| 1 |
\def\fileversion{v5.9} \def\filedate{2026/03/08} |
| 2 |
|
| 3 |
\title{\texttt{\itshape |
| 4 |
lineno.sty \ \fileversion\ \filedate |
| 5 |
\unskip}\\\ \\ |
| 6 |
A \LaTeX\ package to attach |
| 7 |
\\ line numbers to paragraphs |
| 8 |
\unskip}\author{% |
| 9 |
Stephan I. Böttcher |
| 10 |
\\ Uwe Lück |
| 11 |
\\ Karl Wette |
| 12 |
\unskip}\date{% |
| 13 |
\url{https://github.com/latex-lineno/lineno} |
| 14 |
\\} |
| 15 |
|
| 16 |
\documentclass[a4paper,12pt]{article}%D |
| 17 |
\usepackage{lineno}%D |
| 18 |
\usepackage{hyperref}%D |
| 19 |
\catcode`\_\active\let_~ |
| 20 |
\def~{\verb~} |
| 21 |
\let\lessthan< |
| 22 |
\def\angle#1{$\langle${\itshape#1}\/$\rangle$} |
| 23 |
\catcode`\|\active |
| 24 |
\def|#1{\ttfamily\string#1} |
| 25 |
\newenvironment{code} |
| 26 |
{\par\runninglinenumbers |
| 27 |
\modulolinenumbers[1]% |
| 28 |
\linenumbersep.3em |
| 29 |
\footnotesize |
| 30 |
\def\linenumberfont |
| 31 |
{\normalfont\tiny\itshape}} |
| 32 |
{} |
| 33 |
{\makeatletter \gdef\scs#1{\texttt |
| 34 |
{\protect\@backslashchar#1}}} |
| 35 |
\def\old{\par\footnotesize} |
| 36 |
{\catcode`\/\active |
| 37 |
\gdef\path{\begingroup\catcode`\/\active |
| 38 |
\let/\slash\dopath} |
| 39 |
\gdef\dopath#1{\slash\unpenalty#1\endgroup}} |
| 40 |
|
| 41 |
\begin{document}%D |
| 42 |
\pagewiselinenumbers |
| 43 |
\maketitle |
| 44 |
\pagestyle{headings} |
| 45 |
\tableofcontents |
| 46 |
\sloppy |
| 47 |
|
| 48 |
\section{% |
| 49 |
Introductions |
| 50 |
\unskip} |
| 51 |
|
| 52 |
(New v4.00) Parts of former first section |
| 53 |
have been rendered separate subsections for package |
| 54 |
version_v4.00. (/New v4.00) |
| 55 |
|
| 56 |
\subsection{% |
| 57 |
Introduction to versions $\textrm{v}\lessthan4$ |
| 58 |
\unskip} |
| 59 |
|
| 60 |
This package provides line numbers on paragraphs. |
| 61 |
After \TeX\ has broken a paragraph into lines there will |
| 62 |
be line numbers attached to them, with the possibility to |
| 63 |
make references through the \LaTeX\ ~\ref~, ~\pageref~ |
| 64 |
cross reference mechanism. This includes four issues: |
| 65 |
\begin{itemize} |
| 66 |
\item attach a line number on each line, |
| 67 |
\item create references to a line number, |
| 68 |
\item control line numbering mode, |
| 69 |
\item count the lines and print the numbers. |
| 70 |
\end{itemize} |
| 71 |
The first two points are implemented through patches to |
| 72 |
the output routine. The third by redefining ~\par~, ~\@par~ |
| 73 |
and ~\@@par~. The counting is easy, as long as you want |
| 74 |
the line numbers run through the text. If they shall |
| 75 |
start over at the top of each page, the aux-file as well |
| 76 |
as \TeX s memory have to carry a load for each counted line. |
| 77 |
|
| 78 |
I wrote this package for my wife Petra, who needs it for |
| 79 |
transcriptions of interviews. This allows her to |
| 80 |
precisely refer to passages in the text. It works well |
| 81 |
together with ~\marginpar~s, but not too well with displaymath. |
| 82 |
~\footnote~s are a problem, especially when they |
| 83 |
are split, but we may get there. |
| 84 |
(New v4.00 UL) Version v4.00 overcomes the problem, I believe. |
| 85 |
(/UL /New v4.00) |
| 86 |
|
| 87 |
lineno.sty works |
| 88 |
surprisingly well with other packages, for |
| 89 |
example, ~wrapfig.sty~. So please try if it |
| 90 |
works with whatever you need, and if it does, |
| 91 |
please tell me, and if it does not, tell me as |
| 92 |
well, so I can try to fix it. |
| 93 |
|
| 94 |
\subsection{% |
| 95 |
Introduction to versions v4.00ff. (UL) |
| 96 |
\unskip} |
| 97 |
|
| 98 |
~lineno.sty~ has been maintained by Stephan until version_v3.14. |
| 99 |
From version_v4.00 onwards, maintenance is shifting towards |
| 100 |
Uwe Lück (UL), who is the author of v4\dots code and of v4\dots |
| 101 |
changes in documentation. This came about as follows. |
| 102 |
|
| 103 |
Since late 2002, Christian Tapp and Uwe Lück have employed |
| 104 |
~lineno.sty~ for their ~ednotes.sty~, a package supporting |
| 105 |
critical editions, while you find ~ednotes.sty~ and surrounding files in |
| 106 |
CTAN folder \path{macros/latex/contrib/ednotes}. |
| 107 |
|
| 108 |
Soon, some weaknesses of ~lineno.sty~ showed up, mainly since |
| 109 |
Christian's critical editions (using ~ednotes.sty~) needed lots |
| 110 |
of ~\linelabel~s and footnotes. (These weaknesses are due to |
| 111 |
weaknesses of \LaTeX's ~\marginpar~ mechanism that Stephan |
| 112 |
used for ~\linelabel~.) So we changed some ~lineno.sty~ |
| 113 |
definitions in some extra files, which moreover offered new |
| 114 |
features. We sent these files to Stephan, hoping he would take |
| 115 |
the changes into ~lineno.sty~. However, he was too short of time. |
| 116 |
|
| 117 |
Writing a TUGboat article on Ednotes in 2004, we hoped to |
| 118 |
reduce the number of files in the Ednotes bundle and so asked |
| 119 |
Stephan again. Now he generously offered maintenance to me, so |
| 120 |
I could execute the changes on my own. |
| 121 |
|
| 122 |
The improvements are as follows: |
| 123 |
\begin{itemize}\item |
| 124 |
[(i)] Footnotes placement approaches intentions better |
| 125 |
(footnotes formerly liked to pile up at late pages). |
| 126 |
\item |
| 127 |
[(ii)] The number of ~\linelabel~s in one paragraph is no longer |
| 128 |
limited to 18. |
| 129 |
\item |
| 130 |
[(iii)] ~\pagebreak~, ~\nopagebreak~, ~\vspace~, and the star |
| 131 |
and optional versions of ~\\~ work as one would expect |
| 132 |
(section_\ref{s:MVadj}). %% Added for v4.1 |
| 133 |
\item |
| 134 |
[(iv)] A command is offered which chooses the first line number |
| 135 |
to be printed in the margin |
| 136 |
(subsection_\ref{ss:Mod}). %% Added for v4.1 |
| 137 |
\item |
| 138 |
[(v)] (New v4.1) \LaTeX\ tabular environments (optionally) |
| 139 |
get line numbers as well, and you can refer to them in the |
| 140 |
usual automatic way. (It may be considered a shortcoming that, |
| 141 |
precisely, \emph{rows} are numbered, not lines.---See |
| 142 |
subsection_\ref{ss:Tab}.) |
| 143 |
\item |
| 144 |
[(vi)] We are moving towards referring to math items |
| 145 |
(subsection_\ref{ss:MathRef} and the hooks in |
| 146 |
subsection_\ref{ss:LL}). |
| 147 |
(/New v4.1) |
| 148 |
\end{itemize} |
| 149 |
(Thanks to Stephan for making this possible!) |
| 150 |
|
| 151 |
|
| 152 |
Ednotes moreover profits from Stephan's offer with regard |
| 153 |
to the documentation of our code which yielded these |
| 154 |
improvements formerly. This documentation now becomes |
| 155 |
printable, being part of the ~lineno.sty~ documentation. |
| 156 |
|
| 157 |
Of course, Stephan's previous ~lineno.sty~ versions were a great |
| 158 |
and ingenious work and exhibit greatest \TeX pertise. I never |
| 159 |
could have done this. I learnt a lot in studying the code when |
| 160 |
Christian pointed out strange output results and error |
| 161 |
messages, and there are still large portions of ~lineno.sty~ |
| 162 |
which I don't understand (consider only pagewise numbering of |
| 163 |
lines). Fortunately, Stephan has offered future help if |
| 164 |
needed.---My code for attaching line numbers to \emph{tabular |
| 165 |
environments} (as mentioned above, now still in |
| 166 |
~edtable.sty~) %% %% TODO |
| 167 |
developed from macros which Stephan and Christian experimented |
| 168 |
with in December 2002. Stephan built the basics. |
| 169 |
(However, I then became too proud to follow his advice only to |
| 170 |
use and modify ~longtable.sty~.) |
| 171 |
|
| 172 |
There are some issues concerning use of counters on which I |
| 173 |
don't agree with Stephan and where I would like to change the |
| 174 |
code if ~lineno.sty~ is ``mine'' as Stephan offered. However, |
| 175 |
Stephan is afraid of compatibility problems from which, in |
| 176 |
particular, his wife could suffer in the near future. So he |
| 177 |
demanded that I change as little as possible for my first |
| 178 |
version. Instead of executing changes that I plan I just offer |
| 179 |
my opinions at the single occasions. I hope to get in touch |
| 180 |
this way with users who consider subtle features vital which I |
| 181 |
consider strange. |
| 182 |
|
| 183 |
On the other hand, the sections on improvements of the |
| 184 |
implementation have been blown up very much and may be tiring |
| 185 |
and litte understandable for mere \emph{users}. These users |
| 186 |
may profit from the present presentation just by jumping to |
| 187 |
sections_\ref{s:Opts} and_\ref{s:UserCmds}. There is a user's |
| 188 |
guide ulineno.tex which may be even more helpful, but it has |
| 189 |
not been updated for a while. %% TODO |
| 190 |
|
| 191 |
\subsection{% |
| 192 |
Availability |
| 193 |
\unskip} |
| 194 |
|
| 195 |
In case you have found the present file otherwise than from |
| 196 |
CTAN: A recent version and documentation of this package |
| 197 |
should be available from CTAN folder |
| 198 |
\path{macros/latex/contrib/lineno}. |
| 199 |
Or mail to one of the addresses at top of file. |
| 200 |
|
| 201 |
\subsection{% |
| 202 |
Introductory code |
| 203 |
\unskip} |
| 204 |
|
| 205 |
This style option is written for \LaTeXe, November 1994 or later, |
| 206 |
since we need the ~\protected@write~ macro. |
| 207 |
|
| 208 |
(New v4.00) And we use ~\newcommand*~ for |
| 209 |
controlling length of user macro arguments, which has been |
| 210 |
available since December 1994. |
| 211 |
(/New v4.00) |
| 212 |
\begin{code}\begin{verbatim} |
| 213 |
\NeedsTeXFormat{LaTeX2e}[1994/12/01] |
| 214 |
\ProvidesPackage{lineno} |
| 215 |
[\filedate\space line numbers on paragraphs \fileversion] |
| 216 |
\RequirePackage{etoolbox} |
| 217 |
\RequirePackage{kvoptions} |
| 218 |
\end{verbatim} |
| 219 |
\end{code} |
| 220 |
|
| 221 |
|
| 222 |
\section{% |
| 223 |
Put the line numbers to the lines |
| 224 |
\unskip} |
| 225 |
|
| 226 |
(New v4.00) This section contained the most |
| 227 |
basic package code previously. For various purposes of |
| 228 |
version_4\dots, much of these basics have been to be modified. |
| 229 |
Much of my (UL's) reasoning on these modifications has been to |
| 230 |
be reported. Sorry, the present section has been blown up |
| 231 |
awfully thus and contains ramifications that may be difficult |
| 232 |
to trace. We add some ~\subsection~ commands in order to cope |
| 233 |
with the new situation. (/New v4.00) |
| 234 |
|
| 235 |
\subsection{% |
| 236 |
Basic code of \texttt{lineno.sty} \scs{output} |
| 237 |
\unskip}\label{ss:output} |
| 238 |
|
| 239 |
The line numbers have to be attached by the output |
| 240 |
routine. We simply set the ~\interlinepenalty~ to $-100000$. |
| 241 |
The output routine will be called after each line in the |
| 242 |
paragraph, except the last, where we trigger by ~\par~. |
| 243 |
The ~\linenopenalty~ is small enough to compensate a bunch of |
| 244 |
penalties (e.g., with ~\samepage~). |
| 245 |
|
| 246 |
(New v3.04) Longtable uses |
| 247 |
~\penalty~$-30000$. The lineno penalty range was |
| 248 |
shrunk to $-188000 \dots -32000$. (/New v3.04) |
| 249 |
(New v4.00) New values are listed below (11111f.). (/New v4.00) |
| 250 |
\begin{code}\begin{verbatim} |
| 251 |
\newcount\linenopenalty\linenopenalty=-100000 |
| 252 |
\end{verbatim} |
| 253 |
\end{code} |
| 254 |
(UL) Hm. It is never needed below |
| 255 |
that this is a counter. ~\def\linenopenalty{-100000\relax}~ |
| 256 |
would do. (I guess this consumes more memory, but it |
| 257 |
is more important to save counters than to save memory.) |
| 258 |
I was frightened by ~-\linenopenalty~ below, but indeed |
| 259 |
\TeX\ interprets the string ~--100000~ as 100000. |
| 260 |
Has any user or extension package writer ever called |
| 261 |
~\linenopenalty=xxx~, or could I really change this?---The |
| 262 |
counter is somewhat faster than the macro. Together with the |
| 263 |
compatibility question this seems to support keeping the |
| 264 |
counter. (???) |
| 265 |
(/UL) |
| 266 |
\begin{code}\begin{verbatim} |
| 267 |
\mathchardef\linenopenaltypar=32000 |
| 268 |
\end{verbatim} |
| 269 |
\end{code} |
| 270 |
So let's make a hook to ~\output~, the direct way. The \LaTeX\ |
| 271 |
macro ~\@reinserts~ puts the footnotes back on the page. |
| 272 |
|
| 273 |
(New v3.01) ~\@reinserts~ badly |
| 274 |
screws up split footnotes. The bottom part is |
| 275 |
still on the recent contributions list, and the |
| 276 |
top part will be put back there after the bottom |
| 277 |
part. Thus, since lineno.sty does not play well |
| 278 |
with ~\inserts~ anyway, we can safely experiment |
| 279 |
with ~\holdinginserts~, without making things |
| 280 |
much worse. |
| 281 |
|
| 282 |
Or that's what I thought, but: Just activating |
| 283 |
~\holdinginserts~ while doing the ~\par~ will |
| 284 |
not do the trick: The ~\output~ routine may be |
| 285 |
called for a real page break before all line |
| 286 |
numbers are done, and how can we get control |
| 287 |
over ~\holdinginserts~ at that point? |
| 288 |
|
| 289 |
Let's try this: When the ~\output~ routine is |
| 290 |
run with ~\holdinginserts=3~ for a real page |
| 291 |
break, then we reset ~\holdinginserts~ and |
| 292 |
restart ~\output~. |
| 293 |
|
| 294 |
Then, again, how do we keep the remaining |
| 295 |
~\inserts~ while doing further line numbers? |
| 296 |
|
| 297 |
If we find ~\holdinginserts~=$-3$ we activate it again |
| 298 |
after doing ~\output~. (/New v3.01) |
| 299 |
|
| 300 |
(New v3.02) To work with |
| 301 |
multicol.sty, the original output routine is now |
| 302 |
called indirectly, instead of being replaced. |
| 303 |
When multicol.sty changes ~\output~, it is a |
| 304 |
toks register, not the real thing. (/New v3.02) |
| 305 |
|
| 306 |
(New v4.00) Two further complications are added. |
| 307 |
\begin{itemize}\item |
| 308 |
[(i)] Problems with footnotes formerly resulted from |
| 309 |
\LaTeX's ~\@reinserts~ in ~\@specialoutput~ which Stephan's |
| 310 |
~\linelabel~ called via the ~\marginpar~ mechanism. |
| 311 |
\item |
| 312 |
[(ii)] \LaTeX\ commands using ~\vadjust~ formerly didn't work |
| 313 |
as one would have hoped. The problem is as follows: |
| 314 |
Printing the line number results from |
| 315 |
a box that the output routine inserts at the place of the |
| 316 |
~\interlinepenalty~. ~\vadjust~ items appear \emph{above} the |
| 317 |
~\interlinepenalty~ (\TeX book p._105). So ~\pagebreak~, e.g., |
| 318 |
formerly sent the line number to the next page, while the |
| 319 |
penalty from ~\nopagebreak~ could not tie the following line, |
| 320 |
since it was screened off by the line number box.---Our trick |
| 321 |
is putting the ~\vadjust~ items into a list macro from which |
| 322 |
the output routine transfers them into the vertical list, |
| 323 |
below the line number box. |
| 324 |
\end{itemize} |
| 325 |
In this case_(ii), like in case_(i), footnotes would suffer |
| 326 |
if ~\holdinginserts~ were non-positive. Indeed, in both |
| 327 |
cases_(i) and_(ii) we tackle the footnote problem by extending |
| 328 |
that part of Stephan's output routine that is active when |
| 329 |
~\holdinginserts~ is positive. This extension writes the line |
| 330 |
number ~\newlabel~ to the .aux file (which was formerly done |
| 331 |
under $~\holdinginserts~=-3$) and handles the ~\vadjust~ |
| 332 |
items.---To trigger ~\output~ and its ~\linelabel~ or, resp., |
| 333 |
~\vadjust~ part, the list of signal penalties started |
| 334 |
immediately before is increased here (first for ~\linelabel~, |
| 335 |
second for postponed ~\vadjust~ items): |
| 336 |
\begin{code}\begin{verbatim} |
| 337 |
\mathchardef\@Mllbcodepen=11111 |
| 338 |
\mathchardef\@Mppvacodepen=11112 |
| 339 |
\end{verbatim} |
| 340 |
\end{code} |
| 341 |
(/New v4.00) (New v4.2) David Kastrup urges to use a private |
| 342 |
name instead of ~\the\output~ (LaTeX-L-list). Otherwise an |
| 343 |
~\output~ routine loaded later and using ~\newtoks\output~ |
| 344 |
again may get lost entirely. So we change use of ~\@LN@output~, |
| 345 |
using it for the former purpose. Reference to what appeared |
| 346 |
with the name of ~\output~ here lasts for a few lines and then |
| 347 |
is given away. |
| 348 |
\begin{code}\begin{verbatim} |
| 349 |
\let\@tempa\output |
| 350 |
\newtoks\output |
| 351 |
\let\@LN@output\output |
| 352 |
\output=\expandafter{\the\@tempa} |
| 353 |
\end{verbatim} |
| 354 |
\end{code} |
| 355 |
Now we add two cases to Stephan's output routine. (New v4.00) |
| 356 |
\begin{code}\begin{verbatim} |
| 357 |
\@tempa={% |
| 358 |
\end{verbatim} |
| 359 |
\end{code} |
| 360 |
(/New 4.2) |
| 361 |
\begin{code}\begin{verbatim} |
| 362 |
\LineNoTest |
| 363 |
\if@tempswa |
| 364 |
\end{verbatim} |
| 365 |
\end{code} |
| 366 |
(New v4.00) |
| 367 |
We insert recognition of waiting ~\linelabel~ items--- |
| 368 |
\begin{code}\begin{verbatim} |
| 369 |
\ifnum\outputpenalty=-\@Mllbcodepen |
| 370 |
\WriteLineNo |
| 371 |
\end{verbatim} |
| 372 |
\end{code} |
| 373 |
---and of waiting ~\vadjust~ items: |
| 374 |
\begin{code}\begin{verbatim} |
| 375 |
\else |
| 376 |
\ifnum\outputpenalty=-\@Mppvacodepen |
| 377 |
\PassVadjustList |
| 378 |
\else |
| 379 |
\end{verbatim} |
| 380 |
\end{code} |
| 381 |
(/New v4.00) (New v4.2) Outsource ``Standard'' output |
| 382 |
---which occurs so rarely---to subsection_\ref{ss:LLO}: |
| 383 |
\begin{code}\begin{verbatim} |
| 384 |
\LineNoLaTeXOutput |
| 385 |
\end{verbatim} |
| 386 |
\end{code} |
| 387 |
(/New v4.2) (New v4.00) |
| 388 |
Two new ~\fi~s for the ~\linelabel~ and ~\vadjust~ tests--- |
| 389 |
\begin{code}\begin{verbatim} |
| 390 |
\fi |
| 391 |
\fi |
| 392 |
\end{verbatim} |
| 393 |
\end{code} |
| 394 |
---and the remaining is |
| 395 |
Stephan's code again: |
| 396 |
(/New v4.00) |
| 397 |
\begin{code}\begin{verbatim} |
| 398 |
\else |
| 399 |
\MakeLineNo |
| 400 |
\fi |
| 401 |
} |
| 402 |
\end{verbatim} |
| 403 |
\end{code} |
| 404 |
(New v4.00) Our new macros |
| 405 |
~\WriteLineNo~ and ~\PassVadjustList~ will be dealt with in |
| 406 |
sections_\ref{s:LNref} and_\ref{ss:PVadj}. (/New v4.00) |
| 407 |
