The PC-SIG Library 9

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

/ The PC-SIG Library 9 / The PC-SIG Library on CD ROM - Ninth Edition.iso / 901_1000 / DISK0977 / DISK0977.ZIP / PROINDEX.DOC < prev next >

Wrap

Text File | 1987-01-15 | 124KB | 3,763 lines

P P O I N D E X Version 2.23 Copyright 1985, 1986, 1987 by Gary Elfring INTRODUCTION Do you need to create an index for the back of your book, thesis, report, or software manual? Do your legal briefs or court re- ports need indexes? Now there is an easy way to index an word processed document. PROINDEX will perform all the work necessary to make an index. PROINDEX takes your text file along with a list of keywords or phrases and looks for matches between the two. It produces an index in any one of numerous different styles, including those suggested by "The Chicago Manual of Style". There is no limit on the length of your text file or the number of keywords in your list. The resulting index can be saved as a text file or sent to the printer. PROINDEX will run automatically or interactively. In the inter- active mode it shows you each match to a keyword or phrase in context. You control which entries are marked in your index. PROINDEX will produce an index that references absolute line numbers, page numbers, or page and line numbers. The program is compatible with all major word processors. True secondary refer- ences are also possible using PROINDEX in the interactive mode. Other features include: indexing sessions can be halted at any time and then restarted where you left off, full path support, a configuration file controls your indexing style. User select- able: starting page number of text, multiple or single references per page, string to preceed each page number, page / line / or page and line references, lines per page, case conversion, 7 or 8 bit ASCII, interactive or batch mode. SHAREWARE This program is provided on a "shareware" basis! Try the program out. If you don't like it- erase it. If you do like it we hope you will pay for it. The shareware contribution mentioned here is 30% less than the retail price of this package. Thus shareware users get the exact same program as commercial users only they pay less. NOTE that the utilities mentioned further in this document do NOT come with the "shareware" version of the program. Only register- ed users get the utilities programs mentioned here. (This is the incentive to send in your registration fee.) To register send $65.00 to: Gary Elfring 4N899 West Mary Drive St. Charles, Illinois, 60174 PLEASE mention the product name, (PROINDEX), and where you got it from. You will then receive: the latest version of the program (including all the missing utilities and any new ones we dream up), improved typeset documentation, telephone support, and one free update. WARRANTY NONE!! No warranty is provided to anyone who has not contributed for this program. Gary Elfring specifically disclaims all other warranties, expressed or implied, including but not limited to, implied warranties of merchantability and fitness for a particu- lar purpose with respect to defects in the disk or documentation, and the program license granted herein. In no event shall Gary Elfring be liable for any loss of profit or any other commercial damage, including but not limited to special, incidental, conse- quential, or other damages. Table of Contents INTRODUCTION................................................. 1 CREATING AN INDEX............................................ 3 Index Styles............................................... 4 PREPARING YOUR TEXT.......................................... 7 General.................................................... 7 Non-ASCII Indexes.......................................... 8 MAKING A KEYWORD FILE........................................ 10 General.................................................... 10 Creating the Keyword File.................................. 11 INDEXING TECHNIQUES.......................................... 12 Interactive Control........................................ 12 Page Breaks................................................ 13 Keyword Tips............................................... 13 Secondary References....................................... 14 CONFIGURATION................................................ 16 OPERATION.................................................... 19 Getting Started............................................ 19 Opening Screen............................................. 19 Data Entry................................................. 19 Program Operation.......................................... 20 Executing MS-DOS Commands.................................. 23 PROINDEX Control Parameters................................ 24 Indexing................................................... 29 OUTPUT....................................................... 31 PROGRAM SPEED................................................ 34 UTILITY PROGRAMS............................................. 36 COMMON.TXT................................................. 36 MAKEWORD................................................... 36 MERGE...................................................... 37 PAGE....................................................... 39 REMOVE..................................................... 41 SORT....................................................... 41 STRIP...................................................... 42 PROBLEMS..................................................... 43 Getting Started............................................ 43 PROINDEX Misses References................................. 43 Paging Problems............................................ 44 Error Messages............................................. 44 DEMONSTRATION FILES.......................................... 47 Table of Contents MENU SCREENS................................................. 49 Opening Screen............................................. 49 Input File Names Entry..................................... 49 Main Menu.................................................. 50 Index Control Parameters................................... 50 Interactive Indexing Menu.................................. 51 MS-DOS Command Menu........................................ 51 The Configuration File..................................... 52 INTRODUCTION The PROINDEX program is a software tool designed to produce an index of your word processed documents. An index is an alphabetical list of names, subject, references, etc. together with the position each appears in the text. This index usually appears at the end of a book or publication. The exact format or style of this index can be selected by the user. PROINDEX will create an index from almost any kind of text file. All that is needed is your original text file and a list of keywords or phrases. PROINDEX compares this reference list with your original text file and marks each keyword it finds in an index. Utilities are provided that will automatically prepare a reference list for you if you don't want to create your own. The resulting index can be paged by a utility supplied along with PROINDEX or it can be customized by you with your word processor. Please note that PROINDEX does require some work on your part before you can create an index. Your original text file must be paged and you must create a list of keywords or phrases. In addition, text files created by certain word processors must be converted to an ASCII format. During the indexing process you can exercise control over which references are included in your index by running the program in interactive mode. In this mode, PROINDEX shows you each reference word in context and lets you chose whether to mark it or not. If you want all references included in your index automatically, you can run PROINDEX in the automatic mode. PROINDEX will handle almost any type of text file, including those produced by WordStar and most other word processors. There is no limit on the size of a text file to be indexed or on the number of references to search for. PROINDEX is designed to perform as much of the indexing task as possible automatically, since that is what a computer is good at. In operation, PROINDEX generates temporary working copies of your input text and keyword files. PROINDEX requires an amount of free disk space equal to your text file size plus keyword file size. A little more space is required when the output of PROINDEX is directed to the disk. The program checks the amount of free disk space before it makes these temporary work files and lets you know if there is not enough room. You can specify a separate work drive if you need more disk space. Temporary files are always erased before the end of a PROINDEX run. PROINDEX has several limitations. No keyword or phrase may be longer than 70 characters in length. Also, PROINDEX will not locate phrases that wrap from one line to the next. An index references absolute pages in your text, and thus the input to PROINDEX must be paged in some way. With most word processors you can simply print your text to a disk file. PROINDEX will then count lines to determine what page it is on. An alternate paging 1 method is to mark the end of each page with any single 8 bit character you chose. This gives you indexing compatibility with documents that do not follow the paging format of your word processor. PROINDEX will not correctly display text in the interactive mode if the text lines are longer than 80 characters. This problems does not effect the operation of PROINDEX or the quality of your index. It only concerns the display. If lines are longer than 80 characters the IBM BIOS will wrap the lines on the screen and PROINDEX will lose track of where it is on the screen. The only fix for this problem significantly slows down the operation of PROINDEX. Since the problem does not effect the final index we have left it in to avoid slowing PROINDEX down. PROINDEX requires an IBM PC or compatible computer to run. It needs a minimum of 192K of memory, one disk drive, and MS-DOS version 2.1 or higher. PROINDEX performs BIOS calls to interrupt 10H, video control, and thus may not work on all IBM compatibles. A special version of PROINDEX is available that uses the ANSI screen driver to bypass this problem. The ANSI version is slower in operation. The maximum file size that can be indexed is limited only by the amount of available disk space. Hard disk or large RAM disks are recommended when indexing files greater than 180K bytes long. 2 CREATING AN INDEX Creating an index from scratch is not as easy a task as it seems. The quality of an index is related to two different items. It depends on how good the reference list is and how well this reference list is compared to the text. PROINDEX will take care of the later problem for you. This only leaves the problem of creating a good reference list. There are two basic approaches to creating a reference list. One approach is to have the author of the document make up a list of keywords or phrases. This list can be added to as the indexing process continues. This approach has the disadvantage that it can miss entire subjects that should be in the index. That is because this approach relies on human memory. A second approach is to use a set of software tools to turn the original text file into a sorted list of unique words. We supply two utilities, MAKEWORD and REMOVE, to do this work automatical- ly. This list is then used as the basis for a reference list. It can be added to and modified by the document author. PROINDEX supports this approach by supplying a set of indexing tools that will build and maintain this sorted list of unique words for you. Once a reference list exists the style of the index must be chosen. The simplest index style lists all references in alphabe- tical order. All chapters in the document are numbered sequen- tially. The completed index simply lists the page number each reference can be found on. When a citation is found more than once on a page only one reference to that page will be put into the index. A typical example of this style index looks as follows: All command, 15 BIOS calls, 1 Bold display, 14, 34 Data entry, 13, 23 Delete key, 6, 9, 14 IBM PC, 1, 5, 7 Line numbers, 3, 5, 7 A slightly more advanced form of index would illustrate when a citation covers more than one page. Assume that the reference Ape is discussed in your text on pages 12 through 15. PROINDEX would produce the following citation: Ape, 12, 13, 14, 15 Why doesn't PROINDEX automatically turn this citation into the more correct form, Ape, 12 - 15 ? The answer is, PROINDEX does not know the context of the text. It is possible that there were four separate references to Ape on those four pages. In that case listing each page number separately would be the correct form. Thus, you must decide which form is correct and edit in the change, if any, with your word processor. 