%Copyright (c) Denis Howe 1992 %You may copy this file freely but you must not charge for it and %you must not remove this notice. Please send me a copy of any %modifications you make. Denis Howe \documentstyle{slides} \begin{document} \pagestyle{plain} %The rest up to before \end{document} should really go %in a separate file, say "slides.tex", included with %\blackandwhite{slides} \begin{slide}{} \centering Introduction to GNU Emacs \vspace{3cm} Denis Howe {\tt dbh@doc.ic.ac.uk} \end{slide} \begin{slide}{} GNU Emacs The {\em Free Software Foundation}'s editor \begin{itemize} \item extensible -- add new functions \item customizable -- rebind keys, hooks \item self-documenting -- on-line context-sensitive help \item real-time display -- WYSIWYG ({\it cf} YAFIYGI) \end{itemize} \end{slide} \begin{slide}{} \fbox{\parbox{16cm}{\tt This is an example screen. \\ The user has just typed \\ \^{}X\^{}F \\ \rule{0cm}{10cm} -\,-\,**\,-Emacs:~prog.c~~~~(C)\,-\,-\,-\,23\%-\,-\,-\,-\,-\,-\,- \\ Find file: /homes/dbh/src/\rule[-0.2ex]{0.4em}{2ex}}} \end{slide} \begin{slide}{} Basic concepts \begin{description} \item[Buffer] - A collection of lines (from a file) stored in memory \item[Window] - An area of the screen showing the contents of a buffer \item[Mode line] - Status line at bottom of window showing file name, position, modification status, buffer modes, time \item[Minibuffer/echo area] - A one-line buffer \\ below mode line for displaying messages and reading parameters \item[Point] - Current (cursor) position in a buffer \end{description} \end{slide} \begin{slide}{} Key sequences Eg. C-x C-f = Control-x Control-f = find-file Keys are: \begin{description} \item[Control characters] : \\ C-@ C-a .. C-z TAB LFD RET ESC DEL \item[Printing characters] : \\ SPC ! .. 0 .. 9 .. @ A~..~Z .. a .. z .. \~{} \item[Meta key] $\Diamond$ on Sun \end{description} \end{slide} \begin{slide}{} Key sequences \vspace{3cm} {\em Prefix} keys: ESC C-x C-h C-c \\ start a key sequence Eg. ESC f (two key sequence) \\ or equivalently, M-f (hold Meta key and hit F) Key sequences execute {\em commands} \\ eg. forward-word Characters are displayed as, eg. \\ \^{}@ \^{}A \^{}[ \^{}? $\backslash$ 234 \end{slide} \begin{slide}{} Starting Emacs from the shell Try one of {\tt epoch}, {\tt emacs}, {\tt gnu}, {\tt gnuemacs} \\ or {\tt uemacs} (micro-emacs -- not GNU emacs). Command line arguments \begin{description} \item[-nw] ignore window system (eg. X-windows) \item[-q] ignore .emacs file \item[\em files] to load and edit \end{description} Exiting Emacs C-x C-c (save-buffers-kill-emacs) \end{slide} \begin{slide}{} Basic editing commands Printing characters execute the command \\ self-insert-command Moving around the buffer \begin{verbatim} top of buffer M-< previous screen M-v previous line C-p C-a M-b C-b C-f M-f C-e start back back forw forw end of word char char word of line line next line C-n next screen C-v end of buf M-> \end{verbatim} \end{slide} \begin{slide}{} Advanced editing commands \vspace{5cm} \centering \LARGE C-h \vspace{5cm} \end{slide} \begin{slide}{} More advanced editing commands Numeric prefix argument. \\ Eg. ESC 7 C-n \\ Repeat C-n 7 times -- down 7 lines \begin{description} \item[M-x] (execute-extended-command) \\ Read a command name from the \\ minibuffer and execute it \end{description} Eg. M-x goto-line \\ Move cursor to line number $N$. Line 1 is the beginning of the buffer Line number is either prefix argument (ESC $N$ M-x goto-line) or read from minibuffer \end{slide} \begin{slide}{} The Minibuffer For input and editing of parameters, eg.