Usenet 1994 January

home *** CD-ROM | disk | FTP | other *** search

/ Usenet 1994 January / usenetsourcesnewsgroupsinfomagicjanuary1994.iso / sources / misc / volume3 / mg2a / part01 next >

Wrap

Internet Message Format | 1989-02-03 | 53.1 KB

Path: xanth!mcnc!uvaarpa!umd5!ames!necntc!ncoast!allbery From: BLARSON@ECLA.USC.EDU (Bob Larson) Newsgroups: comp.sources.misc Subject: v03i025: Mg 2a part 1 of 15 Message-ID: <12401298994.47.BLARSON@ECLA.USC.EDU> Date: 26 May 88 04:49:58 GMT Sender: allbery@ncoast.UUCP Reply-To: BLARSON@ECLA.USC.EDU (Bob Larson) Lines: 1500 Approved: allbery@ncoast.UUCP comp.sources.misc: Volume 3, Issue 25 Submitted-By: "Bob Larson" <BLARSON@ECLA.USC.EDU> Archive-Name: mg2a/Part1 # This is a shell archive. # Remove everything above and including the cut line. # Then run the rest of the file through sh. #----cut here-----cut here-----cut here-----cut here----# #!/bin/sh # shar: Shell Archiver # Run the following text with /bin/sh to create: # README # tutorial # mg2a.change # random.c # regex.h # This archive created: Wed May 25 19:20:50 1988 # By: blarson cat << \SHAR_EOF > README Mg 2a README May 15, 1988 Mg (mg) is a Public Domain EMACS style editor. It is "broadly" compatible with GNU Emacs, the latest creation of Richard M. Stallman, Chief GNUisance and inventor of Emacs. GNU Emacs (and other portions of GNU as they are released) are essentially free, (there are handling charges for obtaining it) and so is Mg. You may never have to learn another editor. (But probably will, at least long enough to port Mg...) Mg was formerly named MicroGnuEmacs, the name change was done at the request of Richard Stallman. Mg is not associated with the GNU project, and most of it does not have the copyright restrictions present in GNU Emacs. (However, some of the system dependent modules and the regular expression module do have copyright notices, specificly the VMS/primos termcap routines and the amiga specific routines. Look at the source code for exact copyright restrictions.) The Mg authors individually may or may not agree with the opinions expressed by Richard Stallman in "The GNU Manifesto". To avoid GNU copyright restrictions, replace the re_search.c, regex.h and regex.c files with empty files. Documentation of Mg is in the TeX file mg.tex. This should be formatted with the TeX text formatter and printed. A start twords a mg programmers guied in in mgprog.doc, and some of the changes from 1b are mentioned briefly in mg2a.change. This program is intended to be a small, fast, and portable editor for people who can't (or don't want to) run real Emacs thing for one reason or another. It is compatible with GNU because there shouldn't be any reason to learn more than one Emacs flavor. We have excised most MicroEMACS features that were incompatible with the big brother, and added missing features that seemed essential. There are at least two other major versions of MicroEMACS in circulation. One comes from Daniel Lawrence, (based on an old version from Dave Conroy) and is several versions have been posted to usenet. It uses a 3.x version numbering scheme, and the latest I know about is 3.9i. It has some features not found in Mg, missing others, is bigger, and is incompatible with GNU Emacs. It might be a better choice for you if you *must* have something not present here and can't run GNU. Another variety uses a different numbering scheme, and is up to v30. This also comes from mod.sources, and is the latest version from the original MicroEMACS author Dave Conroy. Mg is derived from this version, and for the most part has replaced it. Mg is continuing to diverge from other MicroEmacs varients. Significant modifacations would me nessisary to adapt code from either the 3.x strains or v30. Command functions and key mapping, for instance, are completely different. This is the third distribution release of Mg. (It went through four beta releases to iron out the changes made by the various authors.) Prior releases were known as MicroGnuEmacs 1a and MicroGnuEmacs 1b. Beyond the work of Dave Conroy, author of the original public domain v30, the current version contains the work of: blarson@ecla.usc.edu Bob Larson mic@emx.utexas.edu Mic Kaczmarczik mwm@violet.berkeley.edu Mike Meyer sandra@cs.utah.edu Sandra Loosemore mp1u+@andrew.cmu.edu Michael Portuesi RCKG01M@CALSTATE.BITNET Stephen Walton hakanson@mist.cs.orst.edu Marion Hakanson People who have worked on previos versions of Mg: rtech!daveb@sun.com Dave Brower These systems are known to work in the current version: 4.2 & 4.3 BSD Unix, SunOs 3.2, Ultrix-32 System V OS9/68k VMS Amiga Primos Atari ST Ms-Dos support is planned, but did not get done in time for this release. (Jeff Siegal <jbs@eddie.mit.edu> was the one doing it.) The Ms-Dos files will probably be distributed seperatly when it becomes available. Cpm/68k support was dropped due to compiler bugs. Eunice support was dropped because of lack of interest. Mg 1b does support those systems. One change to late to make it into mg.tex is readding bsmap-mode (only if BSMAP is #defined when compiling). This is a toggle that controls input mapping to exchange the ^H (backspace) and DEL characters. Like GNU emacs input keymaps, it is not displayed on the mode line and will cause them to be treated as each other for echoing. (With bsmap-mode enabled, DEL will echo ^H in the echo line.) How to Make a Mg --------------------------- On UNIX at least, it's easy. (Note that even on these systems you may want to change a compile time option.) If you have BSD UNIX, do: ln sys/bsd/Makefile . make For System V, do: ln sys/sysv/Makefile . make There are several other directories under sys: osk, vms, amiga, atari, prime. You should follow the directions contained therein to make one of those versions. For most systems (everyting except the amiga, and atari currently), the termcap terminal definition is used. There is a readme file in the default subdirectory of the sys directory explaining what entries are used and how. (Termcap is a way to do display manipulation in a terminal independent manner.) Besides the normal startup file (usually .mg) terminal specific initialization files may be used. (For example, in .mg.vt100 you may want to (global-set-key "\e[A" 'previous-line) to have the up arrow key work.) Some changes made to make this version more like Gnu Emacs may break startup files. Gnu Emacs 18 has both backward-delete-char and delete-backward-char that apperently do the same thing. This version has only the latter because that is what is documented in my manual (version 