3 Other index styles are also supported by PROINDEX. Some indexes assume each chapter of text starts with page number one. The page references in this type of index must delineate which chapter they represent. This is done by assigning the chapters individual numbers, characters, or names. Assuming each chapter was assigned a sequential number, chapter one gets number 1, this index style would look as follows: Remarks, 3-39 Root directory, 5-7, 7-6 Screen, 2-5 Segments, 4-6, 4-9 System devices, 2-5 Text files, 7-123, 8-5 A legal index style might be completely different. These indexes typically must show both page and line numbers of a citation or reference. PROINDEX lets you chose any character you want to separate page numbers from line numbers. In addition, multiple references per page must be allowed. Assuming you chose the ":" character the index in this style would look as follows. Atwell vs Drumond, 6:3, 6:14 Automobile, 9:23, 13:56, 19:7 Ford Pinto, 1:3, 3:5, 3:29, 3:46 Supreme Court, 19:7 Once an index style has been selected, you can run the PROINDEX program. From the main menu of the program you can select the index control parameter menu. Here you can configure PROINDEX to produce an index in the style you want. Once the questions on this menu have been answered you can begin the actual indexing process. PROINDEX typically sends its output to a disk file. Once PROINDEX has completed indexing your text you will have to polish or finish the resulting file. This can be done in one of two ways. Either you can import the index into your word processed text and edit it to what ever form you desire, or you can use one of the utilities we provide to do this work for you. For instance, if the chapters were indexed separately you would run the MERGE utility to combine the separate indexes into one master index. Then use SORT to ensure your index is in alphabetical order. Finally, the PAGE routine will put top and bottom margins in your index. For details on the operation of these utilities see the chapters that describe each utility. Index Styles To help you choose an indexing style a number of different sample indexes are shown below. Each sample index also shows the settings of the index control parameters. These control 4 parameters are fully explained in the OPERATION section of this manual. Since there are 14 different control parameters only the values needed to create each sample index are shown below. All values are shown in the order they appear on the index control parameters menu. Primary References with Standard Page Numbers All command, 15 Backslash character, 10 BIOS calls, 1 Bold display, 14, 34 Data entry, 13, 23 Delete key, 6, 9, 14 [1] [N] [N] [0C] [ ] [P] [P] [:] [66] [Y] [N] [D] [T] [Y] Primary and Secondary References, Page Numbers Inclusive Options, 1, 9 - 15 Page, ending, 5, 12 length, 5, 11 numbers, 2, 4, 5, 11 Path names, 6, 8 - 11, 13 Quit, command, 15 program, 14, 19 RAM disk, 1, 8 [1] [N] [N] [0C] [ ] [P] [P] [:] [66] [Y] [N] [D] [T] [Y] Numeric Chapter with Page References Remarks, 7-39 Root directory, 5-7, 7-6 Screen, 2-5 Segments, 9-6, 9-9 System devices, 2-5 Text files, 7-123, 8-15 [1] [N] [N] [0C] [9- ] [P] [P] [:] [66] [Y] [N] [D] [T] [Y] 5 Alphabetic Chapter with Page References close, A.6, B.9 graph, B.9, D.17 pages, A.5, A.9, A.23 read, C.4, D.9 sort, D.5, F.172 wait, A.1 [1] [N] [N] [0C] [F. ] [P] [P] [:] [66] [Y] [N] [D] [T] [Y] Absolute Line References big, 23, 145, 876 cute, 1254, 3478 end, 12, 534, 95687 funny, 45, 897 little, 142, 23754 small, 9832 [1] [N] [N] [0C] [ ] [L] [P] [:] [66] [Y] [N] [D] [T] [Y] Page and Line References with Multiple References per Page at, 6:3, 6:14, 8:52 ape, 3:4, 5:15 dog, 7:1, 7:4, 7:23 egg, 5:5, 5:8 fire, 2:3, 4:14 gold, 8:9, 8:56 [1] [Y] [N] [0C] [ ] [B] [T] [:] [66] [Y] [N] [D] [T] [Y] 6 PREPARING YOUR TEXT General Some preparatory work must be done on your text file before PROINDEX will be able to index it. The amount and type of work to be done depends on the particular word processor that you are using. This section gives a general view of what must be done for most word processors. Specific instructions for some popular word processors can be found at the end of this section. In an ideal world there would be a single word processor and PROINDEX would handle its files without any work on your part. Unfortunately, there are many different word processors and each seems to save its files in its own particular way. To index text properly PROINDEX needs to see text in a specific format. How you get your text into the right format depends on your particular word processor. Word processors can be classified into one of two different groups. Each group saves its files in a different format. The first group saves its text files as ASCII, or something close, with no special or header information. The second group appends special, usually binary, information to the beginning of a text file. How can you tell what group your word processor belongs to? Generally, if you can use the MS-DOS TYPE command to display a text file on your screen you have a Group 1 word processor. Note that even though your text looks weird WordStar is a Group 1 word processor. Examples include: PC-Write, WordStar. If strange things happen when you try this test you have a Group 2 word processor. Examples include: MultiMate, WordStar 2000. You need to know what type or group of word processor you have to prepare your text for indexing. PROINDEX requires that the text to be indexed must be paged in some manner. PROINDEX does not understand any particular word processor's paging commands. How do you page your text? With almost all modern word processors this is an easy task. Simply instruct your word processor to print the text file to disk. The resulting disk file is properly paged and header information will have been removed. What if you can't print the text file to disk? PROINDEX supports a second option for determining where page breaks occur. You can select any single 8 bit character and use it to mark the end of each page. Using this method you first go through your original text file and insert a special character at the end of each page. Next you must know what group your word processor belongs to. If you have a Group 1 word processor skip the rest of this paragraph. If you have a Group 2 word processor you will still need to print your text file to disk before it can be used. Doing 7 this removes the header information appended to your text file. It also adds extra lines and page breaks but these will be ignored by PROINDEX. You are now just about ready to index your text. Before you can index it you will need a keyword file- see the section on these files that follows. You must also tell PROINDEX how it will determine where page breaks occur. If you want an index that matches your word processed document you must tell PROINDEX how many lines of text constitute one page. Typically this is 66 lines for a standard 8 A by 11 inch page. However, this number can be varied for different sizes of documents. If you are using special characters to mark the end of a page you must indicate this for PROINDEX. You will have to tell PROINDEX this is the way to determine page breaks and what the special character will be. The number of lines per page will not matter in this case. All this information will be entered on the index parameters menu page. PC-Write. Group 1, No special instructions are required for PC- Write. This word processor saves its files in straight ASCII. Simply print the text file to disk to page it or mark the end of each page with a special character and use the file as it stands. When printing a file to disk PC-Write typically appends a form feed, control-L or i, to the end of each page. You can use this as your special character to end each page. WordStar. Group 1, WordStar uses the most significant bit, MSB, of each character for its own purposes. While this makes your text look strange, PROINDEX can handle this text without any problems. Simply print the text file to disk to page it. Optionally use the special character to mark the end of a page and PROINDEX can handle the original text file directly. Microsoft Word, MultiMate, Pfs: Write, WordStar 2000. Group 2, These word processor save special header information at the beginning of a file. All files must be printed to disk before they can be used. Display Writer. Group 2, This word processor saves special header information at the beginning of a file. In addition, the files are saved in a non-ASCII form. All files must be printed to disk with the convert to ASCII option before they can be used. Non-ASCII Indexes A number of people have been using PROINDEX to index files that are not in ASCII. This includes things like mathematical symbols, Hebrew, and early bronze age hieroglyphics. In all these cases they equate an ASCII character to one of their own special characters. PROINDEX will index non-ASCII files just as it does ASCII files. 8 However, there are certain restrictions and requirements that must be observed for the special case of non-ASCII files. First, you should not use PROINDEX in the interactive mode with a non- ASCII file. PROINDEX sends characters straight to the screen in this mode and non-ASCII characters will not look like much on your screen. Second, if you are using a Group 2 word processor you must convert your disk file by printing it to disk. This removes header information from the beginning of the file. Finally, indexing non-ASCII files implies you are printing these files in some special way. You might be typesetting them or using a program like Fancy Font for instance. Remember that these printing programs work like typesetters and thus there is no correlation between lines of text in the file and the printed output. You must use a special character to mark the end of each page of text. Important notes. When you are trying to index a file that is not in ASCII it is important to remember to turn off the case conversion feature. Otherwise all "a" characters will be converted and compared to all "A" characters. This is probably not what you wanted. If you are using the full eight bits available for a character you must also remember to disable the "strip 8th bit" feature. Otherwise a character represented by the number 0FFH, 255 decimal, will seem the same as one represented by 7FH, 127 decimal to PROINDEX. 9 MAKING A KEYWORD FILE General PROINDEX needs a list of keywords or phrases to run. You must create this list before you attempt to index your text. We feel creating a keyword file is easier than marking all the references in your text file. So, how do you create this keyword file? The format of the keyword files is very important. The PROINDEX program expects to see a single keyword or phrase on each line of the keyword file. A line ends with a carriage return / line feed, f, combination. PROINDEX searches for an exact match to the keyword. Thus to find plurals or slight variations in spelling of a keyword or phrase it is best to shorten the keyword down to the lowest common set of characters possible. See the previous section on Indexing Techniques for more details. A keyword file would look something like this: Accuracy Assembly language Basic (space)C(space) (space)debug EPROM ejump High level language Input Key phrase Macro Macro Macro PL/M Some quick comments about this file will probably help you to understand it a little better. First, case is usually ignored by the program its only use is for your convenience. The program will pay no attention to it, unless you instruct it to. Locating certain words is quite difficult for a program. For example, finding C, as in the C programming language, can cause a problem. If just the single letter C is to be searched for the majority of your matches will be incorrect. The easy way around this is to add a space before and after the letter C in your keywords list. This forces the program to search for eCe, a more reliable indicator of the word C. Finally, to catch all references to debugging a program a keyword of edebug might be used. This will catch all words that start with the letters DEBUG. This includes debug, debugs, debugging, etc. The same rule applies to plurals. All extra letters should be left off of keywords. Note that no keyword or phrase may be longer than 70 characters. Keywords or phrases longer than this will automatically be truncated to 70 characters in length. The remaining characters 10 from that keyword will be used as the next keyword. No keywords will be lost by this process. One keyword will be shortened, and an additional junk keyword will be created. Each time PROINDEX completes a search through your entire text file for a keyword it will produce a pleasant warbling sound on your computer speaker. This lets you know that you have finished a given keyword search without having to shift your view to the keyword list. This feature is especially important when using a number of identical keywords to search for slightly different references, secondary references. Creating the Keyword File The format of a keyword file is more critical than that of a text file. PROINDEX expects to find a disk file containing a number of lines of text. Each line of text has one keyword or phrase followed by a carriage return / line feed pair, f. Keyword files must be written in ASCII unless you run intend to run PROINDEX in the non-interactive mode. If you are using a Group 1 word processor your keyword file should be acceptable to PROINDEX. If you are using a Group 2 word processor you must convert the file to ASCII. Most Group 2 word processors offer this option. For an explanation of Group 1 & 2 word processors see the previous section. Note that a keyword file that has been printed to disk with page breaks will not work correctly with PROINDEX. If your word processor will not let you convert your file to ASCII, or you can not use it to create a straight ACSII file, you can use the EDLIN text editor that came with your version of the MS-DOS operating system. This program will produce an ASCII keyword file. Some word processors end a text file with the control-Z character while others do not. Note that PROINDEX does not care whether a file ends with a control-Z. This character will simply be ignored if it is present. 