NAME

webster - access to the Merriam-Webster database

SYNOPSIS

#include <text/webster.h>

DESCRIPTION

These functions provide programmatic access to the Merriam-Webster database used by /NextApps/Webster.  They constitute the Merriam-Webster database access package of the text library, libtext.  The link editor searches this library under the "-ltext" option.  Declarations for these functions may be obtained either from the include file <text/webster.h>, or from the master include file <text/text.h>. 
 
The following program fragment prints a simple, unformatted definition for a word on the standard output:
 

#include "webster.h"
 void
main()
{
ReferenceBook ∗book;
Definition ∗d;
int exact = 1;
 book = referenceOpen("Webster-Dictionary");
dictionaryFoldSenses = 80; /∗ fold output lines to 80 chars ∗/
if (d = getDefinition("amphora", book, exact))
{
putDefinition(d, 0);
printf("0);
freeDefinition(d);
}
}
 

The Webster databases include the Ninth Collegiate Dictionary (tm) and the Collegiate Thesaurus (tm).  They are found in /NextLibrary/References/Webster-{Dictionary,Thesaurus}.  Each book consists of a source file and an index file; the dictionary also may have a full-index file, artwork in pictures/..., and some ancillary material in info/... - the “front matter” from the paper book. 
 
The conventions documented here are subject to change as multilingual dictionary formats emerge.  The source file is currently a sequential list of tags followed by data. A tag is a one byte code describing an element of an entry (e.g., an entry, a function, a pronunciation, etc.).  The formatting of the associated data is type-specific.  A series of one or two byte codes marking special glyphs and font changes is also used.  The index files are simple, alphabetically sorted files, each line containing a word and its byte offset location in the source file. 
 
The three principal data structures are SenseList, which is a linked list of the elements in an entry; Definition, which is a full entry; and ReferenceBook, which is a descriptor holding open file descriptors and other material describing a reference book:

 typedef struct SenseList
/∗ A SenseList is a tagged list of fields: s[0] is the tag,
 ∗ s[1] is the data. ∗/
{
char ∗s;
int n;
struct SenseList ∗next;
} SenseList;
 typedef struct
/∗ A Definition entry contains any of a number of fields, all
 ∗optional. The field pointers are taken from the SenseList,
 ∗ and point to the first of 0 or more fields of the appropriate
 ∗ type which may be present in the entry. ∗/
{
char ∗entry;/∗ the main entry word ∗/
char ∗dotted;/∗ the dotted form of the word ∗/
char ∗pronunciation;/∗ the pronunciation field ∗/
char ∗function;/∗ the grammatical function ∗/
char ∗date;/∗ date word was first used ∗/
char ∗etymology;/∗ etymological description ∗/
char ∗inflection;/∗ inflected forms (-ed, -ies, -ing, etc.) ∗/
char ∗variant;/∗ variant forms and inflections ∗/
char ∗picture;/∗ pathname of file containing picture, if any ∗/
char ∗subject;/∗ internal subject codes, per-sense ∗/
SenseList ∗l;/∗ linked list of senses comprising the entry ∗/
int section;/∗ one of: dMain, dAbbr, dForeign, dBio, dGeo, etc. ∗/
long offset;/∗ where in the book it came from ∗/
struct ReferenceBook ∗book;/∗ what book it came from ∗/
} Definition;
  typedef struct ReferenceBook
/∗ A reference book includes a tagged source file, an index,
 ∗ and possibly a full-text index. This data structure also
 ∗ contains pointers to book-specific i/o routines. ∗/
{
FILE ∗source, ∗index, ∗fullIndex;
Definition ∗(∗getDef)();
int (∗putDef)(), (∗freeDef)();
} ReferenceBook;

SUMMARY

 
ReferenceBook ∗referenceOpen(char ∗name);

Opens reference book name for searching and returns a pointer to the open descriptor, or zero in the event of failure.  Currently name may be one of Webster-Dictionary or Webster-Thesaurus. 

int referenceClose(ReferenceBook ∗r);

Closes r, and frees the associated storage. 

Definition ∗nextDefinition(ReferenceBook ∗book);

Reads and returns the next definition from book.  Steps through the list of definitions in the book. 

Definition ∗getDefinition(char ∗word, ReferenceBook ∗book, int match);

Gets the first definition for word in book, and returns a pointer to the open descriptor, or zero if no definition was found.  If match is true, the word must match exactly, otherwise prefix matches are accepted. 

Definition ∗ getNextDefinition(char ∗word, ReferenceBook ∗book, int match);

Returns the next definition matching word, or zero if no there are more.  This routine may be called after an initial call to getDefinition(). 

int freeDefinition(Definition ∗d);

Frees the storage used by d. 

int putDefinition(Definition ∗d, int (∗output)());

Generates the definition d, calling the function output for each character processed.  The information is preconverted at lookup time to some output format (dictionaryOutputFormat, one of W_ASCII, W_PS, etc).  If no output function is given the information is written to the standard output. 
 
The W_ASCII output format is the default, as it is likely to be the most useful for computational linguistic applications. When W_ASCII is used, the global variable dictionaryFoldSenses should be set to the desired line width in characters. 

SEE ALSO

text(3)

4th Berkeley Distribution  —  July 7, 1989