1.ig 2Copyright (C) 1989-2000, 2001, 2004, 2005 Free Software Foundation, Inc. 3 4Permission is granted to make and distribute verbatim copies of 5this manual provided the copyright notice and this permission notice 6are preserved on all copies. 7 8Permission is granted to copy and distribute modified versions of this 9manual under the conditions for verbatim copying, provided that the 10entire resulting derived work is distributed under the terms of a 11permission notice identical to this one. 12 13Permission is granted to copy and distribute translations of this 14manual into another language, under the above conditions for modified 15versions, except that this permission notice may be included in 16translations approved by the Free Software Foundation instead of in 17the original English. 18.. 19. 20. 21.ie \n(.V<\n(.v \ 22. ds tx T\h'-.1667m'\v'.224m'E\v'-.224m'\h'-.125m'X 23.el \ 24. ds tx TeX 25. 26. 27.\" Like TP, but if specified indent is more than half 28.\" the current line-length - indent, use the default indent. 29.de Tp 30. ie \\n(.$=0:((0\\$1)*2u>(\\n(.lu-\\n(.iu)) .TP 31. el .TP "\\$1" 32.. 33. 34. 35.de TQ 36. br 37. ns 38. TP \\$1 39.. 40. 41. 42.\" The BSD man macros can't handle " in arguments to font change macros, 43.\" so use \(ts instead of ". 44.tr \(ts" 45. 46. 47.TH @G@EQN @MAN1EXT@ "@MDATE@" "Groff Version @VERSION@" 48. 49. 50.SH NAME 51@g@eqn \- format equations for troff 52. 53. 54.SH SYNOPSIS 55.nr a \n(.j 56.ad l 57.nr i \n(.i 58.in +\w'\fB@g@eqn 'u 59.ti \niu 60.B @g@eqn 61.de OP 62. ie \\n(.$-1 .RI "[\ \fB\\$1\fP" "\\$2" "\ ]" 63. el .RB "[\ " "\\$1" "\ ]" 64.. 65.OP \-rvCNR 66.OP \-d xy 67.OP \-T name 68.OP \-M dir 69.OP \-f F 70.OP \-s n 71.OP \-p n 72.OP \-m n 73.RI "[\ " files\|.\|.\|. "\ ]" 74.br 75.ad \na 76. 77.LP 78It is possible to have whitespace between a command line option and its 79parameter. 80. 81. 82.SH DESCRIPTION 83This manual page describes the GNU version of 84.BR eqn , 85which is part of the groff document formatting system. 86.B eqn 87compiles descriptions of equations embedded within 88.B troff 89input files into commands that are understood by 90.BR troff . 91Normally, it should be invoked using the 92.B \-e 93option of 94.BR groff . 95The syntax is quite compatible with Unix eqn. 96The output of GNU 97.B eqn 98cannot be processed with Unix troff; 99it must be processed with GNU troff. 100If no files are given on the command line, the standard input 101will be read. 102A filename of 103.B \- 104will cause the standard input to be read. 105. 106.LP 107.B eqn 108searches for the file 109.B eqnrc 110in the directories given with the 111.B \-M 112option first, then in 113.BR @SYSTEMMACRODIR@ , 114.BR @LOCALMACRODIR@ , 115and finally in the standard macro directory 116.BR @MACRODIR@ . 117If it exists, 118.B eqn 119will process it before the other input files. 120The 121.B \-R 122option prevents this. 123. 124.LP 125GNU 126.B eqn 127does not provide the functionality of neqn: 128it does not support low-resolution, typewriter-like devices 129(although it may work adequately for very simple input). 130. 131. 132.SH OPTIONS 133.TP 134.BI \-d xy 135Specify delimiters 136.I x 137and\~\c 138.I y 139for the left and right end, respectively, of in-line equations. 140Any 141.B delim 142statements in the source file overrides this. 143. 144.TP 145.B \-C 146Recognize 147.B .EQ 148and 149.B .EN 150even when followed by a character other than space or newline. 151.TP 152.B \-N 153Don't allow newlines within delimiters. 154This option allows 155.B eqn 156to recover better from missing closing delimiters. 157. 158.TP 159.B \-v 160Print the version number. 161. 162.TP 163.B \-r 164Only one size reduction. 165. 166.TP 167.BI \-m n 168The minimum point-size is\~\c 169.IR n . 170.B eqn 171will not reduce the size of subscripts or superscripts to 172a smaller size than\~\c 173.IR n . 174. 175.TP 176.BI \-T name 177The output is for device 178.IR name . 179The only effect of this is to define a macro 180.I name 181with a value of\~\c 182.BR 1 . 