11 INDEXING TECHNIQUES Using a computer to generate an index is much faster than indexing by hand. Computer indexing is also a significantly different process than hand indexing. The human brain does a wonderful job of selecting references by context. A computer selects citations based on an exact pattern match. These differences in indexing technique are significant. The following section details a number of techniques that are useful in computer indexing. Interactive Control PROINDEX will work in either an automatic or interactive mode. The automatic mode is only suitable for finding exact matches to your references. In general, the automatic mode is used by lawyers, court reporters, and researchers. These people are only interested in precise matches to a reference. For everyone else the interactive mode will probably produce a better index. Why should a computer program, that is supposed to do your work for you, require interactive input? The answer is quite simple. The process of creating an index requires human input concerning the context each reference is found in. There is an entire profession of people that make indexes for authors. If it were possible to replace these people with a computer program it would have been done a long while ago. So what does PROINDEX do for you? It performs all the mechanical aspects of indexing. PROINDEX will find all references to a keyword or phrase for you. In the interactive mode, however it is up to you, the author, to decide whether or not to include each reference in your index. You make this decision by viewing each keyword or phrase in context. How about a simple minded, but well-intentioned, example? Suppose you have written a book on English usage. Now you want an index entry for the section on prepositions where the word "the" is discussed. If your keyword list simply says find all references to "the" your index will be useless, and exceedingly long. You want to find specific references to the word "the". PROINDEX locates all references to a given keyword or phrase and asks you whether or not to include each one. Of course you would have to be out of your mind to put the word "the" in a keyword list. You would spend the rest of your life answering PROINDEX's "Mark this reference?" question. The interactive control of PROINDEX gives you four different options each time it locates a reference. You can mark or bypass each specific citation. You can also skip all further references to that keyword or mark all further references. These options give you complete control over the indexing process. 12 Page Breaks Another problem with indexing concerns the form of the text you want to index. If your index page or line references are to have any meaning they must refer to real page or line numbers. The easiest way to do this is to instruct your word processor to print your text file to disk. Use the printed text file as input to PROINDEX and tell PROINDEX how many lines there are per page of text. Normally 66 lines per page. Other word processors mark the end of each page with a special character. With this type of text simply tell PROINDEX what character each page ends with. PROINDEX will then correctly find the page breaks in your document. There are also times when you will want to mark page breaks in text yourself. Suppose you are going to typeset a word processed document. The page breaks for the typeset document will not match the page breaks for the word processed document. Yet you want the finished index to match the paging of the typeset document. In this case you should typeset the document before indexing it. Once the document is typeset you can use it to mark the page breaks in your original word processed document. Chose any single 8 bit character to mark the end of each page. Once these page break indicators are in place, run PROINDEX. The resulting index will match the typeset document. Keyword Tips Constructing a good keyword or phrase list is not as easy as it seems. The quality of your index depends on this list. Thus, making sure your list is as complete as possible is important. The computer is extremely picky about what you keywords look like. PROINDEX always looks for an exact match for your keyword or phrase. This presents some problems when building a keyword list. For instance, suppose you are writing a computer program manual and you want to catch all references to "debugging" a program. If your reference list contains the word "debugging" you are going to miss a number of possible citations. All references to "debug", "debugged", and "debugs" will be skipped! The way to solve this problem is to always use the root of a keyword in the reference list. In the previous example the keyword should have been "debug". This will catch all references to any usage of "debug". In some cases getting the root of a keyword will require you to drop a few letters from the end of your keyword. This is preferable to getting too specific with a keyword and missing entries. Also remember that the ASCII space character is useful in delimiting a keyword. If you are looking for the keyword "out" 13 you will find a number of false references inside words like "shout". To avoid this problem put a space in front of the keyword "out". This forces PROINDEX to search for the sequence eout. You can also add a space to the end of the word, but remember this rules out finding plurals and other variations on the keyword. See the chapter on keyword files for more information. Secondary References It is possible to create an index with secondary references in it using PROINDEX. This will require some work on your part. A secondary reference can be thought of as an and function. You want a reference to X and Y. Since secondary references require interpretation of document context, PROINDEX must be run in the interactive mode to generate them. A typical index with secondary references might look something like this: Keyword, date, 2, 7, 9 files, 2, 5, 6 size, 5, 7, 12 If your keyword list simply contains the three phrases: keyword date, keyword file, and keyword size you will run into problems. Your document might be talking about "keyword lengths" but not refer to them in that exact way. For instance as: "the length of a keyword". So how do you create secondary references? The easiest way is to shorten the original phrase down to a single word. If we do this for the previous example we get three keywords that are exactly the same. Original Keyword Revised Keyword Keyword date Keyword Keyword files Keyword Keyword size Keyword The draw back to this approach is that you now have three keywords that are exactly the same. How do you know which is which when PROINDEX shows them in context? To solve this problem keep two different reference word lists. The first lists complete keywords or phrases. We keep a printed version of this list next to us while indexing. The second list is the abbreviated list that will be used by PROINDEX. The first time PROINDEX shows you the reference "keyword" in context, you look at your printed list to see that you want "keyword dates". When PROINDEX finishes with this reference it will beep and move to the next similar reference. Move down your printed list to find that the next reference is "keyword files". You repeat this process for all three references. The resulting index will look as follows: 14 Keyword, 2, 7, 9 Keyword, 2, 5, 6 Keyword, 5, 7, 12 Note that although each keyword is the same the page references are different. You have created true secondary references. You must now edit the output index file produced by PROINDEX to a form that reflects the true meaning of these references. 15 CONFIGURATION Not everyone has the same type of disk files or uses the same word processor. In addition, everyone seems to want a slightly different index style or format. The individual user thus needs some way to customize this program. That way you can make the program work just the way you want it to, and we don't have to supply numerous different versions of the program. PROINDEX uses a configuration file to control its operation. It creates this file the first time you use it. An example of this file can be seen in figure 7. The configuration file is usually called PROINDEX.CNF. This file initially contains all the default values for control of PROINDEX operation. These default values are the ones shown below. While this file is in ASCII and can be modified with any text editor, there is no reason to do so! The PROINDEX program automatically changes the configuration file anytime you make a change on the index control parameters menu screen, as seen in figure 4. Changing any answer on this menu alters the configuration file automatically, and changes the way PROINDEX operates. This file gives you control over most of the internal functions of the index program. These functions control the following features: Starting page number of text Multiple index references per page Should a special character end a page String to precede each index reference Should the index reference pages, lines, or both Number of lines per page Conversion of lower to upper case Should the 8th bit be stripped from text Index sent to disk or printer Path for indexes sent to disk Run the program in interactive mode NOTE: If at any time the configuration file can not found by PROINDEX, then a new one will be created with the following default values: 16 Temporary work: 0 Starting page number: 1 Allow multiple references/page: N Special character ends page: N End of page code: 0C Page reference code: none Reference page/lines/both: P Lines start at: P Character to separate pages from lines: : Lines per page: 66 Convert to upper case: Y Strip 8th bit: Y Index sent to disk/printer: D Disk index path: T Run in interactive mode: Y The configuration file that PROINDEX uses has to have a name and reside somewhere on a disk. It is normally called PROINDEX.CNF, however, you can change this name in one of two ways. This allows you to maintain different configuration files for different types of text. If you invoke PROINDEX by typing: PROINDEX c:\path\filename.ext The program will assume that the configuration file must have the name you specified, "filename.ext" and be found on the path name, "c:\path" you gave it. If it is not found there it will be created there. This method of supplying a name for the configuration file takes precedence over all others. If no argument is passed to PROINDEX when the program is started the program environment strings are examined. See your DOS manual for an explanation of the "SET" command. If PROINDEX finds an environment string of the form: PROINDEX.CNF=d:\path\yourfile.nam it will use that name and path for the configuration file. Once again, if the file can not be found it will be created. The environment string area will not be examined if you pass an argument to the program when it is started. MS-DOS environment strings are easy to set up. Just remember that environment strings established by the SET command only remain in effect until the next time you reboot your system. If you want to permanently set an environment variable you must include the proper SET command in your autoexec.bat file. A typical SET command looks like: SET PROINDEX.CNF=c:\path\yourfile.ext If you do not pass an argument to PROINDEX and you have not used the DOS SET command to redefine the configuration file name, then PROINDEX will automatically use the name PROINDEX.CNF for the configuration file. If you are running under MS-DOS 3.0 or higher 17 then this configuration file will be created, saved, and retrieved from the same subdirectory that the program PROINDEX.EXE is saved in. If you are running MS-DOS 2.X then this configuration file will end up on the default disk drive. Sorry, but path support under this version of DOS is incomplete. If you are now totally confused don't panic! The special options mentioned above for renaming the configuration file are advanced features of the program. You do not have to use these features. Just invoke the program by typing: PROINDEX and let PROINDEX take care of the configuration file all by itself. 