|
| 408 |
\subsection{% |
| 409 |
\scs{LineNoTest} |
| 410 |
\unskip} |
| 411 |
|
| 412 |
The float mechanism inserts ~\interlinepenalty~s during |
| 413 |
~\output~. So carefully reset it before going on. Else |
| 414 |
we get doubled line numbers on every float placed in |
| 415 |
horizontal mode, e.g, from ~\linelabel~. |
| 416 |
|
| 417 |
Sorry, neither a ~\linelabel~ nor a ~\marginpar~ should |
| 418 |
insert a penalty, else the following line number |
| 419 |
could go to the next page. Nor should any other |
| 420 |
float. So let us suppress the ~\interlinepenalty~ |
| 421 |
altogether with the ~\@nobreak~ switch. |
| 422 |
|
| 423 |
Since (ltspace.dtx, v1.2p)[1996/07/26], the ~\@nobreaktrue~ does |
| 424 |
it's job globally. We need to do it locally here. |
| 425 |
\begin{code}\begin{verbatim} |
| 426 |
\def\LineNoTest{% |
| 427 |
\let\@@par\@@@par |
| 428 |
\ifnum\interlinepenalty<-\linenopenaltypar |
| 429 |
\advance\interlinepenalty-\linenopenalty |
| 430 |
\@LN@nobreaktrue |
| 431 |
\fi |
| 432 |
\@tempswatrue |
| 433 |
\ifnum\outputpenalty>-\linenopenaltypar\else |
| 434 |
\ifnum\outputpenalty>-188000\relax |
| 435 |
\@tempswafalse |
| 436 |
\fi |
| 437 |
\fi |
| 438 |
} |
| 439 |
|
| 440 |
\def\@LN@nobreaktrue{\let\if@nobreak\iftrue} % renamed v4.33 |
| 441 |
\end{verbatim} |
| 442 |
\end{code} |
| 443 |
(UL) I thought here were |
| 444 |
another case of the save stack problem explained in \TeX book, |
| 445 |
p._301, namely through both local and global changing |
| 446 |
~\if@nobreak~. However, ~\@LN@nobreak~ is called during |
| 447 |
~\@LN@output~ only, while ~\@nobreaktrue~ is called by \LaTeX's |
| 448 |
~\@startsection~ only. The latter never happens during |
| 449 |
~\@LN@output~. So there is no local value of ~\if@nobreak~ on |
| 450 |
save stack when ~\@nobreaktrue~ acts, since ~\the\@LN@output~ |
| 451 |
(where ~\@LN@output~ is a new name for the original ~\output~) |
| 452 |
is executed within a group (\TeX book p._21). |
| 453 |
(/UL) |
| 454 |
|
| 455 |
\subsection{% |
| 456 |
Other output routines (v4.2) |
| 457 |
\unskip}\label{ss:LLO} |
| 458 |
|
| 459 |
I had thought of dealing with bad interference of footnotes |
| 460 |
(and ~\enlargethispage~) with (real) ~\marginpar~s and floats |
| 461 |
\emph{here}. Yet this is done in |
| 462 |
\[ |
| 463 |
~http://~\mbox{[CTAN]} |
| 464 |
~/macros/latex/contrib/tamefloats/tameflts.sty~ |
| 465 |
\] |
| 466 |
now, and I prefer striving for compatibility with the latter. |
| 467 |
(See there for expanding on the problem.) |
| 468 |
This requires returning the special absolute value of |
| 469 |
~\holdinginserts~ that ~lineno.sty~ finds at the end of a newly |
| 470 |
typeset paragraph---now done in subsection_\ref{ss:calls} |
| 471 |
(~\linenumberpar~). |
| 472 |
The former ~\LineNoHoldInsertsTest~ has been filled into here. |
| 473 |
Note: when the following code is invoked, we have |
| 474 |
~\if@tempswa~_ =_~\iftrue~. |
| 475 |
WARNING: I am still not sure whether the present code is good |
| 476 |
for cooperating with other packages that use ~\holdinginserts~. |
| 477 |
\begin{code}\begin{verbatim} |
| 478 |
\def\LineNoLaTeXOutput{% |
| 479 |
\ifnum \holdinginserts=\thr@@ % v4.33 without \@tempswafalse |
| 480 |
\global\holdinginserts-\thr@@ |
| 481 |
\unvbox\@cclv |
| 482 |
\ifnum \outputpenalty=\@M \else \penalty\outputpenalty \fi |
| 483 |
\else |
| 484 |
\if@twocolumn \let\@makecol\@LN@makecol \fi |
| 485 |
\the\@LN@output % finally following David Kastrup's advice. |
| 486 |
\ifnum \holdinginserts=-\thr@@ |
| 487 |
\global\holdinginserts\thr@@ \fi |
| 488 |
\fi |
| 489 |
} |
| 490 |
\end{verbatim} |
| 491 |
\end{code} |
| 492 |
\textit{More on dealing with output routines from other |
| 493 |
packages:} |
| 494 |
Since ~lineno.sty~'s output routine is called at least once |
| 495 |
for each output line, I think it should be in \TeX's |
| 496 |
original ~\output~, while output routines dealing with |
| 497 |
building pages and with floats etc.\ should be filled into |
| 498 |
registers addressed by ~\output~ after ~\newtoks\output~. |
| 499 |
Therefore \begin{enumerate} |
| 500 |
\item |
| 501 |
~tameflts.sty~ should be loaded \emph{after} ~lineno.sty~; |
| 502 |
\item |
| 503 |
if a class changes ~\output~ (APS journal class revtex4, |
| 504 |
e.g.), ~lineno.sty~ should be loaded by ~\RequirePackage~ |
| 505 |
[here presumably following some options in |
| 506 |
brackets]~{lineno}~ \emph{preceding} ~\documentclass~. |
| 507 |
\item |
| 508 |
If you actually maintain such a class, please consider |
| 509 |
loading ~lineno.sty~ on some draft option. The bunch of |
| 510 |
lineno's package options may be a problem, but perhaps the |
| 511 |
purpose of your class is offering only very few of lineno's |
| 512 |
options anyway, maybe just one. |
| 513 |
\end{enumerate} |
| 514 |
The latter may also be needed with classes that don't follow |
| 515 |
David Kastrup's rule on changing ~\output~. |
| 516 |
|
| 517 |
\subsection{% |
| 518 |
\scs{MakeLineNo}: Actually attach line number |
| 519 |
\unskip}\label{ss:MLN} |
| 520 |
|
| 521 |
We have to return all the page to the current page, and |
| 522 |
add a box with the line number, without adding |
| 523 |
breakpoints, glue or space. The depth of our line number |
| 524 |
should be equal to the previous depth of the page, in |
| 525 |
case the page breaks here, and the box has to be moved up |
| 526 |
by that depth. |
| 527 |
|
| 528 |
The ~\interlinepenalty~ comes after the ~\vadjust~ from a |
| 529 |
~\linelabel~, so we increment the line number \emph{after} |
| 530 |
printing it. The macro ~\makeLineNumber~ produces the |
| 531 |
text of the line number, see section \ref{appearance}. |
| 532 |
|
| 533 |
(UL) I needed a while to understand |
| 534 |
the sentence on incrementing. Correctly: writing the |
| 535 |
~\newlabel~ to the .aux file is triggered by the signal |
| 536 |
penalty that ~\end@float~ inserts via ~\vadjust~. |
| 537 |
However, this could be changed by our new ~\PostponeVadjust~. |
| 538 |
After ~\c@linenumber~ has been introduced as a \LaTeX\ |
| 539 |
counter, it might be preferable that it behaved like standard |
| 540 |
\LaTeX\ counters which are incremented shortly before printing. |
| 541 |
But this may be of little practical relevance in this case, |
| 542 |
as ~\c@linenumber~ is driven in a very non-standard |
| 543 |
way.---However still, this behaviour of ~\c@linenumber~ |
| 544 |
generates a problem with our ~edtable.sty~. |
| 545 |
(/UL). |
| 546 |
|
| 547 |
Finally we put in the natural ~\interlinepenalty~, except |
| 548 |
after the last line. |
| 549 |
|
| 550 |
(New v3.10) Frank Mittelbach points out that box255 may be |
| 551 |
less deep than the last box inside, so he proposes to |
| 552 |
measure the page depth with ~\boxmaxdepth=\maxdimen~. |
| 553 |
(/New v3.10) |
| 554 |
|
| 555 |
(UL, New v4.00) We also resume the matter of |
| 556 |
~\vadjust~ items that was started in section_\ref{ss:output}. |
| 557 |
|
| 558 |
\TeX\ puts only nonzero interline |
| 559 |
penalties into the vertical list (\TeX book p._105), while |
| 560 |
~lineno.sty~ formerly replaced the signal interline penalty by |
| 561 |
something closing with an explicit penalty of the value that |
| 562 |
the interline penalty would have without ~lineno.sty~. |
| 563 |
This is usually 0. Now, explicit vertical penalties can be |
| 564 |
very nasty with respect to ~\nopagebreak~, e.g., a low (even |
| 565 |
positive) ~\widowpenalty~ may force a widow where you |
| 566 |
explicitly tried to forbid it by ~\nopagebreak~ |
| 567 |
(see explanation soon below). |
| 568 |
The ~\nopagebreak~ we create here would never work if all |
| 569 |
those zero penalties were present.---On |
| 570 |
the other hand, we cannot just omit Stephan's zero penalties, |
| 571 |
because \TeX\ puts a penalty of 10000 after what ~lineno.sty~ |
| 572 |
inserts (\TeX book p._125). This penalty must be overridden |
| 573 |
to allow page breaks between ordinary lines. To revive |
| 574 |
~\nopagebreak~, we therefore replace those zero (or low) |
| 575 |
penalties by penalties that the user demanded by |
| 576 |
~\nopagebreak~.---This mechanism is not perfect and does not |
| 577 |
exactly restore the original \LaTeX\ working of ~\pagebreak~ |
| 578 |
and ~\nopagebreak~. Viz., if there are several vertical |
| 579 |
penalties after a line which were produced by closely sitting |
| 580 |
~\[no]pagebreak~s, without ~lineno.sty~ the lowest penalty would |
| 581 |
be effective (cf._\TeX book exercise_14.10). Our mechanism, by |
| 582 |
contrast, chooses the \emph{last} user-set penalty of the line |
| 583 |
as the effective one. It would not be very difficult to come |
| 584 |
more close to the original mechanism, but until someone urges |
| 585 |
us we will cling to the present simple way. You may consider an |
| 586 |
advantage of the difference between our mechanism and the |
| 587 |
original one that the user here can actually override low |
| 588 |
penalties by ~\nopagebreak~, which may be what a lay \LaTeX\ |
| 589 |
user would expect. |
| 590 |
(/UL, /New v4.00) |
| 591 |
\begin{code}\begin{verbatim} |
| 592 |
\def\MakeLineNo{% |
| 593 |
\@LN@maybe@normalLineNumber % v4.31 |
| 594 |
\boxmaxdepth\maxdimen\setbox\z@\vbox{\unvbox\@cclv}% |
| 595 |
\@tempdima\dp\z@ \unvbox\z@ |
| 596 |
\sbox\@tempboxa{\hb@xt@\z@{\makeLineNumber}}% |
| 597 |
\end{verbatim} |
| 598 |
\end{code} |
| 599 |
(New v4.00) Previously, |
| 600 |
\begin{old}\begin{verbatim} |
| 601 |
% \stepcounter{linenumber}% |
| 602 |
\end{verbatim} |
| 603 |
\end{old} |
| 604 |
followed. (Of course, there was no |
| 605 |
comment mark; I put it there to make |
| 606 |
reading the actual code easy.) |
| 607 |
|
| 608 |
(New v4.22: improved) Why not just |
| 609 |
\[~\global\advance\c@linenumber\@ne~?\] |
| 610 |
~\stepcounter~ additionally resets ``subordinate'' |
| 611 |
counters, but which could these (usefully) be? |
| 612 |
Again, may be column counters with ~edtable.sty~!? |
| 613 |
|
| 614 |
But then, our ~edtable.sty~ and its ~longtable~ option |
| 615 |
should use it as well. So use a shorthand supporting |
| 616 |
uniformity. You can even use it as a hook for choosing |
| 617 |
~\global\advance\c@linenumber\@ne~ instead of our choice. |
| 618 |
(/New v4.22) |
| 619 |
\begin{code}\begin{verbatim} |
| 620 |
\stepLineNumber |
| 621 |
\end{verbatim} |
| 622 |
\end{code} |
| 623 |
(New v4.4) Now |
| 624 |
\begin{code}\begin{verbatim} |
| 625 |
\ht\@tempboxa\z@ \@LN@depthbox |
| 626 |
\end{verbatim} |
| 627 |
\end{code} |
| 628 |
appends the box containing the line number without changing |
| 629 |
~\prevdepth~---see end of section. |
| 630 |
Now is the time for inserting the $\dots$ (/New v4.4) |
| 631 |
~\vadjust~ items. We cannot do this much later, because |
| 632 |
their right place is above the artificial interline |
| 633 |
penalty which Stephan's code will soon insert |
| 634 |
(cf._\TeX book p._105). The next command is just ~\relax~ |
| 635 |
if no ~\vadjust~ items have been accumulated for the |
| 636 |
current line. Otherwise it is a list macro inserting |
| 637 |
the ~\vadjust~ items and finally resetting itself. |
| 638 |
(This is made in section_\ref{ss:PVadj} below.) |
| 639 |
If the final item is a penalty, it is stored so it can |
| 640 |
compete with other things about page breaking. |
| 641 |
\begin{code}\begin{verbatim} |
| 642 |
\@LN@do@vadjusts |
| 643 |
\count@\lastpenalty |
| 644 |
\end{verbatim} |
| 645 |
\end{code} |
| 646 |
At this place, |
| 647 |
\begin{old}\begin{verbatim} |
| 648 |
% \ifnum\outputpenalty=-\linenopenaltypar\else |
| 649 |
\end{verbatim} |
| 650 |
\end{old} |
| 651 |
originally followed. We need something \emph{before} the |
| 652 |
~\else~: |
| 653 |
\begin{code}\begin{verbatim} |
| 654 |
\ifnum\outputpenalty=-\linenopenaltypar |
| 655 |
\ifnum\count@=\z@ \else |
| 656 |
\end{verbatim} |
| 657 |
\end{code} |
| 658 |
So final ~\pagebreak[0]~ or ~\nopagebreak[0]~ has no |
| 659 |
effect---but this will make a difference after headings only, |
| 660 |
where nobody should place such a thing anyway. |
| 661 |
\begin{code}\begin{verbatim} |
| 662 |
\xdef\@LN@parpgbrk{% |
| 663 |
\penalty\the\count@ |
| 664 |
\global\let\noexpand\@LN@parpgbrk |
| 665 |
\noexpand\@LN@screenoff@pen}% v4.4 |
| 666 |
\end{verbatim} |
| 667 |
\end{code} |
| 668 |
That penalty will replace former ~\kern\z@~ in |
| 669 |
~\linenumberpar~, see subsection_\ref{ss:calls}.---A |
| 670 |
few days earlier, I tried to send just a penalty value. |
| 671 |
However, the ~\kern\z@~ in ~\linenumberpar~ is crucial, |
| 672 |
as I then found out. See below.---The final penalty is |
| 673 |
repeated, but this does no harm. (It would not be very |
| 674 |
difficult to avoid the repeating, but it may even be |
| 675 |
less efficient.) It may be repeated due to the previous |
| 676 |
~\xdef~, but it may be repeated as well below in the |
| 677 |
present macro where artificial interline penalty is to |
| 678 |
be overridden. |
| 679 |
\begin{code}\begin{verbatim} |
| 680 |
\fi |
| 681 |
\else |
| 682 |
\end{verbatim} |
| 683 |
\end{code} |
| 684 |
(/New v4.00) |
| 685 |
\begin{code}\begin{verbatim} |
| 686 |
\@tempcnta\outputpenalty |
| 687 |
\advance\@tempcnta -\linenopenalty |
| 688 |
\end{verbatim} |
| 689 |
\end{code} |
| 690 |
(New v4.00) |
| 691 |
\begin{old}\begin{verbatim} |
| 692 |
% \penalty\@tempcnta |
| 693 |
\end{verbatim} |
| 694 |
\end{old} |
| 695 |
followed previously. To give ~\nopagebreak~ a chance, |
| 696 |
we do |
| 697 |
\begin{code}\begin{verbatim} |
| 698 |
\penalty \ifnum\count@<\@tempcnta \@tempcnta \else \count@ \fi |
| 699 |
\end{verbatim} |
| 700 |
\end{code} |
| 701 |
instead.---In ~linenox0.sty~, the ~\else~ thing once was omitted. |
| 702 |
Sergei Mariev's complaint (thanks!) showed that it is vital |
| 703 |
(see comment before ~\MakeLineNo~). |
| 704 |
The remaining ~\fi~ from previous package version closes the |
| 705 |
~\ifnum\outputpenalty~\dots |
| 706 |
(/New v4.00) |
| 707 |
\begin{code}\begin{verbatim} |
| 708 |
\fi |
| 709 |
} |
| 710 |
\end{verbatim} |
| 711 |
\end{code} |
| 712 |
(New v4.00) |
| 713 |
\begin{code}\begin{verbatim} |
| 714 |
\newcommand\stepLineNumber{\stepcounter{linenumber}} |
| 715 |
\end{verbatim} |
| 716 |
\end{code} |
| 717 |
For reason, see use above. (/New v4.00) |
| 718 |
|
| 719 |
(New v4.4) The depth preserving trick is drawn here from |
| 720 |
~\MakeLineNo~ because it will be used again in |
| 721 |
section_\ref{ss:calls}. |
| 722 |
(v5.3) Handle special value of ~\prevdepth=-1000pt~. (/v5.3) |
| 723 |
\begin{code}\begin{verbatim} |
| 724 |
\def\@LN@depthbox{% |
| 725 |
\ifdim\@tempdima=-1000pt |
| 726 |
% \nointerlineskip is already set so we don't need set it again |
| 727 |
% (and we shouldn't back up) |
| 728 |
\else |
| 729 |
\dp\@tempboxa=\@tempdima |
| 730 |
\nointerlineskip |
| 731 |
\kern-\@tempdima |
| 732 |
\fi |
| 733 |
\box\@tempboxa} |
| 734 |
\end{verbatim} |
| 735 |
\end{code} |
| 736 |
(/New v4.4) |
| 737 |
|
| 738 |
\section{% |
| 739 |
Control line numbering |
| 740 |
\unskip} |
| 741 |
\subsection{% |
| 742 |
Inserting \scs{output} calls %% own subsec. v4.4. |
| 743 |
\unskip}\label{ss:calls} |
| 744 |
The line numbering is controlled via ~\par~. \LaTeX\ |
| 745 |
saved the \TeX-primitive ~\par~ in ~\@@par~. We push it |
| 746 |
one level further out, and redefine ~\@@par~ to insert |
| 747 |
the ~\interlinepenalty~ needed to trigger the |
| 748 |
line numbering. And we need to allow pagebreaks after a |
| 749 |
paragraph. |
| 750 |
|
| 751 |
New (2.05beta): the prevgraf test. A paragraph that ends with a |
| 752 |
displayed equation, a ~\noindent\par~ or ~wrapfig.sty~ produce empty |
| 753 |
paragraphs. These should not get a spurious line number via |
| 754 |
~\linenopenaltypar~. |
| 755 |
\begin{code}\begin{verbatim} |
| 756 |
\let\@@@par\@@par |
| 757 |
\newcount\linenoprevgraf |
| 758 |
\end{verbatim} |
| 759 |
\end{code} |
| 760 |
(UL) And needs ~\linenoprevgraf~ |
| 761 |
to be a counter? Perhaps there may be a paragraph having |
| 762 |
thousands of lines, so ~\mathchardef~ doesn't suffice (really??). |
| 763 |
A macro ending on ~\relax~ might suffice, but would be |
| 764 |
somewhat slow. I think I will use ~\mathchardef~ next time. |
| 765 |
Or has any user used ~\linenoprevgraf~? (/UL) |
| 766 |
\begin{code}\begin{verbatim} |
| 767 |
\def\linenumberpar{% |
| 768 |
\ifvmode \@@@par \else |
| 769 |
\ifinner \@@@par \else |
| 770 |
\xdef\@LN@outer@holdins{\the\holdinginserts}% v4.2 |
| 771 |
\advance \interlinepenalty \linenopenalty |
| 772 |
\linenoprevgraf \prevgraf |
| 773 |
\global \holdinginserts \thr@@ |
| 774 |
\@@@par |
| 775 |
\ifnum\prevgraf>\linenoprevgraf |
| 776 |
\penalty-\linenopenaltypar |
| 777 |
\fi |
| 778 |
\end{verbatim} |
| 779 |
\end{code} |
| 780 |
(New v4.00) |
| 781 |
\begin{old}\begin{verbatim} |
| 782 |
% \kern\z@ |
| 783 |
\end{verbatim} |
| 784 |
\end{old} |
| 785 |
was here previously. What for? |
| 786 |
According to \TeX book p._125, Stephan's |
| 787 |
interline penalty is changed into 10000. At the end of a |
| 788 |
paragraph, the ~\parskip~ would follow that penalty of 10000, |
| 789 |
so there could be a page break neither at the |
| 790 |
~\parskip~ nor at the ~\baselineskip~ (\TeX book p._110)---so |
| 791 |
there could never be a page break between two paragraphs. |
| 792 |
So something must screen off the 10000 penalty. |
| 793 |
Indeed, the ~\kern~ is a place to break. |
| 794 |
(Stephan once knew this: see `allow pagebreaks' above.) |
| 795 |
|
| 796 |
Formerly, I tried to replace ~\kern\z@~ by |
| 797 |
\begin{old}\begin{verbatim} |
| 798 |
% \penalty\@LN@parpgpen\relax |
| 799 |
\end{verbatim} |
| 800 |
\end{old} |
| 801 |
---but this allows a page break after heading. So: |
| 802 |
\begin{code}\begin{verbatim} |
| 803 |
\@LN@parpgbrk |
| 804 |
\end{verbatim} |
| 805 |
\end{code} |
| 806 |
|
| 807 |
These and similar changes were formerly done by ~linenox1.sty~. |
| 808 |
(/New v4.00) |
| 809 |
|
| 810 |
(New v4.4) |
| 811 |
A ~\belowdisplayskip~ may precede the previous when the paragraph |