183Typically 184.B eqnrc 185will use this to provide definitions appropriate for the output device. 186The default output device is 187.BR @DEVICE@ . 188. 189.TP 190.BI \-M dir 191Search 192.I dir 193for 194.B eqnrc 195before the default directories. 196. 197.TP 198.B \-R 199Don't load 200.BR eqnrc . 201. 202.TP 203.BI \-f F 204This is equivalent to a 205.BI gfont\ F 206command. 207. 208.TP 209.BI \-s n 210This is equivalent to a 211.BI gsize\ n 212command. 213This option is deprecated. 214.B eqn 215will normally set equations at whatever the current point size 216is when the equation is encountered. 217. 218.TP 219.BI \-p n 220This says that subscripts and superscripts should be 221.I n\~\c 222points smaller than the surrounding text. 223This option is deprecated. 224Normally 225.B eqn 226makes sets subscripts and superscripts at 70% 227of the size of the surrounding text. 228. 229. 230.SH USAGE 231Only the differences between GNU 232.B eqn 233and Unix eqn are described here. 234. 235.LP 236Most of the new features of GNU 237.B eqn 238are based on \*(tx. 239There are some references to the differences between \*(tx and GNU 240.B eqn 241below; 242these may safely be ignored if you do not know \*(tx. 243. 244.SS Automatic spacing 245.B eqn 246gives each component of an equation a type, and adjusts the spacing 247between components using that type. 248Possible types are: 249. 250.RS 251.TP \w'punctuation'u+2n 252ordinary 253an ordinary character such as `1' or `\c 254.IR x '; 255. 256.TP 257operator 258a large operator such as 259.ds Su `\s+5\(*S\s0' 260.if \n(.g .if !c\(*S .ds Su the summation operator 261\*(Su; 262. 263.TP 264binary 265a binary operator such as `\(pl'; 266. 267.TP 268relation 269a relation such as `='; 270. 271.TP 272opening 273a opening bracket such as `('; 274. 275.TP 276closing 277a closing bracket such as `)'; 278. 279.TP 280punctuation 281a punctuation character such as `,'; 282. 283.TP 284inner 285a subformula contained within brackets; 286.TP 287suppress 288spacing that suppresses automatic spacing adjustment. 289.RE 290. 291.LP 292Components of an equation get a type in one of two ways. 293. 294.TP 295.BI type\ t\ e 296This yields an equation component that contains\~\c 297.I e 298but that has type\~\c 299.IR t , 300where 301.I t 302is one of the types mentioned above. 303For example, 304.B times 305is defined as 306. 307.RS 308.IP 309.B 310type "binary" \e(mu 311.RE 312. 313.IP 314The name of the type doesn't have to be quoted, but quoting protects 315from macro expansion. 316. 317.TP 318.BI chartype\ t\ text 319Unquoted groups of characters are split up into individual characters, 320and the type of each character is looked up; 321this changes the type that is stored for each character; 322it says that the characters in 323.I text 324from now on have type\~\c 325.IR t . 326For example, 327. 328.RS 329.IP 330.B 331chartype "punctuation" .,;: 332.RE 333. 334.IP 335would make the characters `.,;:' have type punctuation 336whenever they subsequently appeared in an equation. 337The type\~\c 338.I t 339can also be 340.B letter 341or 342.BR digit ; 343in these cases 344.B chartype 345changes the font type of the characters. 346See the 347.B Fonts 348subsection. 349. 350.SS New primitives 351.TP 352.IB e1\ smallover\ e2 353This is similar to 354.BR over ; 355.B smallover 356reduces the size of 357.I e1 358and 359.IR e2 ; 360it also puts less vertical space between 361.I e1 362or 363.I e2 364and the fraction bar. 365The 366.B over 367primitive corresponds to the \*(tx 368.B \eover 369primitive in display styles; 370.B smallover 371corresponds to 372.B \eover 373in non-display styles. 374. 375.TP 376.BI vcenter\ e 377This vertically centers 378.I e 379about the math axis. 380The math axis is the vertical position about which characters 381such as `\(pl' and `\(mi' are centered; also it is the vertical position 382used for the bar of fractions. 383For example, 384.B sum 385is defined as 386. 387.RS 388.IP 389.B 390{ type "operator" vcenter size +5 \e(*S } 391.RE 392. 393.TP 394.IB e1\ accent\ e2 395This sets 396.I e2 397as an accent over 398.IR e1 . 