18 OPERATION Getting Started Before you do anything at all copy the files from your PROINDEX master disk to a work disk or a subdirectory on your hard disk. Take the original PROINDEX disk and put it away. No installation is needed for either program. Simply copy those two files to the disk drive you want to work on. They will work on a floppy, RAM, or hard disk. If this is an update of PROINDEX that you are installing, please delete the old version of PROINDEX and ALL copies of the old configuration file, PROINDEX.CNF. If the new version of PROINDEX finds an old configuration file it will return back to DOS. You must delete the old configuration file before the new version will run. No work you have done will be lost. The first time you run PROINDEX it will create a configuration file. Any changes you make to the configuration of PROINDEX will automatically be saved in this file. See the previous section for more information on this configuration file. The PROINDEX program is easy to use. Your first step is to create a text file to index. Next you need a list of keywords or phrases. See the sections on text and keyword files for more details on creating these files. To start the program type: PROINDEX or PROINDEX b:\path\config.fil NOTE: {cr} stands for the newline/enter/carriage return key. The optional file name and path in the second example instruct PROINDEX to use that name and path for its configuration file. Opening Screen PROINDEX starts by clearing the screen and printing an opening message. This message includes the version number of the program and a copyright notice, as seen in figure 1. The program then attempts to read the configuration file. If the file can not be found a default configuration file is created and read. See the previous section on configuration for more details. This initial screen will remain up for about three seconds. The program will then automatically move to the next screen. Data Entry Before explaining the index control parameters screen some information about entering and correcting data is necessary. All keyboard entry in PROINDEX uses a special screen handler. This screen handler lets you easily enter and correct data entries. 19 Most data entry in PROINDEX is done based on something called a data field. The program only allows data to be entered in certain places, fields, and verifies the entries as they are being made. A field begins with a "[" character and ends with a "]" character. You may enter data into a field as long as the cursor is present there. To enter information simply type on your IBM PC keyboard. You can use either the backspace or the delete key to correct typing errors. Note that you can not delete your way past the left hand "[" character. If default data is present in a data field you can leave it there by going to another data field or type over that information. Once you start typing over data the screen handler removes all data in that field. The screen will still show the old data as a typing aid. There are several different ways to move from one data field to another. Hitting a return key after entering data moves you to the next sequential, lower, field. You can also use the IBM PC's cursor control keys to move between fields. The a key, or a Control-A, will move the cursor to the previous field, (up). The b key, or Control-F, will move the cursor down one field. Note this is the same as hitting the return key. Finally, the g key, or Control-W, performs a special data entry function. Hitting this key enters all data in all fields as it stands. When you hit the g key all data will be entered as it appears on the screen. This feature lets you avoid answering all the questions on the screen each time you use the program. You need only answer those questions that you wish to change. Then use g to advance to the next menu. Most data fields will already have information in them when the cursor appears there for the first time. This information is the normal default value for that question. You may leave this information as it is by hitting return, a, or b. This will enter the default data and move to another data field. You may also change the default data by typing over it and then moving to another field. Program Operation Figure 2 shows what the second screen will look like. This screen or menu is designed to collect the name of the text file to index and the name of the keyword list. You can also tell PROINDEX what disk drive and path to use for a work drive. Entries on this screen work as follows: The text file. First you must enter the path and name of the file to index in response to the question, "Path & name of file to index?". It can be any valid path and file name up to 43 characters long. Note that the file name itself can only be eight characters with a three character extension. The path name may include a drive specifier. Use of a drive and path is optional. The following examples show correct responses to this question: 20 [filename.ext ] [c:filename.ext ] [\work\text\filename.ext ] [d:\text\filename.txt ] The keyword file. Next you must enter the path and name of the keyword list. The same restrictions apply to this path and file name as to the index file name. Examples include: [keyword.dat ] [c:keyword.dat ] [\work\text\keyword.dat ] [d:\text\keyword.dat ] Note: If either the file to index or the keyword list is left blank PROINDEX will not be able to index your text. You can still access the control parameters menu or go back and edit the file names. However, if you have no text or keyword file name the program will not be able to index. PROINDEX will display an error message if you try to run an index and return to the main menu. You can then correct the file name or quit the program. Temporary work path. This question is optional. If not answered the displayed value will be used. The temporary work path describes what disk drive and path should be used for work files. If you leave this field blank the default drive, where ever you are at the moment, will be used. If you need to specify only a disk drive simply enter a single letter followed by a colon specifying the drive name. Full path names may also be specified. Note that PROINDEX does not care whether your path name ends with a backslash character. If this character is not found PROINDEX will add it when it needs to. All work files are temporary and will be erased by PROINDEX when the indexing job is done. Examples include: [ ] [d: ] [\work\temp ] [\work\temp\ ] [c:\temporary\ ] Once this information is entered the screen will display the main control menu shown in figure 3. All program operation starts at this menu and returns back to it when done. This menu gives you a choice of six different control options. Simply type the letter that corresponds with what you want to do next, followed by a carriage return. The choices are as follows: 21 C=Continue a previous indexing session E=Edit file names and paths above I=Index the text P=Edit the index control parameters Q=Quit the program X=eXecute an MS-DOS command At this point you can go back and edit the file names and paths previously described with the E option. You can start the indexing of your text with the I option or you can continue a previously halted indexing session with the C option. You can go change the index control parameterswith the P option. You can switch to a menu that will let you invoke MS-DOS commands from within PROINDEX by using the X option. Or you can quit the program and return to DOS, Q option. The program will ask you to chose one of these six options. Enter the letter corresponding to your choice and hit a return. When you are ready to start indexing hit the I or C key. This will take you to the index menu shown in figure 5. Finally, to leave the program type the Q key. This will return you back to the MS-DOS operating system. 22 Executing MS-DOS Commands It would often be nice to be able to run an MS-DOS command without leaving the PROINDEX program. For example, suppose you quit indexing in the middle of a session and want to see what you have done so far. Select the X option on the main menu and you will go to the MS-DOS command sub-menu. This menu is shown in figure 6. It lets you run MS-DOS commands without leaving PROINDEX. When this menu appears on the screen the cursor will be positioned at the beginning of a large blank data field that follows the word "Command". It looks as follows: Command [ ] Simply type in any valid MS-DOS command up to 64 characters in length. For our previous example we wanted to see our index. So type in: "type indxfile.ind", followed by a return. This will display your index file just as if you had exited PROINDEX and done the same command on your PC. You can use the Control-S key to start and stop the printing of a long index. Command [type indxfile.ind ] You might also want to display a directory of disk files while running PROINDEX. Use the same disk directory command you would normally use to see what files are available. For example: Command [dir b:*.txt ] will display all disk files on drive B: that end with the extension ".txt". Once an MS-DOS command has run PROINDEX will prompt you with a message: Hit any key to return to PROINDEX Striking any key will return you to the main menu of PROINDEX. If you enter the MS-DOS command menu by mistake and do not wish to execute any command simple leave the data field blank and hit return. This will bring you back to the main menu. 23 PROINDEX Control Parameters The style of index created by PROINDEX depends on how a number of control parameters are set. You can change these parameters at any time to change the look or style of your index. All changes to these parameters are made on a separate menu page. Figure 4 shows the index control parameters menu. This screen is used to modify the index control parameters. All changes made to this menu will be remembered by PROINDEX. They are saved in the disk configuration file. You never have to return here unless you want to change a value. This screen controls the following options: Starting page number. Indexing very large documents can result in problems. You may not have room for the necessary temporary files. Or, perhaps, it is just easier to break a long file up into several shorter ones. How can you handle an index in this case? PROINDEX allows you to specify the starting page number of a document. Suppose you had a document broken up into two pieces. The index for the first section assumes that it starts at page one. When you go to index the second file you can tell PROINDEX that this file starts at page 41, for example. All you need to know is the starting page number of each section of text you index. The program asks you, "Starting page # of text (1-9999)?". You may respond with any number from 1 to 9999. The default value is one. Examples include: [1 ] - start at page 1 [195 ] - this file begins at page 195 Multiple references/page. A normal index contains only a single reference for each keyword per page. Thus if a keyword is found five times on one page, only one reference to that page will be found in the index. This feature can be defeated if required. The, "Multiple references per page (Y/N)?", question turns the feature on and off. Examples include: [Y] - Let multiple references per page be flagged [N] - Only 1 reference per page to a keyword End of page. Sometimes it is inconvenient to create disk files that resemble the printed output of your text. Yet the index program needs to know what page each reference it finds is on. PROINDEX lets you use any eight bit character to mark the end of a page. If this feature is turned on the program will treat each special character as an end of page indicator. Why would you want to mark the end of a page by manually inserting a special character? Suppose your word processed text is to be typeset. The page breaks of your word processed document will not match the page breaks of your typeset document. PROINDEX 24 will handle this problem if you mark the page endings in your word processed document with a special character. You can use any character you want. There are two questions on the Index Parameters menu that control this feature as follows: Should a special character end a page (Y/N): [N] Page ending character code in hex: [0C] If you reply with a "Y" to the first question, then PROINDEX will use ONLY the special character defined in the second question to define the end of a page. If you reply with an "N" to the first question then PROINDEX will count lines to determine when a page ends. The special character that marks the end of a page can be defined, IN HEX, in the second question. A table of hex verses ASCII is provided below to help you select a character. Note that PROINDEX defaults to the i character, Control-L or hex 0C. Hex Char Hex Char Hex Char Hex Char 00 NUL 10 ^P 20 SP 30 0 01 ^A 11 ^Q 21 ! 