[#S] -- Align next word or group per tab stop S. E.g. [#3]word tabs "word" according to the current setting for stop 3. (See [ctb] & [stb] for clearing and setting stops.)

[#S:F~X] -- Align next word or group per tab stop S with intervening fill type F, character X. E.g. [#3:p~"]word tabs "word" per stop 3, with partial fill using fill character (") (in braille, dot 5).

[/] -- Translation code, to prevent contractions straddling the code. E.g. line[/]age would assure that no "ea" contraction is used in the word lineage, as appropriate when the word has the unusual meaning "quantity of lines"

[:]-- Start word group to be treated as one (for alignment). E.g. [fr][:]several words[;] will set "several words" flush right on the line.

[;] -- End word group to be treated as one (for alignment). See [:] (above)

[<] -- Force new line (hard return); similar in effect to [sl1] and [hl]. See [ki] regarding interruption of a protected block.

[>] -- Tab to next stop, according to its alignment. E.g. [>]word tabs "word" according to the setting for the next tab stop to the right on the current line.

[°...] -- Synonym for [#...]

[ab] -- Equivalent to [g2] (French "abrégé")

[aln...] -- Synonym for [ind...]

[atxN] -- The [atx1] code initiates a mode wherein all ordinary (printable) ASCII characters are considered to be literal text, even those characters and sequences that are defined as special in-text controls according to the translation table in use. An example of such a control would be the two-character sequence &+, which forces a "letter sign" when certain tables are in use. An [atx0] command turns the "all ASCII is text" mode off, restoring the normal condition wherein any in-text controls that are defined for the table in use are recognized and processed as such. [atxN] is defined only for certain tables, as it is not relevant for those tables for which there are no special in-text controls defined. (Click here to see list of special in-text controls)

[bar] -- Translation code, providing a short way to signal that the previous letter or digit has a "bar" over it. See also the [bolim] and [bulim] codes. Note that more generally [e]...[os]―[oe] can be used for "bar" over any expression, e.g. [e]x+y[os]―[oe] would mean a bar over the entire expression "x+y". (Note: The "―" symbol shown above is a Horizontal Bar - Unicode U+2015 - DUSCI designation D+ec45)

[be] -- Base/subscript end; see [bs].

[bline] -- Translation code, used in Nemeth code only, where it is necessary to force inclusion of a baseline/multipurpose indicator (dot 5). (Most such indicators are produced automatically in translation.) See also [tcs] and [ts].

[bolim] -- Translation code, used for "lim" with bar over (upper limit).

[bs] -- Translation code, to mark the beginning of a base or subscript level. The [bs] ... [be] pair may include any expression, including further levels of subscripts; i.e. the codes may be nested. E.g. log[bs]2[be] x[bs]i[be]means "log to the base 2 of x sub i". See also [tcs] and [ts].

[bsfe] -- Translation code (British tables only); terminates [bsfs]...

[bsfs] is a translation code, presently applicable only in the British tables, which can be used to begin the name of a special math function, in the case of uncommon functions that are not directly recognized (most are recognized). For example, [bsfs]tr[bsfe](x) would cause the "tr" in "tr(x)" to be treated as a function name rather than the product of t and r.

[bulim] -- Translation code, used for "lim" with bar under (lower limit).

[cap-invert] -- Translation code applicable only in computer notation to be transcribed according to BANA Computer Braille Code (CBC); causes "inverted" treatment of capitals (i.e. lowercase is indicated, uppercase is not).

[cap-normal] -- Reverses the effect of [cap-invert] (see above).

[caplvN] -- Translation code to determine whether capital letters are to be shown in braille and if so, under what circumstances. This code is applicable to certain tables only; the documentation for the particular table within the translation tables usage guide should be consulted as to what values of N are valid, and their meanings.

[caplv1] -- Suppress capitals except in technical notation (math) and BCN computerese)

[caplv3] -- Use capitals everywhere -- which is the default, that is initial, condition.
These are actually effective only when using "British with capitals" translation table selection; "- without capitals" suppresses all capitals in all notations regardless of the "caplv" setting.

[cb] -- Translation code; when supported by a particular translation table, this coded marks the start of text that is to be transcribed into a form of computer notation. The might include CBC (for BANA rules), British Computer Braille, or French Computer Braille. The computer notation is terminated either a [tx] or [cz] code (each of which initiates a different form of translation. See also [cbi] and [cbn].

[cb-&] -- Translation code applicable only in computer notation to be transcribed according to BANA Computer Braille Code (CBC) (as is true of all the [cb-...] codes following in this list). Explicitly inserts a CBC "continuation indicator" into the braille. (See also the [wb-cb] for a more automatic, hence generally better way to provide for continuation indicators.)

[cb-de] -- Translation code similar to [cb-&] above; explicitly inserts ending indicator for CBC "halfline down" mode.

[cb-ds] -- Translation code similar to [cb-&] above; explicitly inserts starting indicator for CBC "halfline down" mode.

[cb-ee] -- Translation code similar to [cb-&] above; explicitly inserts ending indicator for CBC "emphasized" mode.

[cb-es] -- Translation code similar to [cb-&] above; explicitly inserts starting indicator for CBC "emphasized" mode.

[cb-ne] -- Translation code similar to [cb-&] above; explicitly inserts ending indicator for CBC "Nemeth" mode.

[cb-ns] -- Translation code similar to [cb-&] above; explicitly inserts starting indicator for CBC "Nemeth" mode.

[cb-se] -- Translation code similar to [cb-&] above; explicitly inserts ending indicator for CBC "shape" mode.

[cb-ss] -- Translation code similar to [cb-&] above; explicitly inserts starting indicator for CBC "shape" mode.

[cb-t1] -- Translation code similar to [cb-&] above; explicitly inserts CBC "transcriber's primary option symbol".

[cb-t2] -- Translation code similar to [cb-&] above; explicitly inserts CBC "transcriber's secondary option symbol".

[cb-ue] -- Translation code similar to [cb-&] above; explicitly inserts ending indicator for CBC "halfline up" mode.

[cb-us] -- Translation code similar to [cb-&] above; explicitly inserts starting indicator for CBC "halfline up" mode.

[cbc...] -- Synonym for [hds...].

[cbi] -- Translation code, a variant of [cb], which has the same effect and also unconditionally puts the CBC start-CBC indicator into the braille.

[cbn] -- Translation code, a variant of [cb], which has the same effect and also unconditionally will NOT put the CBC start-CBC indicator into the braille.

[cbp] -- Synonym for [hde].

[cd...] -- Synonym for [fr...].

[cpN] -- (Conditional Page Eject) Position on new line (if necessary), then eject page if fewer than N+1 lines remain. E.g. [cp4] would start a new page if fewer than 5 blank lines remained, and otherwise would be equivalent to [l].

