On the Network and DOS based versions of Unibase by DMAC, a long drun may be gracefully terminated by pressing the {SCROLL LOCK} key. If you forget to turn off the {SCROLL LOCK} key, the next time you execute a drun, you will get the message 'Terminating at Operator Request'.

Return to Questions
 

Yes, if output is done by the backup server. The environment variable ETDUAL2 has been added. When ETDUAL2 points to the location of the files on the other server, then Unibase by DMAC will simultaneously mark files on the primary and backup as having been output.  On a Novell network, it usually requires attaching to the other server and mapping a drive letter to the other system. 

Example: map p:=server1_name\sys:\unibase\files
Then set ETDUAL2=p: 

On a UNIX system, doing this will usually require setting ETDUAL2 to the full path of the files on the other server including the server name. 

Set ETDUAL2=server_name:user/unibase/files

Return to Questions
 

The environment variable TABSZ may be used to set the width of the tab stops in ez_edit.

Return to Questions
 

Yes, ez_edit will now handle very large files and a counter has been added to show that it is reading in a large file.

Return to Questions
 

Yes, the high and low ranges in the check box edit portion of the record format generator will now accept negative numbers provided the field type is numeric.

Return to Questions
 

The environment variable TATT (Tag Screen ATTribute) is used to set the non-gui default tag screen attribute used in the "paint the screen" portion of the record format generator. TATT=A, B, C, D, or E respectively gives normal characters, highlighted characters, underlined characters, blinking characters, or reverse characters for the tag display.

Return to Questions
 

The environment variable FATT (Field Screen ATTribute) is used to set the default field screen attribute used in the "paint the screen" portion of the record format generator. FATT=A, B, C, D, or E respectively gives a normal underline, a highlighted underline, an "underlined" underline (which looks like a solid line), a blinking underline, or a reverse underline for the field display.

Return to Questions
 

Yes. Environment variable UBDATE alters the order of the month, day, and year on the login menu, when displaying file information, when displaying status information and when displaying the system date. UBDATE set to 0 gives mmddyy. UBDATE set to 1 gives ddmmyy. UBDATE set to 2 gives yymmdd.

Return to Questions
 

Yes. The path description for devices in the device table may now include the control function <OPID>, any global variable ($var01-$var99), or any environment variable (%variable%). If global variables are going to be used, the environment variable UBPARENT=Y should be set to ensure that global variables retain their values. CAUTION. Any programs that use global variables should be SURE to initialize them properly as global variables retain values unless erased by a parent program.

Return to Questions
 

Yes. Color choices from the Operator Log-In can now be aborted by keying 1 space 1 space 1 space 1 space 1 space {FLD REL}.

Return to Questions
 

All hex characters in the range of 00 to fd may be output using the control functions <hex nn> when nn is the hex character you wish to output. The hex characters in the range of fc to ff may be output by setting the environment variable ALLHEX=Y in the Unibase.ini file in the Unibase\bin directory. However, fc through ff may not be used with spooled devices.

Return to Questions
 

Unibase by DMAC can output an unlimited number of files in the DOS/Novell version.

Return to Questions
 

Yes, starting with Unibase version 7.4, it is compatible with all Windows versions up to Windows 98, Release 2.

Return to Questions
 

Yes, by setting the environment variable UBSESN. Each session of Unibase will require a DIFFERENT (UBSESN) session number. Bat files should be created with different session numbers (up to 50). The format is UBSESN=n where n is a number from 1 to 50. The config.sys file needs to be modified to increase the number of files and buffers allowed so that it will accommodate multiple sessions of Unibase.

Return to Questions
 

If you are going to use sessions, then operators in data entry should be instructed to key {RCD} {S} before switching to another session. This command, which is normally used to suspend the clock, also tells Unibase to close any files that data entry might have open/locked. When the operator returns to the session, simply pressing any key will resume operations for the session.

Return to Questions
 

Beginning with version 7.2, the keyboard program defaults to the newer extended keyboards and expects KBEXT=Y to be set in the Unibase.ini file in the Unibase\bin directory. To use an old keyboard, you must remove KBEXT=Y from the setup.bat file in the Unibase\bin directory and map the arrow keys on the numeric keypad accordingly. Even if you have a separate arrow keypad, it will be inoperable unless you have an extended keyboard.