| 812 |
ends on a display-math; or there may be a ~\topsep~ from a list, etc. |
| 813 |
~\addvspace~ couldn't take account for it with ~\kern\z@~ |
| 814 |
here. v4.32 therefore moved the space down -- with at least two |
| 815 |
bad consequences. |
| 816 |
Moreover, David Josef Dev observes that ~\kern\z@~ may |
| 817 |
inappropriately yield column depth 0pt. |
| 818 |
For these reasons, we introduce ~\@LN@screenoff@pen~ below. |
| 819 |
(/New v4.4) |
| 820 |
\begin{code}\begin{verbatim} |
| 821 |
\global\holdinginserts\@LN@outer@holdins % v4.2 |
| 822 |
\advance\interlinepenalty -\linenopenalty |
| 823 |
\fi % from \ifinner ... \else |
| 824 |
\fi} % from \ifvmode ... \else |
| 825 |
\end{verbatim} |
| 826 |
\end{code} |
| 827 |
(New v4.00, v4.4) Initialize ~\@LN@parpgbrk~, accounting |
| 828 |
for earlier space and for appropriate columndepth. |
| 829 |
We use former ~\MakeLineNo~'s depth-preverving trick |
| 830 |
~\@LN@depthbox~ again: |
| 831 |
\begin{code}\begin{verbatim} |
| 832 |
\def\@LN@screenoff@pen{% |
| 833 |
\ifdim\lastskip=\z@ |
| 834 |
\@tempdima\prevdepth \setbox\@tempboxa\null |
| 835 |
\@LN@depthbox \fi} |
| 836 |
|
| 837 |
\global\let\@LN@parpgbrk\@LN@screenoff@pen |
| 838 |
\end{verbatim} |
| 839 |
\end{code} |
| 840 |
(/New v4.4, v4.00) |
| 841 |
\subsection{% |
| 842 |
Turning on/off %% own subsec. v4.4. |
| 843 |
\unskip}\label{ss:OnOff} |
| 844 |
The basic commands to enable and disable line numbers. |
| 845 |
~\@par~ and ~\par~ are only touched, when they are ~\let~ |
| 846 |
to ~\@@@par~/~\linenumberpar~. The line number may be |
| 847 |
reset to 1 with the star-form, or set by an optional |
| 848 |
argument ~[~\angle{number}~]~. |
| 849 |
|
| 850 |
(New v4.00) We add ~\ifLineNumbers~ etc.\ since |
| 851 |
a number of our new adjustments need to know whether |
| 852 |
linenumbering is active. This just provides a kind of |
| 853 |
shorthand for ~\ifx\@@par\linenumberpar~; moreover it is |
| 854 |
more stable: who knows what may happen to ~\@@par~?---A |
| 855 |
caveat: ~\ifLineNumbers~ may be wrong. E.g., it may be |
| 856 |
~\iffalse~ where it acts, while a ~\linenumbers~ a few |
| 857 |
lines below---in the same paragraph---brings about that |
| 858 |
the line where the ~\ifLineNumbers~ appears gets a |
| 859 |
marginal number. |
| 860 |
(New v4.3) Just noticed: Such tricks have been |
| 861 |
disallowed with v4.11, see subsections_\ref{ss:LL} |
| 862 |
and_\ref{ss:OnOff}.---Moreover, the switching between |
| 863 |
meanings of ~\linelabel~ for a possible error message |
| 864 |
as of v4.11 is removed. Speed is difficult to esteem |
| 865 |
and also depends on applications. Just use the most |
| 866 |
simple code you find. (/New v4.3) |
| 867 |
\begin{code}\begin{verbatim} |
| 868 |
\newif\ifLineNumbers \LineNumbersfalse |
| 869 |
\end{verbatim} |
| 870 |
\end{code} |
| 871 |
(/New v4.00) |
| 872 |
\begin{code}\begin{verbatim} |
| 873 |
\def\linenumbers{% |
| 874 |
\LineNumberstrue % v4.00 |
| 875 |
\xdef\@LN@outer@holdins{\the\holdinginserts}% v4.3 |
| 876 |
\end{verbatim} |
| 877 |
\end{code} |
| 878 |
(New v4.3) The previous line is for ~{linenomath}~ |
| 879 |
in a first numbered paragraph. (/New v4.3) |
| 880 |
\begin{code}\begin{verbatim} |
| 881 |
\let\@@par\linenumberpar |
| 882 |
% \let\linelabel\@LN@linelabel % v4.11, removed v4.3 |
| 883 |
\ifx\@par\@@@par\let\@par\linenumberpar\fi |
| 884 |
\ifx\par\@@@par\let\par\linenumberpar\fi |
| 885 |
\@LN@maybe@moduloresume % v4.31 |
| 886 |
\@ifnextchar[{\resetlinenumber}%] |
| 887 |
{\@ifstar{\resetlinenumber}{}}% |
| 888 |
} |
| 889 |
|
| 890 |
\def\nolinenumbers{% |
| 891 |
\LineNumbersfalse % v4.00 |
| 892 |
\let\@@par\@@@par |
| 893 |
% \let\linelabel\@LN@LLerror % v4.11, removed v4.3 |
| 894 |
\ifx\@par\linenumberpar\let\@par\@@@par\fi |
| 895 |
\ifx\par\linenumberpar\let\par\@@@par\fi |
| 896 |
} |
| 897 |
\end{verbatim} |
| 898 |
\end{code} |
| 899 |
(New v4.00) Moreover, it is useful to switch to |
| 900 |
~\nolinenumbers~ in ~\@arrayparboxrestore~. We postpone this |
| 901 |
to section_\ref{ss:ReDef} where we'll have an appending macro |
| 902 |
for doing this. (/New v4.00) |
| 903 |
|
| 904 |
What happens with a display math? Since ~\par~ is not executed, |
| 905 |
when breaking the lines before a display, they will not get |
| 906 |
line numbers. Sorry, but I do not dare to change |
| 907 |
~\interlinepenalty~ globally, nor do I want to redefine |
| 908 |
the display math environments here. |
| 909 |
\begin{displaymath} |
| 910 |
display \ math |
| 911 |
\end{displaymath} |
| 912 |
See the subsection below, for a wrapper environment to make |
| 913 |
it work. But that requires to wrap each and every display |
| 914 |
in your \LaTeX\ source %%. |
| 915 |
(see option ~displaymath~ in subsections_\ref{ss:v3opts} |
| 916 |
and_\ref{ss:display} for some relief [UL]). |
| 917 |
|
| 918 |
The next two commands are provided to turn on line |
| 919 |
numbering in a specific mode. Please note the difference: |
| 920 |
for pagewise numbering, ~\linenumbers~ comes first to |
| 921 |
inhibit it from seeing optional arguments, since |
| 922 |
re-/presetting the counter is useless. |
| 923 |
\begin{code}\begin{verbatim} |
| 924 |
\def\pagewiselinenumbers{\linenumbers\setpagewiselinenumbers} |
| 925 |
\def\runninglinenumbers{\setrunninglinenumbers\linenumbers} |
| 926 |
\end{verbatim} |
| 927 |
\end{code} |
| 928 |
Finally, it is a \LaTeX\ style, so we provide for the use |
| 929 |
of environments, including the suppression of the |
| 930 |
following paragraph's indentation. |
| 931 |
|
| 932 |
(UL) I am drawing the following |
| 933 |
private thoughts of Stephan's to publicity so that others may |
| 934 |
think about them---or to remind myself of them in an efficient |
| 935 |
way. (/UL) |
| 936 |
\begin{old}\begin{verbatim} |
| 937 |
% TO DO: add \par to \linenumbers, if called from an environment. |
| 938 |
% To DO: add an \@endpe hack if \linenumbers are turned on |
| 939 |
% in horizontal mode. {\par\parskip\z@\noindent} or |
| 940 |
% something. |
| 941 |
\end{verbatim} |
| 942 |
\end{old} |
| 943 |
(UL) However, I rather think that ~\linenumbers~ and %% v4.31 |
| 944 |
~\nolinenumbers~ should execute a ~\par~ already. (Then the |
| 945 |
~\par~s in the following definitions should be removed.) (/UL) |
| 946 |
\begin{code}\begin{verbatim} |
| 947 |
\@namedef{linenumbers*}{\par\linenumbers*} |
| 948 |
\@namedef{runninglinenumbers*}{\par\runninglinenumbers*} |
| 949 |
|
| 950 |
\def\endlinenumbers{\par\@endpetrue} |
| 951 |
\let\endrunninglinenumbers\endlinenumbers |
| 952 |
\let\endpagewiselinenumbers\endlinenumbers |
| 953 |
\expandafter\let\csname endlinenumbers*\endcsname\endlinenumbers |
| 954 |
\expandafter\let\csname endrunninglinenumbers*\endcsname\endlinenumbers |
| 955 |
\let\endnolinenumbers\endlinenumbers |
| 956 |
\end{verbatim} |
| 957 |
\end{code} |
| 958 |
|
| 959 |
\subsection{% |
| 960 |
Display math |
| 961 |
\unskip}\label{ss:DM} |
| 962 |
|
| 963 |
Now we tackle the problem to get display math working. |
| 964 |
There are different options. |
| 965 |
\begin{enumerate}\item[ |
| 966 |
1.] Precede every display math with a ~\par~. |
| 967 |
Not too good. |
| 968 |
\item[ |
| 969 |
2.] Change ~\interlinepenalty~ and associates globally. |
| 970 |
Unstable. |
| 971 |
\item[ |
| 972 |
3.] Wrap each display math with a ~{linenomath}~ |
| 973 |
environment. |
| 974 |
\end{enumerate} |
| 975 |
We'll go for option 3. See if it works: |
| 976 |
\begin{linenomath} |
| 977 |
\begin{equation} |
| 978 |
display \ math |
| 979 |
\end{equation} |
| 980 |
\end{linenomath} |
| 981 |
The star form ~{linenomath*}~ should also number the lines |
| 982 |
of the display itself, |
| 983 |
\begin{linenomath*} |
| 984 |
\begin{eqnarray} |
| 985 |
multi && line \\ |
| 986 |
display && math \\ |
| 987 |
& |
| 988 |
\begin{array}{c} |
| 989 |
with \\ |
| 990 |
array |
| 991 |
\end{array} |
| 992 |
& |
| 993 |
\end{eqnarray} |
| 994 |
\end{linenomath*} |
| 995 |
including multline displays. |
| 996 |
|
| 997 |
First, here are two macros to turn |
| 998 |
on linenumbering on paragraphs preceeding displays, with |
| 999 |
numbering the lines of the display itself, or without. |
| 1000 |
The ~\ifx..~ tests if line numbering is turned on. It |
| 1001 |
does not harm to add these wrappers in sections that are |
| 1002 |
not numbered. Nor does it harm to wrap a display |
| 1003 |
twice, e.q, in case you have some ~{equation}~s wrapped |
| 1004 |
explicitely, and later you redefine ~\equation~ to do it |
| 1005 |
automatically. |
| 1006 |
|
| 1007 |
(New v4.3) To avoid the spurious line number above a |
| 1008 |
display in vmode, I insert ~\ifhmode~. (/New v4.3) |
| 1009 |
\begin{code}\begin{verbatim} |
| 1010 |
\newcommand\linenomathNonumbers{% |
| 1011 |
\ifLineNumbers |
| 1012 |
\ifnum\interlinepenalty>-\linenopenaltypar |
| 1013 |
\global\holdinginserts\thr@@ |
| 1014 |
\advance\interlinepenalty \linenopenalty |
| 1015 |
\ifhmode % v4.3 |
| 1016 |
\advance\predisplaypenalty \linenopenalty |
| 1017 |
\fi |
| 1018 |
\fi |
| 1019 |
\fi |
| 1020 |
\ignorespaces |
| 1021 |
} |
| 1022 |
|
| 1023 |
\newcommand\linenomathWithnumbers{% |
| 1024 |
\ifLineNumbers |
| 1025 |
\ifnum\interlinepenalty>-\linenopenaltypar |
| 1026 |
\global\holdinginserts\thr@@ |
| 1027 |
\advance\interlinepenalty \linenopenalty |
| 1028 |
\ifhmode % v4.3 |
| 1029 |
\advance\predisplaypenalty \linenopenalty |
| 1030 |
\fi |
| 1031 |
\advance\postdisplaypenalty \linenopenalty |
| 1032 |
\advance\interdisplaylinepenalty \linenopenalty |
| 1033 |
\fi |
| 1034 |
\fi |
| 1035 |
\ignorespaces |
| 1036 |
} |
| 1037 |
\end{verbatim} |
| 1038 |
\end{code} |
| 1039 |
The ~{linenomath}~ environment has two forms, with and |
| 1040 |
without a star. The following two macros define the |
| 1041 |
environment, where the stared/non-stared form does/doesn't number the |
| 1042 |
lines of the display or vice versa. |
| 1043 |
\begin{code}\begin{verbatim} |
| 1044 |
\newcommand\linenumberdisplaymath{% |
| 1045 |
\def\linenomath{\linenomathWithnumbers}% |
| 1046 |
\@namedef{linenomath*}{\linenomathNonumbers}% |
| 1047 |
} |
| 1048 |
|
| 1049 |
\newcommand\nolinenumberdisplaymath{% |
| 1050 |
\def\linenomath{\linenomathNonumbers}% |
| 1051 |
\@namedef{linenomath*}{\linenomathWithnumbers}% |
| 1052 |
} |
| 1053 |
|
| 1054 |
\def\endlinenomath{% |
| 1055 |
\ifLineNumbers % v4.3 |
| 1056 |
\global\holdinginserts\@LN@outer@holdins % v4.21 |
| 1057 |
\fi |
| 1058 |
\global % v4.21 support for LaTeX2e earlier than 1996/07/26. |
| 1059 |
\@ignoretrue |
| 1060 |
} |
| 1061 |
\expandafter\let\csname endlinenomath*\endcsname\endlinenomath |
| 1062 |
\end{verbatim} |
| 1063 |
\end{code} |
| 1064 |
The default is not to number the lines of a display. But |
| 1065 |
the package option ~mathlines~ may be used to switch |
| 1066 |
that behavior. |
| 1067 |
\begin{code}\begin{verbatim} |
| 1068 |
\nolinenumberdisplaymath |
| 1069 |
\end{verbatim} |
| 1070 |
\end{code} |
| 1071 |
|
| 1072 |
\section{% |
| 1073 |
Line number references |
| 1074 |
\unskip}\label{s:LNref} |
| 1075 |
\subsection{% |
| 1076 |
Internals %% New subsec. v4.3. |
| 1077 |
\unskip} |
| 1078 |
The only way to get a label to a line number in a |
| 1079 |
paragraph is to ask the output routine to mark it. |
| 1080 |
|
| 1081 |
(New v4.00) The following two paragraphs don't hold any |
| 1082 |
longer, see below. (/New v4.00) |
| 1083 |
\begin{old}\begin{verbatim} |
| 1084 |
% We use the marginpar mechanism to hook to ~\output~ for a |
| 1085 |
% second time. Marginpars are floats with number $-1$, we |
| 1086 |
% fake marginpars with No $-2$. Originally, every negative |
| 1087 |
% numbered float was considered to be a marginpar. |
| 1088 |
% |
| 1089 |
% The float box number ~\@currbox~ is used to transfer the |
| 1090 |
% label name in a macro called ~\@LNL@~\angle{box-number}. |
| 1091 |
\end{verbatim} |
| 1092 |
\end{old} |
| 1093 |
A ~\newlabel~ is written to the aux-file. The reference |
| 1094 |
is to ~\theLineNumber~, \emph{not} ~\thelinenumber~. |
| 1095 |
This allows to hook in, as done below for pagewise line |
| 1096 |
numbering. |
| 1097 |
|
| 1098 |
(New v3.03) The ~\@LN@ExtraLabelItems~ are added for a hook |
| 1099 |
to keep packages like ~{hyperref}~ happy. (/New v3.03) |
| 1100 |
|
| 1101 |
(New v4.00) |
| 1102 |
We fire the ~\marginpar~ mechanism, so we leave \LaTeX's |
| 1103 |
~\@addmarginpar~ untouched. |
| 1104 |
\begin{old}\begin{verbatim} |
| 1105 |
% \let\@LN@addmarginpar\@addmarginpar |
| 1106 |
% \def\@addmarginpar{% |
| 1107 |
% \ifnum\count\@currbox>-2\relax |
| 1108 |
% \expandafter\@LN@addmarginpar |
| 1109 |
% \else |
| 1110 |
% \@cons\@freelist\@currbox |
| 1111 |
% \protected@write\@auxout{}{% |
| 1112 |
% \string\newlabel |
| 1113 |
% {\csname @LNL@\the\@currbox\endcsname}% |
| 1114 |
% {{\theLineNumber}{\thepage}\@LN@ExtraLabelItems}}% |
| 1115 |
% \fi} |
| 1116 |
\end{verbatim} |
| 1117 |
\end{old} |
| 1118 |
OK, we keep Stephan's ~\@LN@ExtraLabelItems~: |
| 1119 |
(/New v4.00) |
| 1120 |
\begin{code}\begin{verbatim} |
| 1121 |
\let\@LN@ExtraLabelItems\@empty |
| 1122 |
\end{verbatim} |
| 1123 |
\end{code} |
| 1124 |
(New v4.00) |
| 1125 |
We imitate the ~\marginpar~ mechanism without using the |
| 1126 |
~\@freelist~ boxes. ~\linelabel~ will indeed place a signal |
| 1127 |
penalty (~\@Mllbcodepen~, new), and it will put a label into |
| 1128 |
some list macro ~\@LN@labellist~. A new part of the output |
| 1129 |
routine will take the labels from the list and will write |
| 1130 |
~\newlabel~s to the .aux file. |
| 1131 |
|
| 1132 |
The following is a version of \LaTeX's ~\@xnext~. |
| 1133 |
\begin{code}\begin{verbatim} |
| 1134 |
\def\@LN@xnext#1\@lt#2\@@#3#4{\def#3{#1}\gdef#4{#2}} |
| 1135 |
\end{verbatim} |
| 1136 |
\end{code} |
| 1137 |
This takes an item ~#1~ from a list ~#4~ into ~#3~; |
| 1138 |
to be used as ~\expandafter\@LN@xnext#4\@@#3#4~. |
| 1139 |
Our lists use ~\@lt~ after each item for separating. |
| 1140 |
Indeed, there will be another list macro which can |
| 1141 |
appear as argument ~#4~, this will be used for moving |
| 1142 |
~\vadjust~ items (section_\ref{ss:PVadj}). |
| 1143 |
The list for ~\linelabel~s is the following: |
| 1144 |
\begin{code}\begin{verbatim} |
| 1145 |
\global\let\@LN@labellist\@empty |
| 1146 |
\end{verbatim} |
| 1147 |
\end{code} |
| 1148 |
The next is the new part of the output routine writing the |
| 1149 |
~\newlabel~ to the .aux file. Since it is no real page output, |
| 1150 |
the page is put back to top of the main vertical list. |
| 1151 |
\begin{code}\begin{verbatim} |
| 1152 |
\def\WriteLineNo{% |
| 1153 |
\unvbox\@cclv |
| 1154 |
\expandafter \@LN@xnext \@LN@labellist \@@ |
| 1155 |
\@LN@label \@LN@labellist |
| 1156 |
\protected@write\@auxout{}{\string\newlabel{\@LN@label}% |
| 1157 |
{{\theLineNumber}{\thepage}\@LN@ExtraLabelItems}}% |
| 1158 |
} |
| 1159 |
\end{verbatim} |
| 1160 |
\end{code} |
| 1161 |
(/New v4.00) |
| 1162 |
|
| 1163 |
\subsection{% |
| 1164 |
The \scs{linelabel} command |
| 1165 |
\unskip}\label{ss:LL} |
| 1166 |
To refer to a place in line ~\ref{~\angle{foo}~}~ at page |
| 1167 |
~\pageref{~\angle{foo}~}~ you place a ~\linelabel{~\angle{foo}~}~ at |
| 1168 |
that place. |
| 1169 |
|
| 1170 |
\linelabel{demo} |
| 1171 |
\marginpar{\tiny\raggedright |
| 1172 |
See if it works: This paragraph |
| 1173 |
starts on page \pageref{demo}, line |
| 1174 |
\ref{demo}. |
| 1175 |
\unskip}% |
| 1176 |
(New v4.11) |
| 1177 |
\begin{old}\begin{verbatim} |
| 1178 |
% If you use this command outside a ~\linenumbers~ |
| 1179 |
% paragraph, you will get references to some bogus |
| 1180 |
% line numbers, sorry. But we don't disable the command, |
| 1181 |
% because only the ~\par~ at the end of a paragraph may |
| 1182 |
% decide whether to print line numbers on this paragraph |
| 1183 |
% or not. A ~\linelabel~ may legally appear earlier than |
| 1184 |
% ~\linenumbers~. |
| 1185 |
\end{verbatim} |
| 1186 |
\end{old} |
| 1187 |
This trick is better not allowed---see subsections_\ref{ss:LL} |
| 1188 |
and_\ref{ss:OnOff}. |
| 1189 |
(/New v4.11) |
| 1190 |
|
| 1191 |
~\linelabel~ |
| 1192 |
\begin{old}\begin{verbatim} |
| 1193 |
%, via a fake float number $-2$, %% new mechanism v4.00 |
| 1194 |
\end{verbatim} |
| 1195 |
\end{old} |
| 1196 |
puts a |
| 1197 |
~\penalty~ into a ~\vadjust~, which triggers the |
| 1198 |
pagebuilder after putting the current line to the main |
| 1199 |
vertical list. A ~\write~ is placed on the main vertical |
| 1200 |
list, which prints a reference to the current value of |
| 1201 |
~\thelinenumber~ and ~\thepage~ at the time of the |
| 1202 |
~\shipout~. |
| 1203 |
|
| 1204 |
A ~\linelabel~ is allowed only in outer horizontal mode. |
| 1205 |
In outer vertical mode we start a paragraph, and ignore |
| 1206 |
trailing spaces (by fooling ~\@esphack~). |
| 1207 |
|
| 1208 |
(New v4.00) We aim at relaxing the previous condition. |
| 1209 |
We insert a hook ~\@LN@mathhook~ and a shorthand |
| 1210 |
~\@LN@postlabel~ to support the ~mathrefs~ option which |
| 1211 |
allows ~\linelabel~ in math mode. |
| 1212 |
|
| 1213 |
The next paragraph is no longer valid. |
| 1214 |
\begin{old}\begin{verbatim} |
| 1215 |
% The argument of ~\linelabel~ is put into a macro with a |
| 1216 |
% name derived from the number of the allocated float box. |
| 1217 |
% Much of the rest is dummy float setup. |
| 1218 |
\end{verbatim} |
| 1219 |
\end{old} |
| 1220 |
(/New v4.00) |
| 1221 |
|
| 1222 |
(New v4.11) |
| 1223 |
\begin{old}\begin{verbatim} |
| 1224 |
% \def\linelabel#1{% |
| 1225 |
\end{verbatim} |
| 1226 |
\end{old} |
| 1227 |
I forgot ~\linenumbers~ today, costed me hours or so. |
| 1228 |
\begin{code}\begin{verbatim} |
| 1229 |
\def\@LN@LLerror{\PackageError{lineno}{% |
| 1230 |
\string\linelabel\space without \string\linenumbers}{% |
| 1231 |
Just see documentation. (New feature v4.11)}\@gobble} |
| 1232 |
\end{verbatim} |
| 1233 |
\end{code} |
| 1234 |
(New v4.3) Here some things have changed for v4.3. |
| 1235 |
The previous ~#1~ has been replaced by ~\@gobble~. |
| 1236 |