[cpN:L] -- Eject to new page under the same conditions as [cpN], and otherwise skip L lines. E.g. [cp4:1].

[cs] -- Translation code applicable only in computer notation to be transcribed according to BANA Computer Braille Code (CBC); causes explicitly indicated "countable spaces" to be used where appropriate (i.e. where long sequences of spaces occur) in computer notation.

[cs-off] -- Reverses the effect of [cs]. (see above).

[ctb] -- Clear all tab stop settings (usually prior to a series of [stb...] settings). E.g. [ctb][stb1:l:5][stb2:d:22] would clear tabs, then set stop 1 for left alignment at position 5 and stop 2 for decimal alignment at position 22.

[cu...] -- Synonym for [pi...].

[cz] -- Translation code, initiates material already directly in braille (North American ASCII braille code), sometimes also called "computer grade 0" braille. See also [tx] and [cb], which initiate alternate modes, terminating the effect of [cz]; and also [d...] and [q...] for other ways of setting short braille sequences directly.

[d~W] -- Set word W directly (unchanged) in output, followed by space. May be used for forcing braille. E.g. [d~,8x0'] would et ",8x0'" (braille dots 6,2-3-6,1-3-4-6,3-5-6,3). See also [q...] (which is now generally preferred).

[e] -- Translation code, to mark the beginning of an expression modified by a directly-under subscript expression and/or a directly-over superscript expression. Note: when the same base expression is affected by both directly-under and directly-over indices, the former should be given first. E.g. [e]Σ[us]i = 1[ue][os]n[oe]x[bs]i[be] means "the sum (signified by a capital Greek sigma) from i = 1 to n of x sub i". See also [tcs] and [ts]. (Note: A Greek capital Sigma "Σ" is used above - Unicode designation U+03a3 - DUSCI D+e253)

[ecane] -- End of cancelled expression; see [ecans].

[ecans] -- Translation code, to mark the beginning of a cancelled expression. E.g. [ecans]x+1[ecane] signifies that the expression x+1 is shown as cancelled, i.e. has a line through it. See also [tcs] and [ts].

[ee~E] -- End text element (style) E. See [es...].

[enclis] -- Translation code, used in Nemeth code only, to signify that a list of items qualifies as an "enclosed list" and therefore is to be translated in the more efficient manner reserved for such lists. See Nemeth manual for a full definition of such lists. This code is placed before the opening enclosure symbol (e.g. left parenthesis) that starts the list. E.g. [enclis](a, a+1, ...) signifies that the entire list meets the definition of an "enclosed list". See also [tcs] and [ts].

[eng] -- Translation code, initiate English text (in Swahili tables only). See [lng~...] and [swa].

[eqvn] -- Translation code, used primarily in Nemeth code, to signify that a letter stands for a digit when it occurs first in a number. E.g. [eqvn]t2e4 signifies that the t stands for a digit, e.g. 3. See also [tcs] and [ts].

[es~E] -- Start text element (style) E. E.g. [es~h1.]A Heading[ee~h1.] would mark the text "A Heading" as an element of type "h1.".

[etb...] -- Synonym for [stb...].

[evb] -- Translation code, to stand in for an extended vertical bar, as in a matrix (not to be confused with the ASCII vertical bar, which is an ordinary character). The code is repeated on each line as necessary, aligned with those above and below that form a single vertical line. See also [tcs] and [ts].

[fc...] -- Synonym for [sd...].

[fe] -- End of fraction; see [fs].

[fl] -- Line within fraction; see [fs].

[fl-lifg] -- Translation code, initiates mode in which grade 1 is presumed to be Latin, Italian, French or German text (in American tables only). Allowed for historical compatibility; see [lng~...] for currently preferred way to switch languages. See also [fl-span] and [fl-none].

[fl-none] -- Translation code, initiates mode in which grade 1 carries no presumption as to language. (This is now the normal mode.) See also [fl-span] and [fl-lifg], which are cancelled by this code.

[fl-span] -- Translation code, initiates mode in which grade 1 is presumed to be Spanish (in American tables only). Allowed for historical compatibility; see [lng~...] for currently preferred way to switch languages. See also [fl-lifg] and [fl-none].

[foldsN~X] -- Set up for N fold lines, that is lines that will be skipped in the braille to permit folding of the sheet, for example to mail as a letter. The default for N is 2. X is an optional fill or score character. i.e. 3 will produce dots 2 & 5 [folds2~3] will result it 2 fold lines or dots 2 & 5. The ~x parameter may be omitted if preferred.

[fr] -- Align next word or group flush-right on line (ignoring any right margin). See [:].

[fr;F~X] -- Align flush-right, with intervening fill type F using fill character X. If X comprises two characters, the first is used for braille, and the second for print; otherwise character X is used for both media. E.g. [fr;p~".]word would set "word" flush right on the line, with partial fill (i.e. a space at each end of the fill) consisting of (") (braille dot 5) in the braille, and "." in the print.

[fs] -- Translation code, to mark the beginning of a fraction. E.g. [fs]x+1[fl]b[fe] means the fraction whose numerator is x+1 and whose denominator is b. See also [tcs] and [ts].

[ftb...] -- Synonym for [stb...].

[fte~X] -- End of font/typeform passage; see [fts~X].

[fts~X] -- Translation code, to signify the beginning of a bold (X=b),italic (X=i) or underline (X=u) typeform passage. These should be used for passages (words) only; see the separate individual typeform indicators under "Special Print-Text Codes" below. E.g. L.C.D. stands for [fts~b]Least Common Denominator[fte~b] means that the words "Least Common Denominator" are in bold. (See [ftlvN] for additional details)

[ftlvN] -- Translation code to determine whether emphasized text (as marked by [fts~...] ...[fte~...] codes) are to be indicated in braille and if so, under what circumstances. This code is applicable to certain tables only, and the table usage guide should be consulted as to what values of N are valid, and their meanings.

[gN] -- Translation code, initiate translation in grade N (N = 1, 1.5 or 2), according to the translation table in use. Grade 2 is normal except for tables designed for uncontracted translation, in which this code is ignored. (See [gNL] regarding "locking")

[gNL] -- Translation code to "lock" (L = l) or "unlock" (L = u) the grade of translation in the same manner as the corresponding [gN] code except for the concept of "locking," which works as follows. When a "lock" code is given, any subsequent [gN] or [gNL] code EXCEPT for the exactly corresponding "unlocking" form is ignored. For example, after a [g1l] code, only a [g1u] code "unlocks" the grade (which nevertheless remains in grade 1). These codes are provided so that ordinary [gN] forms may be used for switching levels within styles, yet it is possible to override their effect and, for example, do an entire file in grade 1. These codes are applicable to certain tables only, and the table usage guide should be consulted as to what values of N are valid, and their meanings.