Return to Questions
 

Yes, if UBRSET=Y is in the Unibase.ini file in the Unibase\bin directory, then the {FIELD REL} key no longer resets an error message. Only the designated {RESET} key will do so.

Return to Questions
 

Yes. If KEYBUK=Y is set in the Unibase.ini file in the Unibase\bin directory, then Unibase accesses SHOWKEYU instead of SHOWKEY and CHNGKEYU instead of CHANGKEY. These display the keyboard layout from the United Kingdom. Also, there is a file KEYBOARD.UK in the misc directory, which when renamed to KEYBOARD, displays the characters on a UK keyboard for the typewriter overlay.

Return to Questions
 

Yes, Unibase will support a Computer Logics Nine Track tape drive controller. The environment variable UBTPCL=Y must be set in the Unibase.ini file in the Unibase\bin directory to use this controller and TDRIVER.EXE, OUTRIGHT.CFG, LCA.BMD, and LCA.SGL files from the Computer Logics programs must be in the Unibase bin directory. Unibase by DMAC will load TDRIVER as needed. (Computer Logic now calls its program API.EXE and this program must be renamed to TDRIVER.EXE before Unibase can use it. This works in DOS only.

Return to Questions
 

Yes, packed decimal output to tape is available. You MUST use the EBCDIC codeset when you output and then you MUST use the ASCII mode when making the tape. Also, for every packed decimal field, you must allow two extra bytes in the record length in the standard job. The two extra bytes are used by Unibase and are not output.

Return to Questions
 

Yes. If NOCHBKFL=Y is set in the Unibase.ini file in the Unibase\bin directory, then a keyer cannot character back into another field. Once character back reaches the beginning of a field, then it no longer has an effect. The keyer MUST use field back to get to the prior field.

Return to Questions
 

Yes. When FLDCORBG=Y and RCDCORBG=Y are set in the Unibase.ini file in the Unibase\bin directory, doing a {FLD} {CORR} / {RCD} {CORR} will always go to the beginning of the field/record.

Return to Questions
 

Yes, this is done by pressing {RCD}{G} or by mapping a key to "BE " which stands for bypass error. {RCD}{F} also works as a way to flag an error.

Return to Questions
 

Yes. See Manual 4, Appendix D, Page 23.

Return to Questions
 

Yes. With ETSWAP you can specify any pathname (not over 21 characters) in the Unibase.ini file in the Unibase\bin directory.

Return to Questions
 

Yes, as long as UBKEYR1=Y is set in the Unibase.ini file in the Unibase\bin directory. This allows the top row of keys to function independently of the data type (A,N,L,U,M,B,T) when the operator is in TYP (non 029) mode.

Return to Questions

Yes, by setting the environment variable UBNMULG=Y in the Unibase.ini file in the Unibase\bin directory.

Return to Questions
 

Yes, the "Purge Deleted Records" function under Utilities will handle files specified with a wild card.

Return to Questions
 

Yes. The verbs "setupper", "setnormal", "setlower" and "setuplower" (sentence case), control the case of records output from an output program.

Return to Questions
 

Yes, global variables may be used in the define file statements.

Return to Questions
 

Yes, global variables may be used as datafile and index names in open statements.

Return to Questions
 

Yes. The "equate" verb tells Unibase to declare a variable ("variable") so that every time it appears in this program, it is read as the source. It can also be used for constants in the program. For instance, it is useful for equating names to field numbers or foreign names to Unibase verbs.

Return to Questions
 

Yes, if you set environment variable UBOVNR=Y in the Unibase.ini file in the Unibase\bin directory, it will NOT release the field in data entry when the oversign key is used.

