This specification is intended as an aid to preparing Bible Texts for use with various Bible search programs. While it is possible to parse and use these files directly, these files may be further processed by a Bible search program by indexing and/or converting to another format, such as the XML Scripture Encoding Model (XSEM) or OSIS. Software tools to do this are available.  GBF is not XML, because it doesn't use the same conventions for tags and because strict nesting is not required.  Conversions to other formats should be relatively simple. This format is designed to leave most of the detailed formatting up to the Bible search program, which may be running in DOS, Windows, Linux, or some other platform.

Since this format is still under development, some things might change, especially with respect to footnote and cross reference handling. There are still some things that are a bit ambiguous in this specification, and which may be clarified later with a reference implementation of a GBF reader. Check the master copy of this document before writing programs to take advantage of Bibles distributed in this format. This document was last updated 1 February 2004. Recent changes are marked in bold font.

General Bible Format (GBF) files are plain ASCII text files, with lines limited to no more than 254 characters. Lines are ended with the DOS standard CR-LF pair of characters. GBF files consist of a series of tokens, which may be tags, words, spaces, or punctuation. These tokens may not cross line boundaries. When read, line endings are the same as a single space, except when they immediately follow the <CM> token, in which case they are ignored. It is up to the program displaying or converting the text to another format to insert soft line breaks at the appropriate places. For ASCII values greater than 127, the U. S. Windows ANSI character set is assumed. (Currently, the only use for these characters is for typographic quotation marks.) GBF readers targeting DOS or Unix should convert characters in this range to the closest equivalents on those platforms.

There are several kinds of tags in GBF files, indicated by the first character after the "<" character. These include: file header tags, text type tags, file tail tags, font attribute tags, paragraph attribute tags, informational tags, sync marks, special characters, etc. All tags start with the less-than symbol (<) and two characters that identifies the tag and end with the greater-than symbol (>). Tag identifiers are case sensitive, and may be extended in the future to include additional letters, numbers, and other characters. Some tags take a parameter (like sync marks). This parameter is inserted just before the ending ">". Unrecognized tags (not in this specification) should be ignored by GBF readers aimed at this version of the GBF specification, preferably with a warning. In tags with start and stop versions, the second letter of the tag is upper case in the start tag, and lower case in the stop tag.

The text type mutually exclusive tag group specifies the logical contents of the following text and implies a default set of character and paragraph settings. The actual fonts and paragraph attributes assigned to each of these kinds of text is determined by the displaying program, preferably with some allowance for customization from the user.

File header tags are presented in the following order (except allowed repetitions noted below), starting at the beginning of the file.

File body tags are used as necessary to mark the type of content of the text.

File tail tags must be in the following order:

Text attribute tag pairs are inserted into the text as necessary to indicate text attributes like italics. These may or may not be properly represented in all Bible viewers due to platform limitations. All of these text attributes are assumed to be off at the beginning of each chapter book unless the start tag is explicitly repeated after the chapter sync mark. The stop tag for each font attribute should be explicitly inserted before the end of each book, and if appropriate, the start tag reasserted after the next book begins.

These tags describe attributes of paragraphs, like justification, indenting, etc. All of these attributes are assumed to be in the default state (non-indented prose, left justified, direction left-to-right) at the beginning of each chapter unless the appropriate start tag is repeated after the chapter sync mark. Justification tags are mutually exclusive, as are direction tags.

These tags provide various additional pieces of information about the text that are intended to be displayed on demand with a right mouse click, in a separate window, or at the bottom of a page. The start and stop tags indicate a range of words over which the footnote or reference apply. The sequence indicators shown as "seq" below are used to match the start and stop tags in case of overlapping references, and is a short string of numbers or letters that are guaranteed to be unique in the range of text they cover.

Word information tags:

Verse sync marks identify the current book, chapter, and verse. Each kind of sync mark is required at the beginning of the section (book, chapter, or verse) that it identifies. Sync marks may optionally be repeated within a section. If no specific number is specified in a sync mark, the number is assumed to be one more than the previous sync mark of the same kind.

Book markers may be either numeric or an English name or abbreviation as follows. The STEP # is not used in this format, but is listed for reference in conversions.

These tags indicate just a single character in the text.

External note and highlight files may either be prepared commentaries or notations made by the individual Bible student as he or she studies the Bible using a Bible search program. When a note and highlight file is open along with a Bible text, the notes are made available to the reader by a pop-up mechanism or a separate window, or printed at the bottom of a page as footnotes. The highlights are applied to the Bible text background as it is displayed. The note and highlight files are plain ASCII text, using the same tags as the Bible text for font and paragraph characteristics, plus the following:

[Home]