[gd~...] -- This code is for inclusion of graphics data. Normally, this command is entered in a file as a result of using DBT's Layout/Picture menu item. It is not practical to enter this command manually, e.g. by the direct code list, because the graphic image information is included with the command as a parameter, and that information is specially coded and usually lengthy. For the same reason, the command parameter is not shown in DBT's coded view.

[hde] -- End centered heading. See [hds].

[hds] -- Start centered heading, with possible line skip or page eject according to current default settings. (For braille files, this is usually equivalent to [hds1:0]; for print files, to [hds2:1].) E.g. [hds]Some Heading[hde] would cause "Some Heading" to be centered on a new line.

[hdsN] -- Start centering, as with [hds], but first eject page if fewer than N+1 blank lines remain.

[hdsN:L] -- Start centering, with conditional new page as with [hdsN], but skip L lines (before heading) if no page eject. E.g. [hds3:1]Some Heading[hde]

[hiL:V:R:I:J:K:M:N:O:P] -- Initiate (or terminate) hierarchy. Sets the current hierarchy level to 1, the left margin to L, the overflow (runover) margin to V, and the right margin offset to R (defaulting to 1, 1 and 0 respectively). Also sets the characteristic increment for each of these margins to I, J and K respectively. (The increments may be negative, and all default to 0). This command determines the effect of subsequent [hlN] commands, which declare a given hierarchy level and set the margins correspondingly. Limits M, N, O and P may optionally be supplied, and if they are they work as follows: M sets an upper limit on subsequent levels that may be declared; that is, any higher value as a parameter to an [hlN] command is treated as M. Likewise N, O and P set limits on the left and runover margin sites and right margin indentation amount. In each of those cases, if the increment is positive, the limit is an upper limit; otherwise it is a lower limit. All limits that are supplied are imposed together; that is, the more stringent limits will be observed. In all cases, a negative limit value means "no limit," and all the values default to -1. E.g. [hi1:5:0:2:2] sets up margin treatment for an outline where level 1 items are to start at position 1 and run over to 5, and both of those margins move up by 2 for each higher level.

[hlN] -- Set current hierarchy level to N, and the left, runover and right margin accordingly (see the description of command [hi], above). The default for N is the current level, i.e. [hl] is synonymous with [<]. See [ki] regarding interruption of a protected block.

[htbfmtN:FFF] -- Set Table Layout

N = 0 means to use BANA rules for table layouts.

N = 1 means to use RNIB rules.

N = 2+ is reserved for future use.

FFF is a list of formats to try, in the order they should be tried.

FFF might be something like “csl” to try first Columnar format, then Stairstep format, then Listed format. (Implicitly, Linear – or “paragraph” format is always a fallback for “Auto-format” tables; thus, we could specify “cslp” instead of “csl”. The two are functionally equivalent.)

[htbfmt] is generally intended for use within the <initial> style definition, though it could be found anywhere. It applies only to tables that are to be “auto” formatted. If X/Y swap is allowed when more compact, then the FFF sequence is followed independently for the swapped and unswapped cases, and the more compact presentation is chosen. In other words, we might be comparing the size of an unswapped paragraph table to a swapped columnar table to choose the more compact form.

[htbs:1:2:3:4:5:6] -- Set Table Properties.

Parameter 1: "r" for "Related Columns" (or "Regular Table" if you prefer), "u" for "Unrelated Columns", "m" for "Matrix"

Parameter 2: Number of row headings

Parameter 3: Number of column headings

Parameter 4: This parameter has no function.

Parameter 5: n/N/0 to disallow X/Y swap. y/Y/1 to require it. a/A to allow it (swap if more compact)

Parameter 6: Requested layout:
a (or blank) = Auto
c = Columnar
2 = Two-page
s = Stairstep
l = Listed
p = Linear (The mnemonic is for "Paragraph", which is a misnomer for "Linear")

[htt] -- Synonym for [tle].

[i] -- Translation code that initiates grade 1 treatment for the following word only. E.g. In But [i]NOW voted for the proposal. the word NOW will be translated in grade 1 ("intégral" in French), while the surrounding words will be translated according to the prevailing grade. See also [ii] and [g...].

[iN] -- Translation code, a generalized form of the [i] code that allows for the opposite effect. N = 1 is synonymous with an ordinary [i], that is, it marks the following word as to be translated in grade 1. N = 0 means that the following word is to be translated according to the prevailing grade, that is NOT to be forcibly translated in grade 1, even though the general logic may otherwise cause that effect. For example, words in contracted French that end in the letters "tz" are normally forced to grade 1 (and marked as such) because that combination would be read back as "tez". However, certain words regarded as commonly recognizable are exempted from this rule and can be forced to be treated normally by prefixing the [i0] code -- for example, [i0]chintz would be treated normally (with "ch" and "in" contractions) in contracted French. These codes are applicable to certain tables only; consult the table usage guide as to whether they are available.

[idle] -- Does nothing (may be used as "placeholder") See also [x] and [z]

[ifbrl] -- Include next ordinary word or format control in braille output only. E.g. This is [ifprt]print [ifbrl]braille text. would set a sentence describing the type of type of text in which it resides. See also [ifprt]below.

[ifprt] -- Include next ordinary word or format control in print output only. E.g. [ifbrl][fr;p~"][ifprt][fr;f~."]123 would set "123" flush right on the line, with intervening space partially filled with dot-5 in the braille and completely filled with "." in the print.

[ii] -- Translation code, acts like [i], that is it causes a word to be translated in grade 1, and moreover prefixes an appropriate grade 1 indicator, according to the language (e.g. dots 5-6 in English).

[im...] -- Synonym for [pi...].

[in] -- Equivalent to [g1] (French "integral")

[ind]-- Restore left margin to normal, i.e. to the full width for pages in this document; equivalent to [ind0].

[indP] -- Position on a new line (if not already on a new line), at position P, and set left margin to P. E.g. [ind5]This ...

[inmP] -- Set left margin to position P, but do not alter current position.

[ixrtd] -- Delimiter between index and root in general indexed root expression; see [ixrts].

[ixrte] -- End of general indexed root expression; see [ixrts].

[ixrts~X] -- Translation code, used to mark the start of a general indexed root expression, i.e. a root that has an explicit index, such as 3 for a cube root. When there is no index, i.e. the expression is a square root, see [sqrts] instead. E.g. [ixrts]n[ixrtd]x+1[ixrte] means "the nth root of the expression x+1". See also [tcs] and [ts].