17) and bound by Gnu Emacs to DEL. ---------------------------------------------------------------------- Known limitaions: Recursive bindings may cause help and key rebinding code to go into an infinite loop, aborting with a stack overflow. Overwrite mode does not work in macros. (Characters are inserted rather than overwriting.) Dired mode has some problems: Rename does not update the buffer. Doing a dired again will update the buffer (whether it needs it or not) and will lose any marks for deletion. .. and . are not recognized as special cases. On systems with 16 bit integers, the kill buffer cannot exceed 32767 bytes. New implementation oddities: insert and define-key are new commands corresponding to the mocklisp functions in Gnu Emacs. (Mg does not have non-command functions.) (Mg's insert will only insert one string.) The display wrap code does not work at all like that of GNU emacs. ------------------------------------------------------------------------ If you have a change to make that you think should be incorporated into the next version of Mg, send it the mg-support mail list. Addresses are: mg-support%ais1@ecla.usc.edu {cit-vax,sdcrdcf,trwrb}!oberon!ais1!mg-support Support for additional systems and terminals should include being available for beta testing as other changes are made. (Send a short note to mg-support.) Currently, beta test copies of Mg are made available via Internet ftp, so beta testers need access to the Internet. (UUCP sites that are customers of uunet can get it via them. Contact uunet!uunet-request for details.) If you can't reach one of us via a computer network, I suppose you could send a change to my snail mail address below on 5" os9 format disks or 9 track tape (ANSI variable label or Prime magsav format), but this effectivly rules you out as a potential beta tester. (Don't expect the disk or tape back unless you inculude a SASE with sufficent postage.) I will not be sending out copies on magnetic media, so please don't ask. If you somehow got an incomplete or non-standard copy, (i.e. missing one of the sys directories mentioned here as working) complain to who you got it from not to me. Robert Larson 309 S. Alexandria Ave. Apt. 117 Los Angeles, CA 90020 Alternatively, and under the same conditions, you can send either a 3" AmigaDOS format disk or a 9 track tape (Unix tar format) to: Mike Meyer P.O. Box 4730 Berkeley, CA 94704 SHAR_EOF cat << \SHAR_EOF > tutorial Copyright (c) 1985 Richard M. Stallman. See end for copying conditions. You are looking at the Emacs tutorial. Emacs commands generally involve the CONTROL key or the META (ESC) key. Rather than write out META or CONTROL each time we want you to prefix a character, we'll use the following abbreviations: C-<chr> means hold the CONTROL key while typing the character <chr> Thus, C-f would be: hold the CONTROL key and type f. M-<chr> means type <ESC>, release it, then type the character <chr>. The characters ">>" at the left margin indicate directions for you to try using a command. For instance: >> Now type C-v (View next screen) to move to the next screen. (go ahead, do it by depressing the control key and v together). From now on, you'll be expected to do this whenever you finish reading the screen. Note that there is an overlap when going from screen to screen; this provides some continuity when moving through the file. The first thing that you need to know is how to move around from place to place in the file. You already know how to move forward a screen, with C-v. To move backwards a screen, type M-v (type <ESC>v). >> Try typing M-v and then C-v to move back and forth a few times. SUMMARY ------- The following commands are useful for viewing screenfuls: C-v Move forward one screenful M-v Move backward one screenful C-l Clear screen and redisplay everything putting the text near the cursor at the center. (That's control-L, not control-1. There is no such character as control-1.) >> Find the cursor and remember what text is near it. Then type a C-l. Find the cursor again and see what text is near it now. BASIC CURSOR CONTROL -------------------- Getting from screenful to screenful is useful, but how do you reposition yourself within a given screen to a specific place? There are several ways you can do this. One way (not the best, but the most basic) is to use the commands previous, backward, forward and next. As you can imagine these commands (which are given to Emacs as C-p, C-b, C-f, and C-n respectively) move the cursor from where it currently is to a new place in the given direction. It is also possible to move the cursor with the arrow keys, but this requires you move your hand from the keyboard, it is also not supported on other machines that do support Emacs. Emacs runs on everything from a CP/M machine to large mainframes. Here then, in a more graphical form are the commands: Previous line, C-p : : Backward, C-b .... Current cursor position .... Forward, C-f : : Next line, C-n >> Move the cursor to the line in the middle of that diagram and type C-l to see the whole diagram centered in the screen. You'll probably find it easy to think of these by letter. P for previous, N for next, B for backward and F for forward. These are the basic cursor positioning commands and you'll be using them ALL the time so it would be of great benefit if you learn them now. >> Do a few C-n's to bring the cursor down to this line. >> Move into the line with C-f's and then up with C-p's. See what C-p does when the cursor is in the middle of the line. >> Try to C-b at the beginning of a line. Do a few more C-b's. Then do C-f's back to the end of the line and beyond. When you go off the top or bottom of the screen, the text beyond the edge is shifted onto the screen so that your instructions can be carried out while keeping the cursor on the screen. >> Try to move the cursor off the bottom of the screen with C-n and see what happens. If moving by characters is too slow, you can move by words. M-f (ESC-f) moves forward a word and M-b moves back a word. >> Type a few M-f's and M-b's. Intersperse them with C-f's and C-b's. Notice the parallel between C-f and C-b on the one hand, and M-f and M-b on the other hand. Very often Meta characters are used for operations related to English text whereas Control characters operate on the basic textual units that are independent of what you are editing (characters, lines, etc). C-a and C-e move to the beginning or end of a line. >> Try a couple of C-a's, and then a couple of C-e's. See how repeated C-a's do nothing. Two other simple cursor motion commands