31 1 02 ^B 12 ^R 22 " 32 2 03 ^C 13 ^S 23 # 33 3 04 ^D 14 ^T 24 $ 34 4 05 ^E 15 ^U 25 % 35 5 06 ^F 16 ^V 26 & 36 6 07 ^G 17 ^W 27 ` 37 7 08 ^H 18 ^X 28 ( 38 8 09 TAB 19 ^Y 29 ) 39 9 0A LF 1A ^Z 2A * 3A : 0B ^K 1B ESC 2B + 3B ; 0C FF 1C ^\ 2C , 3C < 0D CR 1D ^] 2D - 3D = 0E ^N 1E RS 2E . 3E > 0F ^O 1F US 2F / 3F ? Hex Char Hex Char Hex Char Hex Char 40 @ 50 P 60 ' 70 p 41 A 51 Q 61 a 71 q 42 B 52 R 62 b 72 r 43 C 53 S 63 c 73 s 44 D 54 T 64 d 74 t 45 E 55 U 65 e 75 u 46 F 56 V 66 f 76 v 47 G 57 W 67 g 77 w 48 H 58 X 68 h 78 x 49 I 59 Y 69 i 79 y 4A J 5A Z 6A j 7A z 4B K 5B [ 6B k 7B { 4C L 5C \ 6C l 7C | 4D M 5D ] 6D m 7D } 4E N 5E ^ 6E n 7E ~ 4F O 5F _ 6F o 7F DEL 25 Chapter / section references. Index style is often a matter of individual choice. Some people prefer indexes that clearly separate chapter entries from each other. To this end, each individual chapter will start with page number one and the index must specify which chapter each page reference is referring to. A typical example might look as follows: ASCII,1.1, 1.7, 2.5 bull,2.3, 3.7 This index example tells us that a reference to "ASCII" occurs in chapter one on pages one and seven, and in chapter two on page five. PROINDEX gives you the ability to build your indexes in this fashion. A single question on the index parameters menu controls this option: String to precede each index reference: [ ] Up to five separate characters may be specified here. As long as this field is blank, no string will be added to each page reference. Leave the field blank for a straight numerical index. If you want to mark separate chapters put in an appropriate string in place of the blank one. To generate the example index above the string was set to: String to precede each index reference: [1. ] for all indexing work done in chapter one. The string was then changed to "2. " while indexing chapter two. When all chapters were done the MERGE utility was used to produce a single master index. WARNING: the string must only contain printable ASCII characters. The first blank character found when going from left to right terminates the string. To remove an old string type over it with the space bar. Page, line, or page & line numbers. PROINDEX can create indexes that reference page numbers, line numbers, or a combination of page and line numbers. A page number is simply the page that a reference was found on. An index that shows only line numbers shows the number of lines from the beginning of the text file for each reference. An index created with both page and line numbers will tell you what page and what line the reference can be found. What is meant by line number? After all, line numbers could be counted from the beginning of a page and include all blank lines, or they could be counted based on only those lines that have text on them. PROINDEX allows you to select either case with a second question. Finally, page numbers and line numbers have to be separated from each other. The choice of a character to separate them is up to you in the third question. The three questions relating to this option look as follows: 26 Should index reference pages/lines/both (P/L/B): [P] If both, line numbers begin with page or text (P/T): [P] If both, character to separate pages from lines: [:] The second and third questions only apply when you have selected an index with both page and line numbers. See the first question. If both page and line numbers are selected you must make a choice concerning what the line numbers will reference. Entering a "P" will tell PROINDEX to count lines from the beginning of the page. If you are using the standard 66 lines per page, then the line numbers will run from 1 - 66. Entering a "T" tells PROINDEX to count only those lines that have text on them. Thus the first line of text will be line number 1, even if it is located half way down the page. Blank lines will not be counted. The third question asks for an ASCII character to separate page numbers from line numbers. The default is the colon ":". With this default character page and line numbers in the index will look as follows: ASCII, 23:7 test, 1:24, 3:1, 15:47 zebra, 99:65 Note that you can use any character you want, except for a space or a "~", to separate pages from lines. However, there are some very bad choices. Do not, for instance, use the comma character "," since this would make it difficult to see what was a page number and what was a line number. Lines per page. Most text contains 66 lines per page. This is not always the case, however. This data field lets you change the number of lines per page for your text. The program asks, "Lines per page (1 - 999)?". You can respond with any number from 1 to 999. [1 ] - each line is a new page [132] - each new page begins after 132 lines Text case. Text files usually contain both upper and lower case letters. This can cause some interesting problems when indexing a file. If the program pays attention to case then the words "Test" and "test" do not match. Thus this question asks, "Convert lower to upper case (Y/N)?" Normally the default value of "Y" is used. Examples include: [Y] - yes, do case conversion [N] - no, leave case of letters as it is 8 to 7 bit ASCII. Some word processors, WordStar for one, use the 8th data bit to indicate special things. This makes matching text difficult. This feature lets you strip the extra bit from every character before comparisons are made. The program asks, "Strip 8th bit from text (Y/N)?". Legal answers are: 27 [Y] - yes, remove extra bit [N] - no, leave text alone Output. The actual index produced by PROINDEX has to go somewhere. This question asks, "Index sent to Disk or Printer, (D/P)?" Normally it is sent to the disk, as indicated by the default answer "D" to this question. You can, however, redirect this output to the printer, LPT1, if you so desire. Simply answer the question with a "P" for printer. [D] - output to the disk [P] - output to the printer Indexes that are sent to the disk use the same file name as the keyword file, with an extension of ".IND". The related "disk index path" question supplies the path for this index file. You can use the path for the text, keyword, or work files for the index file. Thus if the file to index was called "filename.txt" its index would be the file "filename.ind". See the "disk index path" section below for an example of what path would be used with this file. Disk index path. To keep the amount of information needed to run this program to a minimum the path name for an index saved to disk is taken from one of three places. Thus, you don't have to specify another path name but you do have to identify the source of the path. A disk index file will always be saved in one of three possible paths. These are the paths specified for the: text file, keyword file, or work files. The normal default is the text file path. Thus your index will end up in the same place as your text file. You can change this path by responding to this question as follows: [T] - use the text file path [K] - use the keyword file path [W] - use the work area path Interactive mode. The program normally runs in an interactive mode. That is, it requires responses from you before it will include a reference in the index. Indexing normally requires human thought. If you wish to circumvent this feature the program can be made to run all by itself. When it asks, "Run in interactive mode (Y/N)?", change the answer to an "N". [Y] - yes, run in interactive mode [N] - no, program runs by itself 28 Indexing Once you have told PROINDEX to start indexing your text it replies by printing the "Creating temporary work files" message. After the program has created the work files, this could take up to several seconds depending on the speed of your computer, it will proceed to the actual text file indexing. When PROINDEX is ready to begin the indexing process it will change the screen from that shown in figure 3 to the one shown in figure 5. This screen tells you what page of text you are on, what the present keyword is, and what number keyword it is using. The keyword number gives you an indication of how far through your index you have progressed. When PROINDEX is finished looking for a given keyword it will make a pleasant warbling sound. In the automatic mode the bottom half of this screen will remain blank. In the interactive mode PROINDEX will display from 8 to 10 lines of text surrounding each occurrence of the keyword it finds. Double spaced text cuts this down to 4 - 5 lines. The keyword itself is always shown in bold face type. The program searches progressively through your text file looking for all occurrences of the first keyword, then the second, etc. If the keyword can not be found anywhere in your text file PROINDEX displays the message on the screen, "No reference to keyword: XXXXXXXX". The program then pauses, waiting for you to hit a key before it will continue the indexing process. When a keyword is located in your text file the program stops and waits for you to tell it what to do. As shown in figure 5, it asks you, "What would you like to do?". You respond by hitting one of five possible keys. NOTE: No return key entry is needed here. The program responds immediately to any key that is typed. Error handling here has been sacrificed for speed of execution. If you hit any key other than "A", "B", "M", "Q", or "S" that key is ignored and your computer beeps. The five single letter responses correspond to one of the following options: B - Bypass. Bypass this one occurrence of the keyword in text. This option has no effect on further occurrences of this keyword in text. M - Mark. Mark this occurrence of the keyword in the index. This option may or may not have an effect on other occurrences of this keyword in text. Note that if the "more than one reference per page" option has not been enabled then once a reference has been marked no other occurrences of it on this page will be found. A - All. All further references found for this keyword will automatically by marked in the index without asking. The "more than one reference per page" option still controls whether multiple references per page are allowed. Once the program 29 finishes searching the entire text file for this keyword it will return to the normal interactive mode. Q - Quit. Stop the indexing process and return to the operating system. When a "Q" is typed the program asks, "Are you sure?". If you type a "Y" for yes the program will exit to MS-DOS. Otherwise the program continues where you left off. You may continue an indexing session you ended with this command at a later time. S - Skip. Skip all further references to this keyword in your text file without asking. Once the program finishes searching the entire text file for this keyword it will return to the interactive mode. 30 OUTPUT The exact form of the index created by PROINDEX depends on the options you have chosen on the Index Control Parameters menu. In general, this index will always have a basic form. Each line of the index will start with a matching line from your keyword or phrase list. This information will be followed by a comma, one space, and page or line references. Each page or line reference will be separated from the previous one by a comma and one space. The last page or line reference will not have a trailing comma. A typical line would look as follows: keyword or phrase, XXXX, YYYYYY, ZZZ The end result of an indexing session should be an index. Where does that index go? With PROINDEX