[kbeL:M] -- End block-protect on a character block basis. This is like [kpeL:M] but without the implied [l]. See [kbs].

[kbs] -- Start block-protect on a character block basis. This is like [kps] but without the implied [l]. E.g. The [kbs]protected text ... will be on one[kbe] page.

[kbs1] (note that the numeric parameter 1, not the letter l, follows the "s") -- Start block-protect on a character block basis, as for [kbs], but consider the beginning of this actual line (possibly caused by a soft return) to be the beginning of the block. This can be used, for example, to assure that the last line of the last paragraph of the body of a letter is on the same page with the signature block.

[kiN] -- Declares whether [hl] and [<] commands cause "interruption" of a protected (kept) block, i.e. whether there is an implied [kpe] before and [kps] after these commands. N=0 (the default) for no, N=1 for yes.

[kpe] -- End keeping text on one page (block protect). There is an implied [l] even if a page break is not necessary. See also [kps].

[kpeL:M] -- End keeping text on one page after L lines. That is, an additional L lines are effectively added to the protected block. M may be omitted and defaults to 0, but otherwise must be in the range 0 < M <= L, and in that case any new block-protect beginning within M lines is effectively attached to the current block, so that both are kept together. See also [kps].

[kpfL:M] -- Finish all keeping-text blocks on one page after L lines, attaching any new blocks opened within M lines. This is equivalent to giving as many [kpe] commands as may be needed to terminate any [kps] ... blocks opened within the outermost one, and then a [kpeL:M] command. See also [kps].

[kps] -- Start keeping text on one page (block protect). There is an implied [l] (new line if necessary), even if a page break is not necessary. E.g. [kps]These[l]three small[l]lines[kpe] will keep the text: "These ... lines" together on the same page.

[l] -- Begin a new line if necessary (but do nothing if already on a new line). This is therefore a "conditional" hard return; see also [<], the unconditional hard return.

[ld...] -- Synonym for [sd...].

[lea] -- Accept the next word or group as the "reference document page number". (In a braille file, this would be called the "print" or "inkprint" page number.) E.g. [lea]26 would be given at the point in the text where page 26 begins in the reference edition of the document. The command is equivalent to [lea;f~-]. The actual treatment of the page number depends on current default settings. See [svles...].

[lea;F~X] -- Accept the "reference document page number" as with [lea], but use variant fill type F and fill character X in the output treatment.

[lec] -- Accept the next word or group as the reference document page number for purposes of display on continuation pages only (overriding the setting of the most recent [lea] command). E.g. [lea]123-127 word [lec]127 would accept "123-127" as the reference page number at the point of the reference page break (a form that might be used when reference pages 123-126 are blank), but then, for use on any continuation pages, replace that range with just "127".

[led] -- Discontinue using reference page number, i.e. stop showing the reference page number on continuation pages.

[linenum] or [linenumC:A] -- Handle the next word as a line number for line-numbered poetry or prose.

[linenum] treats next word from the document for use as a line number, and may cause a line break in the text. If a line number comprises more than one word, then the words may be grouped using [:] and [;], e.g.

[linenum][:]48 B[;]

Ordinarily, [linenum] will cause the word that is to be treated as a line number to be placed on the right margin, adjusting the right margin automatically as necessary, without otherwise interfering with text flow.

If, since the last forced line break, there has been text in the document (i.e. if the line number comes in the middle or at the end of a "paragraph") then [linenum] implies [pm3:3], which can result in a line break, so that the line number itself and the text following it will be placed on the next line. This is intended to avoid having two line numbers occupy the same physical space in the braille layout, without requiring the transcriber to plan ahead. A right margin set implicitly by [linenum] will override any later attempt on the page to reduce the right margin using [rm], unless two parameters are given to the [rm] code.

[linenum] will ordinarily adjust the right margin to leave room for the widest line number on the page (counting only line numbers set with [linenum] itself, plus two spaces. [linenum] "looks ahead" in this respect, when determining how to set the right margin. The right margin set by [linenum] is ordinarily reset at the end of each page. However, it can be reset earlier, to separate a page into different sections for line-numbering margin computations, or carried across a page boundary, by using the "override margin" feature of [rm].

[linenum] is unique among DBT codes in that it does not intefere with pending effects of other codes. So, if [linenum] and a line number appear after a [tab] code, but before the word to be placed at a tabstop, the tabstop placement will be correct even so.

[linenum] can take two optional parameters. The first specifies a column for alignment purposes. Here, -1 can be used to specify the rightmost cell on the page. The second parameter, if specified, should be l, r, or c, to specify left, right, or center alignment on this cell. Note that the paramters control where the line number will be placed. They do no affect other text on the line. Without any parameters, [linenum] is equivalent to [linenum-1:r]. [linenum] will not have any effect on any margins unless the parameters are in fact [linenum-1:r] or, equivalently, [linenum] or [linenum-1].

[lnb~X] -- switches the base translation table from the current table to the one corresponding to the Table Designator X. Table Designators for each language are shown under Help: Language Translation Tables for each language.

A detailed explanation of this code is also given in the Language Switching Topic)

[lng] -- Translation code, to revert to the base (initial or default) language according to the table in use.

[lng~X] -- Translation code, to switch to a new language. The initial (base or default) language is always the main one supported by the table, E.g. English in the American tables, French in the French tables. The [lng~X] code is preferred for language switching thereafter. X indicates the new language as specified by the ISO 639 standard 2-letter language code. (In some cases, an older 3-letter DBT code is also accepted.) The codes currently accepted in the American tables, and at least partially in some other tables, are as follows: de or deu = German; en or eng = English; es or esp = Spanish; fi = Finnish; fr or fra = French; it or ita = Italian; la or lat = Latin; mi or mao = Maori; nl = Dutch; pt = Portuguese; sv = Swedish; sw or swa = Kiswahili (Swahili).

[lpr] -- Restore line position per last save. See [lps].

[lps] -- Save current line position. This is generally to allow temporary repositioning for placement of text at a special place on the line, to be followed by resumption of the current position. E.g. worda [lps][taa36]123[lpr]wordb would cause "123" to be placed in column 36 of the same line in which "worda" appears, and then "wordb" would follow "worda" normally.

[md] -- Synonym for [rm].

[mec] -- Accept the next word as the "continued" text for guide words. (Without this command, there is no special text for "continued" pages.) E.g. [mec](cont.) would declare "(cont.)" as the text for continuation pages, a commonly-used setting for English.

[meg] -- Accept next word as "guide" word. E.g. abcedarian [meg]abcedarianThis word ...would set "abcedarian" as text, then declare that this point in the text is associated with the guideword "abcedarian".