Ensuing, the ~\linelabel~ error message is re-implemented. |
| 1237 |
I find it difficult to compare efficiency of slight |
| 1238 |
alternatives---so choose an easy one. Explicit switching |
| 1239 |
in ~\linenumbers~ and ~\nolinenumbers~ is an additional |
| 1240 |
command that may better be avoided. |
| 1241 |
\begin{code}\begin{verbatim} |
| 1242 |
\newcommand\linelabel{% |
| 1243 |
\ifLineNumbers \expandafter \@LN@linelabel |
| 1244 |
\else \expandafter \@LN@LLerror \fi} |
| 1245 |
|
| 1246 |
\gdef\@LN@linelabel#1{% |
| 1247 |
\end{verbatim} |
| 1248 |
\end{code} |
| 1249 |
~\gdef~ for hyperref ``symbolically''. (/New v4.11) |
| 1250 |
\begin{code}\begin{verbatim} |
| 1251 |
\ifx\protect\@typeset@protect |
| 1252 |
\end{verbatim} |
| 1253 |
\end{code} |
| 1254 |
$\gets$ And a ~\linelabel~ should never be replicated in a |
| 1255 |
mark or a TOC entry. (/New v4.3) |
| 1256 |
\begin{code}\begin{verbatim} |
| 1257 |
\ifvmode |
| 1258 |
\ifinner \else |
| 1259 |
\leavevmode \@bsphack \@savsk\p@ |
| 1260 |
\fi |
| 1261 |
\else |
| 1262 |
\@bsphack |
| 1263 |
\fi |
| 1264 |
\ifhmode |
| 1265 |
\ifinner |
| 1266 |
\@parmoderr |
| 1267 |
\else |
| 1268 |
\end{verbatim} |
| 1269 |
\end{code} |
| 1270 |
(New v4.00) |
| 1271 |
\begin{code}\begin{verbatim} |
| 1272 |
\@LN@postlabel{#1}% |
| 1273 |
\end{verbatim} |
| 1274 |
\end{code} |
| 1275 |
\begin{old}\begin{verbatim} |
| 1276 |
% \@floatpenalty -\@Mii |
| 1277 |
% \@next\@currbox\@freelist |
| 1278 |
% {\global\count\@currbox-2% |
| 1279 |
% \expandafter\gdef\csname @LNL@\the\@currbox\endcsname{#1}}% |
| 1280 |
% {\@floatpenalty\z@ \@fltovf \def\@currbox{\@tempboxa}}% |
| 1281 |
% \begingroup |
| 1282 |
% \setbox\@currbox \color@vbox \vbox \bgroup \end@float |
| 1283 |
% \endgroup |
| 1284 |
% \@ignorefalse \@esphack |
| 1285 |
\end{verbatim} |
| 1286 |
\end{old} |
| 1287 |
(/New v4.00) |
| 1288 |
\begin{code}\begin{verbatim} |
| 1289 |
\@esphack |
| 1290 |
\end{verbatim} |
| 1291 |
\end{code} |
| 1292 |
(New v4.00) |
| 1293 |
The ~\@ignorefalse~ was appropriate before because the |
| 1294 |
~\@Esphack~ in ~\end@float~ set ~\@ignoretrue~. Cf._\LaTeX's |
| 1295 |
~\@xympar~. (/New v4.00) |
| 1296 |
\begin{code}\begin{verbatim} |
| 1297 |
\fi |
| 1298 |
\else |
| 1299 |
\end{verbatim} |
| 1300 |
\end{code} |
| 1301 |
(New v4.00) |
| 1302 |
\begin{code}\begin{verbatim} |
| 1303 |
\@LN@mathhook{#1}% |
| 1304 |
\end{verbatim} |
| 1305 |
\end{code} |
| 1306 |
\begin{old}\begin{verbatim} |
| 1307 |
% \@parmoderr |
| 1308 |
\end{verbatim} |
| 1309 |
\end{old} |
| 1310 |
Instead of complaining, you may just do your job. |
| 1311 |
(/New v4.00) |
| 1312 |
\begin{code}\begin{verbatim} |
| 1313 |
\fi |
| 1314 |
\fi |
| 1315 |
} |
| 1316 |
\end{verbatim} |
| 1317 |
\end{code} |
| 1318 |
(New v4.00) The shorthand just does what happened |
| 1319 |
with ~linenox0.sty~ before ~ednmath0.sty~ (New v4.1: |
| 1320 |
now ~mathrefs~ option) appeared, and |
| 1321 |
the hook is initialized to serve the same purpose. |
| 1322 |
So errors come just where Stephan had built them in, |
| 1323 |
and this is just the \LaTeX\ ~\marginpar~ behaviour. |
| 1324 |
\begin{code}\begin{verbatim} |
| 1325 |
\def\@LN@postlabel#1{\g@addto@macro\@LN@labellist{#1\@lt}% |
| 1326 |
\vadjust{\penalty-\@Mllbcodepen}} |
| 1327 |
\def\@LN@mathhook#1{\@parmoderr} |
| 1328 |
\end{verbatim} |
| 1329 |
\end{code} |
| 1330 |
(/New v4.00) |
| 1331 |
|
| 1332 |
\modulolinenumbers[3] |
| 1333 |
\firstlinenumber{1} |
| 1334 |
\section{% |
| 1335 |
The appearance of the line numbers |
| 1336 |
\unskip}\label{appearance} |
| 1337 |
\subsection{% |
| 1338 |
Basic code %% own subsec. v4.2. |
| 1339 |
\unskip} |
| 1340 |
|
| 1341 |
The line numbers are set as ~\tiny\sffamily\arabic{linenumber}~, |
| 1342 |
$10pt$ left of the text. With options to place it |
| 1343 |
right of the text, or . . . |
| 1344 |
|
| 1345 |
. . . here are the hooks: |
| 1346 |
\begin{code}\begin{verbatim} |
| 1347 |
\def\makeLineNumberLeft{% |
| 1348 |
\hss\linenumberfont\LineNumber\hskip\linenumbersep} |
| 1349 |
|
| 1350 |
\def\makeLineNumberRight{% |
| 1351 |
\linenumberfont\hskip\linenumbersep\hskip\columnwidth |
| 1352 |
\hb@xt@\linenumberwidth{\hss\LineNumber}\hss} |
| 1353 |
|
| 1354 |
\def\linenumberfont{\normalfont\tiny\sffamily} |
| 1355 |
|
| 1356 |
\newdimen\linenumbersep |
| 1357 |
\newdimen\linenumberwidth |
| 1358 |
|
| 1359 |
\linenumbersep=10pt |
| 1360 |
\linenumberwidth=10pt |
| 1361 |
\end{verbatim} |
| 1362 |
\end{code} |
| 1363 |
Margin switching requires ~pagewise~ numbering mode, but |
| 1364 |
choosing the left or right margin for the numbers always |
| 1365 |
works. |
| 1366 |
\begin{code}\begin{verbatim} |
| 1367 |
\def\switchlinenumbers{\@ifstar |
| 1368 |
{\let\makeLineNumberOdd\makeLineNumberRight |
| 1369 |
\let\makeLineNumberEven\makeLineNumberLeft}% |
| 1370 |
{\let\makeLineNumberOdd\makeLineNumberLeft |
| 1371 |
\let\makeLineNumberEven\makeLineNumberRight}% |
| 1372 |
} |
| 1373 |
|
| 1374 |
\def\setmakelinenumbers#1{\@ifstar |
| 1375 |
{\let\makeLineNumberRunning#1% |
| 1376 |
\let\makeLineNumberOdd#1% |
| 1377 |
\let\makeLineNumberEven#1}% |
| 1378 |
{\ifx\c@linenumber\c@runninglinenumber |
| 1379 |
\let\makeLineNumberRunning#1% |
| 1380 |
\else |
| 1381 |
\let\makeLineNumberOdd#1% |
| 1382 |
\let\makeLineNumberEven#1% |
| 1383 |
\fi}% |
| 1384 |
} |
| 1385 |
|
| 1386 |
\def\leftlinenumbers{\setmakelinenumbers\makeLineNumberLeft} |
| 1387 |
\def\rightlinenumbers{\setmakelinenumbers\makeLineNumberRight} |
| 1388 |
|
| 1389 |
\leftlinenumbers* |
| 1390 |
\end{verbatim} |
| 1391 |
\end{code} |
| 1392 |
~\LineNumber~ is a hook which is used for the modulo stuff. |
| 1393 |
It is the command to use for the line number, when you |
| 1394 |
customize ~\makeLineNumber~. Use ~\thelinenumber~ to |
| 1395 |
change the outfit of the digits. |
| 1396 |
|
| 1397 |
|
| 1398 |
We will implement two modes of operation: |
| 1399 |
\begin{itemize} |
| 1400 |
\item numbers ~running~ through (parts of) the text |
| 1401 |
\item ~pagewise~ numbers starting over with one on top of |
| 1402 |
each page. |
| 1403 |
\end{itemize} |
| 1404 |
Both modes have their own count register, but only one is |
| 1405 |
allocated as a \LaTeX\ counter, with the attached |
| 1406 |
facilities serving both. |
| 1407 |
\begin{code}\begin{verbatim} |
| 1408 |
\newcounter{linenumber} |
| 1409 |
\newcount\c@pagewiselinenumber |
| 1410 |
\let\c@runninglinenumber\c@linenumber |
| 1411 |
\end{verbatim} |
| 1412 |
\end{code} |
| 1413 |
Only the running mode counter may be reset, or preset, |
| 1414 |
for individual paragraphs. The pagewise counter must |
| 1415 |
give a unique anonymous number for each line. |
| 1416 |
|
| 1417 |
(New v4.3) ~\newcounter{linenumber}~ |
| 1418 |
was the only ~\newcounter~ in the whole package, and |
| 1419 |
formerly I was near using ~\newcount~ instead. Yet |
| 1420 |
~\newcounter~ may be quite useful for ~\includeonly~. |
| 1421 |
It also supports resetting ``subcounters'', but what |
| 1422 |
could these be? Well, ~edtable~ might introduce a |
| 1423 |
subcounter for columns. |
| 1424 |
(Note that \LaTeX's setting commands would work with |
| 1425 |
~\newcount\c@linenumber~ already, apart from this. |
| 1426 |
And perhaps sometimes ~\refstepcounter{linenumber}~ |
| 1427 |
wouldn't work---cf._my discussion of ~\stepcounter~ in |
| 1428 |
subsection_\ref{ss:MLN}, similarly ~\refstep...~ would |
| 1429 |
be quite useless. |
| 1430 |
Even the usual redefinitions of ~\thelinenumber~ would |
| 1431 |
work. It is nice, on the other hand, that |
| 1432 |
~\thelinenumber~ is predefined here. \LaTeX's |
| 1433 |
initialization of the value perhaps just serves making |
| 1434 |
clear \LaTeX\ counters should always be changed |
| 1435 |
globally.---Shortened and improved the discussion here.) |
| 1436 |
(/New v4.3) |
| 1437 |
|
| 1438 |
(New v4.22) |
| 1439 |
~\c@linenumber~ usually is---globally---incremented by |
| 1440 |
~\stepcounter~ (at present), so resetting it locally would |
| 1441 |
raise the save stack problem of \TeX book p._301, moreover |
| 1442 |
it would be is useless, there is no hope of keeping the |
| 1443 |
values local (but see subsection_\ref{ss:ILN}). So I insert |
| 1444 |
~\global~: (/New v4.22) |
| 1445 |
\begin{code}\begin{verbatim} |
| 1446 |
\newcommand*\resetlinenumber[1][\@ne]{% |
| 1447 |
\global % v4.22 |
| 1448 |
\c@runninglinenumber#1\relax} |
| 1449 |
\end{verbatim} |
| 1450 |
\end{code} |
| 1451 |
(New v4.00) |
| 1452 |
\begin{old}\begin{verbatim} |
| 1453 |
% \newcommand\resetlinenumber[1][1]{\c@runninglinenumber#1} |
| 1454 |
\end{verbatim} |
| 1455 |
\end{old} |
| 1456 |
Added ~\relax~, being quite sure that this does no harm |
| 1457 |
and is quite important, as with ~\setcounter~ etc. |
| 1458 |
I consider this a bug fix (although perhaps no user has |
| 1459 |
ever had a problem with this). (/New v4.00) |
| 1460 |
|
| 1461 |
(v4.22: I had made much fuss about resetting subordinate |
| 1462 |
counters here---removed, somewhat postponed.) |
| 1463 |
|
| 1464 |
|
| 1465 |
\subsection{% |
| 1466 |
Running line numbers |
| 1467 |
\unskip} |
| 1468 |
|
| 1469 |
Running mode is easy, ~\LineNumber~ and ~\theLineNumber~ |
| 1470 |
produce ~\thelinenumber~, which defaults to |
| 1471 |
~\arabic{linenumber}~, using the ~\c@runninglinenumber~ |
| 1472 |
counter. This is the default mode of operation. |
| 1473 |
\begin{code}\begin{verbatim} |
| 1474 |
\def\makeRunningLineNumber{\makeLineNumberRunning} |
| 1475 |
|
| 1476 |
\def\setrunninglinenumbers{% |
| 1477 |
\def\theLineNumber{\thelinenumber}% |
| 1478 |
\let\c@linenumber\c@runninglinenumber |
| 1479 |
\let\makeLineNumber\makeRunningLineNumber |
| 1480 |
} |
| 1481 |
|
| 1482 |
\setrunninglinenumbers\resetlinenumber |
| 1483 |
\end{verbatim} |
| 1484 |
\end{code} |
| 1485 |
|
| 1486 |
\subsection{% |
| 1487 |
Pagewise line numbers |
| 1488 |
\unskip}\label{ss:PW} |
| 1489 |
|
| 1490 |
Difficult, if you think about it. The number has to be |
| 1491 |
printed when there is no means to know on which page it |
| 1492 |
will end up, except through the aux-file. My solution |
| 1493 |
is really expensive, but quite robust. |
| 1494 |
|
| 1495 |
With version ~v2.00~ the hashsize requirements are |
| 1496 |
reduced, because we do not need one controlsequence for |
| 1497 |
each line any more. But this costs some computation time |
| 1498 |
to find out on which page we are. |
| 1499 |
|
| 1500 |
~\makeLineNumber~ gets a hook to log the line and page |
| 1501 |
number to the aux-file. Another hook tries to find out |
| 1502 |
what the page offset is, and subtracts it from the counter |
| 1503 |
~\c@linenumber~. Additionally, the switch |
| 1504 |
~\ifoddNumberedPage~ is set true for odd numbered pages, |
| 1505 |
false otherwise. |
| 1506 |
\begin{code}\begin{verbatim} |
| 1507 |
\def\setpagewiselinenumbers{% |
| 1508 |
\let\theLineNumber\thePagewiseLineNumber |
| 1509 |
\let\c@linenumber\c@pagewiselinenumber |
| 1510 |
\let\makeLineNumber\makePagewiseLineNumber |
| 1511 |
} |
| 1512 |
|
| 1513 |
\def\makePagewiseLineNumber{\logtheLineNumber\getLineNumber |
| 1514 |
\ifoddNumberedPage |
| 1515 |
\makeLineNumberOdd |
| 1516 |
\else |
| 1517 |
\makeLineNumberEven |
| 1518 |
\fi |
| 1519 |
} |
| 1520 |
\end{verbatim} |
| 1521 |
\end{code} |
| 1522 |
Each numbered line gives a line to the aux file |
| 1523 |
\begin{verse} |
| 1524 |
~\@LN{~\angle{line}~}{~\angle{page}~}~ |
| 1525 |
\end{verse} |
| 1526 |
very similar to the ~\newlabel~ business, except that we need |
| 1527 |
an arabic representation of the page number, not what |
| 1528 |
there might else be in ~\thepage~. |
| 1529 |
\begin{code}\begin{verbatim} |
| 1530 |
\def\logtheLineNumber{\protected@write\@auxout{}{% |
| 1531 |
\end{verbatim} |
| 1532 |
\end{code} |
| 1533 |
(New v4.00) (UL) |
| 1534 |
As Daniel Doherty observed, the earlier line |
| 1535 |
\begin{old}\begin{verbatim} |
| 1536 |
% \string\@LN{\the\c@linenumber}{\noexpand\the\c@page}}} |
| 1537 |
\end{verbatim} |
| 1538 |
\end{old} |
| 1539 |
here may lead into an infinite loop when the user resets |
| 1540 |
the page number (think of ~\pagenumbering~, e.g.). |
| 1541 |
Stephan and I brief\/ly discussed the matter and decided |
| 1542 |
to introduce a ``physical''-page counter to which |
| 1543 |
~\logtheLineNumber~ refers. It was Stephan's idea to use |
| 1544 |
~\cl@page~ for reliably augmenting the ``physical''-page |
| 1545 |
counter. However, this relies on the output routine once |
| 1546 |
doing ~\stepcounter{page}~. Before Stephan's |
| 1547 |
suggestion, I had thought of appending the stepping to |
| 1548 |
\LaTeX's ~\@outputpage~.---So the macro definition ends |
| 1549 |
as follows. |
| 1550 |
\begin{code}\begin{verbatim} |
| 1551 |
\string\@LN{\the\c@linenumber}{% |
| 1552 |
\end{verbatim} |
| 1553 |
\end{code} |
| 1554 |
(New v4.2) |
| 1555 |
The `truepage' counter must start with ~\c@~ so it works |
| 1556 |
with ~\include~, and the ~\@addtoreset~ below is needed |
| 1557 |
for the same purpose. |
| 1558 |
\begin{code}\begin{verbatim} |
| 1559 |
\noexpand\the\c@LN@truepage}}} |
| 1560 |
|
| 1561 |
\newcount\c@LN@truepage |
| 1562 |
\g@addto@macro\cl@page{\global\advance\c@LN@truepage\@ne} |
| 1563 |
\@addtoreset{LN@truepage}{@ckpt} |
| 1564 |
\end{verbatim} |
| 1565 |
\end{code} |
| 1566 |
(/New v4.2) I had thought of offering more |
| 1567 |
features of a \LaTeX\ counter. However, the user should |
| 1568 |
better \emph{not} have access to this counter. ~\c@page~ |
| 1569 |
should suffice as a pagewise master counter.---To be sure, |
| 1570 |
along the present lines the user \emph{can} manipulate |
| 1571 |
~\c@LN@truepage~ by ~\stepcounter{page}~. E.g., she might |
| 1572 |
do this in order to manually insert a photograph. Well, |
| 1573 |
seems not to harm. |
| 1574 |
|
| 1575 |
The above usage of ~\g@addto@macro~ and ~\cl@page~ may be |
| 1576 |
not as stable as Stephan intended. His proposal used |
| 1577 |
~\xdef~ directly. But he used ~\cl@page~ as well, and who |
| 1578 |
knows \dots{} And as to ~\g@addto@macro~, I have introduced |
| 1579 |
it for list macros anyway. |
| 1580 |
(/UL) (/New v4.00) |
| 1581 |
|
| 1582 |
From the aux-file we get one macro ~\LN@P~\angle{page} for each |
| 1583 |
page with line numbers on it. This macro calls four other |
| 1584 |
macros with one argument each. These macros are |
| 1585 |
dynamically defined to do tests and actions, to find out |
| 1586 |
on which page the current line number is located. |
| 1587 |
|
| 1588 |
We need sort of a pointer to the first page with line |
| 1589 |
numbers, initiallized to point to nothing: |
| 1590 |
\begin{code}\begin{verbatim} |
| 1591 |
\def\LastNumberedPage{first} |
| 1592 |
\def\LN@Pfirst{\nextLN\relax} |
| 1593 |
\end{verbatim} |
| 1594 |
\end{code} |
| 1595 |
The four dynamic macros are initiallized to reproduce |
| 1596 |
themselves in an ~\xdef~ |
| 1597 |
\begin{code}\begin{verbatim} |
| 1598 |
\let\lastLN\relax % compare to last line on this page |
| 1599 |
\let\firstLN\relax % compare to first line on this page |
| 1600 |
\let\pageLN\relax % get the page number, compute the line number |
| 1601 |
\let\nextLN\relax % move to the next page |
| 1602 |
\end{verbatim} |
| 1603 |
\end{code} |
| 1604 |
During the end-document run through the aux-files, we |
| 1605 |
disable ~\@LN~. I may put in a check here later, to give |
| 1606 |
a rerun recommendation. |
| 1607 |
\begin{code}\begin{verbatim} |
| 1608 |
\AtEndDocument{\let\@LN\@gobbletwo} |
| 1609 |
\end{verbatim} |
| 1610 |
\end{code} |
| 1611 |
Now, this is the tricky part. First of all, the whole |
| 1612 |
definition of ~\@LN~ is grouped, to avoid accumulation |
| 1613 |
on the save stack. Somehow ~\csname~\angle{cs}~\endcsname~ pushes |
| 1614 |
an entry, which stays after an ~\xdef~ to that \angle{cs}. |
| 1615 |
|
| 1616 |
If ~\LN@P~\angle{page} is undefined, initialize it with the |
| 1617 |
current page and line number, with the |
| 1618 |
\emph{pointer-to-the-next-page} pointing to nothing. And |
| 1619 |
the macro for the previous page will be redefined to point |
| 1620 |
to the current one. |
| 1621 |
|
| 1622 |
If the macro for the current page already exists, just |
| 1623 |
redefine the \emph{last-line-number} entry. |
| 1624 |
|
| 1625 |
Finally, save the current page number, to get the pointer to the |
| 1626 |
following page later. |
| 1627 |
\begin{code}\begin{verbatim} |
| 1628 |
\def\@LN#1#2{{\expandafter\@@LN |
| 1629 |
\csname LN@P#2C\@LN@column\expandafter\endcsname |
| 1630 |
\csname LN@PO#2\endcsname |
| 1631 |
{#1}{#2}}} |
| 1632 |
|
| 1633 |
\def\@@LN#1#2#3#4{\ifx#1\relax |
| 1634 |
\ifx#2\relax\gdef#2{#3}\fi |
| 1635 |
\expandafter\@@@LN\csname LN@P\LastNumberedPage\endcsname#1% |
| 1636 |
\xdef#1{\lastLN{#3}\firstLN{#3}% |
| 1637 |
\pageLN{#4}{\@LN@column}{#2}\nextLN\relax}% |
| 1638 |
\else |
| 1639 |
\def\lastLN##1{\noexpand\lastLN{#3}}% |
| 1640 |
\xdef#1{#1}% |
| 1641 |
\fi |
| 1642 |
\xdef\LastNumberedPage{#4C\@LN@column}} |
| 1643 |
\end{verbatim} |
| 1644 |
\end{code} |
| 1645 |
The previous page macro gets its pointer to the |
| 1646 |
current one, replacing the ~\relax~ with the cs-token |
| 1647 |
~\LN@P~\angle{page}. |
| 1648 |
\begin{code}\begin{verbatim} |
| 1649 |
\def\@@@LN#1#2{{\def\nextLN##1{\noexpand\nextLN\noexpand#2}% |
| 1650 |
\xdef#1{#1}}} |
| 1651 |
\end{verbatim} |
| 1652 |
\end{code} |
| 1653 |
Now, to print a line number, we need to find the page, |
| 1654 |
where it resides. This will most probably be the page where |
| 1655 |
the last one came from, or maybe the next page. However, it can |
| 1656 |
be a completely different one. We maintain a cache, |
| 1657 |
which is ~\let~ to the last page's macro. But for now |
| 1658 |
it is initialized to expand ~\LN@first~, where the poiner |
| 1659 |
to the first numbered page has been stored in. |
| 1660 |
\begin{code}\begin{verbatim} |
| 1661 |
\def\NumberedPageCache{\LN@Pfirst} |
| 1662 |
\end{verbatim} |
| 1663 |
\end{code} |
| 1664 |
To find out on which page the current ~\c@linenumber~ is, |
| 1665 |
we define the four dynamic macros to do something usefull |
| 1666 |