Return to Questions
 

The environment variable DMACI (which must be set in the Unibase.ini file in the Unibase\bin directory) for image entry has a sub-variable V:, which refers to the VGA mode to be used. V:LOW is mode: 12h, displays 640 pixels x 480 pixels x 16 colors for an image. V:HIGH is mode: 102h, displays 800 x 600 pixels x 16 colors. V:HIGH1 is mode 104h, displays 1024 x 768 pixels x 16 colors. V:SUPER is mode 106h, displays 1280 x 1024 pixels x 16 colors. Remember that as more of the image is displayed, the smaller the text/picture becomes. If the monitor cannot display the specified resolution, Unibase by DMAC will drop back to a display that the monitor can handle.

Return to Questions
 

Yes, by using the code {0}|af, where {0} refers to the image in the work file record and |af indicates to convert it to an ASCII file that represents a binary file.

Return to Questions
 

Yes, by using the control function <ilist 0,cntr>.

Return to Questions
 

Yes, by using the control function <isource 0>.

Return to Questions
 

By using control function <ipathidx 0>. It gives the number of the path definition in the imgpath.idx file used with the current record of the current work file.

Return to Questions
 

By using control function <ipath 0>. It gives the current record's image path.

Return to Questions
 

By using control function <iname 0>. It gives the name of the image being used with the current record of the current work file, including the first directory.

Return to Questions
 

A.

You need to use the control function <ientry 0>. It tells if images are used in the current workfile. If <ientry 0> is 0, then the workfile does NOT use images. If it is 1, then the workfile USES images.

Return to Questions
 

Yes, by using the control function <icount 0>. It gives the number of images in the current work file.

Return to Questions

By using the control function <vtime 0>, it displays the elapsed verify time as hhmmss.

Return to Questions
 

The control function <voper 0> contains the name of the most recent operator to access the current workfile in verify mode.

Return to Questions
 

The control function <recfmts 0> returns a 2 digit number of the record formats in the Standard Job used to create the file. Also, you can open the Standard Job and view the Record Format list.

Return to Questions
 

Yes, by using the control function <outdev 0;1>. If the current workfile has been output, this is the name of the output device where n=1 is the most recent outdev. Also found under file function in display file info.

Return to Questions
 

Yes, by using the control function <lmode 0>. It is a one letter code for last mode of access. The codes are 0-entry, resume or update; 1-verify; 2-examine; 3-correct. Another way is to use a file edit, using the control function <mode>.

Return to Questions
 

Yes, by using the control function <etime 0>. It displays the elapsed entry time as hhmmss.

Return to Questions
 

Yes, by using the control function <eoper 0>. It contains the name of the most recent operator to access the current work file in entry mode.

Return to Questions
 

Yes, the control function <cdate 0> gives the creation date for the current workfile.

Return to Questions
 

The control function <accum 0;xx> gives the current value of accumulator xx for the current workfile.

Return to Questions
 

Yes, if the environment variable UBTYPW=Y is set in the Unibase.ini file in the Unibase\bin directory, and only if the operator is also in typewriter mode.

Return to Questions
 

Yes, but it must be done from the DOS prompt. The syntax is: ubupgvar nn "data"nn is the global variable number and the data to be moved into it is in DOUBLE quotes. This program allows you to initialize or change global variables by system calls to a .bat file.

Return to Questions
 

Yes, using the F: parameter in the environment variable DMACI (which is used for imaging data entry only and must be set in the Unibase.ini file in the Unibase\bin directory). F:SMALL will use 8x8 pixel characters in the Unibase screen in image entry. F:MED will use 8x14 pixel characters. F:BIG will use 8x16 pixel characters. F:BIG1 will use 10x18 pixel characters. F:BIG2 will use 12x30 pixel characters. This parameter does NOT affect the image display portion of image entry. In order to avoid wrapping the screen display around on top of itself, parameter F:BIG1 is meant to be used with V:HIGH1 or SUPER and parameter F:BIG2 is meant to be used with V:SUPER.

Return to Questions
 

Set the environment variables UBMXX and UBMXY in the Unibase.ini file in the Unibase\bin directory. Unibase Image Entry allocates memory to handle approximately an 8 1/2" x 11" page image. This is an image width and length of about 2550 by 3300. If your image width and/or length exceed these defaults, use UBMXX and UBMXY to set larger sizes so Unibase knows to allocate more memory for the image.