are M-< (Meta Less-than), which moves to the beginning of the file, and M-> (Meta Greater-than), which moves to the end of the file. You probably don't need to try them, since finding this spot again will be boring. On most terminals the "<" is above the comma and you must use the shift key to type it. On these terminals you must use the shift key to type M-< also; without the shift key, you would be typing M-comma. The location of the cursor in the text is also called "point". To paraphrase, the cursor shows on the screen where point is located in the text. Here is a summary of simple moving operations including the word and sentence moving commands: C-f Move forward a character C-b Move backward a character M-f Move forward a word M-b Move backward a word c-n Move to next line C-p Move to previous line C-a Move to beginning of line C-e Move to end of line M-< Go to beginning of file M-> Go to end of file >> Try all of these commands now a few times for practice. Since the last two will take you away from this screen, you can come back here with M-v's and C-v's. These are the most often used commands. Like all other commands in Emacs, these commands can be given arguments which cause them to be executed repeatedly. The way you give a command a repeat count is by typing C-u and then the digits before you type the command. For instance, C-u 8 C-f moves forward eight characters. >> Try giving a suitable argument to C-n or C-p to come as close as you can to this line in one jump. The only apparent exception to this is the screen moving commands, C-v and M-v. When given an argument, they scroll the screen up or down by that many lines, rather than screenfuls. This proves to be much more useful. >> Try typing C-u 8 C-v now. Did it scroll the screen up by 8 lines? If you would like to scroll it down you can give an argument to M-v. WHEN EMACS IS HUNG ----------------- If Emacs gets into an infinite (or simply very long) computation which you don't want to finish, you can stop it safely by typing C-g. You can also use C-g to discard a numeric argument or the beginning of a command that you don't want to finish. >> Type C-u 100 to make a numeric arg of 100, then type C-g. Now type C-f. How many characters does it move? If you have typed an <ESC> by mistake, you can get rid of it with a C-g. WINDOWS ------- Emacs can have several windows, each displaying its own text. At this stage it is better not to go into the techniques of using multiple windows. But you do need to know how to get rid of extra windows that may appear to display help or output from certain commands. It is simple: C-x 1 One window (i.e., kill all other windows). That is Control-x followed by the digit 1. C-x 1 makes the window which the cursor is in become the full screen, by getting rid of any other windows. >> Move the cursor to this line and type C-l (Control-L). >> Type M-x. The cursor will move to the bottom of the screen. >> Type the words "describe-bindings" and hit return. See how this window shrinks, while a new one appears to display which functions are connected to which keys. >> Type C-x 1 and see the documentation listing window disappear. INSERTING AND DELETING ---------------------- If you want to insert text, just type it. Characters which you can see, such as A, 7, *, etc. are taken by Emacs as text and inserted immediately. Type <Return> (the carriage-return key) to insert a Newline character. You can delete the last character you typed by typing <DEL>. More generally, <DEL> deletes the character immediately before the current cursor position. >> Do this now, type a few characters and then delete them by typing <DEL> a few times. Don't worry about this file being changed; you won't affect the master tutorial. This is just a copy of it. >> Now start typing text until you reach the right margin, and keep typing. When a line of text gets too big for one line on the screen, the line of text is "continued" off the edge of the screen. The dollar sign at the right margin indicates a line which has been continued. >> Use <DEL>s to delete the text until the line fits on one screen line again. The continuation mark goes away. >> Move the cursor to the beginning of a line and type <DEL>. This deletes the newline before the line and merges the line onto the previous line. The resulting line may be too long to fit, in which case it has a continuation mark. >> Type <Return> to reinsert the Newline you deleted. Remember that most Emacs commands can be given a repeat count; this includes characters which insert themselves. >> Try that now -- type C-u 8 * and see what happens. You've now learned the most basic way of typing something in Emacs and correcting errors. You can delete by words or lines as well. Here is a summary of the delete operations: <DEL> delete the character just before the cursor C-d delete the next character after the cursor M-<DEL> kill the word immediately before the cursor M-d kill the next word after the cursor C-k kill from the cursor position to end of line Notice that <DEL> and C-d vs M-<DEL> and M-d extend the parallel started by C-f and M-f (well, <DEL> isn't really a control character, but let's not worry about that). Now suppose you kill something, and then you decide that you want to get it back? Well, whenever you kill something bigger than a character, Emacs saves it for you. To yank it back, use C-y. You can kill text in one place, move elsewhere, and then do C-y; this is a good way to move text around. Note that the difference between "Killing" and "Deleting" something is that "Killed" things can be yanked back, and "Deleted" things cannot. Generally, the commands that can destroy a lot of text save it, while the ones that attack only one character, or nothing but blank lines and spaces, do not save. For instance, type C-n a couple times to postion the cursor at some line on this screen. >> Do this now, move the cursor and kill that line with C-k. Note that a single C-k kills the contents of the line, and a second C-k kills the line itself, and make all the other lines move up. If you give C-k a repeat count, it kills that many lines AND their contents. The text that has just disappeared is saved so that you can retrieve it. To retrieve the last killed text and put it where the cursor currently is, type C-y. >> Try it; type C-y to yank the text back. Think of C-y as if you were yanking something back that someone took away from you. Notice that if you do several C-k's in a row the text that is killed is all saved together so that one C-y will yank all of the lines. >> Do this