399.I e2 400is assumed to be at the correct height for a lowercase letter; 401.I e2 402will be moved down according if 403.I e1 404is taller or shorter than a lowercase letter. 405For example, 406.B hat 407is defined as 408. 409.RS 410.IP 411.B 412accent { "^" } 413.RE 414. 415.IP 416.BR dotdot , 417.BR dot , 418.BR tilde , 419.BR vec , 420and 421.B dyad 422are also defined using the 423.B accent 424primitive. 425. 426.TP 427.IB e1\ uaccent\ e2 428This sets 429.I e2 430as an accent under 431.IR e1 . 432.I e2 433is assumed to be at the correct height for a character without a descender; 434.I e2 435will be moved down if 436.I e1 437has a descender. 438.B utilde 439is pre-defined using 440.B uaccent 441as a tilde accent below the baseline. 442. 443.TP 444.BI split\ \(ts text \(ts 445This has the same effect as simply 446. 447.RS 448.IP 449.I text 450.RE 451. 452.IP 453but 454.I text 455is not subject to macro expansion because it is quoted; 456.I text 457will be split up and the spacing between individual characters 458will be adjusted. 459. 460.TP 461.BI nosplit\ text 462This has the same effect as 463. 464.RS 465.IP 466.BI \(ts text \(ts 467.RE 468. 469.IP 470but because 471.I text 472is not quoted it will be subject to macro expansion; 473.I text 474will not be split up 475and the spacing between individual characters will not be adjusted. 476. 477.TP 478.IB e\ opprime 479This is a variant of 480.B prime 481that acts as an operator on\~\c 482.IR e . 483It produces a different result from 484.B prime 485in a case such as 486.BR A\ opprime\ sub\ 1 : 487with 488.B opprime 489the\~\c 490.B 1 491will be tucked under the prime as a subscript to the\~\c 492.B A 493(as is conventional in mathematical typesetting), 494whereas with 495.B prime 496the\~\c 497.B 1 498will be a subscript to the prime character. 499The precedence of 500.B opprime 501is the same as that of 502.B bar 503and 504.BR under , 505which is higher than that of everything except 506.B accent 507and 508.BR uaccent . 509In unquoted text a\~\c 510.B ' 511that is not the first character will be treated like 512.BR opprime . 513. 514.TP 515.BI special\ text\ e 516This constructs a new object from\~\c 517.I e 518using a 519.BR @g@troff (@MAN1EXT@) 520macro named 521.IR text . 522When the macro is called, 523the string 524.B 0s 525will contain the output for\~\c 526.IR e , 527and the number registers 528.BR 0w , 529.BR 0h , 530.BR 0d , 531.BR 0skern , 532and 533.BR 0skew 534will contain the width, height, depth, subscript kern, and skew of\~\c 535.IR e . 536(The 537.I "subscript kern" 538of an object says how much a subscript on that object should be tucked in; 539the 540.I skew 541of an object says how far to the right of the center of the object an 542accent over the object should be placed.) 543The macro must modify 544.B 0s 545so that it will output the desired result with its origin at the current 546point, and increase the current horizontal position by the width 547of the object. 548The number registers must also be modified so that they correspond to the 549result. 550. 551.IP 552For example, suppose you wanted a construct that `cancels' an expression 553by drawing a diagonal line through it. 554. 555.RS 556.IP 557.ft B 558.if t .ne 6+\n(.Vu 559.br 560\&.EQ 561.br 562define cancel 'special Ca' 563.br 564\&.EN 565.br 566\&.de Ca 567.br 568\&.\ \ ds 0s \e 569.br 570\eZ'\e\e*(0s'\e 571.br 572\ev'\e\en(0du'\e 573.br 574\eD'l \e\en(0wu -\e\en(0hu-\e\en(0du'\e 575.br 576\ev'\e\en(0hu' 577.br 578\&.. 579.ft 580.RE 581. 582.IP 583Then you could cancel an expression\~\c 584.I e 585with 586.BI \%cancel\ {\ e\ } 587. 588.IP 589Here's a more complicated construct that draws a box round an expression: 590. 591.RS 592.IP 593.ft B 594.if t .ne 11+\n(.Vu 595\&.EQ 596.br 597define box 'special Bx' 598.br 599\&.EN 600.br 601\&.de Bx 602.br 603\&.\ \ ds 0s \e 604.br 605\eZ'\eh'1n'\e\e*(0s'\e 606.br 607\eZ'\e 608.br 609\ev'\e\en(0du+1n'\e 610.br 611\eD'l \e\en(0wu+2n 0'\e 612.br 613\eD'l 0 -\e\en(0hu-\e\en(0du-2n'\e 614.br 615\eD'l -\e\en(0wu-2n 0'\e 616.br 617\eD'l 0 \e\en(0hu+\e\en(0du+2n'\e 618.br 619\&'\e 620.br 621\eh'\e\en(0wu+2n' 622.br 623\&.