\ name of file to load, and for displaying messages Provides {\em completion} of file-names, command names, Lisp symbols etc. \begin{description} \item[C-g] Quit minibuffer input or other activity \item[TAB, RET] Complete as much as possible \item[SPC] Complete current word \item[?] Show all completions \end{description} \end{slide} \begin{slide}{} On-line help \vspace{3cm} C-h followed by: \begin{tabular}{cll} A & apropos & search for help on a word \\ F & function & describe a Lisp function \\ I & info & documentation browser \\ K & key & describe a key sequence \\ V & variable & describe a Lisp variable \\ T & tutorial & read about and try commands \end{tabular} \end{slide} \begin{slide}{} Correcting text {\em Killed} (deleted) text is saved for later \\ re-insertion \begin{description} \item[DEL, C-d] -- Delete character before/after \\ cursor \item[M-DEL, M-d] -- Kill word before/after \\ cursor \item[C-k] -- Kill to end of line \item[C-t, M-t, C-x C-t] -- Swap characters/ \\ words/lines before and after the cursor \item[M-l, M-u, M-c] -- Convert word to lower case, upper case, initial capital \item[C-\_] -- Undo (a sequence of) changes \end{description} \end{slide} \begin{slide}{} Search \begin{description} \item[C-s] (isearch-forward) \item[C-r] (isearch-backward) \end{description} Search happens as you type the string \begin{description} \item[DEL] Return to last position matching what's left \item[ESC] Stop search, stay at found string \item[C-g] Quit and return to start of search \item[C-w] Add rest of found word to search string \item[C-q] Search for control char \end{description} \end{slide} \begin{slide}{} Search Search string is remembered \begin{description} \item[C-s C-s] searches again \item[M-C-s] (isearch-forward-regex) \\ Search for a {\em regular expression} \end{description} Replace \begin{description} \item[M-\%] (query-replace) Type {\tt y} or {\tt n} for \\ each occurrence of $FROM$ string \item[M-x query-replace-regexp] Search for \\ regular expression to replace \end{description} \end{slide} \begin{slide}{} The Mark and the Region \begin{description} \item[C-@, C-SPC] (set-mark-command) set \\ {\em mark} at current position \end{description} Text between mark and point is {\em the region} \begin{description} \item[C-w] (kill-region) kills text between point and mark ("cut") \item[C-y] (yank) Insert last killed text at point \\ ("paste") \item[M-g] (fill-region) join/split short/long lines \end{description} \end{slide} \begin{slide}{} Spelling correction \begin{description} \item[M-\$] (spell-word) Correct spelling of word \\ before point \item[M-x spell-buffer] Correct spelling of each \\ word in the buffer \item[M-/] (dabbrev-expand) search buffer for an expansion of the abbreviation before point \end{description} \end{slide} \begin{slide}{} Files and buffers \begin{description} \item[C-x C-f] (find-file) {\em Visit} a file in its own buffer. Set default directory for completion to file's directory. \item[C-x C-s] (save-buffer) Save the current buffer in the file it is visiting. \item[M-x revert-buffer] Re-read file from disc \item[C-x b $BUFFER$] (switch-to-buffer) Select or create a buffer named $BUFFER$ \item[C-x k] (kill-buffer) Kill one buffer \item[C-x C-b] (list-buffers) \end{description} \end{slide} \begin{slide}{} Dired mode \vspace{3cm} C-x C-f (find-file) on a directory shows output of \ {\tt ls -l} \ in a buffer. Provides commands to visit, delete, copy, \\ rename directory entries. \end{slide} \begin{slide}{} Windows You can have multiple windows in one emacs screen. Each window has its own \begin{itemize} \item buffer \item point \item mode line \item mode \item local variables (eg. {\tt fill-column}) \end{itemize} One window is selected (has the cursor) \end{slide} \begin{slide}{} Window commands \vspace{3cm} \begin{description} \item[C-x 2] (split-window-vertically) Split the \\ selected window into two windows, \\ one above the