now, type C-k several times. Now to retrieve that killed text: >> Type C-y. Then move the cursor down a few lines and type C-y again. You now see how to copy some text. FILES ----- In order to make the text you edit permanent, you must put it in a file. Otherwise, it will go away when your invocation of Emacs goes away. You put your editing in a file by "finding" the file. What finding means is that you see the contents of the file in your Emacs; and, loosely speaking, what you are editing is the file itself. However, the changes still don't become permanent until you "save" the file. This is so you can have control to avoid leaving a half-changed file around when you don't want to. If you look near the bottom of the screen you will see a line that begins and ends with dashes, and contains the string: "Mg: TUTORIAL" Your copy of the Emacs tutorial is called "TUTORIAL". Whatever file you find, that file's name will appear in that precise spot. The commands for finding and saving files are unlike the other commands you have learned in that they consist of two characters. They both start with the character Control-x. There is a whole series of commands that start with Control-x; many of them have to do with files, buffers, and related things, and all of them consist of Control-x followed by some other character. Another thing about the command for finding a file is that you have to say what file name you want. We say the command "reads an argument from the terminal" (in this case, the argument is the name of the file). After you type the command C-x C-f Find a file Emacs asks you to type the file name. It echoes on the bottom line of the screen. When you type <Return> to end the file name it disappears. >> Type C-x C-f, then type C-g. This cancels the C-x C-f command that was using the minibuffer. So you do not find any file. In a little while the file contents appear on the screen. You can edit the contents. When you wish to make the changes permanent, issue the command C-x C-s Save the file The contents of Emacs are written into the file. When saving is finished, Emacs prints the name of the file written. You should save fairly often, so that you will not lose very much work if the system should crash. >> Type C-x C-s, saving your copy of the tutorial. This should print "Wrote TUTORIAL" at the bottom of the screen. To make a new file, just find it "as if" it already existed. Then start typing in the text. When you ask to "save" the file, Emacs will really create the file with the text that you have inserted. From then on, you can consider yourself to be editing an already existing file. BUFFERS ------- If you find a second file with C-x C-f, the first file remains inside Emacs. This way you can get quite a number of files inside Emacs. The object inside Emacs which holds the text read from one file is called a "buffer." Finding a file makes a new buffer inside Emacs. To see a list of the buffers that exist in Emacs, type C-x C-b List buffers >> Try C-x C-b now. See how each buffer has a name, and it may also have a file name for the file whose contents it holds. Some buffers do not correspond to files. For example, the buffer named "*Buffer List*" does not have any file. It is the buffer which contains the buffer list that was made by C-x C-b. ANY text you see in an Emacs window has to be in some buffer. >> Type C-x 1 to get rid of the buffer list. If you make changes to the text of one file, then find another file, this does not save the first file. Its changes remain inside Emacs, in that file's buffer. The creation or editing of the second file's buffer has no effect on the first file's buffer. This is very useful, but it also means that you need a convenient way to save the first file's buffer. It would be a nuisance to have to switch back to it with C-x C-f in order to save it with C-x C-s. So we have C-x s Save some buffers C-x s goes through the list of all the buffers you have and finds the ones that contain files you have changed. For each such buffer, C-x s asks you whether to save it. EXTENDING THE COMMAND SET ------------------------- There are many, many more Emacs commands than could possibly be put on all the control and meta characters. Emacs gets around this with the X (eXtend) command. This comes in two flavors: C-x Character eXtend. Followed by one character. M-x Named command eXtend. Followed by a long name. These are commands that are generally useful but used less than the commands you have already learned about. You have already seen two of them: the file commands C-x C-f to Find and C-x C-s to Save. Another example is the command to tell Emacs that you'd like to stop editing and get rid of Emacs. The command to do this is C-x C-c. (Don't worry; it offers to save each changed file before it kills the Emacs.) C-z is the usual way to exit Emacs, because it is always better not to kill the Emacs if you are going to do any more editing. On systems which allow it, C-z exits from Emacs to a CLI but does not destroy the Emacs; you can resume editing by ending that CLI or depth arranging. You would use C-x C-c if you were running out of memory. You would also use it to exit an Emacs invoked under mail handling programs and other random utilities, since they may not believe you have really finished using the Emacs if it continues to exist. There are many C-x commands. The ones you know are: C-x C-f Find file. C-x C-s Save file. C-x C-b List buffers. C-x C-c Quit Emacs. Named eXtended commands are commands which are used even less frequently, or commands which are used only in certain modes. These commands are usually called "functions". An example is the function replace-string, which globally replaces one string with another. When you type M-x, Emacs prompts you at the bottom of the screen with M-x and you should type the name of the function you wish to call; in this case, "query-replace". Just type "que<TAB>" and Emacs will complete the name. End the command name with <Return>. Then type the two "arguments"--the string to be replaced, and the string to replace it with--each one ended with a Return. >> Move the cursor to the blank line two lines below this one. Then type M-x repl s<Return>changed<Return>altered<Return>. Notice how this line has changed: you've replaced the word c-h-a-n-g-e-d with "altered" wherever it occured after the cursor. MODE LINE --------- If Emacs sees that you are typing commands slowly it shows them to you at the bottom of the screen in an area called the "echo area." The echo area contains the bottom line of the screen. The line