\ \ nr 0w +2n 624.br 625\&.\ \ nr 0d +1n 626.br 627\&.\ \ nr 0h +1n 628.br 629\&.. 630.ft 631.RE 632. 633.TP 634.BI space\ n 635A positive value of the integer\~\c 636.I n 637(in hundredths of an em) sets the vertical spacing before the equation, 638a negative value sets the spacing after the equation, replacing the 639default values. 640This primitive provides an interface to 641.BR groff 's 642.B \ex 643escape (but with opposite sign). 644. 645.IP 646This keyword has no effect if the equation is part of a 647.B pic 648picture. 649. 650.SS Extended primitives 651.TP 652.BI col\ n\ {\ .\|.\|.\ } 653.TQ 654.BI ccol\ n\ {\ .\|.\|.\ } 655.TQ 656.BI lcol\ n\ {\ .\|.\|.\ } 657.TQ 658.BI rcol\ n\ {\ .\|.\|.\ } 659.TQ 660.BI pile\ n\ {\ .\|.\|.\ } 661.TQ 662.BI cpile\ n\ {\ .\|.\|.\ } 663.TQ 664.BI lpile\ n\ {\ .\|.\|.\ } 665.TQ 666.BI rpile\ n\ {\ .\|.\|.\ } 667The integer value\~\c 668.I n 669(in hundredths of an em) increases the vertical spacing between rows, 670using 671.BR groff 's 672.B \ex 673escape. 674Negative values are possible but have no effect. 675If there is more than a single value given in a matrix, the biggest one 676is used. 677. 678.SS Customization 679The appearance of equations is controlled by a large number of parameters. 680These can be set using 681the 682.B set 683command. 684. 685.TP 686.BI set\ p\ n 687This sets parameter\~\c 688.I p 689to value\~\c 690.IR n ; 691.I n\~\c 692is an integer. 693For example, 694. 695.RS 696.IP 697.B 698set x_height 45 699.RE 700. 701.IP 702says that 703.B eqn 704should assume an x\~height of 0.45\~ems. 705. 706.RS 707.LP 708Possible parameters are as follows. 709Values are in units of hundredths of an em unless otherwise stated. 710These descriptions are intended to be expository rather than 711definitive. 712. 713.ie t \ 714. TP \w'\fBdefault_rule_thickness'u+2n 715.el \ 716. TP 717.B minimum_size 718.B eqn 719will not set anything at a smaller point-size than this. 720The value is in points. 721. 722.TP 723.B fat_offset 724The 725.B fat 726primitive emboldens an equation 727by overprinting two copies of the equation 728horizontally offset by this amount. 729. 730.TP 731.B over_hang 732A fraction bar will be longer by twice this amount than 733the maximum of the widths of the numerator and denominator; 734in other words, it will overhang the numerator and 735denominator by at least this amount. 736. 737.TP 738.B accent_width 739When 740.B bar 741or 742.B under 743is applied to a single character, 744the line will be this long. 745Normally, 746.B bar 747or 748.B under 749produces a line whose length is the width of the object to which it applies; 750in the case of a single character, 751this tends to produce a line that looks too long. 752. 753.TP 754.B delimiter_factor 755Extensible delimiters produced with the 756.B left 757and 758.B right 759primitives will have a combined height and depth of at least this many 760thousandths of twice the maximum amount by which the sub-equation that 761the delimiters enclose extends away from the axis. 762. 763.TP 764.B delimiter_shortfall 765Extensible delimiters produced with the 766.B left 767and 768.B right 769primitives will have a combined height and depth 770not less than the difference of 771twice the maximum amount by which the sub-equation that 772the delimiters enclose extends away from the axis 773and this amount. 774. 775.TP 776.B null_delimiter_space 777This much horizontal space is inserted 778on each side of a fraction. 779. 780.TP 781.B script_space 782The width of subscripts and superscripts is increased by this amount. 783. 784.TP 785.B thin_space 786This amount of space is automatically inserted after punctuation 787characters. 788. 789.TP 790.B medium_space 791This amount of space is automatically inserted on either side 792of binary operators. 793. 794.TP 795.B thick_space 796This amount of space is automatically inserted on either side of 797relations. 798. 799.TP 800.B x_height 801The height of lowercase letters without ascenders such as `x'. 