions.  Simple routines that do completion or
lookups based on the results from the Analyzer.  These routines are
meant as examples for application writers, but are quite useful as
they are.

@item semantic-ia-sb.el
Speedbar support for the analyzer, displaying context info, and
completion lists.

@end table

@node Tools
@section Tools

These files contain various tools a user can use.

@table @file
@item semantic-idle.el
Idle scheduler for @semantic{}.  Manages reparsing buffers after
edits, and large work tasks in idle time.  Includes modes for showing
summary help and pop-up completion.

@item senator.el
The @semantic{} navigator.  Provides many ways to move through a
buffer based on the active tag table.

@item semantic-decorate.el
A minor mode for decorating tags based on details from the parser.
Includes overlines for functions, or coloring class fields based on
protection.

@item semantic-decorate-include.el
A decoration mode for include files, which assists users in setting up
parsing for their includes.

@item semantic-complete.el
Advanced completion prompts for reading tag names in the minibuffer, or
inline in a buffer.

@item semantic-imenu.el
Imenu support for using @semantic{} tags in imenu.

@item semantic-mru-bookmark.el
Automatic bookmarking based on tags.  Jump to locations you've been
before based on tag name.

@item semantic-sb.el
Support for @semantic{} tag usage in Speedbar.

@item semantic-util-modes.el
A bunch of small minor-modes that exposes aspects of the semantic
parser state.  Includes @code{semantic-stickyfunc-mode}.

@item document.el
@itemx document-vars.el
Create an update comments for tags.

@item semantic-adebug.el
Extensions of @file{data-debug.el} for @semantic{}.

@item semantic-chart.el
Draw some charts from stats generated from parsing.


@item semantic-elp.el
Profiler for helping to optimize the @semantic{} analyzer.


@end table

@node Tests
@section Tests

@table @file

@item semantic-utest.el
Basic testing of parsing and incremental parsing for most supported
languages.

@item semantic-ia-utest.el
Test the semantic analyzer's ability to provide smart completions.

@item semantic-utest-c.el
Tests for the C parser's lexical pre-processor.

@item semantic-regtest.el
Regression tests from the older Semantic 1.x API.

@end table

@node Glossary
@appendix Glossary

@table @keyword
@item BNF
In semantic 1.4, a BNF file represented ``Bovine Normal Form'', the
grammar file used for the 1.4 parser generator.  This was a play on
Backus-Naur Form which proved too confusing.

@item bovinate
A verb representing what happens when a bovine parser parses a file.

@item bovine lambda
In a bovine, or LL parser, the bovine lambda is a function to execute
when a specific set of match rules has succeeded in matching text from
the buffer.

@item bovine parser
A parser using the bovine parser generator.  It is an LL parser
suitible for small simple languages.

@item context

@item LALR

@item lexer
A program which converts text into a stream of tokens by analyzing
them lexically.  Lexers will commonly create strings, symbols,
keywords and punctuation, and strip whitespaces and comments.

@item LL

@item nonterminal
A nonterminal symbol or simply a nonterminal stands for a class of
syntactically equivalent groupings.  A nonterminal symbol name is used
in writing grammar rules.

@item overloadable
Some functions are defined via @code{define-overload}.
These can be overloaded via ....

@item parser
A program that converts @b{tokens} to @b{tags}.

@item tag
A tag is a representation of some entity in a language file, such as a
function, variable, or include statement.  In semantic, the word tag is
used the same way it is used for the etags or ctags tools.

A tag is usually bound to a buffer region via overlay, or it just
specifies character locations in a file.

@item token
A single atomic item returned from a lexer.  It represents some set
of characters found in a buffer.

@item token stream
The output of the lexer as well as the input to the parser.

@item wisent parser
A parser using the wisent parser generator.  It is a port of bison to
Emacs Lisp.  It is an LALR parser suitable for complex languages.
@end table


@node GNU Free Documentation License
@appendix GNU Free Documentation License
@include doclicense.texi

@node Index
@unnumbered Index
@printindex cp

@iftex
@contents
@summarycontents
@end iftex

@bye

@c Following comments are for the benefit of ispell.

@c LocalWords: alist API APIs arg argc args argv asis assoc autoload Wisent
@c LocalWords: backquote bnf bovinate bovinates LALR
@c LocalWords: bovinating bovination bovinator bucketize
@c LocalWords: cb cdr charquote checkcache cindex CLOS
@c LocalWords: concat concocting const constantness ctxt Decl defcustom
@c LocalWords: deffn deffnx defun defvar destructor's dfn diff dir
@c LocalWords: doc docstring EDE EIEIO elisp emacsman emph enum
@c LocalWords: eq Exp EXPANDFULL expresssion fn foo func funcall
@c LocalWords: ia ids iff ifinfo imenu imenus init int isearch itemx java kbd
@c LocalWords: keymap keywordtable lang languagemode lexer lexing Ludlam
@c LocalWords: menubar metaparent metaparents min minibuffer Misc mode's
@c LocalWords: multitable NAvigaTOR noindent nomedian nonterm noselect
@c LocalWords: nosnarf obarray OLE OO outputfile paren parsetable POINT's
@c LocalWords: popup positionalonly positiononly positionormarker pre
@c LocalWords: printf printindex Programmatically pt punctuations quotemode
@c LocalWords: ref regex regexp Regexps reparse resetfile samp sb
@c LocalWords: scopestart SEmantic semanticdb setfilename setq
@c LocalWords: settitle setupfunction sexp sp SPC speedbar speedbar's
@c LocalWords: streamorbuffer struct subalist submenu submenus
@c LocalWords: subsubsection sw sym texi texinfo titlefont titlepage
@c LocalWords: tok TOKEN's toplevel typemodifiers uml unset untar
@c LocalWords: uref usedb var vskip xref yak

@ignore
   arch-tag: cbc6e78c-4ff1-410e-9fc7-936487e39bbf
@end ignore