immediately above it is called the MODE LINE. The mode line says something like --**-Mg: TUTORIAL (fundamental)------------------------ This is a very useful "information" line. The stars near the front mean that you have made changes to the text. Right after you visit or save a file, there are no stars, just dashes. The part of the mode line inside the parentheses is to tell you what modes you are in. The default mode is fundamental which is what you are in now. It is an example of a "mode". There are several modes in Emacs for editing different styles of text, such as indent, bsmap, fill, etc. Each mode makes a few commands behave differently. One mode which is very useful, especially for editing English text, is Auto Fill mode. When this mode is on, Emacs breaks the line in between words automatically whenever the line gets too long. You can turn this mode on by doing M-x auto-fill-mode<Return>. When the mode is on, you can turn it off by doing M-x auto-fill-mode<Return>. >> Type M-x auto-fill-mode<Return> now. Then insert a line of "asdf " over again until you see it divide into two lines. You must put in spaces between them because Auto Fill breaks lines only at spaces. The margin is usually set at 70 characters, but you can change it with the C-x f command. You should give the margin setting you want as a numeric argument. >> Type C-x f with an argument of 20. (C-u 2 0 C-x f). Then type in some text and see Emacs fill lines of 20 characters with it. Then set the margin back to 70 using C-x f again. If you make changes in the middle of a paragraph, Auto Fill mode does not re-fill it for you. To re-fill the paragraph, type M-q (Meta-q) with the cursor inside that paragraph. >> Move the cursor into the previous paragraph and type M-q. SEARCHING --------- Emacs can do searches for strings (these are groups of contiguous characters or words) either forward through the file or backward through it. To search for the string means that you are trying to locate it somewhere in the file and have Emacs show you where the occurrences of the string exist. This type of search is somewhat different from what you may be familiar with. It is a search that is performed as you type in the thing to search for. The command to initiate a search is C-s for forward search, and C-r for reverse search. BUT WAIT! Don't do them now. When you type C-s you'll notice that the string "I-search" appears as a prompt in the echo area. This tells you that Emacs is in what is called an incremental search waiting for you to type the thing that you want to search for. <ESC> terminates a search. >> Now type C-s to start a search. SLOWLY, one letter at a time, type the word 'cursor', pausing after you type each character to notice what happens to the cursor. >> Type C-s to find the next occurrence of "cursor". >> Now type <DEL> four times and see how the cursor moves. >> Type <ESC> to terminate the search. Did you see what happened? Emacs, in an incremental search, tries to go to the occurrence of the string that you've typed out so far. To go to the next occurrence of 'cursor' just type C-s again. If no such occurrence exists Emacs beeps and tells you that it is a failing search. C-g would also terminate the search. If you are in the middle of an incremental search and type <DEL>, you'll notice that the last character in the search string is erased and the search backs up to the last place of the search. For instance, suppose you currently have typed 'cu' and you see that your cursor is at the first occurrence of 'cu'. If you now type <DEL>, the 'u' on the search line is erased and you'll be repositioned in the text to the occurrence of 'c' where the search took you before you typed the 'u'. This provides a useful means for backing up while you are searching. If you are in the middle of a search and happen to type a control character (other than a C-s or C-r, which tell Emacs to search for the next occurrence of the string), the search is terminated. The C-s starts a search that looks for any occurrence of the search string AFTER the current cursor position. But what if you want to search for something earlier in the text? To do this, type C-r for Reverse search. Everything that applies to C-s applies to C-r except that the direction of the search is reversed. GETTING MORE HELP ----------------- In this tutorial we have tried to supply just enough information to get you started using Emacs. There is so much available in Emacs that it would be impossible to explain it all here. However, you may want to learn more about Emacs since it has numerous desirable features that you don't know about yet. CONCLUSION ---------- Remember, to exit Emacs permanently use C-x C-c. To exit to a shell temporarily, so that you can come back in, use C-z. This tutorial is meant to be understandable to all new users, so if you found something unclear, don't sit and blame yourself - complain! COPYING ------- This tutorial, like all of GNU Emacs, is copyrighted, and comes with permission to distribute copies on certain conditions: Copyright (c) 1985 Richard M. Stallman Permission is granted to anyone to make or distribute verbatim copies of this document as received, in any medium, provided that the copyright notice and permission notice are preserved, and that the distributor grants the recipient permission for further redistribution as permitted by this notice. Permission is granted to distribute modified versions of this document, or of portions of it, under the above conditions, provided also that they carry prominent notices stating who last altered them. The conditions for copying Emacs itself are slightly different but in the same spirit. Please read the file COPYING and then do give copies of GNU Emacs to your friends. Help stamp out ownership of software by using, writing, and sharing free software! Mg itself is public domain, and may be given away freely. See the README file about differences from GNU emacs, and why Mg exists. ******************************************************************************* *** This document heavily cut by Randy M. Spencer to apply to *** *** Mg written my Mike Meyer and gang. It was released *** *** at the AAA users group meeting in Lafayette CA, an Amiga Users *** *** Group. My profound thanks to Richard Stallman for his work, I *** *** am proud to carry his initials. *** *** Additional modifacations were done by Robert A. Larson for Mg *** *** version 2a, mainly the name change from MicroGnuEmacs to Mg. *** ******************************************************************************* See other