and execute the current cache macro. ~\lastLN~ is run |
| 1667 |
first, testing if the line number in question may be on a |
| 1668 |
later page. If so, disable ~\firstLN~, and go on to the |
| 1669 |
next page via ~\nextLN~. |
| 1670 |
\begin{code}\begin{verbatim} |
| 1671 |
\def\testLastNumberedPage#1{\ifnum#1<\c@linenumber |
| 1672 |
\let\firstLN\@gobble |
| 1673 |
\fi} |
| 1674 |
\end{verbatim} |
| 1675 |
\end{code} |
| 1676 |
Else, if ~\firstLN~ finds out that we need an earlier |
| 1677 |
page, we start over from the beginning. Else, ~\nextLN~ |
| 1678 |
will be disabled, and ~\pageLN~ will run |
| 1679 |
~\gotNumberedPage~ with four arguments: the first line |
| 1680 |
number on this column, the page number, the column |
| 1681 |
number, and the first line on the page. |
| 1682 |
\begin{code}\begin{verbatim} |
| 1683 |
\def\testFirstNumberedPage#1{\ifnum#1>\c@linenumber |
| 1684 |
\def\nextLN##1{\testNextNumberedPage\LN@Pfirst}% |
| 1685 |
\else |
| 1686 |
\let\nextLN\@gobble |
| 1687 |
\def\pageLN{\gotNumberedPage{#1}}% |
| 1688 |
\fi} |
| 1689 |
\end{verbatim} |
| 1690 |
\end{code} |
| 1691 |
We start with ~\pageLN~ disabled and ~\nextLN~ defined to |
| 1692 |
continue the search with the next page. |
| 1693 |
\begin{code}\begin{verbatim} |
| 1694 |
\long\def \@gobblethree #1#2#3{} |
| 1695 |
|
| 1696 |
\def\testNumberedPage{% |
| 1697 |
\let\lastLN\testLastNumberedPage |
| 1698 |
\let\firstLN\testFirstNumberedPage |
| 1699 |
\let\pageLN\@gobblethree |
| 1700 |
\let\nextLN\testNextNumberedPage |
| 1701 |
\NumberedPageCache |
| 1702 |
} |
| 1703 |
\end{verbatim} |
| 1704 |
\end{code} |
| 1705 |
When we switch to another page, we first have to make |
| 1706 |
sure that it is there. If we are done with the last |
| 1707 |
page, we probably need to run \TeX\ again, but for the |
| 1708 |
rest of this run, the cache macro will just return four |
| 1709 |
zeros. This saves a lot of time, for example if you have |
| 1710 |
half of an aux-file from an aborted run, in the next run |
| 1711 |
the whole page-list would be searched in vain again and |
| 1712 |
again for the second half of the document. |
| 1713 |
|
| 1714 |
If there is another page, we iterate the search. |
| 1715 |
\begin{code}\begin{verbatim} |
| 1716 |
\def\testNextNumberedPage#1{\ifx#1\relax |
| 1717 |
\global\def\NumberedPageCache{\gotNumberedPage0000}% |
| 1718 |
\PackageWarningNoLine{lineno}% |
| 1719 |
{Line number reference failed, re-run to get it right}% |
| 1720 |
\else |
| 1721 |
\global\let\NumberedPageCache#1% |
| 1722 |
\fi |
| 1723 |
\testNumberedPage |
| 1724 |
} |
| 1725 |
\end{verbatim} |
| 1726 |
\end{code} |
| 1727 |
\linelabel{demo2} |
| 1728 |
\marginpar{\tiny\raggedright |
| 1729 |
Let's see if it finds the label |
| 1730 |
on page \pageref{demo}, |
| 1731 |
line \ref{demo}, and back here |
| 1732 |
on page \pageref{demo2}, line |
| 1733 |
\ref{demo2}. |
| 1734 |
\unskip}% |
| 1735 |
To separate the official hooks from the internals there is |
| 1736 |
this equivalence, to hook in later for whatever purpose: |
| 1737 |
\begin{code}\begin{verbatim} |
| 1738 |
\let\getLineNumber\testNumberedPage |
| 1739 |
\end{verbatim} |
| 1740 |
\end{code} |
| 1741 |
So, now we got the page where the number is on. We |
| 1742 |
establish if we are on an odd or even page, and calculate |
| 1743 |
the final line number to be printed. |
| 1744 |
\begin{code}\begin{verbatim} |
| 1745 |
\newif\ifoddNumberedPage |
| 1746 |
\newif\ifcolumnwiselinenumbers |
| 1747 |
\columnwiselinenumbersfalse |
| 1748 |
|
| 1749 |
\def\gotNumberedPage#1#2#3#4{\oddNumberedPagefalse |
| 1750 |
\ifodd \if@twocolumn #3\else #2\fi\relax\oddNumberedPagetrue\fi |
| 1751 |
\advance\c@linenumber\@ne |
| 1752 |
\ifcolumnwiselinenumbers |
| 1753 |
\subtractlinenumberoffset{#1}% |
| 1754 |
\else |
| 1755 |
\subtractlinenumberoffset{#4}% |
| 1756 |
\fi |
| 1757 |
} |
| 1758 |
\end{verbatim} |
| 1759 |
\end{code} |
| 1760 |
You might want to run the pagewise mode with running line |
| 1761 |
numbers, or you might not. It's your choice: |
| 1762 |
\begin{code}\begin{verbatim} |
| 1763 |
\def\runningpagewiselinenumbers{% |
| 1764 |
\let\subtractlinenumberoffset\@gobble |
| 1765 |
} |
| 1766 |
|
| 1767 |
\def\realpagewiselinenumbers{% |
| 1768 |
\def\subtractlinenumberoffset##1{\advance\c@linenumber-##1\relax}% |
| 1769 |
} |
| 1770 |
|
| 1771 |
\realpagewiselinenumbers |
| 1772 |
\end{verbatim} |
| 1773 |
\end{code} |
| 1774 |
For line number references, we need a protected call to |
| 1775 |
the whole procedure, with the requested line number stored |
| 1776 |
in the ~\c@linenumber~ counter. This is what gets printed |
| 1777 |
to the aux-file to make a label: |
| 1778 |
\begin{code}\begin{verbatim} |
| 1779 |
\def\thePagewiseLineNumber{\protect |
| 1780 |
\getpagewiselinenumber{\the\c@linenumber}}% |
| 1781 |
\end{verbatim} |
| 1782 |
\end{code} |
| 1783 |
And here is what happens when the label is refered to: |
| 1784 |
\begin{code}\begin{verbatim} |
| 1785 |
\def\getpagewiselinenumber#1{{% |
| 1786 |
\c@linenumber #1\relax\testNumberedPage |
| 1787 |
\thelinenumber |
| 1788 |
}} |
| 1789 |
\end{verbatim} |
| 1790 |
\end{code} |
| 1791 |
% |
| 1792 |
A summary of all per line expenses: |
| 1793 |
\begin{description}\item |
| 1794 |
[CPU:] The ~\output~ routine is called for each line, |
| 1795 |
and the page-search is done. |
| 1796 |
\item |
| 1797 |
[DISK:] One line of output to the aux-file for each |
| 1798 |
numbered line |
| 1799 |
\item |
| 1800 |
[MEM:] One macro per page. Great improvement over v1.02, |
| 1801 |
which had one control sequence per line in |
| 1802 |
addition. It blew the hash table after some five |
| 1803 |
thousand lines. |
| 1804 |
\end{description} |
| 1805 |
|
| 1806 |
\subsection{% |
| 1807 |
Twocolumn mode (New v3.06) |
| 1808 |
\unskip} |
| 1809 |
|
| 1810 |
Twocolumn mode requires another patch to the ~\output~ |
| 1811 |
routine, in order to print a column tag to the .aux |
| 1812 |
file. |
| 1813 |
\begin{code}\begin{verbatim} |
| 1814 |
\AtBeginDocument{% v4.2, revtex4.cls (e.g.). |
| 1815 |
% <- TODO v4.4+: Or better in \LineNoLaTeXOutput!? |
| 1816 |
\let\@LN@orig@makecol\@makecol |
| 1817 |
} |
| 1818 |
|
| 1819 |
\def\@LN@makecol{% |
| 1820 |
\@LN@orig@makecol |
| 1821 |
\setbox\@outputbox \vbox{% |
| 1822 |
\boxmaxdepth \@maxdepth |
| 1823 |
\protected@write\@auxout{}{% |
| 1824 |
\string\def\string\@LN@column{\if@firstcolumn1\else2\fi}% |
| 1825 |
}% |
| 1826 |
\box\@outputbox |
| 1827 |
}% \vbox |
| 1828 |
} %% TODO cf. revtexln.sty. |
| 1829 |
|
| 1830 |
\def\@LN@column{1} |
| 1831 |
|
| 1832 |
\ifdefined\IfFormatAtLeastTF |
| 1833 |
\IfFormatAtLeastTF{2025-06-01}{ |
| 1834 |
% disable \@LN@makecol |
| 1835 |
\AtBeginDocument{% |
| 1836 |
\let\@LN@makecol\@makecol |
| 1837 |
} |
| 1838 |
% replace with hook code |
| 1839 |
\AddToHook{build/column/after}{% |
| 1840 |
\setbox\@outputbox \vbox to\ht\@outputbox{% |
| 1841 |
\boxmaxdepth \@maxdepth |
| 1842 |
\protected@write\@auxout{}{% |
| 1843 |
\string\def\string\@LN@column{\if@firstcolumn1\else2\fi}% |
| 1844 |
}% |
| 1845 |
\unvbox\@outputbox |
| 1846 |
}% |
| 1847 |
} |
| 1848 |
}{ |
| 1849 |
} |
| 1850 |
\fi |
| 1851 |
\end{verbatim} |
| 1852 |
\end{code} |
| 1853 |
|
| 1854 |
\subsection{% |
| 1855 |
Numbering modulo $m$, starting at $f$ |
| 1856 |
\unskip}\label{ss:Mod} |
| 1857 |
|
| 1858 |
Most users want to have only one in five lines numbered. |
| 1859 |
~\LineNumber~ is supposed to produce the outfit of the |
| 1860 |
line number attached to the line, while ~\thelinenumber~ |
| 1861 |
is used also for references, which should appear even if |
| 1862 |
they are not multiples of five. |
| 1863 |
|
| 1864 |
(New v4.00) Moreover, some users want to |
| 1865 |
control which line number should be printed first. Support |
| 1866 |
of this is now introduced here---see ~\firstlinenumber~ |
| 1867 |
below.---~numline.sty~ by Michael Jaegermann and |
| 1868 |
James Fortune offers controlling which \emph{final} |
| 1869 |
line numbers should not be printed. What is |
| 1870 |
it good for? We ignore this here until some user demands |
| 1871 |
it.---Peter Wilson's ~ledmac.sty~ offers much different |
| 1872 |
choices of line numbers to be printed, due to Wayne Sullivan. |
| 1873 |
(/New v4.00) |
| 1874 |
|
| 1875 |
(New v4.22) ~\c@linenumbermodulo~ is rendered a |
| 1876 |
fake counter, as discussed since v4.00. So it can |
| 1877 |
no longer be set by ~\setcounter~. ~\modulolinenumbers~ |
| 1878 |
serves this purpose. Well, does anybody want to do |
| 1879 |
what worked with ~\addtocounter~? (Then please tell |
| 1880 |
me.)---At least, ~\value~ still works. For the same |
| 1881 |
purpose I rename the fake `firstlinenumber' counter |
| 1882 |
~\n@...~ to ~\c@...~. (/New v4.22) |
| 1883 |
\begin{old}\begin{verbatim} |
| 1884 |
% \newcount\c@linenumbermodulo % removed for v4.22 |
| 1885 |
\end{verbatim} |
| 1886 |
\end{old} |
| 1887 |
|
| 1888 |
|
| 1889 |
(New v4.00) \par |
| 1890 |
~\themodulolinenumber~ waits for being declared |
| 1891 |
~\LineNumber~ by ~\modulolinenumbers~. (This has |
| 1892 |
been so before, no change.) Here is how it |
| 1893 |
looked before: |
| 1894 |
\begin{old}\begin{verbatim} |
| 1895 |
% \def\themodulolinenumber{{\@tempcnta\c@linenumber |
| 1896 |
% \divide\@tempcnta\c@linenumbermodulo |
| 1897 |
% \multiply\@tempcnta\c@linenumbermodulo |
| 1898 |
% \ifnum\@tempcnta=\c@linenumber\thelinenumber\fi |
| 1899 |
% }} |
| 1900 |
\end{verbatim} |
| 1901 |
\end{old} |
| 1902 |
(UL) This was somewhat slow. This arithmetic |
| 1903 |
happens at every line. This time I tend to declare an extra |
| 1904 |
line counter (as opposed to my usual recommendations to use |
| 1905 |
counters as rarely as possible) which is stepped every line. |
| 1906 |
It could be incremented in the same way as ~\c@LN@truepage~ |
| 1907 |
is incremented via ~\cl@page~! This is another point in favour |
| 1908 |
of ~{linenumber}~ being a \LaTeX\ counter! |
| 1909 |
When this new counter equals ~\c@linenumbermodulo~, it is reset, |
| 1910 |
and ~\thelinenumber~ is executed.---It gets much slower by my |
| 1911 |
support of controlling the first line number below. I should |
| 1912 |
improve this.---On |
| 1913 |
the other hand, time expense means very little nowadays, |
| 1914 |
while the number of \TeX\ counters still is limited. |
| 1915 |
|
| 1916 |
For the same purpose, moreover, attaching the line number |
| 1917 |
box could be intercepted earlier (in ~\MakeLineNo~), |
| 1918 |
without changing ~\LineNumber~. However, this may be |
| 1919 |
bad for the latter's announcement as a wizard interface |
| 1920 |
in section_\ref{s:UserCmds}. |
| 1921 |
(/UL) |
| 1922 |
|
| 1923 |
Here is the new code. It is very near to my ~lnopatch.sty~ |
| 1924 |
code which introduced the first line number feature |
| 1925 |
before.---I add starting with a ~\relax~ which is so often |
| 1926 |
recommended---without understanding this really. At least, |
| 1927 |
it will not harm.---Former group braces appear as |
| 1928 |
~\begingroup~/~\endgroup~ here. |
| 1929 |
\begin{code}\begin{verbatim} |
| 1930 |
\def\themodulolinenumber{\relax |
| 1931 |
\ifnum\c@linenumber<\c@firstlinenumber \else |
| 1932 |
\begingroup |
| 1933 |
\@tempcnta\c@linenumber |
| 1934 |
\advance\@tempcnta-\c@firstlinenumber |
| 1935 |
\divide\@tempcnta\c@linenumbermodulo |
| 1936 |
\multiply\@tempcnta\c@linenumbermodulo |
| 1937 |
\advance\@tempcnta\c@firstlinenumber |
| 1938 |
\ifnum\@tempcnta=\c@linenumber \thelinenumber \fi |
| 1939 |
\endgroup |
| 1940 |
\fi |
| 1941 |
} |
| 1942 |
\end{verbatim} |
| 1943 |
\end{code} |
| 1944 |
(/New v4.00) |
| 1945 |
|
| 1946 |
The user command to set the modulo counter: |
| 1947 |
(New v4.31) \dots\ a star variant is introduced to implement |
| 1948 |
Hillel Chayim Yisraeli's idea to print the first line number |
| 1949 |
after an interruption of the edited text by some editor's |
| 1950 |
text, regardless of the modulo. If it is 1, it is printed only |
| 1951 |
with ~\firstlinenumber{1}~. I.e., you use ~\modulolinenumbers*~ |
| 1952 |
for the new feature, without the star you get the simpler |
| 1953 |
behaviour that we have had so far. And you can switch back |
| 1954 |
from the refined behaviour to the simple one by using |
| 1955 |
~\modulolinenumbers~ without the star.---This enhancement |
| 1956 |
is accompanied by a new package option ~modulo*~ which just |
| 1957 |
executes ~\modulolinenumbers*~ |
| 1958 |
(subsection_\ref{ss:v3opts}).---`With ~\firstlinenumber{1}~' |
| 1959 |
exactly means: `1' is printed if and only if the last |
| 1960 |
~\firstlinenumber~ before or in the paragraph that follows |
| 1961 |
the ``interruption'' has argument `1' (or something |
| 1962 |
\emph{expanding} to `1', or (to) something that \TeX\ |
| 1963 |
``reads'' as 1, e.g.: a \TeX\ count register storing |
| 1964 |
1).---At present, this behaviour may be unsatisfactory with |
| 1965 |
pagewise line-numbering $\dots$ I'll make an experimental |
| 1966 |
extra package if someone complains \dots |
| 1967 |
\begin{code}\begin{verbatim} |
| 1968 |
\newcommand\modulolinenumbers{% |
| 1969 |
\@ifstar |
| 1970 |
{\def\@LN@maybe@moduloresume{% |
| 1971 |
\global\let\@LN@maybe@normalLineNumber |
| 1972 |
\@LN@normalLineNumber}% |
| 1973 |
\@LN@modulolinenos}% |
| 1974 |
{\let\@LN@maybe@moduloresume\relax \@LN@modulolinenos}% |
| 1975 |
} |
| 1976 |
|
| 1977 |
\global\let\@LN@maybe@normalLineNumber\relax |
| 1978 |
\let\@LN@maybe@moduloresume\relax |
| 1979 |
\gdef\@LN@normalLineNumber{% |
| 1980 |
\ifnum\c@linenumber=\c@firstlinenumber \else |
| 1981 |
\ifnum\c@linenumber>\@ne |
| 1982 |
\def\LineNumber{\thelinenumber}% |
| 1983 |
\fi |
| 1984 |
\fi |
| 1985 |
\end{verbatim} |
| 1986 |
\end{code} |
| 1987 |
~\def~ instead of ~\let~ enables taking account of a |
| 1988 |
redefinition of ~\thelinenumber~ in a present numbering |
| 1989 |
environment (e.g.). |
| 1990 |
\begin{code}\begin{verbatim} |
| 1991 |
\global\let\@LN@maybe@normalLineNumber\relax} |
| 1992 |
\end{verbatim} |
| 1993 |
\end{code} |
| 1994 |
Instead of changing ~\LineNumber~ directly by |
| 1995 |
~LN@moduloresume~, these tricks enable ~\modulolinenumbers*~ |
| 1996 |
to act as locally as I can make it. I don't know how to |
| 1997 |
avoid that the output routine switches back to the normal |
| 1998 |
modulo behaviour by a global change. (An ~\aftergroup~ may |
| 1999 |
fail in admittedly improbable cases.) |
| 2000 |
\begin{code}\begin{verbatim} |
| 2001 |
\newcommand*\@LN@modulolinenos[1][\z@]{% |
| 2002 |
\end{verbatim} |
| 2003 |
\end{code} |
| 2004 |
The definition of this macro is that of the former |
| 2005 |
~\modulolinenumbers~. (/New v4.31) |
| 2006 |
\begin{code}\begin{verbatim} |
| 2007 |
\let\LineNumber\themodulolinenumber |
| 2008 |
\ifnum#1>\@ne |
| 2009 |
\chardef % v4.22, note below |
| 2010 |
\c@linenumbermodulo#1\relax |
| 2011 |
\else\ifnum#1=\@ne |
| 2012 |
\end{verbatim} |
| 2013 |
\end{code} |
| 2014 |
\begin{old}\begin{verbatim} |
| 2015 |
% \def\LineNumber{\thelinenumber}% |
| 2016 |
\end{verbatim} |
| 2017 |
\end{old} |
| 2018 |
(New v4.00) I am putting something here to enable |
| 2019 |
~\firstlinenumber~ with $~\c@linenumbermodulo~=1$. |
| 2020 |
With ~lnopatch.sty~, a trick was offered for this purpose. |
| 2021 |
It is now obsolete. |
| 2022 |
|
| 2023 |
\begin{code}\begin{verbatim} |
| 2024 |
\def\LineNumber{\@LN@ifgreat\thelinenumber}% |
| 2025 |
\end{verbatim} |
| 2026 |
\end{code} |
| 2027 |
(/New v4.00) |
| 2028 |
\begin{code}\begin{verbatim} |
| 2029 |
\fi\fi |
| 2030 |
} |
| 2031 |
\end{verbatim} |
| 2032 |
\end{code} |
| 2033 |
(New v4.00) The default of ~\@LN@ifgreat~ is |
| 2034 |
\begin{code}\begin{verbatim} |
| 2035 |
\let\@LN@ifgreat\relax |
| 2036 |
\end{verbatim} |
| 2037 |
\end{code} |
| 2038 |
The previous changes as soon as ~\firstlinenumber~ is used: |
| 2039 |
\begin{code}\begin{verbatim} |
| 2040 |
\newcommand*\firstlinenumber[1]{% |
| 2041 |
\chardef\c@firstlinenumber#1\relax |
| 2042 |
\end{verbatim} |
| 2043 |
\end{code} |
| 2044 |
No counter, little values allowed only---OK?---(UL) |
| 2045 |
The change is local---OK? The good thing is that |
| 2046 |
~\global\firstlinenumber{~\angle{number}~}~ works. Moreover, |
| 2047 |
~\modulolinenumbers~ acts locally as well. (/UL) |
| 2048 |
|
| 2049 |
(New v4.31) |
| 2050 |
\begin{code}\begin{verbatim} |
| 2051 |
\let\@LN@ifgreat\@LN@ifgreat@critical} |
| 2052 |
|
| 2053 |
\def\@LN@ifgreat@critical{% |
| 2054 |
\ifnum\c@linenumber<\c@firstlinenumber |
| 2055 |
\expandafter \@gobble |
| 2056 |
\fi}% |
| 2057 |
\end{verbatim} |
| 2058 |
\end{code} |
| 2059 |
(/New v4.31) |
| 2060 |
|
| 2061 |
The default |
| 2062 |
value of ~\c@firstlinenumber~ %% v4.31 |
| 2063 |
is 0. This is best for what one would expect from modulo |
| 2064 |
printing. |
| 2065 |
\begin{code}\begin{verbatim} |
| 2066 |
\let\c@firstlinenumber=\z@ |
| 2067 |
\end{verbatim} |
| 2068 |
\end{code} |
| 2069 |
|
| 2070 |
For usage and effects of ~\modulolinenumbers~ and %% v4.31 |
| 2071 |
~\firstlinenumbers~, please consult section_\ref{s:UserCmds}. |
| 2072 |
Two details on ~\firstlinenumbers~ here: |
| 2073 |
(i)_~\firstlinenumber~ acts on a paragraph if and only if |
| 2074 |
(a)_the paragraph is broken into lines ``in line-numbering |
| 2075 |
mode'' (after ~\linenumbers~, e.g.); |
| 2076 |
(b)_it is the last occurrence of a ~\firstlinenumbers~ |
| 2077 |
before or in the paragraph. |
| 2078 |
(The practical applications of this that I can imagine |
| 2079 |
don't seem appealing to me.) |
| 2080 |
Cf._the explanation above of how ~\modulolinenumbers~ and |
| 2081 |
~\firstlinenumbers~ interact---for this and for (ii), |
| 2082 |
which is concerned with possible arguments for |
| 2083 |
~\firstlinenumbers~. |
| 2084 |
|
| 2085 |
Note that the line numbers of the present section |
| 2086 |
demonstrate the two devices. (/New v4.00) |
| 2087 |
\begin{code}\begin{verbatim} |
| 2088 |
\chardef\c@linenumbermodulo=5 % v4.2; ugly? |
| 2089 |
\modulolinenumbers[1] |
| 2090 |
\end{verbatim} |
| 2091 |
\end{code} |
| 2092 |
(New v4.22) The new implementation through ~\chardef~ |