802. 803.TP 804.B axis_height 805The height above the baseline of the center of characters 806such as `\(pl' and `\(mi'. 807It is important that this value is correct for the font 808you are using. 809. 810.TP 811.B default_rule_thickness 812This should set to the thickness of the 813.B \e(ru 814character, or the thickness of horizontal lines produced with the 815.B \eD 816escape sequence. 817. 818.TP 819.B num1 820The 821.B over 822command will shift up the numerator by at least this amount. 823. 824.TP 825.B num2 826The 827.B smallover 828command will shift up the numerator by at least this amount. 829. 830.TP 831.B denom1 832The 833.B over 834command will shift down the denominator by at least this amount. 835. 836.TP 837.B denom2 838The 839.B smallover 840command will shift down the denominator by at least this amount. 841. 842.TP 843.B sup1 844Normally superscripts will be shifted up by at least this amount. 845. 846.TP 847.B sup2 848Superscripts within superscripts or upper limits 849or numerators of 850.B smallover 851fractions 852will be shifted up by at least this amount. 853This is usually less than sup1. 854. 855.TP 856.B sup3 857Superscripts within denominators or square roots 858or subscripts or lower limits will be shifted up by at least 859this amount. 860This is usually less than sup2. 861. 862.TP 863.B sub1 864Subscripts will normally be shifted down by at least this amount. 865. 866.TP 867.B sub2 868When there is both a subscript and a superscript, the subscript 869will be shifted down by at least this amount. 870. 871.TP 872.B sup_drop 873The baseline of a superscript will be no more 874than this much amount below the top of the object on 875which the superscript is set. 876. 877.TP 878.B sub_drop 879The baseline of a subscript will be at least this much below 880the bottom of the object on which the subscript is set. 881. 882.TP 883.B big_op_spacing1 884The baseline of an upper limit will be at least this 885much above the top of the object on which the limit is set. 886. 887.TP 888.B big_op_spacing2 889The baseline of a lower limit will be at least this 890much below the bottom of the object on which the limit is set. 891. 892.TP 893.B big_op_spacing3 894The bottom of an upper limit will be at least this much above the 895top of the object on which the limit is set. 896. 897.TP 898.B big_op_spacing4 899The top of a lower limit will be at least this much below 900the bottom of the object on which the limit is set. 901. 902.TP 903.B big_op_spacing5 904This much vertical space will be added above and below limits. 905. 906.TP 907.B baseline_sep 908The baselines of the rows in a pile or matrix will normally be 909this far apart. 910In most cases this should be equal to the sum of 911.B num1 912and 913.BR denom1 . 914. 915.TP 916.B shift_down 917The midpoint between the top baseline and the bottom baseline 918in a matrix or pile will be shifted down by this much from the axis. 919In most cases this should be equal to 920.BR axis_height . 921. 922.TP 923.B column_sep 924This much space will be added between columns in a matrix. 925. 926.TP 927.B matrix_side_sep 928This much space will be added at each side of a matrix. 929. 930.TP 931.B draw_lines 932If this is non-zero, lines will be drawn using the 933.B \eD 934escape sequence, rather than with the 935.B \el 936escape sequence and the 937.B \e(ru 938character. 939. 940.TP 941.B body_height 942The amount by which the height of the equation exceeds this 943will be added as extra space before the line containing the equation 944(using 945.BR \ex ). 946The default value is 85. 947. 948.TP 949.B body_depth 950The amount by which the depth of the equation exceeds this 951will be added as extra space after the line containing the equation 952(using 953.BR \ex ). 954The default value is 35. 955. 956.TP 957.B nroff 958If this is non-zero, 959then 960.B ndefine 961will behave like 962.B define 963and 964.B tdefine 965will be ignored, 966otherwise 967.B tdefine 968will behave like 969.B define 970and 971.B ndefine 972will be ignored. 