files accompanying this for more system specific information. SHAR_EOF cat << \SHAR_EOF > mg2a.change Here is a list of some of the changes from Mg version 1b to 2a, in reverse semi-cronological order. Much is only of interest if you are modifying Mg. Some things may only apply relitive to beta versions. (I tried to edit these out.) The kill buffer should now avoid growing bigger than possible. (malloc takes an unsigned arg, which may only be 16 bits.) This is a temporary hack until the kill buffer code is rewriten. Bcopy is NOT used properly, the third argument should be declared unsigned but isn't. 16 bit int systems need to treat the third agument as unsigned. Added functions set-default-mode, global-unset-key, and local-unset-key. dired_ functions in system dependent code now need to set bp->b_nmodes to 0. Function key binding should finally work properly. #define FKEYS if you have function keys that don't fit into a char. Termcap XKEYS code completly rewritten. It does not use the FKEYS code. Terminal-dependant initialization files may now be present. System dependent routine startupfile now takes a single char * arguement. It is NULL for the normal startup file, the terminal type name for the terminal dependent file. The BSD version no longer tries to use .emacs if .mg isn't found. BSD, SYS V, and OSK versions can look for a system-wide startup file if STARTUPFILE is #defined to the directory to look in. Extend.c now has a dependancy on key.h. Insert-file now uses adjustname. Optional SYSCLEANUP macro. Various changes to make mg smaller and faster. Portability issuses: Identifiers should now be unique in first 7 chars. (Change dired_makename to d_makename in fileio.c.) Lint found various bugs. Character set dependent portion of def.h moved to sys/default/chrdef.h. Cinfo.c moved there. tty/termcap/* moved there. Bug fixes: insert-file, fill-paragraph, startup code, "make lint" in bsd makefile. "varargs.h" in sys/default is for systems that worked with the non-varargs code on mg1. It does not work on systems that don't happen to pass arguments that way. fncmp is now also used to compare buffer names. Buffer names are now case sensitive iff file names are. The #ifndef ... in some files moved after the #include "def.h". This should be better for systems that define various things in sysdef.h. Let me know what I missed. Minibuffer and search commands modified so ^H does the same as ^?, ^\ as ^Q, and ^] as ^S. bsmap-mode may be done by key bindings, flow-mode still needs to toggle system-dependent capture of ^S/^Q. re_search.c probably needs to change also. dired mode. (requires a significant amount of system dependent code.) Two case, non-case sensitive systems are now fully supported. The routine fncmp is used to compare file names. (strcmp may be used on monocase or case sensitive systems.) If NO_DIR is not defined, adjustname (the replacement for adjustcase) must append the directory name to the filename if it doesn't have one already. Cleanup and rearangement. symbol.c has been renamed to keymap.c. The key name function has been moved to cinfo.c to reduce the number of files dependent on character set. The terminal type determining code has been moved out of sys/default/tty.c. For unix systems, just "#define gettermtype() getenv("TERM")". For VMS and Primos, it's a bit more compicated. I made some changes to the regular expression code to get it to compile under os9/68k. It now compiles, but requires a number of routines not present in the library that would be difficult to implement. lalloc has be rewriten and split into lallocx (for lines not expected to grow) and lalloc. The OFFSET macro may have to be defined in the system dependent file if the normal definition supplied will not work with your compiler. (I considered alternatives, but could not think of one that used as little storage. Use the offsetof macro if your compiler conforms to the ANSI C draft standard.) The MALLOCROUND macro should be defined if you can predict the behavior of your libraries malloc. It takes a single lvalue arguement, and rounds it up to the amount of memory that realy would be allocated by a malloc of that size. (This is most important on memory tight systems.) Apropos is new. It has a few minor problems: slow, displays only global key bindings, dosn't like lots of keys bound to the same function. (Digit-argument is closest to causing problems.) Its output should probably be the basis of future functions files. The file reading and writing code has changed to fix a couple of long-standing bugs. Files no longer get an extra newline at the end of the last line if it doesn't have one. Line length is now only limited by memory and how big a number will fit in a short int. See description below on what needs to change. Umpteen minor bugs fixed. If anyone is doing serious development based on 1b, some should be merged back in. (Kill-buffer had several bugs, help didn't redisplay properly if the *help* buffer was already present, etc.) The display wrap code is better that what was done before, but not at all what GNU emacs does in the same situation. Compile options have been changed to make those things present in GNU emacs the default. Compile-time options to turn on features not in GNU emacs and to turn of features that are in GNU emacs are documented below. SHAR_EOF cat << \SHAR_EOF > random.c /* * Assorted commands. * The file contains the command * processors for a large assortment of unrelated * commands. The only thing they have in common is * that they are all command processors. */ #include "def.h" /* * Display a bunch of useful information about * the current location of dot. The character under the * cursor (in octal), the current line, row, and column, and * approximate position of the cursor in the file (as a percentage) * is displayed. The column position assumes an infinite position * display; it does not truncate just because the screen does. * This is normally bound to "C-X =". */ /*ARGSUSED*/ showcpos(f, n) { register LINE *clp; register long nchar; long cchar; register int nline, row; int cline, cbyte; /* Current line/char/byte */ int ratio; clp = lforw(curbp->b_linep); /* Collect the data. */ nchar = 0; nline = 0; for (;;) { ++nline; /* Count this line */ if (clp == curwp->w_dotp) { cline = nline; /* Mark line */ cchar = nchar + curwp->w_doto; if (curwp->w_doto == llength(clp)) cbyte = '\n'; else cbyte = lgetc(clp, curwp->w_doto); } nchar += llength(clp); /* Now count the chars */ clp = lforw(clp); if (clp == curbp->b_linep) break; nchar++; /* count the newline */ } row = curwp->w_toprow + 1; /* Determine row. */ clp = curwp->w_linep; while (clp!