[mf] -- Synonym for [wb].

[mf-cb] -- Synonym for [wb-cb].

[mf-no] -- Synonym for [wb-no].

[music:{parameter}] -- This code is for music braille that has been brought in from Dancing Dots' GOODFEEL® program.
The music code may take two parameters. The numeric parameter gives a number corresponding to the music file. When a music file is inserted, it is stored in a separate part of the document and is identified by a numeric index. If there is a string parameter, then it is ignored; DBT places the name of the original filename in the string parameter, but it is not used by DBT; it is only stored as an aid to the user. Inserting a music code through ctrl-[ does not cause the music file to be inserted into the document; that can only be done through Edit, Insert File... Music will only be translated if the goodfeel.dll is installed.
The [lg] code is not supported.

[nr] -- Translation code for new row or line that, in the braille, is only indicated by a special symbol rather than an actual new line. For example, may be used in linear poetry or in a binomial coefficient or Christoffel symbol in mathematics.

[oe] -- End of directly-over superscript expression; see [e].

[one] -- End of passage in which numeric indicators are to be omitted; see [ons].

[ons~X] -- Translation code, used to mark the start of a passage where numeric indicators are to be omitted. E.g. [ons]123[one] would cause the numeric indicator to be omitted on "123" in braille. See also [tcs] and [ts].

[os] -- Start of directly-over superscript expression; see [e].

[p] -- Begin new paragraph, with possible line skip and/or indent according to current default settings (see [svprg...]). (In braille files, this is usually a new line, beginning at cell 3. In print files, this is usually a line skip only.)

[pe] -- Power/superscript end; see [ps].

[pg] -- Begin new page unconditionally.

[pg~X] -- Begin new page, and set the page number prefix character to X. E.g. [pg~p]

[pgN] -- Begin new page, and set the page number to N. E.g. [pg123]

[pgN~X] -- Begin new page, and set the page number to N and the page number prefix to character X. E.g. [pg1~p] would start a new page, numbering that page "p1". See also [pv...].

[pi...] -- This former code, for including graphics "pictures", is NOT supported in the same form in the new DBT (from 1995 onward). See the [gd...] (graphics data) command, which replaces it.

[pl] -- Synonym for [sl].

[pm] -- Begin magazine-style new paragraph. This causes the next word to be set on the current line after 3 spaces, or if that is not possible, to be set as after a [p] command. This is equivalent to [pm3:1]. (See [pmN:C].)

[pmN:C] -- [pm] can take two optional parameters that effect its behavior.

The first is the number of spaces to place on the line when a line break will not be necessary. The default is 3. The second parameter affects where the first character of the next line is placed when [pm] does cause a line break. Valid values for the second parameter are:

-1 (next line will be indented as if by [p] -- this is the default)
-2 (next line will be at the prevailing left margin)
-3 (next line will be at the prevailing runover margin)
any positive integer (a specific cell position)

[pnta] -- Set page number type to Arabic.

[pntr] -- Set page number type to Roman.

[po...] -- Synonym for [lea...].

[ps] -- Translation code, to mark the beginning of a power or superscript level. The [ps] ... [pe] pair may include any expression, including further levels of superscripts; i.e. the codes may be nested. E.g. y[ps]x[ps]2[pe][pe] means "y to the power x-squared". See also [tcs] and [ts].

[psac...] -- Synonym for [ptys...].

[psap] -- Synonym for [ptye].

[psc...] -- Synonym for [ptys...].

[pst] -- Synonym for [ptye].

[ptye] -- End poetry mode. (See [ptys...]. This command is actually equivalent to [ptys0].)

[ptys] -- Start poetry mode, i.e. indent runover (word-wrapped) lines by the default amount (usually 2 characters in braille documents, 5 in print.)

[ptysN] -- Start poetry mode, i.e. indent runovers, by N character positions.

[pvN~X] -- Set the page number to N and the page prefix character to X. All the variant forms, corresponding to the [pg...] series, may be used, and have the same meaning except that command does not in itself cause a new page to be started.

[q~W] -- Set word W directly (unchanged) in output. A following space is not automatically forced, as in the case of the [d...] command; thus "[q~W] text" is equivalent to "[d~W]text". E.g. [q~,8]xy would set ",8xy" (braille dots 6,2-3-6,1-3-4-6,1-3-4-5-6).

[rd] -- Synonym for [fr].

[remw] -- Accept the following word or group as a remark. Note that such comments must be quite short, not longer than a word (or line) could be in the formatted document. E.g. [remw]comment and [remw][:]so is this[;] would cause "comment" and "so is this" to be treated as comments -- that is, to be ignored; only the word "and" would be set as text.

[rfd] -- Discontinue using running footer. See [rfs] and [rfe].

[rfe] -- End definition of running footer. See [rfs].

[rfs] -- Start definition of running footer. E.g. [rfs]Section V[rfe] would cause "Section V" to be used as the running footer, from that point in the document until another running footer was similarly declared or the footer was discontinued (see [rfd]). The actual usage of the footer on even, odd, both or neither pages may be separately controlled by the [svrfp...] command.

[rm] -- Restore right margin to normal, i.e. to the full width for pages in this document; equivalent to [rm0] or [rm0:1] Restore Right Margin (but not Override Margin. (See [rmN].

[rmN] -- Indent right margin by N positions from the normal full page width. This affects only the "word wrap" (runover) point, not text that is set by explicit tabulation. E.g. [rm6]

[rmN:O] -- Indent right margin by the greater of N and O (O as in Oscar) positions from the normal full page width, and adjust the right margin implied by [linenum] to O. O in this instance is an override margin; the default for this parameter, -1, will result in no adjustment to the override margin, leaving any previous setting in place. When an override margin is in effect, [rm] and [rmN] (without the additional parameter) cannot be used to reduce the margin to less than the effective override margin or the prevailing automatic margin introduced by [linenum]. Some useful application of the override margin parameter include [rm0:0] to reset the right margin to 0, even if earlier text on the page included line numbers, and [rm8:8] to force a right margin of 8 cells, starting at this point, keeping the right margin impervious to the effects of [rm] or [rm0] codes that may be embedded in style definitions.

[rmn] -- Translation code, used primarily in Nemeth code, to signify that the following letter(s) constitute a Roman numeral. E.g. [rmn]I See also [tcs] and [ts].

[rnc...] -- Synonym for [kps...].

[rnp] -- Synonym for [kpe].

[rnt] -- Synonym for [kpe].

[rpe] -- Terminates [rps]...

[rpsN] -- Marks the beginning of text to be included in the table of contents. N is the "level" of the text within the TOC hierarchy, defaulting to 1 (highest). Use [vss]...[vse] if the text is not also to appear at the current position within the document, i.e. it is to appear ONLY in the TOC, for example: [rps2][vss]This heading appears only in the TOC[vse][rpe]

[rtb] -- Synonym for [ctb].