973The default value is\~0 974(This is typically changed to\~1 by the 975.B eqnrc 976file for the 977.BR ascii , 978.BR latin1 , 979.BR utf8 , 980and 981.B cp1047 982devices.) 983. 984.LP 985A more precise description of the role of many of these 986parameters can be found in Appendix\~H of 987.IR "The \*(txbook" . 988.RE 989. 990.SS Macros 991Macros can take arguments. 992In a macro body, 993.BI $ n 994where 995.I n 996is between 1 and\~9, 997will be replaced by the 998.IR n-th 999argument if the macro is called with arguments; 1000if there are fewer than 1001.I n\~\c 1002arguments, it will be replaced by nothing. 1003A word containing a left parenthesis where the part of the word 1004before the left parenthesis has been defined using the 1005.B define 1006command 1007will be recognized as a macro call with arguments; 1008characters following the left parenthesis 1009up to a matching right parenthesis will be treated as comma-separated 1010arguments; 1011commas inside nested parentheses do not terminate an argument. 1012. 1013.TP 1014.BI sdefine\ name\ X\ anything\ X 1015This is like the 1016.B define 1017command, but 1018.I name 1019will not be recognized if called with arguments. 1020. 1021.TP 1022.BI include\ \(ts file \(ts 1023.TQ 1024.BI copy\ \(ts file \(ts 1025Include the contents of 1026.I file 1027.RB ( include 1028and 1029.B copy 1030are synonyms). 1031Lines of 1032.I file 1033beginning with 1034.B .EQ 1035or 1036.B .EN 1037will be ignored. 1038. 1039.TP 1040.BI ifdef\ name\ X\ anything\ X 1041If 1042.I name 1043has been defined by 1044.B define 1045(or has been automatically defined because 1046.I name 1047is the output device) 1048process 1049.IR anything ; 1050otherwise ignore 1051.IR anything . 1052.I X 1053can be any character not appearing in 1054.IR anything . 1055. 1056.TP 1057.BI undef\ name 1058Remove definition of 1059.IR name , 1060making it undefined. 1061. 1062.LP 1063Besides the macros mentioned above, the following definitions are available: 1064.BR Alpha , 1065.BR Beta , 1066\&.\|.\|., 1067.B Omega 1068(this is the same as 1069.BR ALPHA , 1070.BR BETA , 1071\&.\|.\|., 1072.BR OMEGA ), 1073.B ldots 1074(three dots on the base line), 1075and 1076.BR dollar . 1077. 1078.SS Fonts 1079.B eqn 1080normally uses at least two fonts to set an equation: 1081an italic font for letters, 1082and a roman font for everything else. 1083The existing 1084.B gfont 1085command 1086changes the font that is used as the italic font. 1087By default this is\~\c 1088.BR I . 1089The font that is used as the roman font can be changed 1090using the new 1091.B grfont 1092command. 1093. 1094.TP 1095.BI grfont\ f 1096Set the roman font to\~\c 1097.IR f . 1098. 1099.LP 1100The 1101.B italic 1102primitive uses the current italic font set by 1103.BR gfont ; 1104the 1105.B roman 1106primitive uses the current roman font set by 1107.BR grfont . 1108There is also a new 1109.B gbfont 1110command, which changes the font used by the 1111.B bold 1112primitive. 1113If you only use the 1114.BR roman , 1115.B italic 1116and 1117.B bold 1118primitives to changes fonts within an equation, 1119you can change all the fonts used by your equations 1120just by using 1121.BR gfont , 1122.B grfont 1123and 1124.B gbfont 1125commands. 1126. 1127.LP 1128You can control which characters are treated as letters 1129(and therefore set in italics) by using the 1130.B chartype 1131command described above. 1132A type of 1133.B letter 1134will cause a character to be set in italic type. 1135A type of 1136.B digit 1137will cause a character to be set in roman type. 1138. 1139. 1140.SH FILES 1141.Tp \w'\fB@MACRODIR@/eqnrc'u+2n 1142.B @MACRODIR@/eqnrc 1143Initialization file. 1144. 1145. 1146.SH BUGS 1147Inline equations will be set at the point size that is current at the 1148beginning of the input line. 1149. 1150. 1151.SH "SEE ALSO" 1152.BR groff (@MAN1EXT@), 1153.BR @g@troff (@MAN1EXT@), 1154.BR @g@pic (@MAN1EXT@), 1155.BR groff_font (@MAN5EXT@), 1156.I The\ \*(txbook 1157. 1158.\" Local Variables: 1159.\" mode: nroff 1160.\" End: 1161