=curbp->b_linep && clp!=curwp->w_dotp) { ++row; clp = lforw(clp); } /*NOSTRICT*/ ratio = nchar ? (100L*cchar) / nchar : 100; ewprintf("Char: %c (0%o) point=%ld(%d%%) line=%d row=%d col=%d", cbyte, cbyte, cchar, ratio, cline, row, getcolpos()); return TRUE; } getcolpos() { register int col, i, c; col = 1; /* Determine column. */ for (i=0; i<curwp->w_doto; ++i) { c = lgetc(curwp->w_dotp, i); if (c == '\t' #ifdef NOTAB && !(curbp->b_flag & BFNOTAB) #endif ) { col |= 0x07; ++col; } else if (ISCTRL(c) != FALSE) ++col; ++col; } return col; } /* * Twiddle the two characters on either side of * dot. If dot is at the end of the line twiddle the * two characters before it. Return with an error if dot * is at the beginning of line; it seems to be a bit * pointless to make this work. This fixes up a very * common typo with a single stroke. Normally bound * to "C-T". This always works within a line, so * "WFEDIT" is good enough. */ /*ARGSUSED*/ twiddle(f, n) { register LINE *dotp; register int doto; register int cr; VOID lchange(); dotp = curwp->w_dotp; doto = curwp->w_doto; if(doto==llength(dotp)) { if(--doto<=0) return FALSE; } else { if(doto==0) return FALSE; ++curwp->w_doto; } cr = lgetc(dotp, doto--); lputc(dotp, doto+1, lgetc(dotp, doto)); lputc(dotp, doto, cr); lchange(WFEDIT); return TRUE; } /* * Open up some blank space. The basic plan * is to insert a bunch of newlines, and then back * up over them. Everything is done by the subcommand * procerssors. They even handle the looping. Normally * this is bound to "C-O". */ /*ARGSUSED*/ openline(f, n) { register int i; register int s; if (n < 0) return FALSE; if (n == 0) return TRUE; i = n; /* Insert newlines. */ do { s = lnewline(); } while (s==TRUE && --i); if (s == TRUE) /* Then back up overtop */ s = backchar(f | FFRAND, n); /* of them all. */ return s; } /* * Insert a newline. * If you are at the end of the line and the * next line is a blank line, just move into the * blank line. This makes "C-O" and "C-X C-O" work * nicely, and reduces the ammount of screen * update that has to be done. This would not be * as critical if screen update were a lot * more efficient. */ /*ARGSUSED*/ newline(f, n) { register LINE *lp; register int s; if (n < 0) return FALSE; while (n--) { lp = curwp->w_dotp; if (llength(lp) == curwp->w_doto && lforw(lp) != curbp->b_linep && llength(lforw(lp)) == 0) { if ((s=forwchar(FFRAND, 1)) != TRUE) return s; } else if ((s=lnewline()) != TRUE) return s; } return TRUE; } /* * Delete blank lines around dot. * What this command does depends if dot is * sitting on a blank line. If dot is sitting on a * blank line, this command deletes all the blank lines * above and below the current line. If it is sitting * on a non blank line then it deletes all of the * blank lines after the line. Normally this command * is bound to "C-X C-O". Any argument is ignored. */ /*ARGSUSED*/ deblank(f, n) { register LINE *lp1; register LINE *lp2; register RSIZE nld; lp1 = curwp->w_dotp; while (llength(lp1)==0 && (lp2=lback(lp1))!=curbp->b_linep) lp1 = lp2; lp2 = lp1; nld = (RSIZE) 0; while ((lp2=lforw(lp2))!=curbp->b_linep && llength(lp2)==0) ++nld; if (nld == 0) return (TRUE); curwp->w_dotp = lforw(lp1); curwp->w_doto = 0; return ldelete((RSIZE)nld, KNONE); } /* * Delete any whitespace around dot, then insert a space. */ justone(f, n) { (VOID) delwhite(f, n); return linsert(1, ' '); } /* * Delete any whitespace around dot. */ /*ARGSUSED*/ delwhite(f, n) { register int col, c, s; col = curwp->w_doto; while (((c = lgetc(curwp->w_dotp, col)) == ' ' || c == '\t') && col < llength(curwp->w_dotp)) ++col; do { if (curwp->w_doto == 0) { s = FALSE; break; } if ((s = backchar(FFRAND, 1)) != TRUE) break; } while ((c = lgetc(curwp->w_dotp, curwp->w_doto)) == ' ' || c == '\t'); if (s == TRUE) (VOID) forwchar(FFRAND, 1); (VOID) ldelete((RSIZE)(col - curwp->w_doto), KNONE); return TRUE; } /* * Insert a newline, then enough * tabs and spaces to duplicate the indentation * of the previous line. Assumes tabs are every eight * characters. Quite simple. Figure out the indentation * of the current line. Insert a newline by calling * the standard routine. Insert the indentation by * inserting the right number of tabs and spaces. * Return TRUE if all ok. Return FALSE if one * of the subcomands failed. Normally bound * to "C-J". */ /*ARGSUSED*/ indent(f, n) { register int nicol; register int c; register int i; if (n < 0) return (FALSE); while (n--) { nicol = 0; for (i=0; i<llength(curwp->w_dotp); ++i) { c = lgetc(curwp->w_dotp, i); if (c!=' ' && c!='\t') break; if (c == '\t') nicol |= 0x07; ++nicol; } if (lnewline() == FALSE || (( #ifdef NOTAB curbp->b_flag&BFNOTAB) ? linsert(nicol, ' ') == FALSE : ( #endif ((i=nicol/8)!=0 && linsert(i, '\t')==FALSE) || ((i=nicol%8)!=0 && linsert(i, ' ')==FALSE)))) return FALSE; } return TRUE; } /* * Delete forward. This is real * easy, because the basic delete routine does * all of the work. Watches for negative arguments, * and does the right thing. If any argument is * present, it kills rather than deletes, to prevent * loss of text if typed with a big argument. * Normally bound to "C-D". */ /*ARGSUSED*/ forwdel(f, n) { if (n < 0) return backdel(f | FFRAND, -n); if (f & FFARG) { /* Really a kill. */ if ((lastflag&CFKILL) == 0) kdelete(); thisflag |= CFKILL; } return ldelete((RSIZE) n, (f & FFARG) ? KFORW : KNONE); } /* * Delete backwards. This is quite easy too, * because it's all done with other functions. Just * move the cursor back, and delete forwards. * Like delete forward, this actually does a kill * if presented with an argument. */ /*ARGSUSED*/ backdel(f, n) { register int s; if (n < 0) return forwdel(f | FFRAND, -n); if (f & FFARG) { /* Really a kill. */ if ((lastflag&CFKILL) == 0) kdelete(); thisflag |= CFKILL; } if ((s=backchar(f | FFRAND, n)) == TRUE) s = ldelete((RSIZE) n, (f & FFARG) ? KFORW : KNONE); return s; } /* * Kill line. If called without an argument, * it kills from dot to the end of the line, unless it * is at the end of the line, when it kills the newline. * If called with an argument of 0, it kills from the * start of the line to dot. If called with a positive * argument, it kills from dot forward over that number * of newlines. If called with a negative argument it * kills any text before dot on the current line, * then it kills back abs(arg) lines. */ /*ARGSUSED*/ killline(f, n) { register RSIZE chunk; register LINE *nextp; register int i, c; VOID kdelete(); if ((lastflag&CFKILL) == 0) /* Clear kill buffer if */ kdelete(); /* last wasn't a kill. */ thisflag |= CFKILL; if (!