[run] -- Tab to the current runover point. This command is equivalent to [taaN], where N is the current runover position.

[scL:M:N:P] -- Skip lines conditionally, that is skip only if the line just above is not blank. L is the number of lines normally skipped; M is an exceptional value to use following a line that indicates change of reference page; N is an exceptional value to use following a running head. (N or M may be omitted, defaulting to L.) E.g. [sc1:0] would be commonly used for American braille, to cause a skip after any non-blank line except a print page break line.

(From version 11.1 onwards) This now has a 4th parameter (P), giving the number of lines to skip at the very top of page without a running header (which formerly was always treated as a "space above" case). The default for that parameter is 0.

[sdN] -- Start a new line if necessary (like [l]), and then start a new page if the then current page is not even (if N=0) or odd (if N=1). E.g. ... [pg][sd1]Start on odd ... would start a new page in any case, and if necessary advance the page again so that "Start on odd ..." begins on an odd page.

[sibrl] -- Synonym for [ifbrl].

[siimp] -- Synonym for [ifprt].

[skL] -- Skip L lines. This is equivalent to: [l][slL]. E.g. [sk2]

[sk-N] -- Skips to Nth line from end of page (unless already at or beyond that point on the page). E.g. [sk-1] would skip to the last line on the page.

[sknL:M] -- Skipped-line nullify. This command cancels up to L skipped (blank) lines produced as a final effect of the immediately preceding command, plus M lines produced by either the immediately preceding or immediately following command. L defaults to 1 and M defaults to 0. Nonblank lines are not affected

[slL] -- Force L line endings; see [sk...], which is generally preferable for skipping a certain number of lines.

[sqrte] -- End of square root expression; see [sqrts].

[sqrts] -- Translation code, used to mark the start of a square root expression, i.e. a root that has no explicit index. When there is an explicit index, such as 3 for a cube root, see [ixrts] instead. E.g. [sqrts]x+1[sqrte] means "the square root of the expression x+1". See also [tcs] and [ts].

[stL] -- Skips to line L (counting from top line = 1) on page (unless already at or beyond that point).

[stbS:A:P] -- Set tab stop number S for alignment type A, position P. A = l for left, r for right, d for decimal, or c for centering, with P being the focal position for the alignment. E.g. [ctb][stb1:l:5][stb2:d:22][#2]123.45 would clear tabs, then set stop 1 for left alignment at position 5 and stop 2 for decimal alignment at position 22. Then, by the selecting tab stop 2, the text "123.45" would be aligned so that the decimal point falls in position 22.

[svantN] -- Set option for running headers and/or reference page numbers to be anticipatory, i.e. to reflect any changes that take place on the page, rather than the condition at top of page. N = 1 for running-head anticipatory, 2 = reference page number, 3 = both, 0 = neither (which is the assumed value at beginning of file).

[svcbhN~string] where the optional parameter N = 0 (default) - set the computer-braille hyphen (line continuation indicator, applicable to text within [wb-cb]...[wb] pairs) to the string, i.e. the original meaning of [svcbh~string] is unchanged;

1 - set the regular hyphen override to the first character in the string (must be a printable ASCII character whose value is in the range [32,127]);

2 - set both the cb and regular hyphen override as per 0 and 1;

3 - cancel any regular hyphen override (hyphen reverts to the setting given in the .hct table).

So for example, [svcbh2~"] would set the soft hyphen to dot 5 for both computerese and regular text, as would be appropriate for UEB and related braille codes.

[svdac~X] -- Set decimal alignment character to X. This sets the character that is used by the tab commands to control decimal alignment. The default is the character that is common for American usage, i.e. presently [svdac~.] for both print and braille files (= dots 4-6 in the latter).

[svgan~X] -- Set values for generated Arabic numbers. This sets the "numeric indicator" character and, optionally, the characters used for the digits 0-9 in generated Arabic numbers (such as page numbers). X is a string of characters, the first of which is used to represent the "numeric indicator," if any. (Normally this is # [standing for dots 3-4-5-6] in a braille file, blank [meaning there is no indicator] in a print file. (To set this character to blank, use an opening curly brace [{].) The remaining characters of the string, if any, overwrite the characters used for the digits 0 through 9, respectively. Normally these are JABCDEFGHI in a braille file, 0123456789 in a print file. If these characters are not supplied, the current settings are not changed. Ex (note that the brackets within the string "#*<%?:$]\[" are literally brackets, NOT stand-ins for command delimiters): [ifbrl][svgan~,#*<%?:$]\[] would, in a braille file only, set up so that "Antoine digits" are used in the page numbers, prefaced by a dot 6, as is necessary in some French documents.

[svgrn~X] -- Set indicator for generated Roman numbers. This sets any indicator character needed prior to a generated Roman number, E.g. page numbers. X is the indicator character. (Normally this is a semicolon [;], standing for dots 5-6, in a braille file, blank [meaning there is no indicator] in a print file. (To set this character to blank, use an opening curly brace [{].) E.g. [ifbrl][svgrn~,] would, in a braille file only, set up so that dot 6 precedes any Roman page numbers, as is necessary in some French documents.

(From version 11.1) [svgrn~}] -- (that is, with a closing curly brace given as the parameter) is also permitted, and has the special meaning "set Roman numeral page numbers with grade 1 indicators as required for contracted Unified English Braille (UEB)". In most case this would mean that there would be no prefix, but for those Roman numerals that would have a contraction meaning, such as "x" and "cd", a grade 1 indicator (dots 56) prefix would be inserted.

[svlesN:M:L:P] -- Set parameters for treatment of reference page numbers (see [lea] command). N determines the treatment at the point of the page break: 0 for American style (a full line of dots 3-6 to the number flush right), 1 for British (Royal National Institute for the BlindRNIB) style (the number centered, preceded by dot 5, 25). M determines the treatment on continuing pages: 0 for American (a, b ... prefixes on continuation pages), 1 for British (unprefixed). L determines the "bottom tolerance", i.e. the number of lines that must be available at the bottom of the page AFTER the print page break indicator line. P determines the initial text position after the page break: 0 for the left margin, 1 for the runover margin. If all four values are defaulted, the settings are restored to values as at the file beginning, which are N=0, M=0, L=2 and P=0 respectively. If one or more values are specified, any unspecified values remain unchanged.

[svmtsN] -- Set minimum tab spacing to N. This determines the minimum amount that a word must be spaced away from the prior word as a result of a tabulation command (see the [tab...] and [tas...] commands). The default (initial) value is 1.