| 2093 |
decreases the functionality and raises certain compatibility |
| 2094 |
problems. I face this without fear. The maximum modulo value |
| 2095 |
is now ~255~. I expect that this suffices for usual applications. |
| 2096 |
However, some users have ``abused'' ~lineno.sty~ to get |
| 2097 |
~ednotes.sty~ features without line numbers, so have set the |
| 2098 |
modulo to a value beyond the total number of lines in their |
| 2099 |
edition. This ought to be replaced by |
| 2100 |
~\let\makeLineNumber\relax~. (/New v4.22) |
| 2101 |
|
| 2102 |
\section{% |
| 2103 |
Package options |
| 2104 |
\unskip}\label{s:Opts} |
| 2105 |
|
| 2106 |
(New v4.1) |
| 2107 |
The last heading formerly was the heading of what is now |
| 2108 |
subsection_\ref{ss:v3opts}. The options declared there were |
| 2109 |
said to execute user commands only. This was wrong already |
| 2110 |
concerning ~displaymath~ and ~hyperref~. At least, however, |
| 2111 |
these options were no or almost no occasion to skip definitions |
| 2112 |
or allocations. This is different with the options that we now |
| 2113 |
insert. |
| 2114 |
|
| 2115 |
|
| 2116 |
\subsection{% |
| 2117 |
Extended referencing to line numbers. (v4.2) |
| 2118 |
\unskip} |
| 2119 |
This subsection explains and declares package option ~addpageno~. %% v4.31 |
| 2120 |
|
| 2121 |
If a line to whose number you refer by ~\ref~ is not on the |
| 2122 |
present page, it may be useful to add the number of the page |
| 2123 |
on which the line occurs---and perhaps it should not be added |
| 2124 |
otherwise. In general, you could use the Standard \LaTeX\ |
| 2125 |
package varioref for this. However, the latter usually |
| 2126 |
produces verbose output like `on the preceding page'--- |
| 2127 |
unless costumized---, while in critical editions, e.g., one |
| 2128 |
may prefer just adding the page number and some mark on the |
| 2129 |
left of the line number, irrespectively of how far the page is |
| 2130 |
apart etc. To support this, package option ~addpageno~ |
| 2131 |
provides a command ~\vpagelineref~ to be used in place of |
| 2132 |
~\ref~. This produces, e.g., `34.15' when referring to line_15 |
| 2133 |
on page_34 while the present page is not 34. You can customize |
| 2134 |
the outcome, see the package file ~vplref.sty~ where the code |
| 2135 |
and further details are. You may conceive of |
| 2136 |
~\vpagelineref~ as a certain customization of varioref's |
| 2137 |
~\vref~. |
| 2138 |
|
| 2139 |
This implies that option ~addpageno~ requires the files |
| 2140 |
~vplref.sty~ and ~varioref.sty~. ~addpageno~ automatically |
| 2141 |
loads both of them. Yet you can also load ~varioref.sty~ |
| 2142 |
on your own to use its package options. |
| 2143 |
|
| 2144 |
Of course, you might better introduce a shorter command name |
| 2145 |
for ~\vpagelineref~ for your work, while we cannot predict |
| 2146 |
here what shorthand will fit your work. E.g., |
| 2147 |
~\newcommand{\lref}{\vpagelineref}~. |
| 2148 |
|
| 2149 |
If you really want to add the page number in \emph{any} case, |
| 2150 |
use, e.g., some ~\myref~ instead of ~\ref~, after |
| 2151 |
\[~newcommand*{\myref}{\pageref{#1}.\ref{#1}}~\] |
| 2152 |
or what you like. You don't need the ~addpageno~ option in |
| 2153 |
this case. |
| 2154 |
|
| 2155 |
~addpageno~ is due to a suggestion by Sergei Mariev. |
| 2156 |
\begin{code}\begin{verbatim} |
| 2157 |
\DeclareVoidOption{addpageno}{% |
| 2158 |
\AtEndOfPackage{\RequirePackage{vplref}[2005/04/25]}} |
| 2159 |
\end{verbatim} |
| 2160 |
\end{code} |
| 2161 |
\subsection{% |
| 2162 |
\scs{linelabel} in math mode |
| 2163 |
\unskip}\label{ss:MathRef} |
| 2164 |
|
| 2165 |
We have made some first steps towards allowing ~\linelabel~ in |
| 2166 |
math mode. Because our code for this is presently experimental, |
| 2167 |
we leave it to the user to decide for the experiment by calling |
| 2168 |
option ~mathrefs~. We are in a hurry now and thus leave the |
| 2169 |
code, explanations, and discussion in the separate package |
| 2170 |
~ednmath0.sty~. Maybe we later find the time to improve the |
| 2171 |
code and move the relevant content of ~ednmath0.sty~ to here. |
| 2172 |
The optimal situation would be to define ~\linelabel~ from |
| 2173 |
the start so it works in math mode, omitting the ~mathrefs~ |
| 2174 |
option. |
| 2175 |
|
| 2176 |
Actually, this package even provides adjustments for analogously |
| 2177 |
allowing ~ednotes.sty~ commands in math mode. Loading the package |
| 2178 |
is postponed to ~\AtBeginDocument~ when we know whether these |
| 2179 |
adjustments are needed. |
| 2180 |
\begin{code}\begin{verbatim} |
| 2181 |
\DeclareVoidOption{mathrefs}{\AtBeginDocument |
| 2182 |
{\RequirePackage{ednmath0}[2004/08/20]}} |
| 2183 |
\end{verbatim} |
| 2184 |
\end{code} |
| 2185 |
|
| 2186 |
\subsection{% |
| 2187 |
Arrays, tabular environments (Revised v4.11) |
| 2188 |
\unskip}\label{ss:Tab} |
| 2189 |
|
| 2190 |
This subsection explains and declares package options %% v4.31 |
| 2191 |
~edtable~, ~longtable~, and ~nolongtablepatch~. |
| 2192 |
|
| 2193 |
The standard \LaTeX\ tabular environments come as single |
| 2194 |
boxes, so the ~lineno.sty~ versions before v4.00 treated them as |
| 2195 |
(parts of) single lines, printing (at most) one line number |
| 2196 |
beside each and stepping the line number counter once only. |
| 2197 |
Moreover, ~\linelabel~s got lost. Of course, tables are |
| 2198 |
usually so high that you will want to treat each row like a |
| 2199 |
line. (Christian Tapp even desires that the lines of table |
| 2200 |
entries belonging to a single row are treated like ordinary |
| 2201 |
lines.) Footnotes get lost in such environments as well, which |
| 2202 |
was bad for ~ednotes.sty~. |
| 2203 |
|
| 2204 |
We provide adjustments to count lines, print their numbers |
| 2205 |
etc.\ as desired at least for \emph{some} \LaTeX\ tabular |
| 2206 |
environments. (Like with other details, ``some'' is to some |
| 2207 |
extent explained in ~edtable.sty~.) We do this similarly as |
| 2208 |
with option ~mathrefs~ before. We leave code |
| 2209 |
and explanations in the separate package ~edtable.sty~. |
| 2210 |
(For wizards: this package provides adjustments for |
| 2211 |
~ednotes.sty~ as well. However, in the present case we don't try |
| 2212 |
to avoid them unless ~ednotes.sty~ is loaded.) |
| 2213 |
Package option ~edtable~ |
| 2214 |
defines---by loading ~edtable.sty~---an environment ~{edtable}~ |
| 2215 |
which is able to change some \LaTeX\ tabular environments |
| 2216 |
with the desired effects. (v4.11: ~edtable.sty~ v1.3 counts |
| 2217 |
\LaTeX's ~{array}~ [etc.\@] as a ``tabular environment'' as |
| 2218 |
well.) |
| 2219 |
|
| 2220 |
The ~{edtable}~ environment doesn't help with ~longtable.sty~, |
| 2221 |
however. To make up for this, ~{longtable}~ is adjusted in a |
| 2222 |
different way---and this happens only when another ~lineno.sty~ |
| 2223 |
option ~longtable~ is called. In this case, option ~edtable~ |
| 2224 |
needn't be called explicitly: option ~longtable~ works as if |
| 2225 |
~edtable~ had been called. |
| 2226 |
|
| 2227 |
Now, we are convinced that vertical spacing around |
| 2228 |
~{longtable}~ works wrongly---see \LaTeX\ bugs database |
| 2229 |
tools/3180 and 3485, or see explanations in the package |
| 2230 |
~ltabptch.sty~ (which is to be obtained from CTAN folder |
| 2231 |
\path{macros/latex/ltabptch}). Our conviction is so strong |
| 2232 |
that the ~longtable~ option loads---after ~longtable.sty~---the |
| 2233 |
patch package ~ltabptch.sty~. If the user doesn't want this |
| 2234 |
(maybe preferring her own arrangement with the vertical |
| 2235 |
spacing), she can forbid it by calling ~nolongtablepatch~. |
| 2236 |
|
| 2237 |
The following code just collects some choices, which are |
| 2238 |
then executed in section_\ref{ss:ExOpt}. We use an ~\if...~ |
| 2239 |
without ~\newif~ since ~\if...true~ and ~\if...false~ |
| 2240 |
would occur at most two times and only within the present |
| 2241 |
package. (~\AtEndOfClass{\RequirePackage{edtable}}~ |
| 2242 |
could be used instead, I just overlooked this. Now I don't |
| 2243 |
change it because it allows to change the version requirement |
| 2244 |
at one place only.) |
| 2245 |
\begin{code}\begin{verbatim} |
| 2246 |
\let\if@LN@edtable\iffalse |
| 2247 |
|
| 2248 |
\DeclareVoidOption{edtable}{\let\if@LN@edtable\iftrue} |
| 2249 |
|
| 2250 |
\DeclareVoidOption{longtable}{\let\if@LN@edtable\iftrue |
| 2251 |
\PassOptionsToPackage{longtable}{edtable}} |
| 2252 |
|
| 2253 |
\DeclareVoidOption{nolongtablepatch}{% |
| 2254 |
\PassOptionsToPackage{nolongtablepatch}{edtable}} |
| 2255 |
\end{verbatim} |
| 2256 |
\end{code} |
| 2257 |
(/New v4.1) |
| 2258 |
|
| 2259 |
\subsection{% |
| 2260 |
Switch among settings |
| 2261 |
\unskip}\label{ss:v3opts} |
| 2262 |
|
| 2263 |
There is a bunch of package options that execute %% v4.2 |
| 2264 |
user commands only. |
| 2265 |
|
| 2266 |
Options ~left~ (~right~) put the line numbers on the left |
| 2267 |
(right) margin. This works in all modes. ~left~ is the |
| 2268 |
default. |
| 2269 |
\begin{code}\begin{verbatim} |
| 2270 |
\DeclareVoidOption{left}{\leftlinenumbers*} |
| 2271 |
|
| 2272 |
\DeclareVoidOption{right}{\rightlinenumbers*} |
| 2273 |
\end{verbatim} |
| 2274 |
\end{code} |
| 2275 |
Option ~switch~ (~switch*~) puts the line numbers on the |
| 2276 |
outer (inner) margin of the text. This requires running |
| 2277 |
the pagewise mode, but we turn off the page offset |
| 2278 |
subtraction, getting sort of running numbers again. The |
| 2279 |
~pagewise~ option may restore true pagewise mode later. |
| 2280 |
\begin{code}\begin{verbatim} |
| 2281 |
\DeclareVoidOption{switch}{\setpagewiselinenumbers |
| 2282 |
\switchlinenumbers |
| 2283 |
\runningpagewiselinenumbers} |
| 2284 |
|
| 2285 |
\DeclareVoidOption{switch*}{\setpagewiselinenumbers |
| 2286 |
\switchlinenumbers*% |
| 2287 |
\runningpagewiselinenumbers} |
| 2288 |
\end{verbatim} |
| 2289 |
\end{code} |
| 2290 |
In twocolumn mode, we can switch the line numbers to |
| 2291 |
the outer margin, and/or start with number 1 in each |
| 2292 |
column. Margin switching is covered by the ~switch~ |
| 2293 |
options. |
| 2294 |
\begin{code}\begin{verbatim} |
| 2295 |
\DeclareVoidOption{columnwise}{\setpagewiselinenumbers |
| 2296 |
\columnwiselinenumberstrue |
| 2297 |
\realpagewiselinenumbers} |
| 2298 |
\end{verbatim} |
| 2299 |
\end{code} |
| 2300 |
The options ~pagewise~ and ~running~ select the major |
| 2301 |
line number mechanism. ~running~ line numbers refer to a real |
| 2302 |
counter value, which can be reset for any paragraph, |
| 2303 |
even getting multiple paragraphs on one page starting |
| 2304 |
with line number one. ~pagewise~ line numbers get a |
| 2305 |
unique hidden number within the document, but with the |
| 2306 |
opportunity to establish the page on which they finally |
| 2307 |
come to rest. This allows the subtraction of the page |
| 2308 |
offset, getting the numbers starting with 1 on top of each |
| 2309 |
page, and margin switching in twoside formats becomes |
| 2310 |
possible. The default mode is ~running~. |
| 2311 |
|
| 2312 |
The order of declaration of the options is important here |
| 2313 |
~pagewise~ must come after ~switch~, to overide running |
| 2314 |
pagewise mode. ~running~ comes last, to reset the running |
| 2315 |
line number mode, e.g, after selecting margin switch mode |
| 2316 |
for ~pagewise~ running. Once more, if you specify all |
| 2317 |
three of the options ~[switch,pagewise,running]~, the |
| 2318 |
result is almost nothing, but if you later say |
| 2319 |
~\pagewiselinenumbers~, you get margin switching, with |
| 2320 |
real pagewise line numbers. |
| 2321 |
|
| 2322 |
\begin{code}\begin{verbatim} |
| 2323 |
\DeclareVoidOption{pagewise}{\setpagewiselinenumbers |
| 2324 |
\realpagewiselinenumbers} |
| 2325 |
|
| 2326 |
\DeclareVoidOption{running}{\setrunninglinenumbers} |
| 2327 |
\end{verbatim} |
| 2328 |
\end{code} |
| 2329 |
The option ~modulo~ causes only those linenumbers to be |
| 2330 |
printed which are multiples of five. |
| 2331 |
\begin{code}\begin{verbatim} |
| 2332 |
\DeclareVoidOption{modulo}{\modulolinenumbers\relax} |
| 2333 |
\end{verbatim} |
| 2334 |
\end{code} |
| 2335 |
Option ~modulo*~ modifies ~modulo~ in working like |
| 2336 |
~\modulolinenumbers*~---see section_\ref{s:UserCmds}. |
| 2337 |
\begin{code}\begin{verbatim} |
| 2338 |
\DeclareVoidOption{modulo*}{\modulolinenumbers*\relax} |
| 2339 |
\end{verbatim} |
| 2340 |
\end{code} |
| 2341 |
The package option ~mathlines~ switches the behavior of |
| 2342 |
the ~{linenomath}~ environment with its star-form. |
| 2343 |
Without this option, the ~{linenomath}~ environment does |
| 2344 |
not number the lines of the display, while the star-form |
| 2345 |
does. With this option, its just the opposite. |
| 2346 |
|
| 2347 |
\begin{code}\begin{verbatim} |
| 2348 |
\DeclareVoidOption{mathlines}{\linenumberdisplaymath} |
| 2349 |
\end{verbatim} |
| 2350 |
\end{code} |
| 2351 |
~displaymath~ now calls for wrappers of the standard |
| 2352 |
\LaTeX\ display math environment. This was previously |
| 2353 |
done by ~mlineno.sty~. |
| 2354 |
|
| 2355 |
(New v4.3) Option `displaymath' becomes default according |
| 2356 |
to Erik \mbox{Luijten}'s suggestion. I was finally convinced |
| 2357 |
of this as soon as I discovered how to avoid a spurious line |
| 2358 |
number above ~\begin{linenomath}~ (subsection_\ref{ss:DM}). |
| 2359 |
~\endlinenomath~ provides ~\ignorespaces~, so what could go |
| 2360 |
wrong now? |
| 2361 |
\begin{code}\begin{verbatim} |
| 2362 |
\DeclareVoidOption{displaymath}{% |
| 2363 |
\PackageWarningNoLine{lineno}{Option [displaymath] is obsolete}} |
| 2364 |
\end{verbatim} |
| 2365 |
\end{code} |
| 2366 |
(/New v4.3) |
| 2367 |
|
| 2368 |
(New v5.3) Options `sep' and `width' set ~\linenumbersep~ (the separation of |
| 2369 |
the line number to the text) and ~\linenumberwidth~ (the width of the line |
| 2370 |
number box on the right margin) respectively; see section_\ref{s:CustHooks}. |
| 2371 |
\begin{code}\begin{verbatim} |
| 2372 |
\DeclareStringOption[\linenumbersep]{sep} |
| 2373 |
\DeclareStringOption[\linenumberwidth]{width} |
| 2374 |
\AtBeginDocument{% |
| 2375 |
\linenumbersep=\lineno@sep% |
| 2376 |
\linenumberwidth=\lineno@width% |
| 2377 |
} |
| 2378 |
\end{verbatim} |
| 2379 |
\end{code} |
| 2380 |
(/New v5.3) |
| 2381 |
|
| 2382 |
\subsection{% |
| 2383 |
Compatibility with \texttt{hyperref} %% own subsec. v4.3. |
| 2384 |
\unskip} |
| 2385 |
The ~hyperref~ package, via ~nameref~, requires three more |
| 2386 |
groups in the second argment of a ~\newlabel~. Well, why |
| 2387 |
shouldn't it get them? (New v3.07) The presence of the |
| 2388 |
~nameref~ package is now detected automatically |
| 2389 |
~\AtBeginDocument~. (/New v3.07) (Fixed in v3.09) We try |
| 2390 |
to be smart, and test ~\AtBeginDocument~ if the ~nameref~ |
| 2391 |
package is loaded, but ~hyperref~ postpones the loading of |
| 2392 |
~nameref~ too, so this is all in vain. |
| 2393 |
|
| 2394 |
(New v4.3) But we can also test at the first ~\linelabel~. |
| 2395 |
Regarding the error-message for misplaced ~\linelabel~ from v4.11: |
| 2396 |
previously, ~\linenumbers~ rendered ~\linelabel~ the genuine |
| 2397 |
version of ~\linelabel~ from the start on. This doesn't work |
| 2398 |
now, since ~\@LN@linelabel~ may change its meaning after the |
| 2399 |
first ~\linenumbers~ and before a next one (if there is some). |
| 2400 |
(/New v4.3) |
| 2401 |
\begin{code}\begin{verbatim} |
| 2402 |
\DeclareVoidOption{hyperref}{% |
| 2403 |
\PackageWarningNoLine{lineno}{Option [hyperref] is obsolete}} |
| 2404 |
|
| 2405 |
\AtBeginDocument{% |
| 2406 |
\@ifpackageloaded{nameref}{% |
| 2407 |
\end{verbatim} |
| 2408 |
\end{code} |
| 2409 |
(New v4.3) ``Global'' is merely ``symbolic'' ~\AtBeginDoc...~. |
| 2410 |
If ~nameref~ is not detected here, the next ~\@LN@linelabel~ |
| 2411 |
will do almost the same, then globally indeed. |
| 2412 |
\begin{code}\begin{verbatim} |
| 2413 |
\gdef\@LN@ExtraLabelItems{{}{}{}}% |
| 2414 |
}{% |
| 2415 |
\global\let\@LN@@linelabel\@LN@linelabel |
| 2416 |
\gdef\@LN@linelabel{% |
| 2417 |
\end{verbatim} |
| 2418 |
\end{code} |
| 2419 |
~\@ifpackageloaded~ is ``preamble only'', its---very |
| 2420 |
internal---preamble definition is replicated here: |
| 2421 |
\begin{code}\begin{verbatim} |
| 2422 |
\expandafter |
| 2423 |
\ifx\csname [email protected]\endcsname\relax \else |
| 2424 |
\gdef\@LN@ExtraLabelItems{{}{}{}}% |
| 2425 |
\fi |
| 2426 |
\end{verbatim} |
| 2427 |
\end{code} |
| 2428 |
Now aim at the ``usual'' behaviour: |
| 2429 |
\begin{code}\begin{verbatim} |
| 2430 |
\global\let\@LN@linelabel\@LN@@linelabel |
| 2431 |
\global\let\@LN@@linelabel\relax |
| 2432 |
\@LN@linelabel |
| 2433 |
}% |
| 2434 |
}% |
| 2435 |
} |
| 2436 |
\end{verbatim} |
| 2437 |
\end{code} |
| 2438 |
(New 5.4) Option to allow ~\linelabel~ with ~\nolinenumbers~. |
| 2439 |
\begin{code}\begin{verbatim} |
| 2440 |
\DeclareVoidOption{nolinelabelerror}{\def\@LN@LLerror{\relax}} |
| 2441 |
\end{verbatim} |
| 2442 |
\end{code} |
| 2443 |
(/New 5.4) |
| 2444 |
(/New v4.3) |
| 2445 |
|
| 2446 |
(New v4.1) |
| 2447 |
\subsection{% |
| 2448 |
A note on calling so many options |
| 2449 |
\unskip} |
| 2450 |
|
| 2451 |
The number of package options may stimulate worrying about how to |
| 2452 |
\emph{enter} all the options that one would like to use---they may |
| 2453 |
not fit into one line. Fortunately, you can safely break code lines |
| 2454 |
after the commas separating the option names in the ~\usepackage~ |
| 2455 |
command (no comment marks needed). |
| 2456 |
|
| 2457 |
\subsection{% |
| 2458 |
Execute options |
| 2459 |
\unskip}\label{ss:ExOpt} |
| 2460 |
|
| 2461 |
We stop declaring options and execute the ones that are |
| 2462 |
called by the user. (/New v4.1) |
| 2463 |
\begin{code}\begin{verbatim} |
| 2464 |
\ProcessKeyvalOptions* |
| 2465 |
\end{verbatim} |
| 2466 |
\end{code} |
| 2467 |