(f & FFARG)) { for (i = curwp->w_doto; i < llength(curwp->w_dotp); ++i) if ((c = lgetc(curwp->w_dotp, i)) != ' ' && c != '\t') break; if (i == llength(curwp->w_dotp)) chunk = llength(curwp->w_dotp)-curwp->w_doto + 1; else { chunk = llength(curwp->w_dotp)-curwp->w_doto; if (chunk == 0) chunk = 1; } } else if (n > 0) { chunk = llength(curwp->w_dotp)-curwp->w_doto+1; nextp = lforw(curwp->w_dotp); i = n; while (--i) { if (nextp == curbp->b_linep) break; chunk += llength(nextp)+1; nextp = lforw(nextp); } } else { /* n <= 0 */ chunk = curwp->w_doto; curwp->w_doto = 0; i = n; while (i++) { if (lback(curwp->w_dotp) == curbp->b_linep) break; curwp->w_dotp = lback(curwp->w_dotp); curwp->w_flag |= WFMOVE; chunk += llength(curwp->w_dotp)+1; } } /* * KFORW here is a bug. Should be KBACK/KFORW, but we need to * rewrite the ldelete code (later)? */ return (ldelete(chunk, KFORW)); } /* * Yank text back from the kill buffer. This * is really easy. All of the work is done by the * standard insert routines. All you do is run the loop, * and check for errors. The blank * lines are inserted with a call to "newline" * instead of a call to "lnewline" so that the magic * stuff that happens when you type a carriage * return also happens when a carriage return is * yanked back from the kill buffer. * An attempt has been made to fix the cosmetic bug * associated with a yank when dot is on the top line of * the window (nothing moves, because all of the new * text landed off screen). */ /*ARGSUSED*/ yank(f, n) { register int c; register int i; register LINE *lp; register int nline; VOID isetmark(); if (n < 0) return FALSE; nline = 0; /* Newline counting. */ while (n--) { isetmark(); /* mark around last yank */ i = 0; while ((c=kremove(i)) >= 0) { if (c == '\n') { if (newline(FFRAND, 1) == FALSE) return FALSE; ++nline; } else { if (linsert(1, c) == FALSE) return FALSE; } ++i; } } lp = curwp->w_linep; /* Cosmetic adjustment */ if (curwp->w_dotp == lp) { /* if offscreen insert. */ while (nline-- && lback(lp)!=curbp->b_linep) lp = lback(lp); curwp->w_linep = lp; /* Adjust framing. */ curwp->w_flag |= WFHARD; } return TRUE; } #ifdef NOTAB /*ARGSUSED*/ space_to_tabstop(f, n) int f, n; { if(n<0) return FALSE; if(n==0) return TRUE; return linsert((n<<3) - (curwp->w_doto & 7), ' '); } #endif SHAR_EOF cat << \SHAR_EOF > regex.h /* Definitions for data structures callers pass the regex library. Copyright (C) 1985 Richard M. Stallman This program is distributed in the hope that it will be useful, but without any warranty. No author or distributor accepts responsibility to anyone for the consequences of using it or for whether it serves any particular purpose or works at all, unless he says so in writing. Permission is granted to anyone to distribute verbatim copies of this program's source code as received, in any medium, provided that the copyright notice, the nonwarraty notice above and this permission notice are preserved, and that the distributor grants the recipient all rights for further redistribution as permitted by this notice, and informs him of these rights. Permission is granted to distribute modified versions of this program's source code, or of portions of it, under the above conditions, plus the conditions that all changed files carry prominent notices stating who last changed them and that the derived material, including anything packaged together with it and conceptually functioning as a modification of it rather than an application of it, is in its entirety subject to a permission notice identical to this one. Permission is granted to distribute this program (verbatim or as modified) in compiled or executable form, provided verbatim redistribution is permitted as stated above for source code, and A. it is accompanied by the corresponding machine-readable source code, under the above conditions, or B. it is accompanied by a written offer, with no time limit, to distribute the corresponding machine-readable source code, under the above conditions, to any one, in return for reimbursement of the cost of distribution. Verbatim redistribution of the written offer must be permitted. Or, C. it is distributed by someone who received only the compiled or executable form, and is accompanied by a copy of the written offer of source code which he received along with it. Permission is granted to distribute this program (verbatim or as modified) in executable form as part of a larger system provided that the source code for this program, including any modifications used, is also distributed or offered as stated in the preceding paragraph. In other words, you are welcome to use, share and improve this program. You are forbidden to forbid anyone else to use, share and improve what you give them. Help stamp out software-hoarding! */ #ifndef RE_NREGS #define RE_NREGS 10 #endif /* This data structure is used to represent a compiled pattern. */ struct re_pattern_buffer { char *buffer; /* Space holding the compiled pattern commands. */ int allocated; /* Size of space that buffer points to */ int used; /* Length of portion of buffer actually occupied */ char *fastmap; /* Pointer to fastmap, if any, or zero if none. */ /* re_search uses the fastmap, if there is one, to skip quickly over totally implausible characters */ char *translate; /* Translate table to apply to all characters before comparing. Or zero for no translation. The translation is applied to a pattern when it is compiled and to data when it is matched. */ char fastmap_accurate; /* Set to zero when a new pattern is stored, set to one when the fastmap is updated from it. */ char can_be_null; /* Set to one by compiling fastmap if this pattern might match the null string. It does not necessarily match the null string in that case, but if this is zero, it cannot. */ }; /* Structure to store "register" contents data in. Pass the address of such a structure as an argument to re_match, etc., if you want this information back. start[i] and end[i] record the string matched by \( ... \) grouping i, for i from 1 to RE_NREGS - 1. start[0] and end[0] record the entire string matched. */ struct re_registers { int start[RE_NREGS]; int end[RE_NREGS]; }; extern char *re_compile_pattern (); /* Is this really advertised? */ extern VOID re_compile_fastmap (); extern int re_search (), re_search_2 (); extern int re_match (), re_match_2 (); /* 4.2 bsd compatibility (yuck) */ extern char *re_comp (); extern int re_exec (); #ifdef SYNTAX_TABLE extern char *re_syntax_table; #endif SHAR_EOF # End of shell archive exit 0 -------