[svpfdN] -- Set page number first displayed to N. Pages whose page numbers are less than N will not be explicitly numbered. The initial value is normally 2; the default for the command is 1.

[svpnpN:N:N:N] -- Set page number placement values. The N's are given in the order: braille (internal) page number on odd sides, internal number on even sides, print (reference) number on odd sides, reference number on even sides. The possible values for any N are: 0 for no place, 1 for upper left, 2 for upper right, 3 for lower left, 4 for lower right. The default values are 4, 4, 2 and 2 respectively (corresponding to American "textbook format"). E.g. [svpnp2:2:0:0] would establish customary literary format, with braille page numbers at upper right and no print page numbers shown.

[svprgN:L:M] -- Set paragraph treatment values, i.e. the parameters for the [p] command. N determines the position at which to begin a new paragraph; the default value is 3 for braille and 0 for print. L determines how many lines to skip before a paragraph; the default is 0 for braille and 1 for print. M is 0 (the default) if N is to be an absolute position, or is 1 if N is to be relative to the current left margin (i.e. the left margin reckoned as position 1). If all values are omitted, the values are reset to their values at the beginning of the file.

[svrfpN] -- Set running-footer usage value. N is 1 for usage on odd pages only, 2 for even only, 3 for usage on both, 0 on neither. If N is omitted, the initial value is restored, i.e. 3.

[svrfsM:N] -- Set running-footer sideroom. M sets the fixed amount of space reserved on the left for the largest page number expected, and N for a similar amount on the right. (Though specified separately for future compatibility reasons, the two amounts are actually used only in their total, because footers are always centered.) The default is 8:8 (total 16). E.g. [svrfs5:5] would reserve only 5 spaces on either side of the running footer (10 total), permitting a longer footer to fit.

[svrhpN] -- Set running-header usage value. N is 1 for usage on odd pages only, 2 for even only, 3 for usage on both, 0 on neither. If N is omitted, the initial value is restored, i.e. 3.

[svrhsM:N] -- Set running-header sideroom. M sets the fixed amount of space reserved on the left for the largest page number expected, and N for a similar amount on the right. (Though specified separately for future compatibility reasons, the two amounts are actually used only in their total, because running headers are always centered.) The default is 8:8 (total 16). E.g. [svrhs5:5] would reserve only 5 spaces on either side of the running header (10 total), permitting a longer header to fit.

[svrlbN] -- Reserve N lines at the end of each page. All text, including footers and page numbers, will be placed above these reserved lines. This code is useful for allowing for corrections to pages after printing has commenced. Specifying no value, or a value of 0, will disable this behavior.

[svrptN] -- This is for future use, to set a value that determines the treatment of reference points, i.e. the layout of the table of contents; see [rpsN].

[svsblN] -- Set interline spacing value to N. E.g. [svsbl1] starts double-spacing, i.e. 1 blank line between text lines.

[swa] -- Translation code, resume Swahili text (in Swahili tables only). See [lng~...] and [eng].

[taaP:A:F~X] -- Absolute tab with intervening fill. P is the position (column) upon which to align the next word (or words enclosed by [:] ... [;]). A is the type of alignment: L for left, R for right, D for decimal or C for centering. F is the type of fill: F for full (all cells between the current position and the tabbed item are filled) or P (the first and last intervening positions are always spaces, and also the third if there are only three). X is the fill character, E.g. " for braille dot 5. The default for A is L (left), and if F~X is omitted then intervening spaces are not filled. [taa] causes tabulation on the current line regardless of the current position, even if already set text might be overwritten (see also [lps], [tab...] and [tas...]). E.g. [taa38:r]abc would cause the text "abc" to be set so that the "c" would fall into position 38 on the current line. See also the example with [lps].

[tabP:A:F~X] -- Tab with intervening fill. P is the position (column) upon which to align the next word (or words enclosed by [:] ... [;]). A is the type of alignment: L for left, R for right, D for decimal or C for centering. F is the type of fill: F for full (all cells between the current position and the tabbed item are filled) or P (the first and last intervening positions are always spaces, and also the third if there are only three). X is the fill character, E.g. " for braille dot 5. The default for A is L (left), and if F~X is omitted then intervening spaces are not filled. [tab] causes tabulation on the current line provided that the aligned item would fall to the right of the current position, after any minimum spacing (see the [svmts...] command); otherwise, a new line is used. (See also [taa...] and [tas...]). E.g. [tab25:d:p~-]123.45 would cause the text "123.45" to be set so that the decimal point would fall into position 25, with intervening space filled (partially) by "-" (braille dots 3-6). E.g. [tab30]x would place "x" in column 30.

[tasP:A:F~X] -- Tab or space, with intervening fill. This command acts just like the [tab...] command, described above, except that in the case where the aligned item would not fall to the right of the current position after at the least the minimum spacing, it is forced to exactly that position.

[tce] -- End of technical context; see [tcs].

[tcs] -- Translation code, used to mark the start of "technical context", that is material that is to be translated according to special code rules when such rules are applicable to all the text surrounding technical notation, not just the notation itself. This is intended primarily for the Nemeth Braille Code for Mathematics and Science Notation, whose rules have that characteristic. In a work to be translated according to Nemeth Code, the [tcs] ... [tce] codes normally surround practically the whole text, excluding only the title page at the beginning and the numbers associated with print page breaks, which are done according to "literary" conventions. (Note: Print page breaks normally occur frequently, and so for convenience in editing works to be done in Nemeth code, it is advisable to set up the termination and resumption of "technical context" to take place in a style automatically, right along with the declaration of print page change. For example, use a code sequence such as [tce][lea] at the beginning and [l][tcs] at the end of the style.) See also [ts].

[te] -- End of technical notation; see [ts].

[tlc] -- Synonym for [tls].

[tld] -- Discontinue using title (running head). See the [tls] ... [tle] commands.

[tle] -- End definition of title (running head). See [tls].

[tlp] -- Synonym for [tle].

[tls] -- Begin definition of running head. The text between this command and the next [tle] is saved aside and used as the heading for any new pages that may be begun, from the point of definition until either redefined by a new [tls] ... [tle] or discontinued by [tld]. Actual usage on particular pages is governed by [svrhp...]. E.g. [tls]Chapter One[tle] would cause "Chapter One" to be used for a running head until further notice.

[tne] -- Transcriber note ending (see [tns]).

[tns] -- Translation code to set the indicator for the start of a transcriber's note. [tne] sets the corresponding ending indicator. For example: [tns]In print, an upward arrow is shown.[tne] These codes are applicable to certain tables only; consult the table usage guide as to whether they are available.

[top] -- Top of page (no effect if already on fresh page, otherwise equivalent to [pg]).