Return to Questions
 

No, if environment variable UBDZF=Y is set in the Unibase.ini file in the Unibase\bin directory, zero length files will not stop with an error message if they are being processed, they will simply be bypassed. They will also be deleted like regular files.

Return to Questions
 

Yes, by setting environment variable UBDRSZ in the Unibase.ini file in the Unibase\bin directory. If the environment variable is set to a number such as UBDRSZ=4096, then that number is used as the default size. If that number is NOT set, then the default is 80 characters.

Return to Questions
 

Yes, set the environment variable UBPOS1=Y in the Unibase.ini file in the Unibase\bin directory. It will refresh the screen when a position statement is executed in both image entry, as well as data entry as long as there is not a program that uses 'show' statements to display information. If there is a 'show' statement, this data will not be erased whenever a position statement is executed.

Return to Questions
 

Yes, by using the verbs "search" and "searchf". The verb "search" will report occurrences of the string across field boundaries. The verb "searchf" restricts the search to occurrences within a field.

Return to Questions
 

Yes, with the <date4> control function. It returns an eight digit result in the form mmddyyyy. The environment variable UBDATE also affects <date4>.

Return to Questions
 

Yes. Put UBFMTL=Y in the Unibase.ini file in the Unibase\bin directory. You must also assign the format name in every field edit program.

Return to Questions
 

Put UBNOFF=Y in the Unibase.ini file in the Unibase\bin directory. When it is set to Y, output sent to LANPRx devices will NOT perform a top of form at the end of the output.

Return to Questions
 

Yes, by placing the environment variable UBTAB in the Unibase.ini in the Unibase\bin directory. When it is set to Y, tabbing to a field NOT on the screen in data entry will position the "tabbed to" field on the last line of the screen instead of the first line of the screen.

Return to Questions
 

Yes. You need to change your Unibase.ini in your Unibase\bin directory to reflect UBDD=Y. When it is set to that, a change made to a duplicating field while in verify will be carried forward through all succeeding records, even those with a different format number, where the field has been marked as record-to-record dupe and the contents of the succeeding fields were the same as the original field that was changed.

Return to Questions
 

Yes, by setting the environment variable UBIDX in the Unibase.ini file in the Unibase\bin directory. It should point to a directory on the C drive such as "UBIDX=C:\TEMP". Indices will then be stored in that location on the C drive when they are built from the Unibase menu. This technique speeds up index builds.

Return to Questions
 

SCO-UNIX is currently in quality testing. However, Unibase does port to any Unix release V.4.2 or later. We are currently porting to several manufacturers version of Unix and Linux, including HP and Tandem. Please call for current list.

Return to Questions
 

In the Unibase\bin directory there is a file "comname". Deleting this file will allow you to reenter. "User count exceeded" is telling you the full number of licensed users are already logged in and an additional user is trying to log in. Deleting the "comname" file will fix this problem if someone has aborted out of Unibase without logging off.

Return to Questions
 

This is a hardware issue and not a Unibase issue. Contact your IT department for assistance.

Return to Questions
 

Take the set UBCRLF=Y out of your Unibase.ini file in the Unibase\bin directory.

Return to Questions
 

Need to make the output device type "comm". This will append the records, rather than replace each.

Return to Questions
 

Only if you have a backup. Once initialization has been done, the operator stats file is empty. You can also use the Salvage feature of Novell. Contact yoru IT department for assistance on the Novell Salvage feature.

Return to Questions
 

Maximum fields per record format allowed is 999. You can have up to 99 record formats per standard job.

Return to Questions
 

The maximum allowable ASCII string in Unibase is 275 characters at this time. On a Unix system the maximum allowable ASCII string is 128.

Return to Questions
 

The file which that pointer refers to has been deleted. You must rebuild the index.

Return to Questions
 

Your fields are defaulting to spaces, instead of zeros. You need to change the default in check box edits under record formats to reflect zeros instead.

Return to Questions
 

This option is not available in Unibase at this time.

Return to Questions
 

This is a hardware issue, not a Unibase error. Contact your IT department for further assistance.