you have one of two different options. The index can be saved as a disk file, or it can be sent to the printer- as it is created. It is highly recommended that you save your index as a disk file. Sending an index to the printer is a one way trip. There is no way to do anything with that printed index except read it. An index saved as a disk file can be manipulated with a text editor and included into your document. Indexes that are saved to disk must have a file name. PROINDEX always takes the name of your text file, removes its extension, and saves the resulting index with a ".ind" extension. Thus, if your text file was called "bigbook.doc" the resulting disk file index will be named "bigbook.ind". This disk file index will be a pure ASCII file. The path and disk drive that the file is created on are determined by you. One of the Index Control Parameters determines where this disk file will be created. See that section of the manual for more information on where the file will be saved. Saving an index as a disk file does have an inherent problem. What happens if a previous index exists on your disk with the same name? Don't worry- PROINDEX checks for this condition and presents you with the following message and options if it finds a duplicate index file name. PROINDEX is unsure whether you want to erase this old index and start fresh or save the old index. You could also want to continue an old indexing session. Please decide what you want to do and select the appropriate response below. R = Rename old index, last.ind to last.ibk, and start fresh C = Continue a previous indexing session (destroy old index) Q = Quit indexing now (saves old index) Note that PROINDEX gives you several options here. You can stop what you are doing without changing anything by using the Q command. You could tell PROINDEX to rename the old index to a backup file name and then create a new index by using the R 31 option, the most common choice. Or you can instruct PROINDEX to erase the old index and create a new one with the C option. These options are designed to keep your indexes as safe as possible. If you chose to send the index to a printer your printer must be connected to LPT1. If you attempt to send an index to a printer that is not selected or turned off the program will pause and wait for you to enable the printer. When the printer is ready you may hit any key except the "Q" to proceed. Hitting a "Q" will cause the program to return to the main menu. NOTE: If you tell the program to proceed at this point even though the printer is not ready the program will simply repeat the error message. The only way out of this state is to ready your printer or to type a "Q" to return back to the main menu. Indexes produced by PROINDEX can conform to several of the styles suggested by the Chicago Manual of Style. A sample index follows. Note the spaces used to deliniate keywords are still present in the PROINDEX output. Accuracy, 5, 15, 23 Assembly language, Basic, 1, 2, 3, 4, 5, 6, 7, 14, 16, 18, 19, 22 (space)C(space), 4, 5 (space)debug, 1, 2, 5, 6 EPROM, 10 (space)jump, 14, 56 High level language, 12, 13, 15 Input, 1, 3, 7 Key phrase, 12, 13 Macro, 12, 13, 14, 15 Macro, 13, 15 Macro, 15 PL/M, 10 Note that PROINDEX does not attempt to produce a truly finished index. It is assumed that some of your keywords will have to be modified and extra spaces must be removed from the index. This task is left up to you. Simply import the index disk file into your word processor. There you can edit the index to any form your desire. We typically edit the index file shown above to the following form: 32 INDEX Accuracy, 5, 15, 23 Basic, 1-7, 14, 16, 18, 19, 22 C programming language, 4, 5 Debugging, 1-2, 5, 6 EPROMs, 10 Jump instructions, 14, 56 High level languages, 12, 13, 15 Input devices, 1, 3, 7 Key phrases, 12, 13 Macros, 12-15 assembler, 13, 15 commands, 15 PL/M, 10 33 PROGRAM SPEED The amount of time it takes for PROINDEX to index your text depends on the type of computer you have, and the type of disk drives you are using. An IBM AT is much faster than an IBM PC. Hard disks are much faster than floppy disks. Finally, RAM disks are much faster than either floppy or hard disks. PROINDEX is a disk intensive program. It makes extensive use of your disk drives. In addition, PROINDEX must work with four separate disk files at all times. These are: your original text and keyword files, and working copies of the text and keyword files that may have case conversion or bit stripping done on them. Thus, the type of disk you use for a work drive is critical to the speed of operation of PROINDEX. If you have a hard disk drive it is better to use it as the work drive instead of a floppy disk. This will speed the operation of PROINDEX by a factor of 5 to 10 times! Better yet, use a RAM disk drive. This approach is 2 to 4 times faster than a hard disk. A RAM disk driver comes with all MS-DOS operating systems of version 3.0 or higher. If you have a lower version of MS-DOS there are a number of public domain RAM disks available. If you can't find one we will be happy to send you one if you mail us a self addressed stamped disk mailer and a formatted floppy disk. We strongly recommend the use of a RAM disk for most medium to large indexes. An index that takes several hours to run on a floppy disk system can be done in a very few minutes, on that same computer, using a RAM disk. All you need is the enough memory to create the RAM disk. The RAM disk must be big enough to hold work copies of your text and keyword files. PROINDEX will check to make sure there is enough room on its work drive before it starts indexing text. Practically speaking, a RAM disk using 150,000 bytes of memory is adaquate for most small to medium size indexes. It is well worth the effort of creating a RAM disk even if you only use it with PROINDEX on a temporary basis. So, how do you create a RAM disk? Assuming you have version 3.0 or higher of MS-DOS you have all the tools you need to do this. These versions of MS-DOS come with a device driver called VDISK.SYS. To use it you must have or create a disk file called CONFIG.SYS on the disk drive that your system boots up with. This disk must also have a copy of the file VDISK.SYS on it. First copy VDISK.SYS to you system boot disk. For floppy disk drive systems this is the disk that normally resides in drive A:. For hard disk systems this is the hard disk, usually drive C:. If this disk drive already has a CONFIG.SYS file on it you must edit that file, with an ASCII text editor, to include a new line. If your disk does not already have this file you must create it, 34 with an ASCII text editor. You can use the EDLIN program that comes with MS-DOS to do this. You want to include the following line in this CONFIG.SYS file: device=vdisk.sys 150 This tells MS-DOS to build a RAM disk using 150 K bytes of memory. You can change this number to any amount of memory you have free. Note that the RAM disk will not be created until you reboot your system and everything on it will be erased each time you reboot the computer. See pages 4-14 through 4-18 of your MS- DOS manual for more information on the VDISK program. The RAM disk that is created this way will always be the next disk dive letter past your previous last drive. Thus on a two drive floppy disk PC the RAM disk will appear as drive C:. On a PC, XT, or AT with a hard disk the RAM disk will appear as drive D:. When PROINDEX asks for a work drive just enter this disk drive letter followed by a colon. See figure 2. Thats all there is to speeding up PROINDEX! 35 UTILITY PROGRAMS COMMON.TXT The COMMON.TXT file is a list of common words in the English language that you probably don't want to include in your index. It contains about 400 words. Please check these words before using them in conjunction with the REMOVE program. You may want to alter this list before using it. New words can be added to this file, or old ones can be removed. All changes to the file must be done in ASCII. See the STRIP utility if you use WordStar. All file entries must be in lower case and sorted alphabetically. Each entry consists of a single word followed by a carriage return / line feed pair. To add a new entry, type the new word(s) at the beginning of the file, following each word with a carriage return. Once all entries have been made run SORT on the resulting file. This will guarantee the file is in alphabetical order. Make sure there are no duplicate words in this file! You can do this by running MAKEWORD on the file. MAKEWORD This utility will take your text file and turn it into a sorted list of unique words. It will also count the frequency of each word as an option. This is the easy way to make a reference list for PROINDEX. Only words starting with a letter of the alphabet will be put into the list. Single punctuation characters following a word will be removed. The program is run as follows: makeword infile outfile [-ic] The "infile" is your original text file. This file must be in ASCII. WordStar users must first run STRIP on their text to turn it into ASCII. The "outfile" is the name of the disk file where the resulting list of words will be stored. Two options are possible with MAKEWORD. The -i option tells MAKEWORD to ignore case in its output. The output file will be entirely in lower case if this option is used. The -i option should probably always be used. The second option -c tells makeword to count the frequency of each word in the list. makeword data.txt data.lst -i instructs MAKEWORD to find all unique words in data.txt, convert them to lower case, and store them in the file data.lst. makeword data.txt data.lst -ic instructs MAKEWORD to find all unique words in data.txt, count how many times they occur, convert them to lower case, and store them in the file data.lst. The resulting file will look as follows: 36 a 234 at 23 if 12 wonderful l5 zebra 1 MAKEWORD needs disk storage space in order to work. The sorted word file will initially be slightly longer in size than your original text. This will shrink as duplicate words are removed. The SORT utility is invoked by MAKEWORD and it needs additional storage space. On average, you must have 2.1 times the text file size worth of free disk space to invoke MAKEWORD. Thus if your text file is 14,128 bytes long you need 14,128 * 2.1 = 29,669 bytes of free disk space. MAKEWORD invokes an external sort routine called SORT. This routine MUST be on the same disk as the MAKEWORD utility. In addition, the input and output file names must be different! MAKEWORD checks them to make sure they are not the same and prints an error message if they are. MAKEWORD takes time to run! Using a RAM disk and a 6 Mhz AT it takes 51.5 seconds to turn a 30,700 byte document into a sorted unique word list or 113.3 seconds to do this on a 65,200 byte file. This comes to 14 / 28.4 seconds to convert into words, 32.5 / 75.6 seconds for the sort, and 5 / 9.3 seconds for removing duplicate words. As you can see, the larger your file the slower the operation of this program. The program makes three passes at the data. First, it turns your file into a list of words, removing non-valid words and extraneous punctuation. This process is relatively quick. Next, it sorts the list of words. The time this takes is exponentially related to file size! Finally, duplicate words are removed from the list. This is fairly quick. MAKEWORD prints a message on the screen for each pass through your file. This lets you know what the program is doing at any given time. MERGE Most people, and most word processors, do not really like to deal with very long text files. Say, for example, that you have just finished a new book. The double spaced manuscript might run 800 to several thousand pages in length. This size of text file is hard to work with. It is just too cumbersome. The easiest way to deal with this problem is to break the large text file up into numerous separate pieces, probably one file for each chapter. PROINDEX can easily deal with indexing text from several files. You use the starting page number option to tell PROINDEX what page each chapter starts on. All your references will then be correct. 