[ts] -- Translation code, used to mark the start of "technical notation", that is material that is definitely technical in nature, such as mathematics; not words in English or some other natural language. Many braille codes, including Nemeth code, call for such material to be brailled differently than would be the case if the same text were interpreted as a word or other "literary" material, such as an abbreviation. It is not generally necessary to mark material that from the outset is clearly technical, such as numbers, though it does no harm to do so. However, it is advisable to mark any technical notation commencing with letters. E.g. The base of triangle [ts]RST[te] is [ts]x+1[te] inches long
marks RST and x+1 as technical expressions. See also [tcs].

[ttc] -- Synonym for [hds].

[ttt] -- Synonym for [hde].

[tx] -- Translation code, resumes normal text translation, cancelling any special mode for computer notation or direct braille. See also [cb] and[cz], which initiate alternate modes, terminating the effect of [tx] or its variants [txi] and [txn].

[txi] -- Translation code, a variant of [tx], which has the same effect and also unconditionally puts the CBC end-CBC indicator into the braille.

[txn] -- Translation code, a variant of [tx], which has the same effect and also unconditionally will NOT put the CBC end-CBC indicator into the braille.

[u] -- Translation code to mark the following individual symbol (only) as to be translated in grade 1, i.e. as uncontracted. This code is applicable to certain tables only; consult the table usage guide as to whether it is available. See also [ui].

[uclN~X] -- Translation code to turn on, or off, literal treatment of certain characters. N = 0 means to turn off, N = 1 means to turn on, literal treatment. For example, when using the English/Unified table, the command [ucl1~"] causes the ordinary keyboard double quote character (U+0022) to be translated as the UEB "nondirectional" quote, whereas otherwise it is usually sensed as an opening or closing double quote depending on its immediate context. This code is applicable to certain tables only; consult the table usage guide as to whether it is available, the characters (X) to which it may be applied, and any details as what constitutes literal vs. more flexible treatment.

[ue] -- End of directly-under subscript expression; see [e].

[ui] -- Translation code to mark the following individual symbol (only) as to be translated in grade 1, i.e. as uncontracted, and to be so marked using the appropriate indicator (e.g. dots 56, the "grade 1 symbol" indicator, in Unified English Braille). This code is applicable to certain tables only; consult the table usage guide as to whether it is available.

[uoq~X] -- Translation code for setting the "outer quote", i.e. to designate which character is to be regarded as the usual, or outer, quote mark for translation purposes. This code is applicable to certain tables only; consult the table usage guide as to whether it is available, the characters (X) that may be designated, and other pertinent details.

[us] -- Start of directly-under subscript expression; see [e].

[utpN] -- where N=0 (the initial and default condition) means that DBT employs its own "judgment" as to whether a [ts] code should result in an extended g1 mode, N=1 means that an extended g1 mode should always be used (except for those extremely simple cases), and N=2 means that the current grade (normally grade 2) should be maintained -- that is, the [ts] code should effectively be ignored.

[vce] -- Ends vertical centering. See [vcs]

[vcs] -- Starts vertical centering. The vertical centering takes place when a page is finished, so the [vcs] can come anywhere on the first page that is to be vertically centered, that is before any [pg] or [top] command that would eject that page, and the [vce] must come AFTER the command that ejects the last page to be vertically centered.

[vcsM:N] -- Start vertical page alignment by applying the fraction N/M to the number of lines that would otherwise be left empty at the bottom, and adding that many blank lines to the top. M and N default to 2 and 1 respectively. Thus [vcs1] effects approximate bottom alignment. (Bottom alignment may not always be "perfect" because, when a body of text is started later on a page and re-flowed, under some circumstances [such as when the first line as been shortened to accommodate a page number] the line breaks may not always occur at the same positions and hence the text itself may occupy a different number of lines than it did originally.)

[vp] -- Synonym for [wb].

[vp-cb] -- Synonym for [wb-cb].

[vp-no] -- Synonym for [wb-no].

[vrn~X] -- Translation code, peculiar to table in use, to initiate translation according to some "variation" of standard rules. If the parameter is omitted, or the parameter is not supported by the table in use, then standard translation treatment is resumed. Only one variation code may be in effect at a time; a new variation code cancels the prior one. (Notes: A "variation" generally refers to a relatively minor, informal departure whereas officially defined major code dialects are represented by distinct tables or specific switching codes such as [cb]. Variations need not have the same meaning from one table to the next, although in most cases of related tables, a reasonable consistency will be attempted. Variations may be discontinued or changed at any time, without the usual concern for compatibility.) The variation is designated by a three-letter code. Currently defined variations apply only to English in the braille-to-print direction, and are as follows: rlg = traditional "Religious" contractions (for "faith", "grace", etc.) are expanded; quo = dot 4 always "quotes" the following braille character, to be interpreted according to North American ASCII braille code.

(From version 11.1) vrn (Special ~t2 for Mandarin): vrn codes are normally documented with the table documentation, not (except in a generic sense) with the other codes, because by definition the meaning varies from table to table (and may also vary with versions of a table, or be eliminated). The vrn~tN codes applicable to the Chinese (Mandarin) table are already so documented.

[vse] -- Terminates [vss].

[vss] -- Marks the start of text that is not "visible," that is not set as formatted text at the point of encounter, though it may be processed for other purposes. See an example of use with [rpsN]

[wb] -- Restore normal word breaking (per hyphenation table in use). See the [wb-cb] and [wb-no] commands.

[wb-cb] -- Use computer-braille (CBC) continuation indicator and word (string) breaking rules. The normal logic for breaking words that do not fit at the end of line is suspended, and instead the rules applicable to BANA Computer Braille Code, including the automatic supplying of a continuation indicator, are used instead. This effect is terminated by the next [wb] or [wb-no] command.

[wb-no] -- Suspend all word breaking. After this command, a word (that is, a series of non-spaces) will not be broken, even at explicit internal hyphens or dashes. This effect is terminated by the next [wb] or [wb-cb] command. E.g. [wb-no]124-127[wb]

[x] -- Synonym for [idle]

[xcs] -- Cancel capitalized word sequence. Its purpose is to allow manual interruption of capitalized word sequences when it is not desirable that the entire sequence be treated as a single "passage," as for example when the end of one sentence and the beginning of the next contain fully capitalized words, e.g.:

"I assure you, THAT FUSE IS TOO SHORT. BUT IF YOU INSIST, go ahead and light it -- after I have left."

Introduction of an [xcs] after the period in the above sentence would prevent recognition of the two partial sentences as a single capitalized passage. Currently this is only active with the English/British (new), English/Unified, and French/Unified tables.

[z] -- Synonym for [idle]