other \item[C-x o] (other-window) \\ Select the next window \item[C-x 0] (kill-window) \\ Get rid of the selected window \end{description} \end{slide} \begin{slide}{} Major modes {\em Major} modes are mutually exclusive. The \\ current major mode appears in () in the mode-line. Some keys are redefined for each mode. Eg. \begin{description} \item Lisp mode -- bracket matching \item C mode -- understands C syntax \item Mail mode -- read and send mail \item GNUS modes -- read and post news \item Shell mode -- execute shell commands \end{description} \end{slide} \begin{slide}{} Minor modes \vspace{3cm} {\em Minor} modes can be combined \begin{description} \item auto-fill-mode -- Insert newline when you type a SPC beyond {\tt fill-column} \item abbrev-mode -- Expand abbreviations when you type a SPC after one \end{description} \end{slide} \begin{slide}{} Editing programs \begin{description} \item[M-.] (find-tag) \item[M-,] (tags-loop-continue) \\ Visit file and show function definition. \\ Uses pre-compiled TAGS file. \item[M-x compile] and \item[M-x grep] Run a Unix command \\ asynchronously. Save output in a buffer. \item[C-x \`{}] (next-error) Visit the cause of the \\ next error message \end{description} \end{slide} \begin{slide}{} The Init File, .emacs Contains Lisp commands to execute \\ on start-up. Eg. \tt \begin{verbatim} (setq fill-column 60) (display-time) (setq TeX-dvi-print-command "xdvi ") (autoload 'ispell-word "ispell" "" t) (setq text-mode-hook (function (lambda () (auto-fill-mode 1)))) (setq c-mode-hook (function (lambda () (setq comment-column 40)))) (define-key esc-map "[195z" 'undo) (setq search-exit-char 0) \end{verbatim} \end{slide} \begin{slide}{} Defining commands A new command: \begin{verbatim} (defun ;define function kill-whole-line ;its name (n) ;its arguments "Kill line point is on" ;documentation ;begining of body (interactive "*p") ;how to interpret args (beginning-of-line) ;move to start of line (kill-line n)) ;kill n whole lines \end{verbatim} Execute with {\tt M-x kill-whole-line} \\ or bind to a key: {\tt\begin{verbatim} (define-key esc-map "k" 'kill-whole-line) \end{verbatim}} Note ' to quote command name \end{slide} \begin{slide}{} Defining commands {\tt interactive} declares that a function can be executed as an interactive command (via M-x or a key binding) It also says what arguments are passed: \begin{tabular}{ll} \tt (interactive) & no arguments \\ \tt (interactive "p") & prefix argument \\ & converted to a number \\ \tt (interactive "f") & filename read \\ & from minibuffer \\ \tt (interactive "r") & point and mark as \\ & numbers (the region) \end{tabular} See {\tt /usr/local/emacs/lisp/*.el} for examples \end{slide} \begin{slide}{} Some useful functions {\tt (message "It works!")(sit-for 1)} \\ Print a debugging message {\tt (buffer-substring beg end)} \\ Return contents of buffer between \\ two positions {\tt (re-search-forward "e..c[sh]")} \\ Search buffer for regular expression match {\tt (insert "your card please")} \\ Insert some text at point {\tt (y-or-n-p "Reply to Cc:\ recipients?\ ")} \\ Ask a yes/no question \end{slide} \begin{slide}{} Further information Manual page: {\tt man emacs} -- start-up and \\ X-windows options Newsgroup: {\tt gnu.emacs.help} -- general \\ queries about use and programming Lisp Code Direcory (LCD) Database of 500+ Emacs elisp programs FTP from {\tt tut.cis.ohio-state.edu} \\ {\tt pub/gnu/emacs/elisp-archive/LCD-datafile.Z} {\tt lispdir.el} -- an elisp package for locating elisp packages \end{slide} \begin{slide}{} Sed --- {\em S}\/tream {\em ed}\/itor Replace strings matching regular expressions Useful for building tools and reformating data Eg. \begin{verbatim} sed 's/IC/ICSTM/' cv > new_cv \end{verbatim} Replace all