37 So now what do you do with all the separate indexes? They are no fun to merge by hand. If you have 10 chapters you will have 10 separate indexes that all have to be combined together. The MERGE utility, included with PROINDEX, is designed to do this for you automatically. To use MERGE you will need a complete list of all keywords or phrases in your indexes and the actual indexes themselves. MERGE takes each successive keyword or phrase from the keyword file and looks for matches in the index files, in the order you type them. To run MERGE type: merge key.ext newind.ext ind1.ext ind2.ext ... The MERGE program will work with the list of keywords supplied in "key.ext", or any other file name you care to use. It will hunt through each index file, ind1.ext ... indN.ext, for matches to the keywords found in the keywords file and combine these separate indexes into one new index titled "newind.ext", or what ever name you supply. The output of the MERGE program will have the same form that as that generated by PROINDEX. A simple example follows: keywords ind1 ind2 ape ape, 1, 3, 4 camel, 7 camel cow cow, 10 cow dog, 4, 5, 6 dog, 8, 9 dog goose, 7 goat goat gorilla, 4, 5, 6 hen, 9 goose horse, 3, 4 horse, 7, 9 gorilla zebra, 1 zebra, 8 hen horse zebra The MERGE Command merge keywords test ind1 ind2f The New Index (test) ape, 1, 3, 4 camel, 7 cow, 10 dog, 4, 5, 6, 8, 9 goat goose, 7 gorilla, 4, 5, 6 hen, 9 horse, 3, 4, 7, 9 zebra, 1, 8 Some notes about the merging process will probably be of interest now. The merged index will always follow a fixed order. This 38 order is based on the order of index files you supply to MERGE. Thus if index file "ind1" has page references 1 - 100 and index file "ind2" has page references 101 - 200, then file "ind1" should be specified ahead, to the left, of file "ind2". As you can see in the previous example, the separate index files, ind1 & ind2, can use abbreviated keyword lists. Put another way, when you index a separate chapter you can remove those keywords from the keyword list for that chapter that you know will not be referenced. As long as the keyword list for MERGE contains all keywords the final index will be complete. Another point is that you can edit an index by leaving out some keywords from the list supplied to MERGE. If the keyword is not present in MERGE's keyword list it will be ignored in all index files. Finally, MS-DOS limits the number of characters you may type on a command line. Thus if you want to merge 12 separate indexes you can't type a single command to do this. You can, however, run MERGE repeatedly to accomplish the same effect. merge keywords new1 ind1 ind2 ind3 ind4 ind5 merge keywords new2 new1 ind6 ind7 ind8 ind9 merge keywords new3 new2 ind10 ind11 ind12 WARNING: You can not use the same file name for the new index and for a previously made separate index! The MERGE program checks for this possibility and displays an error message if this happens. merge keys index index index2 index3 Results in: "Sorry can not merge the indexes. The new index name 'index' and one of the index files to merge have the same name." PAGE The PAGE utility will take an index list from PROINDEX, break it up into separate pages, and append top and bottom margins. This utility will produce a finished index in the least amount of time. You must supply a disk file that contains the top and bottom margin text. Up to 12 lines of 80 characters each can be used for each margin. Page numbers can automatically be included in the margins if desired. To use PAGE type: page index output margins lines/page page_number where "index" is the input index file. This is typically the file output by PROINDEX. The "output" file is the name of the file you want your finished index to be put in. "Margins" is the name of a disk file that contains the top and bottom margin text, see below for a complete description of this file. The "lines/page" variable is a number telling PAGE how many lines long each page 39 should be including your top and bottom margins. Typically this is 66 lines per page. Note that this number can never be less than the number of lines in your top and bottom margins plus 1. Finally, "page_number" tells PAGE what page number to start the index with, assuming you want a page number in your index. page data.ind newdata.ind margin.txt 66 54 This command tells PAGE to use the index found in "data.ind" and page it based on the margins found in the file "margin.txt". Paging will be based on 66 lines of text per page and the first page of the index will start with page number 54. The margin file describes what the top and bottom margins will look like, along with the position of the page number if any. Each margin may display a page number. The top margin is defined as any text found between the two lines of text that follow: #HEADING #FOOTING The bottom margin is defined as all text after the "#FOOTING" line. If either, or both, of the start margin lines are followed by a space and a single character that character will be replaced in the margin, with the current page number. Thus: #HEADING @ Page -@- Index #FOOTING tells PAGE that the first top margin of the index will start with two blank lines, followed by a line that ends with "Page -54-", followed by a blank line, the word "Index" centered on a line, and two more blank lines. PAGE will only insert one page number in each margin. If no page number is desired, simply use a character in the #HEADING and #FOOTING commands that does not appear in the margins. See the disk file HEADFOOT.MAR for an example margin file. PAGE requires enough free disk space to contain the finished index. Page numbers, and the number of lines per page, can only range up to 32,767. No more than 12 lines of 80 characters can appear in either top or bottom margin. This routine runs very quickly. WARNING: The input and output file names must be different! PAGE checks them to make sure they are not the same and prints an error message if they are. 40 REMOVE This utility removes words from a reference list. It compares the sorted word file produced by MAKEWORD to a list of common words. We supply you with a 400 word list as a starting point. REMOVE simply deletes all words in your reference list that are also found in the common word list. Both files must contain only a single word per line and be sorted in alphabetical order. REMOVE requires free disk space equal in size to the original reference list. To run REMOVE type: remove infile outfile commonwords where "infile" is the sorted list of reference words, "outfile" is the name of the file you want the result stored in, and "commonwords" is the name of a file containing a list of common words you do not want in your reference list. This routine runs very quickly. WARNING: The input and output file names must be different! REMOVE checks them to make sure they are not the same and prints an error message if they are. SORT The SORT utility is an improved version of the MS-DOS SORT command. The MS-DOS SORT command could be used but is much slower! The SORT.EXE file must be located on the same disk as the MAKEWORD utility, since MAKEWORD invokes SORT. SORT can also be used to sort your list of common words or your index into alphabetical order. To use it type: sort filename This will sort the file "filename" into alphabetical order. It requires additional free disk space approximately equal to the size of the original file. This disk space will be released upon completion of the sort. The SORT routine takes an amount of time that grows exponentially based on the size of the file. On an AT running at 6 mhz and using a RAM disk, SORT takes about 32 seconds to sort a 30,000 byte file and 76 seconds to sort a 65,000 byte file. There is no limit on the size of the file to be sorted with the exception of disk space. The SORT utility prints messages to the screen concerning how far it has progressed through the sort. These messages give you an indication of what your current sort status is. Do not attempt to sort indexes that have sub-entries in them. If the sub-entries are indented, SORT will treat them as a new line and they will become intermixed with other entries. The same rule applies to indexes that have multiple lines of page references. 41 STRIP The STRIP utility converts a file, typically from WordStar, into true ASCII. Strip removes: the eighth bit from all data, all extraneous control characters, and soft hyphens in the middle of a line. Soft hyphens at the end of a line are converted to fixed hyphens. STRIP can optionally remove WordStar dot commands and all carriage returns in the middle of a paragraph. To use STRIP type: strip infile outfile [-sd] The "infile" is the name of the file to work on. The "outfile" is the name of the file to save the result in. The "s" option tells STRIP to remove all carriage returns except those at the end of a paragraph or on blank lines. The "d" options tells STRIP to remove all dot commands. STRIP is provided solely as a convenience for WordStar users. WARNING: The input and output file names must be different! STRIP checks them to make sure they are not the same and prints an error message if they are. 42 PROBLEMS Getting Started Most first time problems with PROINDEX come from one of two possible areas. Either your text or keyword file was not properly prepared or you have not set up the index control parameters correctly. If PROINDEX doesn't seem to work at all- it displays junk in the interactive mode and your keywords are also not correctly displayed, then the files from your word processor have not been correctly prepared. Go back and read the section of the manual on preparing text and keyword files. You most likely are using what we call a Group 2 word processor. These word processors do not save text in an ASCII format. The text must be converted to ASCII before it can be used. Demonstration text and keyword files are on your master disk. The next chapter describes how to use and run these demonstration files. Take a look at these files to see how they are constructed. Pay close attention to the organization of the keyword file. Yours should be very similar. PROINDEX Misses References This is a common complaint and is usually an easy problem to fix. The program does what you tell it to. If your keyword file is not exact, or the index control parameters are not set correctly it will appear as if PROINDEX misses some references. Lets start with the keyword file. PROINDEX searches for an exact match to your keyword or phrase. This includes all spaces and punctuation in your keyword file. PROINDEX looks for a match between whatever is on a line in the keyword file and your text. This search includes all characters on a line in the keyword file up to the f character, carriage return / line feed pair. If your keywords have extra spaces, too few spaces, or extra punctuation added to them you will never find a match. See the chapter on preparing your keyword file for more information. If your keyword file is correct there are several other possibilities that can make it appear like PROINDEX is missing references. If you expect to see multiple index references on a single page you must set the index control option telling PROINDEX to function this way. PROINDEX is normally configured to allow only a single reference per page of text. This means that once a match for a keyword is found, PROINDEX will skip the rest of that page! This is the standard way of indexing. If you need to see multiple references per page of text just change the index control parameter. Another reason references may be missed has to do with text case. A keyword with the first letter capitolized will not match non- capitolized text! To get around this problem change the index control parameter that controls text case. This tells PROINDEX to 43 create temporary working copies of your text and keyword files that are all in upper case. This will ensure you don't miss references due to capitolized text. Finally, take note WordStar users, the strip 8th bit for text control parameter can effect your indexing results. If your word processor sets this bit it will effect references in your index. Changing the index control parameter to strip this bit out of your text and keyword files usually corrects the problem. Paging Problems Why don't the page numbers in your index match those in your text file? They really should. If they don't it is because your text file has not been prepared correctly or because an index control parameter has not been set correctly. All text file must be paged by your word processor before being fed to PROINDEX. Print your text file to disk to do this. See the chapter on preparing your text files. Alternately, you must mark the end of each page of text with a special character, and tell PROINDEX what this character is. PROINDEX either looks for this special character or counts the number of lines per page. Both options are selected and controlled on the index control parameters menu. Standard 8 A by 11 inch paper typically has 66 lines per page. PROINDEX defaults to this number. You can change this if you vary the line spacing on your printer or are using a different page size. Error Messages There are a number of different error messages that can be generated while running PROINDEX. These error messages are supposed to be self explanatory. Since screen space is limited these error messages may need a bit of explaining. They are listed here in alphabetical order with a more detailed explanation for those who are still confused. Can't get free disk space on drive X. Does this disk drive exist? PROINDEX always attempts to determine the amount of free disk space on the work disk drive before creating files there. In this case the MS-DOS operating system returned an error code back to PROINDEX which indicates that the specified disk drive, X, does not exist. Are you sure this is a valid disk drive? If it is a valid drive, is a floppy disk in it and is the disk drive door closed? Is the floppy disk formatted? The same questions apply to hard disks. Correct the hardware problem and attempt to index the file again. Or, if you entered an erroneous disk drive name use the E option to edit the disk drive name to a correct on. Then start the indexing process again. Can't make the temporary work file [filename]. 