(New v4.1) Now we know whether ~edtable.sty~ is wanted |
| 2468 |
and (if it is) with which options it is to be called. |
| 2469 |
\begin{code}\begin{verbatim} |
| 2470 |
\if@LN@edtable \RequirePackage{edtable}[2005/03/07] \fi |
| 2471 |
\end{verbatim} |
| 2472 |
\end{code} |
| 2473 |
(/New v4.1) |
| 2474 |
|
| 2475 |
\section{% |
| 2476 |
Package extensions |
| 2477 |
\label{s:Xt}\unskip} |
| 2478 |
|
| 2479 |
Some of the extensions in this section were previously supplied |
| 2480 |
in separate ~.sty~ files. |
| 2481 |
|
| 2482 |
\subsection{% |
| 2483 |
\texttt{displaymath} |
| 2484 |
\unskip}\label{ss:display} |
| 2485 |
(New v4.3) From now on, you no longer need to type |
| 2486 |
the ~{linenomath}~ environment with the ~\[~, ~{equation}~, |
| 2487 |
and ~{eqnarray}~ environments---and you no longer need to |
| 2488 |
use the former package option ~displaymath~ for this feature. |
| 2489 |
(/New v4.3) |
| 2490 |
|
| 2491 |
The standard \LaTeX\ display math environments are |
| 2492 |
wrapped in a ~{linenomath}~ environment. |
| 2493 |
|
| 2494 |
(New 3.05) The ~[fleqn]~ option of the standard |
| 2495 |
\LaTeX\ classes defines the display math |
| 2496 |
environments such that line numbers appear just |
| 2497 |
fine. Thus, we need not do any tricks when |
| 2498 |
~[fleqn]~ is loaded, as indicated by presents of |
| 2499 |
the ~\mathindent~ register. (/New 3.05) |
| 2500 |
|
| 2501 |
(New 3.05a) for ~{eqnarray}~s we rather keep the |
| 2502 |
old trick. (/New 3.05a) |
| 2503 |
|
| 2504 |
(New 3.08) Wrap ~\[~ and ~\]~ into ~{linenomath}~, |
| 2505 |
instead of ~{displaymath}~. Also save the definition |
| 2506 |
of ~\equation~, instead of replicating the current |
| 2507 |
\LaTeX\ definition. (/New 3.08) |
| 2508 |
\begin{code}\begin{verbatim} |
| 2509 |
\@ifundefined{mathindent}{ |
| 2510 |
|
| 2511 |
\let\LN@displaymath\[% |
| 2512 |
\let\LN@enddisplaymath\]% |
| 2513 |
\renewcommand\[{\begin{linenomath}\LN@displaymath}% |
| 2514 |
\renewcommand\]{\LN@enddisplaymath\end{linenomath}}% |
| 2515 |
\end{verbatim} |
| 2516 |
\end{code} |
| 2517 |
|
| 2518 |
\begin{code}\begin{verbatim} |
| 2519 |
\let\LN@equation\equation |
| 2520 |
\let\LN@endequation\endequation |
| 2521 |
\renewenvironment{equation}% |
| 2522 |
{\linenomath\LN@equation}% |
| 2523 |
{\LN@endequation\endlinenomath}% |
| 2524 |
|
| 2525 |
}{}% \@ifundefined{mathindent} -- 3rd arg v4.2, was \par! |
| 2526 |
|
| 2527 |
\let\LN@eqnarray\eqnarray |
| 2528 |
\let\LN@endeqnarray\endeqnarray |
| 2529 |
\renewenvironment{eqnarray}% |
| 2530 |
{\linenomath\LN@eqnarray}% |
| 2531 |
{\LN@endeqnarray\endlinenomath}% |
| 2532 |
\end{verbatim} |
| 2533 |
\end{code} |
| 2534 |
(UL) Indeed. The \LaTeX\ macros are saved for |
| 2535 |
unnumbered mode, which is detected by ~\linenomath~. |
| 2536 |
(/UL) |
| 2537 |
|
| 2538 |
\subsection{% |
| 2539 |
Line numbers in internal vertical mode |
| 2540 |
\unskip}\label{ss:ILN} |
| 2541 |
|
| 2542 |
The command ~\internallinenumbers~ adds line numbers in |
| 2543 |
internal vertical mode, but with limitations: we assume |
| 2544 |
fixed baseline skip. |
| 2545 |
|
| 2546 |
(v4.22) v3.10 provided a global (~\global\advance~) |
| 2547 |
as well as a local version (star-form, using |
| 2548 |
~\c@internallinenumber~). ~\resetlinenumbers~ acted |
| 2549 |
locally and was here used with the global version---save |
| 2550 |
stack danger, \TeX book p._301---in v4.00 I |
| 2551 |
disabled the global version therefore. Now I find that |
| 2552 |
it is better to keep a global version, and the now global |
| 2553 |
~\resetlinenumbers~ is perfect for this. The global version |
| 2554 |
allows continuing the ``internal'' numbers in the ensuing |
| 2555 |
``external'' text, and---unless reset by brackets |
| 2556 |
argument---continuing the above series of line numbers. |
| 2557 |
As with v3.10, the local version always starts with |
| 2558 |
line number one. A new ~\@LN@iglobal~ steps ~\global~ly |
| 2559 |
in the global version, otherwise it is ~\relax~. |
| 2560 |
(I also remove all my stupid discussions as of v4.00. |
| 2561 |
And I use ~\newcommand~.) (v4.22) |
| 2562 |
\begin{code}\begin{verbatim} |
| 2563 |
\let\@LN@iglobal\global % v4.22 |
| 2564 |
|
| 2565 |
\newcommand\internallinenumbers{\setrunninglinenumbers |
| 2566 |
\let\@@par\internallinenumberpar |
| 2567 |
\ifx\@par\@@@par\let\@par\internallinenumberpar\fi |
| 2568 |
\ifx\par\@@@par\let\par\internallinenumberpar\fi |
| 2569 |
\ifx\@par\linenumberpar\let\@par\internallinenumberpar\fi |
| 2570 |
\ifx\par\linenumberpar\let\par\internallinenumberpar\fi |
| 2571 |
\@ifnextchar[{\resetlinenumber}%] |
| 2572 |
{\@ifstar{\let\c@linenumber\c@internallinenumber |
| 2573 |
\let\@LN@iglobal\relax % v4.22 |
| 2574 |
\c@linenumber\@ne}{}}% |
| 2575 |
} |
| 2576 |
|
| 2577 |
\let\endinternallinenumbers\endlinenumbers |
| 2578 |
\@namedef{internallinenumbers*}{\internallinenumbers*} |
| 2579 |
\expandafter\let\csname endinternallinenumbers*\endcsname\endlinenumbers |
| 2580 |
|
| 2581 |
\newcount\c@internallinenumber |
| 2582 |
\newcount\c@internallinenumbers |
| 2583 |
|
| 2584 |
\newcommand\internallinenumberpar{% |
| 2585 |
\ifvmode\@@@par\else\ifinner\@@@par\else\@@@par |
| 2586 |
\begingroup |
| 2587 |
\c@internallinenumbers\prevgraf |
| 2588 |
\setbox\@tempboxa\hbox{\vbox{\makeinternalLinenumbers}}% |
| 2589 |
\ht\@tempboxa\z@ |
| 2590 |
\ifdim\prevdepth=-1000pt |
| 2591 |
% \nointerlineskip is already set so we don't need set it again |
| 2592 |
% (and we shouldn't back up) |
| 2593 |
\else |
| 2594 |
\dp\@tempboxa\prevdepth |
| 2595 |
\nobreak\vskip-\prevdepth |
| 2596 |
\nointerlineskip |
| 2597 |
\fi |
| 2598 |
\box\@tempboxa |
| 2599 |
\endgroup |
| 2600 |
\fi\fi |
| 2601 |
} |
| 2602 |
|
| 2603 |
\newcommand\makeinternalLinenumbers{% |
| 2604 |
\ifnum\c@internallinenumbers>\z@ % v4.2 |
| 2605 |
\hb@xt@\z@{\makeLineNumber}% |
| 2606 |
\@LN@iglobal % v4.22 |
| 2607 |
\advance\c@linenumber\@ne |
| 2608 |
\advance\c@internallinenumbers\m@ne |
| 2609 |
\expandafter\makeinternalLinenumbers\fi |
| 2610 |
} |
| 2611 |
% TODO v4.4+: star: line numbers right!? cf. lnocapt.sty |
| 2612 |
\end{verbatim} |
| 2613 |
\end{code} |
| 2614 |
|
| 2615 |
\subsection{% |
| 2616 |
Line number references with offset |
| 2617 |
\unskip} |
| 2618 |
|
| 2619 |
This extension defines macros to refer to line |
| 2620 |
numbers with an offset, e.g., to refer to a line |
| 2621 |
which cannot be labeled directly (display math). |
| 2622 |
This was formerly knows as ~rlineno.sty~. |
| 2623 |
|
| 2624 |
To refer to a pagewise line number with offset: |
| 2625 |
\begin{quote} |
| 2626 |
~\linerefp[~\angle{OFFSET}~]{~\angle{LABEL}~}~ |
| 2627 |
\end{quote} |
| 2628 |
To refer to a running line number with offset: |
| 2629 |
\begin{quote} |
| 2630 |
~\linerefr[~\angle{OFFSET}~]{~\angle{LABEL}~}~ |
| 2631 |
\end{quote} |
| 2632 |
To refer to a line number labeled in the same mode as currently |
| 2633 |
selected: |
| 2634 |
\begin{quote} |
| 2635 |
~\lineref[~\angle{OFFSET}~]{~\angle{LABEL}~}~ |
| 2636 |
\end{quote} |
| 2637 |
\begin{code}\begin{verbatim} |
| 2638 |
\newcommand\lineref{% |
| 2639 |
\ifx\c@linenumber\c@runninglinenumber |
| 2640 |
\expandafter\linerefr |
| 2641 |
\else |
| 2642 |
\expandafter\linerefp |
| 2643 |
\fi |
| 2644 |
} |
| 2645 |
|
| 2646 |
\newcommand*\linerefp[2][\z@]{{% |
| 2647 |
\let\@thelinenumber\thelinenumber |
| 2648 |
\edef\thelinenumber{\advance\c@linenumber#1\relax |
| 2649 |
\noexpand\@thelinenumber}% |
| 2650 |
\ref{#2}% |
| 2651 |
}} |
| 2652 |
\end{verbatim} |
| 2653 |
\end{code} |
| 2654 |
This goes deep into \LaTeX's internals. |
| 2655 |
\begin{code}\begin{verbatim} |
| 2656 |
\newcommand*\linerefr[2][\z@]{{% |
| 2657 |
\def\@@linerefadd{\advance\c@linenumber#1}% |
| 2658 |
\expandafter\@setref\csname r@#2\endcsname |
| 2659 |
\@linerefadd{#2}% |
| 2660 |
}} |
| 2661 |
|
| 2662 |
\newcommand*\@linerefadd[2]{\c@linenumber=#1\@@linerefadd\relax |
| 2663 |
\thelinenumber} |
| 2664 |
\end{verbatim} |
| 2665 |
\end{code} |
| 2666 |
|
| 2667 |
\subsection{% |
| 2668 |
Numbered quotation environments |
| 2669 |
\unskip} |
| 2670 |
|
| 2671 |
The ~{numquote}~ and ~{numquotation}~ |
| 2672 |
environments are like ~{quote}~ and |
| 2673 |
~{quotation}~, except there will be line |
| 2674 |
numbers. |
| 2675 |
|
| 2676 |
An optional argument gives the number to count |
| 2677 |
from. A star ~*~ (inside or outside the closing |
| 2678 |
~}~) prevent the reset of the line numbers. |
| 2679 |
Default is to count from one. |
| 2680 |
|
| 2681 |
(v4.22: A local version using ~\c@internallinenumber~ |
| 2682 |
might be useful, see subsection_\ref{ss:ILN}.) %% TODO v4.4+ |
| 2683 |
\begin{code}\begin{verbatim} |
| 2684 |
\newcommand\quotelinenumbers |
| 2685 |
{\@ifstar\linenumbers{\@ifnextchar[\linenumbers{\linenumbers*}}} |
| 2686 |
|
| 2687 |
\newdimen\quotelinenumbersep |
| 2688 |
\quotelinenumbersep=\linenumbersep |
| 2689 |
\let\quotelinenumberfont\linenumberfont |
| 2690 |
|
| 2691 |
\newcommand\numquotelist |
| 2692 |
{\leftlinenumbers |
| 2693 |
\linenumbersep\quotelinenumbersep |
| 2694 |
\let\linenumberfont\quotelinenumberfont |
| 2695 |
\addtolength{\linenumbersep}{-\@totalleftmargin}% |
| 2696 |
\quotelinenumbers |
| 2697 |
} |
| 2698 |
|
| 2699 |
\newenvironment{numquote} {\quote\numquotelist}{\endquote} |
| 2700 |
\newenvironment{numquotation} {\quotation\numquotelist}{\endquotation} |
| 2701 |
\newenvironment{numquote*} {\quote\numquotelist*}{\endquote} |
| 2702 |
\newenvironment{numquotation*}{\quotation\numquotelist*}{\endquotation} |
| 2703 |
\end{verbatim} |
| 2704 |
\end{code} |
| 2705 |
|
| 2706 |
\subsection{% |
| 2707 |
Frame around a paragraph |
| 2708 |
\unskip} |
| 2709 |
|
| 2710 |
The ~{bframe}~ environment draws a frame around |
| 2711 |
some text, across page breaks, if necessary. |
| 2712 |
|
| 2713 |
This works only for plain text paragraphs, |
| 2714 |
without special height lines. All lines must be |
| 2715 |
~\baselineskip~ apart, no display math. |
| 2716 |
\begin{code}\begin{verbatim} |
| 2717 |
\newenvironment{bframe} |
| 2718 |
{\par |
| 2719 |
\@tempdima\columnwidth |
| 2720 |
\advance\@tempdima 2\bframesep |
| 2721 |
\setbox\bframebox\hb@xt@\columnwidth{% |
| 2722 |
\hskip-\bframesep |
| 2723 |
\vrule\@width\bframerule\@height\baselineskip\@depth\bframesep |
| 2724 |
\advance\@tempdima-2\bframerule |
| 2725 |
\hskip\@tempdima |
| 2726 |
\vrule\@width\bframerule\@height\baselineskip\@depth\bframesep |
| 2727 |
\hskip-\bframesep |
| 2728 |
}% |
| 2729 |
\hbox{\hskip-\bframesep |
| 2730 |
\vrule\@width\@tempdima\@height\bframerule\@depth\z@}% |
| 2731 |
\nointerlineskip |
| 2732 |
\copy\bframebox |
| 2733 |
\nobreak |
| 2734 |
\kern-\baselineskip |
| 2735 |
\runninglinenumbers |
| 2736 |
\def\makeLineNumber{\copy\bframebox\hss}% |
| 2737 |
} |
| 2738 |
{\par |
| 2739 |
\ifdim\prevdepth=-1000pt \else |
| 2740 |
\kern-\prevdepth |
| 2741 |
\fi |
| 2742 |
\kern\bframesep |
| 2743 |
\nointerlineskip |
| 2744 |
\@tempdima\columnwidth |
| 2745 |
\advance\@tempdima 2\bframesep |
| 2746 |
\hbox{\hskip-\bframesep |
| 2747 |
\vrule\@width\@tempdima\@height\bframerule\@depth\z@}% |
| 2748 |
} |
| 2749 |
|
| 2750 |
\newdimen\bframerule |
| 2751 |
\bframerule=\fboxrule |
| 2752 |
|
| 2753 |
\newdimen\bframesep |
| 2754 |
\bframesep=\fboxsep |
| 2755 |
|
| 2756 |
\newbox\bframebox |
| 2757 |
\end{verbatim} |
| 2758 |
\end{code} |
| 2759 |
|
| 2760 |
\subsection{% |
| 2761 |
\textsf{amsmath} patches |
| 2762 |
\unskip}\label{ss:AMSMath} |
| 2763 |
|
| 2764 |
(New v5.0) Patches ~amsmath~ to work with ~lineno~. These patches used to be |
| 2765 |
supplied by the ~linenoamsmath~ package. See ~linenoamsmathdemo.pdf~ for a |
| 2766 |
demonstration. |
| 2767 |
(/New v5.0) |
| 2768 |
|
| 2769 |
(New v5.1) ~lineno~ tries to use \LaTeX's hook management system to patch |
| 2770 |
~amsmath~, so that the two packages may be loaded independently. This requires |
| 2771 |
the October 2020 release of \LaTeX. As a fallback for older releases, ~lineno~ |
| 2772 |
tests whether ~amsmath~ had already been loaded (by testing for the presence |
| 2773 |
of the ~gather~ command) and if so applies the patches; otherwise if ~amsmath~ |
| 2774 |
has not been loaded, no patches are applied, and a warning is issued. |
| 2775 |
(/New v5.1) |
| 2776 |
|
| 2777 |
(New v5.2) Fix ~lineno~ to work with ~amsmath~'s ~\allowdisplaybreaks~ |
| 2778 |
option. A side effect is that now ~\\*~ suppresses a line number on that |
| 2779 |
line. This is because ~\\*~ prohibits a page break after a given line, and |
| 2780 |
~lineno~ basically works by hijacking page breaks. It's probably not possible |
| 2781 |
to fix this without losing the behaviour of ~\\*~. |
| 2782 |
(/New v5.2) |
| 2783 |
|
| 2784 |
\begin{code}\begin{verbatim} |
| 2785 |
\ifdefined\AddToHook |
| 2786 |
\def\@LN@amsmath@patches#1{\AddToHook{package/amsmath/after}{#1}} |
| 2787 |
\else |
| 2788 |
\ifdefined\endgather |
| 2789 |
\def\@LN@amsmath@patches#1{#1} |
| 2790 |
\else |
| 2791 |
\PackageWarningNoLine{lineno}% |
| 2792 |
{`amsmath' must be loaded before `lineno' for patches to be applied} |
| 2793 |
\def\@LN@amsmath@patches#1{\relax} |
| 2794 |
\fi |
| 2795 |
\fi |
| 2796 |
|
| 2797 |
\@LN@amsmath@patches{ |
| 2798 |
\newcommand*\@LN@amsmath@patch[1]{% |
| 2799 |
\cspreto{#1}{\linenomath}% |
| 2800 |
\cspreto{#1*}{\linenomath}% |
| 2801 |
\csappto{end#1}{\endlinenomath}% |
| 2802 |
\csappto{end#1*}{\endlinenomath}% |
| 2803 |
} |
| 2804 |
\newcount\@LN@amsmath@ams@eqpen |
| 2805 |
\cspreto{math@cr@}{% |
| 2806 |
%%% Uncommenting the following 2 lines restores the line number on a line |
| 2807 |
%%% ended with \\*, by making \\* act just like \\. This is probably |
| 2808 |
%%% undesirable, however, so these lines are disabled. |
| 2809 |
% \global\@eqpen% |
| 2810 |
% \ifnum\dspbrk@lvl <\z@ \interdisplaylinepenalty \else -\@getpen\dspbrk@lvl \fi% |
| 2811 |
\advance\@eqpen\@LN@amsmath@ams@eqpen\relax% |
| 2812 |
} |
| 2813 |
\newcommand*\@LN@amsmath@patch@ams[1]{% |
| 2814 |
\cspreto{#1}{% |
| 2815 |
\linenomath% |
| 2816 |
\postdisplaypenalty=0% |
| 2817 |
\global\@LN@amsmath@ams@eqpen\interdisplaylinepenalty% |
| 2818 |
}% |
| 2819 |
\cspreto{#1*}{% |
| 2820 |
\linenomath% |
| 2821 |
\postdisplaypenalty=0% |
| 2822 |
\global\@LN@amsmath@ams@eqpen\interdisplaylinepenalty% |
| 2823 |
}% |
| 2824 |
\csappto{end#1}{% |
| 2825 |
\global\@LN@amsmath@ams@eqpen\z@% |
| 2826 |
\endlinenomath% |
| 2827 |
}% |
| 2828 |
\csappto{end#1*}{% |
| 2829 |
\global\@LN@amsmath@ams@eqpen\z@% |
| 2830 |
\endlinenomath% |
| 2831 |
}% |
| 2832 |
} |
| 2833 |
\@LN@amsmath@patch{equation} |
| 2834 |
\@LN@amsmath@patch@ams{multline} |
| 2835 |
\@LN@amsmath@patch@ams{gather} |
| 2836 |
\@LN@amsmath@patch@ams{align} |
| 2837 |
\@LN@amsmath@patch@ams{alignat} |
| 2838 |
\@LN@amsmath@patch@ams{flalign} |
| 2839 |
\let\@LN@amsmath@ams@mmeasure\mmeasure@ |
| 2840 |
\def\mmeasure@#1{% |
| 2841 |
\global\@LN@amsmath@ams@eqpen\z@% |
| 2842 |
\begingroup% |
| 2843 |
\interdisplaylinepenalty=0% |
| 2844 |
\@LN@amsmath@ams@mmeasure{#1\\}% |
| 2845 |
\endgroup% |
| 2846 |
\global\@LN@amsmath@ams@eqpen\interdisplaylinepenalty% |
| 2847 |
} |
| 2848 |
} |
| 2849 |
\end{verbatim} |
| 2850 |
\end{code} |
| 2851 |
|
| 2852 |
|
| 2853 |
\section{% |
| 2854 |
Move \scs{vadjust} items (New v4.00) |
| 2855 |
\unskip}\label{s:MVadj} |
| 2856 |
|
| 2857 |
This section completes reviving ~\pagebreak~, ~\nopagebreak~, |
| 2858 |
~\vspace~, and the star and optional form of ~\\~. This was |
| 2859 |
started in section_\ref{ss:output} and resumed in |
| 2860 |
section_\ref{ss:MLN} and subsection_\ref{ss:calls}. |
| 2861 |
The problem was explained in section_\ref{ss:output}: |
| 2862 |
~\vadjust~ items come out at a bad position, and the |
| 2863 |
\LaTeX\ commands named before work with ~\vadjust~ indeed. |
| 2864 |
Our solution was sketched there as well. |
| 2865 |
|
| 2866 |
According to the caveat in subsection_\ref{ss:OnOff} concerning |
| 2867 |
~\ifLineNumbers~, the \LaTeX\ commands enumerated may go |
| 2868 |
wrong if you switch line numbering inside or at the end of |
| 2869 |
a paragraph. |
| 2870 |
|
| 2871 |
\subsection{% |
| 2872 |
Redefining \scs{vadjust} |
| 2873 |
\unskip}\label{ss:PVadj} |
| 2874 |
|
| 2875 |
~\vadjust~ will temporarily be changed into the following |
| 2876 |
command. |
| 2877 |
\begin{code}\begin{verbatim} |
| 2878 |
\def\PostponeVadjust#1{% |
| 2879 |
\global\let\vadjust\@LN@@vadjust |
| 2880 |
\end{verbatim} |
| 2881 |
\end{code} |
| 2882 |
This undoes a ~\global\let\vadjust\PostponeVadjust~ which will |
| 2883 |
start each of the refined \LaTeX\ commands. The ~\global~s |
| 2884 |
are most probably superfluous. They might be useful should one |
| 2885 |
~\vadjust~ appear in a group starting after the change of |
| 2886 |
~\vadjust~ into ~\PostponeVadjust~. |
| 2887 |
(UL) Even the undoing may be superfluous, cf._discussion |
| 2888 |
in section_\ref{ss:ReDef} below. (UL) |
| 2889 |
\begin{code}\begin{verbatim} |
| 2890 |
\vadjust{\penalty-\@Mppvacodepen}% |
| 2891 |
\g@addto@macro\@LN@vadjustlist{#1\@lt}% |
| 2892 |
} |
| 2893 |
\let\@LN@@vadjust\vadjust |
| 2894 |
\global\let\@LN@vadjustlist\@empty |
| 2895 |
\global\let\@LN@do@vadjusts\relax |
| 2896 |
\end{verbatim} |
| 2897 |
\end{code} |
| 2898 |
These ~\global~s are just to remind that |
| 2899 |
all the changes of the strings after ~\let~ should be |
| 2900 |
~\global~ (\TeX book p._301). ~\@LN@vadjustlist~ collects |
| 2901 |
the ~\vadjust~ items of a paragraph. ~\PassVadjustList~ |
| 2902 |
tears one ~\vadjust~ item for the current line out of |
| 2903 |
~\@LN@vadjustlist~ and puts it into ~\@LN@do@vadjusts~. |
| 2904 |
The latter is encountered each line in ~\MakeLineNo~ |
| 2905 |
(section_\ref{ss:MLN}), while those \LaTeX\ ~\vadjust~ |
| 2906 |
commands will come rather rarely. So I decided that |
| 2907 |
~\@LN@do@vadjust~ is ~\relax~ until a ~\vadjust~ item |
| 2908 |
is waiting. In the latter case, ~\@LN@do@vadjusts~ |
| 2909 |
is turned into a list macro which resets itself to |
| 2910 |
~\relax~ when the other contents have been placed in |