occurrences of ``IC'' with ``ICSTM'' \begin{verbatim} sed 's/0\(1-58[0-9]\)/07\1/g' \end{verbatim} Replace ``01-58$d$'' with ``071-58$d$'' where $d$ is any decimal digit \end{slide} \begin{slide}{} Regular Expressions Ordinary characters match themselves, \\ special characters have the following meanings (unless quoted with $\backslash$ ): \begin{description} \item[.] Matches any character except NEWLINE \item[\^{}] At start of RE matches start of line \item[\$] At end of RE matches end of line \item[[abcdef]] Matches any one of the characters {\tt a b c d e} or {\tt f} \item[[\^{}abcdef]] Matches any one character {\em except} {\tt a b c d e} or {\tt f} \item[[A-Z\_{}]] Matches any one character in the range {\tt A \ldots Z} or {\tt \_{}} \end{description} \end{slide} \begin{slide}{} Multi-character REs REs can be joined together. Eg. {\tt [abc][xyz]} matches {\tt a b} or {\tt c} followed by {\tt x y} or {\tt z}, \\ ie. {\tt ax ay az bx by bz cx cy} or {\tt cz} A one-character RE followed by {\tt *} matches zero or more occurrences of the one-character RE.\@ Eg. {\tt ele.*phy\$} matches ``electroencephalography'' \\ or ``telegraphy''. {\tt [A-Za-z0-9]*} matches any alphanumeric sequence. \end{slide} \begin{slide}{} Counted matches \vspace{3cm} $re \backslash \{ m \backslash \}$ matches $re$ $m$ times. Eg. $[a-z] \backslash \{22 \backslash \}$ matches any 22-letter word. $re \backslash \{m,\backslash \}$ matches $re$ at least $m$ times and $re \backslash \{m,n \backslash \}$ matches $re$ between $m$ and $n$ times inclusive. \end{slide} \begin{slide}{} Saved matches The string matched by a RE in $\backslash ($ $\backslash )$ is saved and can be referred to as $\backslash n$. Eg. \begin{verbatim} sed 's/0\(1-58[0-9]\)/07\1/g' \end{verbatim} Replace ``01-58$d$'' with ``071-58$d$''. \\ Note: ``g'' means {\em all} matches, not just first on each line. \begin{verbatim} sed 's/^\(.\{12\}\)\(.\{6\}\)/\2 = \1/' \end{verbatim} Swap round the first 12 characters and the next 6 and insert `` = '' between them \end{slide} \begin{slide}{} Other sed commands \vspace{3cm} Sed also has commands to insert and delete lines of text and even execute {\em loops}. Eg. \begin{verbatim} sed ' #Define a label ``brackets'' : brackets #Replace matching < > with { } s/<\([^<>]*\)>/{\1}/g #Loop until no more replacements t brackets ' \end{verbatim} \end{slide} \begin{slide}{} Awk (Aho, Weinberger and Kernighan) \vspace{3cm} awk breaks each input line into {\em fields} \\ (seperated by white-space). It is programmed with a C-like syntax \\ combined with regular expression matching. \end{slide} \begin{slide}{} Print the longest line in the input \begin{verbatim} pr -te $* | awk ' length > max { max=length; maxline=$0 } END { print maxline }' \end{verbatim} \end{slide} \begin{slide}{} Print the names of all files containing more than 400 lines. {\tt wc -l *} prints \begin{verbatim} 544 Eliot 781 JDR 76 Jim 498 Sandra 248 Scott 422 Steve wc -l * | awk '$1 >= 400 { print $2 }' \end{verbatim} prints \begin{verbatim} Eliot JDR Sandra Steve \end{verbatim} \end{slide} \begin{slide}{} A larger program Reads input lines of the form \begin{tabular}{ccccc} \$1 & \$2 & \$3 & \$4 & ... \\ host & tty & day & time & full name \end{tabular} Prints time, host and full-name if $day$ or $time$ is different from previous line with the same $host$ and $tty$. {\tt /Login/} executes following commands only for lines containing the string ``Login'' lastdone is an {\em associative array} indexed by strings. \end{slide} \begin{slide}{} \begin{verbatim} awk ' /Login/ { next } { tty = $2 $1; when = $3 " " $4; if (when != lastdone[tty]) { lastdone[tty] = when; printf("%s %-8s %s %s %s %s\n", when, $1, $5, $6, $7, $8); } }' \end{verbatim} \end{slide} \end{document}