44 Is this disk write protected? PROINDEX attempted to create a temporary work file and was not allowed to do so by the operating system. This usually indicates that the disk to create the file on was write protected. Remove the write protection and start over. Alternately, a disk drive door was opened during the operation of the program. Don't do that! Fatal Error: can not open disk output file for index, name. PROINDEX attempted to create the disk index file and was not allowed to do so by the operating system. This usually indicates a hardware failure in your computer! By this time PROINDEX has sucessfully worked with all disk drives you have specified. It can not create the last disk file needed! Not enough room on default drive, path \pathname for the temporary files. Chose another disk drive or free more space on it. PROINDEX creates temporary working copies of your text files. These files may have the most significant bit stripped from them and/or their case changed to upper. These work files need disk space equal in length to the size of your original text and keyword files. There is not enough room on the work drive you specified to hold these files. Chose another work drive or free up more disk space for PROINDEX to work with. Sorry, the old index file doesn't have a correct continue marker. You can not continue the index filename. (You probably finished this index. Rather than write over it we will return back to the main menu so you can check this out.) Hit any key return to the main menu. You have attempted to continue an old indexing session with a completed index or with an altered one. Go to the eXecute an MS- DOS command menu and examine the old index. You will have to determine what to do next based on the contents of this index file. System command error When using the eXecute an MS-DOS command menu you attempted to run a non-existant command. Use the DIR command or you MS-DOS manual to find the correct command. (Or correct your spelling.) System command failed When using the eXecute an MS-DOS command menu the command you selected and ran failed. This command returned an error code to PROINDEX instead of returning normally. PROINDEX is simply reporting this fact back to you. You must determine what to do next. The keyword file [filename] can't be found. 45 Is this the right file name and disk drive? PROINDEX can not locate the keyword file you specified at the top of this menu page. Make sure the file name, path, and disk drive are correct. You can edit the keyword file name with the E option on the main menu. If the file name, path, and drive appear correct do a DIR command on that disk, using the eXecute an MS- DOS command menu, to see if the file is really there. The text file [filename] can't be found. Is this the right file name and disk drive? PROINDEX can not locate the text file you specified at the top of this menu page. Make sure the file name, path, and disk drive are correct. You can edit the text file name with the E option on the main menu. If the file name, path, and drive appear correct do a DIR command on that disk, using the eXecute an MS-DOS command menu, to see if the file is really there. There is not enough room on path \pathname for the temporary files. Chose another disk drive or free more disk space. PROINDEX has determined that there is not enough room on the work disk drive and path for all the files it must create. You must have at least as much free disk space here as your text and keyword files occupy. Either change the work drive to a disk that has more room, or make more room on this disk by removing unnecessary files. See below. There is not enough room on the default drive for the temp- orary files. PROINDEX has determined that there is not enough room on the work disk drive for all the files it must create. You must have at least as much free disk space here as your text and keyword files occupy. Either change the work drive to a disk that has more room, or make more room on this disk by removing unnecessary files. See above. This is quite confusing. The program version number and con- figuration file version number don't match. Please erase this configuration file and copy the correct configuration file from your PROINDEX master disk to your work disk! The configuration file that PROINDEX uses to remember the way you want to index your files has a version number saved in it. This ensures that PROINDEX will never use an old configuration file. The configuration file may vary from version to version of PROINDEX. PROINDEX has discovered that its version number does not match that of the configuration file. Simply delete the configuration file. PROINDEX will create a new, correctly identified, configuration file when you run it. Remember to change this configuration file to reflect your indexing style. 46 DEMONSTRATION FILES A demonstration text and keyword file can be found on the PROINDEX master disk. These two files are called PROTEST.TXT and PROTEST.KEY. The text file or document is in the disk file with the '.TXT' extension, while the file with the '.KEY' extension is the keyword or phrase list. This demonstration assumes that the configuration file has not been changed form its default values. These files were created with the WordStar text editor. The text file was made using WordStar in the document mode while the keyword file was created using the non-document mode. The 8th bit, MSB, has been stripped from these files to make them less confusing to people who are unfamiliar with WordStar. To run PROINDEX using these test files first copy both files from this master disk to your PROINDEX work disk or hard disk path. Log on to that disk or path and type: PROINDEX You will see the screen shown in figure 1 followed by the screen from figure 2. You will then be asked for the name of the text file to index. Type: PROTEST.TXT. When asked for the name of the keyword list type: PROTEST.KEY. When asked for the path for temporary work files either give a vaild work drive or hit return to use the current disk for work space. You have now told PROINDEX the name of the keyword and text files and where to put the work files and you will advance to the menu shown by figure 3. Now select the I option, Index the text, from the main menu and hit a return. The program will indicate that it is creating work files and then change to the indexing menu screen, as shown in figure 5. You are now running PROINDEX in the interactive mode. You will be prompted each time PROINDEX finds a match and asked whether to Mark, Skip, Bypass, All, or Quit. This is how PROINDEX runs in the interactive mode. Each keyword or phrase is highlighted in context on your screen. This helps you decide whether to include the reference in your index. You may quit at any time and restart at the point you left off later on. Your index will be saved on the disk under the name PROTEST.IND. To see what your index is like you can quit indexing at any time and return to the main menu, figure 3. From that point select the "X" menu option to eXecute an MS-DOS command. You will move to the MS-DOS execution menu, as shown in figure 6. Enter the MS-DOS command "type protest.ind". This will type out the demonstration index as it presently stands. You can then return back to the main menu, select the "C" option to continue your index, and resume indexing where you left off. If you feel like experimenting you can change some of the variables on the index control parameters menu, figure 4, and see how this effects the resulting index. Or try sending the index to 47 the printer instead of a disk file. 48 Index All command, 2, 12, 29 ASCII, 1, 7 - 9, 11, 13, 16, 25 - 27, 31, 34 - 36, 42, 43 Backslash key, 5, 21 Backspace key, 20 BIOS calls, 2, 3, 5 Bold display, 3, 5, 29 Bypass command, 2, 12, 29 Chapter, 3 - 6, 14, 26, 37 - 39, 43, 44 COMMON.TXT, 36 Configuration, 16 - 19, 24, 46, 47 Control-A, 20 Control-W, 20 Data fields, 20 entry, 3, 5, 12, 19, 20, 29, 36 Default disk, 18 Delete key, 3, 5, 19, 20, 41, 46 End of page, 17, 24 Fancy Font, 9 Floppy disk, 19, 34, 35, 44 Hard disk, 2, 19, 34, 35, 37, 44, 47 HEADFOOT.MAR, 40 IBM PC, 2, 3, 20, 34 Interactive mode, 1, 2, 9, 11, 12, 14, 16, 17, 28 - 30, 43, 47 Interrupt 10H, 2 Line numbers, 3, 4, 13, 26, 27 MAKEWORD, 3, 36, 37, 41 Margins, 4, 39, 40 Mark command, 2, 12, 29 MERGE, 4, 26, 37 - 39 MS-DOS, 2, 7, 11, 17, 18, 22, 23, 30, 34, 35, 39, 41, 44 - 47 Multiple references per page, 6, 24, 29, 43 Opening screen, 19 Options, 5, 12, 18, 21, 22, 24, 29, 31, 32, 36, 42, 44 Output, 1, 4, 9, 15, 24, 28, 31, 32, 36 - 42, 45 Printer2, 16, 17, 28, 31, 32, 44, 48 PROINDEX.CNF, 16, 17, 19 PROINDEX.EXE, 18 Quit, 21 - 23, 31, 47 REMOVE, 3, 7 - 9, 20, 26, 28, 31, 32, 36, 37, 39, 41, 42, 45 Requirements, 9 Screens, 2, 4, 5, 7, 9, 16, 19 - 21, 23, 24, 29, 37, 41, 44, 47 Secondary references, 5, 11, 14, 15 Set Command, 10, 17, 26, 43, 44 Skip command, 12, 13, 30 SORT, 4, 36, 37, 41 Speed, 29, 34, 35 Starting page number, 16, 17, 24, 37 STRIP, 17, 27, 36, 42, 44 Stripping bits, 34 Subdirectories, 18, 19 Temporary file, 1, 45 49 Index Text case, 3, 8 - 10, 13, 16, 17, 24, 26, 27, 34, 36, 43, 44, 45 Utility programs, 1, 4, 26, 36 - 39, 41, 42 WordStar, 1, 7, 8, 27, 36, 42, 44, 47 Work drive, 20, 34, 35, 45 - 47 files, 1, 21, 28, 29, 45, 47 50 DISTRIBUTION NOTICE This program is protected by United States Copyright law and by International Treaty provisions. You may distribute, or share this version of this program with anyone provided the following conditions are met: This version of PROINDEX is being distributed at no charge to all who are interested in it. This version does not include the utility programs mentioned in the documentation. To those of you unfamiliar with this approach, it works like this. PROINDEX is distributed to various BBS systems and user groups. You may copy the program and use it for no charge. Gary Elfring does not handle the actual freeware distribution of this program. Try the BBS version out. If the product does not meet your needs, you pay nothing. If the product does meet your needs, and you can use it, then send us the price of $65.00 U.S. We will then supply you with an updated version of PROINDEX, all the utility programs, improved typeset documentation for that version, telephone support, and one free update. We accept check, and money orders. This offer may be withdrawn at an time without notice. Send check or money order to: Gary Elfring 4N899 West Mary Drive St. Charles, Illinois 60174 312-377-3520 Even if you do not contribute, you are encouraged to copy and distribute the BBS version of the product freely subject to the following restrictions: 1. Do not distribute altered copies. If you have suggestions or fixes you would like to see, send us a written note. We welcome all responses. 2. This product must be distributed complete including the files: PROINDEX.EXE, PROINDEX.DOC, PROTEST.TXT, and PROTEST.KEY. 3. No fee is to be charged for copying or distributing the pro- gram without an express WRITTEN agreement with Gary Elfring The ONLY exception to this rule is: a small charge, ($5 to $10), may be leveled if this program is included on a set of public domain disks. This charge must be reasonable and repre- sent only copying and disk fees. 4. Commercial sale of this product in any manner is prohibited without Gary Elfring's written permission. User Groups/Clubs: Gary Elfring's freeware products may be dis- tributed by Clubs and User Groups subject to the same restric- 51 tions. We would ask only that contributions to our continuing efforts be encouraged by those performing such distribution. Computer Bulletin Boards: Gary Elfring's freeware products may be distributed by BBS's subject to the same restrictions and contribution encouragement. 52