| 2911 |
the vertical list.---~\PassVadjustList~ is invoked by |
| 2912 |
the output routine (section_\ref{ss:output}), so the |
| 2913 |
~\box255~ must be put back. |
| 2914 |
\begin{code}\begin{verbatim} |
| 2915 |
\def\PassVadjustList{% |
| 2916 |
\unvbox\@cclv |
| 2917 |
\expandafter \@LN@xnext \@LN@vadjustlist \@@ |
| 2918 |
\@tempa \@LN@vadjustlist |
| 2919 |
\ifx\@LN@do@vadjusts\relax |
| 2920 |
\gdef\@LN@do@vadjusts{\global\let\@LN@do@vadjusts\relax}% |
| 2921 |
\fi |
| 2922 |
\expandafter \g@addto@macro \expandafter \@LN@do@vadjusts |
| 2923 |
\expandafter {\@tempa}% |
| 2924 |
} |
| 2925 |
\end{verbatim} |
| 2926 |
\end{code} |
| 2927 |
|
| 2928 |
\subsection{% |
| 2929 |
Redefining the \LaTeX\ commands |
| 2930 |
\unskip}\label{ss:ReDef} |
| 2931 |
|
| 2932 |
Now we change ~\pagebreak~ etc.\ |
| 2933 |
so that they use ~\PostponeVadjust~ in place of ~\vadjust~. |
| 2934 |
We try to do this as independently as possible of the |
| 2935 |
implementation of the \LaTeX\ commands to be redefined. |
| 2936 |
Therefore, we don't just copy macro definition code from any |
| 2937 |
single implementation (say, latest \LaTeX) and insert our |
| 2938 |
changes, but attach a conditional |
| 2939 |
~\global\let\vadjust\PostponeVadjust~ |
| 2940 |
to their left ends in a way which should work rather |
| 2941 |
independantly of their actual code. |
| 2942 |
However, ~\vadjust~ should be the primitive again after |
| 2943 |
execution of the command. So the ~\global\let...~ may be used |
| 2944 |
only if it's guaranteed that a ~\vadjust~ is near.---(UL) |
| 2945 |
Sure? In line numbering mode, probably each ~\vadjust~ |
| 2946 |
coming from a \LaTeX\ command should be ~\PostponeVadjust~. |
| 2947 |
~\marginpar~s and floats seem to be the only cases which |
| 2948 |
are not explicitly dealt with in the present section. |
| 2949 |
This would be a way to avoid ~\@LN@nobreaktrue~! |
| 2950 |
Of course, the ~\vadjust~s that the present package uses |
| 2951 |
then must be replaced by ~\@LN@@vadjust~.---Maybe |
| 2952 |
next time. (/UL) |
| 2953 |
|
| 2954 |
The next command and something else will be added to the |
| 2955 |
\LaTeX\ commands we are concerned with here. |
| 2956 |
\begin{code}\begin{verbatim} |
| 2957 |
\DeclareRobustCommand\@LN@changevadjust{% |
| 2958 |
\ifvmode\else\ifinner\else |
| 2959 |
\global\let\vadjust\PostponeVadjust |
| 2960 |
\fi\fi |
| 2961 |
} |
| 2962 |
\end{verbatim} |
| 2963 |
\end{code} |
| 2964 |
(UL) What about math mode? Math display? Warn? (/UL) |
| 2965 |
|
| 2966 |
~\@tempa~ will now become a two place macro which adds first |
| 2967 |
argument (single token), enclosed by ~\ifLineNumbers~\,\dots |
| 2968 |
~\fi~ to the left of second argument. As long as we need it, |
| 2969 |
we can't use the star form of ~\DeclareRobustCommand~ or |
| 2970 |
the like, because AMS-\LaTeX\ uses ~\@tempa~ for ~\@ifstar~. |
| 2971 |
(New v4.41) And for the same reason, that ~\CheckCommand*~ |
| 2972 |
had to be raised! (/New v4.41) |
| 2973 |
\begin{code}\begin{verbatim} |
| 2974 |
\CheckCommand*\@parboxrestore{\@arrayparboxrestore\let\\\@normalcr} |
| 2975 |
|
| 2976 |
\def\@tempa#1#2{% |
| 2977 |
\expandafter \def \expandafter#2\expandafter{\expandafter |
| 2978 |
\ifLineNumbers\expandafter#1\expandafter\fi#2}% |
| 2979 |
} |
| 2980 |
\end{verbatim} |
| 2981 |
\end{code} |
| 2982 |
(UL) This ~\ifLineNumber~ can be fooled by |
| 2983 |
~\linenumbers~ ahead etc. It might be better to place |
| 2984 |
a signal penalty in any case and let the output routine |
| 2985 |
decide what to do. |
| 2986 |
(/UL) |
| 2987 |
|
| 2988 |
We use the occasion to switch off linenumbers where they |
| 2989 |
don't work anyway and where we don't want them, |
| 2990 |
especially in footnotes: |
| 2991 |
\begin{code}\begin{verbatim} |
| 2992 |
\@tempa\nolinenumbers\@arrayparboxrestore |
| 2993 |
\end{verbatim} |
| 2994 |
\end{code} |
| 2995 |
We hope this suffices $\dots$ let's check one thing |
| 2996 |
at least: [(New v4.41) see ~\CheckCommand~ above (/New v4.41)] |
| 2997 |
|
| 2998 |
Now for the main theme of the section. |
| 2999 |
The next lines assume that ~\vspace~, ~\pagebreak~, and |
| 3000 |
~\nopagebreak~ use ~\vadjust~ whenever they occur outside |
| 3001 |
vertical mode; moreover, that they don't directly read |
| 3002 |
an argument. Indeed ~\pagebreak~ and ~\nopagebreak~ first |
| 3003 |
call something which tests for a left bracket ahead, |
| 3004 |
while ~\vspace~ first tests for a star. |
| 3005 |
\begin{code}\begin{verbatim} |
| 3006 |
\@tempa\@LN@changevadjust\vspace |
| 3007 |
\@tempa\@LN@changevadjust\pagebreak |
| 3008 |
\@tempa\@LN@changevadjust\nopagebreak |
| 3009 |
\end{verbatim} |
| 3010 |
\end{code} |
| 3011 |
~\\~, however, uses ~\vadjust~ only in star or optional form. |
| 3012 |
We relax independency of implementation in assuming |
| 3013 |
that ~\@normalcr~ is the fragile version of ~\\~ |
| 3014 |
(and we use ~\@ifstar~!). |
| 3015 |
(Using a copy of ~\\~ would be safer, but an ugly repetition |
| 3016 |
of ~\protect~.) |
| 3017 |
\begin{code}\begin{verbatim} |
| 3018 |
\protected\def\\{% |
| 3019 |
\ifLineNumbers |
| 3020 |
\expandafter \@LN@cr |
| 3021 |
\else |
| 3022 |
\expandafter \@normalcr |
| 3023 |
\fi |
| 3024 |
} |
| 3025 |
\def\@LN@cr{% |
| 3026 |
\@ifstar{\@LN@changevadjust\@normalcr*}% |
| 3027 |
{\@ifnextchar[{\@LN@changevadjust\@normalcr}\@normalcr}% |
| 3028 |
} |
| 3029 |
\end{verbatim} |
| 3030 |
\end{code} |
| 3031 |
Moreover we hope that ~\newline~ never leads to a ~\vadjust~, |
| 3032 |
although names of some commands invoked by ~\\~ contain |
| 3033 |
~newline~. At last, this seems to have been OK since 1989 or |
| 3034 |
even earlier. |
| 3035 |
|
| 3036 |
\modulolinenumbers[1] |
| 3037 |
\firstlinenumber{0} |
| 3038 |
Let's have a few tests.\vspace*{.5\baselineskip} |
| 3039 |
Testing ~\pagebreak~ and ~\nopagebreak~ would be too expensive |
| 3040 |
here, but---oops!---we have just experienced a successful |
| 3041 |
~\vspace*{.5\baselineskip}~. A |
| 3042 |
~\\*[.5\baselineskip]~\\*[.5\baselineskip] may look even more |
| 3043 |
drastical, but this time we are happy about it. Note that the |
| 3044 |
line numbers have moved with the lines. Without our changes, |
| 3045 |
one line number\vadjust{\kern.5\baselineskip} would have |
| 3046 |
``anticipated'' the move of the next line, just as you can |
| 3047 |
observe it now. |
| 3048 |
(/New v4.00) |
| 3049 |
|
| 3050 |
\switchlinenumbers |
| 3051 |
|
| 3052 |
\subsection{% |
| 3053 |
Reminder on obsoleteness |
| 3054 |
\unskip} |
| 3055 |
|
| 3056 |
(New v4.1) We have completed inclusion of the earlier |
| 3057 |
extension packages ~linenox0.sty~, ~linenox1.sty~, ~lnopatch.sty~, |
| 3058 |
and ~linenoamsmath~. If one of them is loaded, though, |
| 3059 |
we produce an error message before something weird happens. |
| 3060 |
We avoid ~\newif~ because the switchings occur so rarely. |
| 3061 |
\begin{code}\begin{verbatim} |
| 3062 |
\AtBeginDocument{% |
| 3063 |
\let\if@LN@obsolete\iffalse |
| 3064 |
\@ifpackageloaded{linenox0}{\let\if@LN@obsolete\iftrue}\relax |
| 3065 |
\@ifpackageloaded{linenox1}{\let\if@LN@obsolete\iftrue}\relax |
| 3066 |
\@ifpackageloaded{lnopatch}{\let\if@LN@obsolete\iftrue}\relax |
| 3067 |
\@ifpackageloaded{linenoamsmath}{\let\if@LN@obsolete\iftrue}\relax |
| 3068 |
\if@LN@obsolete |
| 3069 |
\PackageError{lineno}{Obsolete extension package(s)}{% |
| 3070 |
As of \fileversion, `lineno' includes the functionality of \MessageBreak |
| 3071 |
`linenox0', `linenox1', `lnopatch', and `linenoamsmath'; \MessageBreak |
| 3072 |
these packages are therefore obsolete and must not be loaded.}% |
| 3073 |
\fi |
| 3074 |
} |
| 3075 |
\end{verbatim} |
| 3076 |
\end{code} |
| 3077 |
|
| 3078 |
\modulolinenumbers[1] |
| 3079 |
\section{% |
| 3080 |
The final touch |
| 3081 |
\unskip} |
| 3082 |
|
| 3083 |
There is one deadcycle for each line number. |
| 3084 |
\begin{code}\begin{verbatim} |
| 3085 |
\advance\maxdeadcycles 100 |
| 3086 |
|
| 3087 |
\endinput |
| 3088 |
\end{verbatim} |
| 3089 |
\end{code} |
| 3090 |
|
| 3091 |
\section{% |
| 3092 |
The user commands |
| 3093 |
\unskip}\label{s:UserCmds} |
| 3094 |
|
| 3095 |
The user commands to turn on and off line numbering |
| 3096 |
are |
| 3097 |
\begin{description}\item |
| 3098 |
[|\linenumbers] \ \par |
| 3099 |
Turn on line numbering in the current mode. |
| 3100 |
|
| 3101 |
\item |
| 3102 |
[|\linenumbers*] \ \par$\qquad$ |
| 3103 |
and reset the line number to 1. |
| 3104 |
\def\NL{\angle{number}]}\item |
| 3105 |
[|\linenumbers[\NL] \ \par$\qquad$ |
| 3106 |
and start with \angle{number}. |
| 3107 |
\item |
| 3108 |
[|\nolinenumbers] \ \par |
| 3109 |
Turn off line numbering. |
| 3110 |
\item |
| 3111 |
[|\runninglinenumbers*[\NL] \ \par |
| 3112 |
Turn on ~running~ line numbers, with the same optional |
| 3113 |
arguments as ~\linenumbers~. The numbers are running |
| 3114 |
through the text over pagebreaks. When you turn |
| 3115 |
numbering off and on again, the numbers will continue, |
| 3116 |
except, of cause, if you ask to reset or preset the |
| 3117 |
counter. |
| 3118 |
\item |
| 3119 |
[|\pagewiselinenumbers] \ \par |
| 3120 |
Turn on ~pagewise~ line numbers. The lines on each |
| 3121 |
page are numbered beginning with one at the first |
| 3122 |
~pagewise~ numbered line. |
| 3123 |
\item |
| 3124 |
[|\resetlinenumber[\NL] \ \par |
| 3125 |
Reset ~[~Set~]~ the line number to 1 |
| 3126 |
~[~\angle{number}~]~. |
| 3127 |
\item |
| 3128 |
[|\setrunninglinenumbers] \ \par |
| 3129 |
Switch to ~running~ line number mode. Do \emph{not} |
| 3130 |
turn it on or off. |
| 3131 |
\item |
| 3132 |
[|\setpagewiselinenumbers] \ \par |
| 3133 |
Switch to ~pagewise~ line number mode. Do \emph{not} |
| 3134 |
turn it on or off. |
| 3135 |
\item |
| 3136 |
[|\switchlinenumbers*] \ \par |
| 3137 |
Causes margin switching in pagewise modes. With the |
| 3138 |
star, put the line numbers on the inner margin. |
| 3139 |
\item |
| 3140 |
[|\leftlinenumbers*] \ \par |
| 3141 |
\item |
| 3142 |
[|\rightlinenumbers*] \ \par |
| 3143 |
Set the line numbers in the left/right margin. With the |
| 3144 |
star this works for both modes of operation, without |
| 3145 |
the star only for the currently selected mode. |
| 3146 |
\item |
| 3147 |
[|\runningpagewiselinenumbers] \ \par |
| 3148 |
When using the pagewise line number mode, do not |
| 3149 |
subtract the page offset. This results in running |
| 3150 |
line numbers again, but with the possibility to switch |
| 3151 |
margins. Be careful when doing line number |
| 3152 |
referencing, this mode status must be the same while |
| 3153 |
setting the paragraph and during references. |
| 3154 |
\item |
| 3155 |
[|\realpagewiselinenumbers] \ \par |
| 3156 |
Reverses the effect of ~\runningpagewiselinenumbers~. |
| 3157 |
\item |
| 3158 |
[|\modulolinenumbers[\NL] \ \par |
| 3159 |
Give a number only to lines which are multiples of |
| 3160 |
~[~\angle{number}~]~. If \angle{number} is not specified, the |
| 3161 |
current value in the counter ~linenumbermodulo~ is |
| 3162 |
retained. \angle{number}=1 turns this off without changing |
| 3163 |
~linenumbermodulo~. The counter is initialized to 5. |
| 3164 |
\item |
| 3165 |
[|\modulolinenumbers*[\NL] \ \par |
| 3166 |
Like ~\modulolinenumbers~, the only difference being |
| 3167 |
that the first line number after a ~\linenumbers~ |
| 3168 |
(or ~\runninglinenumbers~, ~\pagewiselinenumbers~, |
| 3169 |
~\quotelinenumbers~) is printed regardless of the |
| 3170 |
modulo---yet `1' is printed only after (or \dots) |
| 3171 |
~\firstlinenumber{1}~. |
| 3172 |
This also applies to the first line of a |
| 3173 |
~{linenumbers}~ or respective environment. |
| 3174 |
See subsection_\ref{ss:Mod} for another explanation. |
| 3175 |
The behaviour may be unsatisfactory with pagewise |
| 3176 |
line-numbering. |
| 3177 |
\item |
| 3178 |
[|\firstlinenumber] \ \par |
| 3179 |
~\firstlinenumber{~\angle{filino}~}~ brings about that |
| 3180 |
(after it) line numbers less than \angle{filino} do |
| 3181 |
\emph{not} appear in the margin. Moreover, with |
| 3182 |
~\modulolinenumbers[~\angle{number}~]~, just the line |
| 3183 |
numbers which are \angle{filino} plus a multiple of |
| 3184 |
\angle{number} are printed.---If you had |
| 3185 |
~\firstlinenumber{~\angle{pos}~}~ with some $\mbox{\angle{pos}}>0$ |
| 3186 |
and want to switch to printing multiples of, e.g., |
| 3187 |
4, you best do ~\modulolinenumbers[4]~ and |
| 3188 |
~\firstlinenumber{0}~. (See subsection_\ref{ss:Mod} |
| 3189 |
for technical details.) |
| 3190 |
\item |
| 3191 |
[|\linenumberdisplaymath] \ \par |
| 3192 |
Number the lines of a display math in a ~{linenomath}~ |
| 3193 |
environment, but do not in a ~{linenomath*}~ |
| 3194 |
environment. This is used by the package option |
| 3195 |
~[mathlines]~. |
| 3196 |
\item |
| 3197 |
[|\nolinenumberdisplaymath] \ \par |
| 3198 |
Do not Number the lines of a display math in a |
| 3199 |
~{linenomath}~ environment, but do in a |
| 3200 |
~{linenomath*}~ environment. This is the default. |
| 3201 |
\item |
| 3202 |
[|\linelabel] \ \par |
| 3203 |
Set a ~\linelabel{~\angle{foo}~}~ to the line number where |
| 3204 |
this commands is in. Refer to it with the \LaTeX\ |
| 3205 |
referencing commands ~\ref{~\angle{foo}~}~ and |
| 3206 |
~\pageref{~\angle{foo}~}~. |
| 3207 |
\end{description} |
| 3208 |
The commands can be used globally, locally within groups |
| 3209 |
or as environments. It is important to know that they |
| 3210 |
take action only when the ~\par~ is executed. The |
| 3211 |
~\end{~\angle{mode}~linenumbers}~ commands provide a ~\par~. |
| 3212 |
Examples: |
| 3213 |
\begin{verse} |
| 3214 |
~{\linenumbers~ \angle{text} ~\par}~ \\ |
| 3215 |
\ \\ |
| 3216 |
~\begin{linenumbers}~ \\ |
| 3217 |
\angle{text} \\ |
| 3218 |
~\end{linenumbers}~ \\ |
| 3219 |
\ \\ |
| 3220 |
\angle{paragraph} ~{\linenumbers\par}~ \\ |
| 3221 |
\ \\ |
| 3222 |
~\linenumbers~ \\ |
| 3223 |
\angle{text} ~\par~ \\ |
| 3224 |
~\nolinenumbers~ \\ |
| 3225 |
\ \\ |
| 3226 |
~\linenumbers~ \\ |
| 3227 |
\angle{paragraph} ~{\nolinenumbers\par}~ \\ |
| 3228 |
\end{verse} |
| 3229 |
(New v4.00) |
| 3230 |
However, the examples containing \angle{paragraph} show what you |
| 3231 |
should \emph{not} do, at least if you use ~\pagebreak~, |
| 3232 |
~\nopagebreak~, ~\vspace~, ~\\*~ or |
| 3233 |
~\\[~\angle{space}~]~---cf._section_\ref{s:MVadj}. |
| 3234 |
|
| 3235 |
The same care should be applied to the ``wizard'' devices |
| 3236 |
~\ifLineNumbers~ (subsection_\ref{ss:OnOff}) and |
| 3237 |
~\PostponeVadjust~ (section_\ref{ss:PVadj}). |
| 3238 |
(/New v4.00) |
| 3239 |
|
| 3240 |
(New v4.11) Oh, and the commands and environments of |
| 3241 |
section_{s:Xt} are missing. Sorry, I am in a hurry now. |
| 3242 |
May be next time.% %% TODO v4.4+ |
| 3243 |
---And the environments ~{linenomath}~ and ~{linenomath*}~should |
| 3244 |
get an own paragraph. In short, each math display, equation, |
| 3245 |
or ~{eqnarray}~ should be ``wrapped'' in one of ~{linenomath}~ |
| 3246 |
and ~{linenomath*}~. |
| 3247 |
|
| 3248 |
\subsection{% |
| 3249 |
Customization hooks |
| 3250 |
\unskip}\label{s:CustHooks} |
| 3251 |
|
| 3252 |
There are several hooks to customize the appearance of the |
| 3253 |
line numbers, and some low level hooks for special |
| 3254 |
effects. |
| 3255 |
\begin{description}\item |
| 3256 |
[|\thelinenumber] \ \par |
| 3257 |
This macro should give the representation of the line |
| 3258 |
number in the \LaTeX-counter ~linenumber~. The |
| 3259 |
default is provided by \LaTeX: \par$\qquad$ |
| 3260 |
~\arabic{linenumber}~ |
| 3261 |
\item |
| 3262 |
[|\makeLineNumberLeft] \ \par |
| 3263 |
This macro is used to attach a line number to the left |
| 3264 |
of the text page. This macro should fill an ~\hbox to 0pt~ |
| 3265 |
which will be placed at the left margin of the |
| 3266 |
page, with the reference point aligned to the line to |
| 3267 |
which it should give a number. Please use the macro |
| 3268 |
~\LineNumber~ to refer to the line number. |
| 3269 |
|
| 3270 |
The default definition is \par$\qquad$ |
| 3271 |
~\hss\linenumberfont\LineNumber\hskip\linenumbersep~ |
| 3272 |
\item |
| 3273 |
[|\makeLineNumberRight] \ \par |
| 3274 |
Like ~\makeLineNumberLeft~, but for line numbers on |
| 3275 |
the right margin. |
| 3276 |
|
| 3277 |
The default definition is \par$\qquad$ |
| 3278 |
~\linenumberfont\hskip\linenumbersep\hskip\columnwidth~ \par$\qquad$ |
| 3279 |
~\hbox to\linenumberwidth{\hss\LineNumber}\hss~ |
| 3280 |
\item |
| 3281 |
[|\linenumberfont] \ \par |
| 3282 |
This macro is initialized to \par$\qquad$ |
| 3283 |
~\normalfont\tiny\sffamily~ |
| 3284 |
\item |
| 3285 |
[|\linenumbersep] \ \par |
| 3286 |
This dimension register sets the separation of the |
| 3287 |
line number to the text. Default value is ~10pt~. |
| 3288 |
\item |
| 3289 |
[|\linenumberwidth] \ \par |
| 3290 |
This dimension register sets the width of the line |
| 3291 |
number box on the right margin. The distance of the |
| 3292 |
right edge of the text to the right edge of the line |
| 3293 |
number is ~\linenumbersep~ + ~\linenumberwidth~. The |
| 3294 |
default value is ~10pt~. |
| 3295 |
\item |
| 3296 |
[|\theLineNumber] (for wizards) \ \par |
| 3297 |
This macro is called for printing a ~\newlabel~ entry |
| 3298 |
to the aux-file. Its definition depends on the mode. |
| 3299 |
For running line numbers it's just ~\thelinenumber~, |
| 3300 |
while in pagewise mode, the page offset subtraction |
| 3301 |
is done in here. |
| 3302 |
\item |
| 3303 |
[|\makeLineNumber] (for wizards) \ \par |
| 3304 |
This macro produces the line numbers. The definition |
| 3305 |
depends on the mode. In the running line numbers |
| 3306 |
mode it just expands ~\makeLineNumberLeft~. |
| 3307 |
\item |
| 3308 |
[|\LineNumber] (for wizards) \ \par |
| 3309 |
This macro is called by ~\makeLineNumber~ to typeset |
| 3310 |
the line number. This hook is changed by the modulo |
| 3311 |
mechanism |
| 3312 |
and by ~\firstlinenumber~. |
| 3313 |
\end{description} |
| 3314 |
\end{document}%D |