Return to Questions
 

If you can access your security features, verify that you have them set the way you want them. If you cannot access the security features, go into Unibase\misc and delete etpasswd file. You will then have to recreate your security features.

Return to Questions
 

Contact your IT department. This is a problem with your internal network and not a Unibase issue.

Return to Questions
 

This is an issue with Novell and not a Unibase issue. Contact your IT department.

Return to Questions
 

You need to delete the index and rebuild.

Return to Questions
 

If you reparse the program, it should work fine.

Return to Questions
 

Check to see how you have your copy set up. It needs to be copy *.*

Return to Questions
 

Yes, as long as no one else has gotten into the file to begin entry.

Return to Questions
 

If the name in the bat file is not in oplog, it is unknown.

Return to Questions
 

You need to rebuild your indexes so that the changes made between the two versions will be reflected in your indexes.

Return to Questions
 

Please contact our sales department for pricing on converting licenses.

Return to Questions
 

Try rebooting your workstation. If that does not work, all operators will need to log off their computers so that the server can be rebooted.

Return to Questions
 

You need to edit the Unibase.ini file which is found in the Unibase\bin directory. At the end of the unibase.ini file, type:

set UBCRLF=Y

or make the printer device type "printer".

Return to Questions
 

Yes, but the copy of the Unibase directory must be activated. Call Technical Support for an activation code.

Return to Questions
 

Global variables must be initialized by executing the file edit "SGVAR" against the file "OPSTWF".

Return to Questions
 

Yes, successfully.

Return to Questions
 

This is normal. You always lose the highlight box on a zoom.

Return to Questions
 

Must have put "Y" in enable edit in check box edits of records format. Check to see if you have listed it in the standard job you are using. It must be in the standard job to work.

Return to Questions
 

The file should have no extension name to it. Rename the file to 'test' and it should work fine.

Return to Questions
 

Your standard job needs to be recreated. Try it again after recreation.

Return to Questions
 

The standard job has output restrictions imposed. If, when you created the standard job, you put a "Y" where it said "Be Verified:", then the output will only recognize those files that have been verified.

Return to Questions
 

Go into miscellaneous directory, delete ".lck" file for the batch which was aborted.

Return to Questions
 

Unibase will accept the following characters as part of a file name:

~ ! @ # $ % ^ & * ( ) - _ { }

The forward slash "/" will be interpreted as a divisor for a subdirectory and a subdirectory will be created using whatever characters precede the "/".

Return to Questions
 

This has to do with memory allocation. Close any other windows you have open, and if possible, allocate more memory.

Return to Questions
 

Run Convert Files under Utilities, Conversion to Unibase 7.4i, Convert from Unibase 7.x.

Return to Questions
 

Make sure no one is using the format first, then use selective delete and follow the on screen prompts.

Return to Questions
 

Yes, it runs fine. We use CL32 Intranetware.

Return to Questions
 

The network drive(s) need(s) to be mapped before network installation can take place.

Return to Questions
 

You need to copy your opstop.scn to Unibase\output. Your modifications will show once you have done this.

Return to Questions
 

The index pointing to the file you are trying to pull information from needs to be rebuilt.

Return to Questions
 

Only one user can key enter in a file at a time.

Return to Questions
 

The file that the index is trying to point to has become corrupted. You need to rebuild the file, then create the index.

Return to Questions
 

In the Utilities Menu there is an option called Recover Corrupted Files. Go into this and follow the on-screen prompts.

Return to Questions
 

Check your key. If there are asterisks (or other symbols other than alpha or numeric), you need to remove them.

Return to Questions
 

To output files that aren't verified you need to take off "Be Verified" restriction from the 2^nd page of Standard Job.

Return to Questions
 

Yes, you can. The File Input/Output Function [E] Read Unformatted File will do short reads. That is, a new record will be started whenever a delimiter is encountered. The delimiter must be entered as a decimal number, usually "010" (code for line feed) for dos-based systems.

Return to Questions
 

Yes. The syntax is: define array arrayname[numrows][numcols]

where numrows is the number of rows in the array and numcols is the number of columns in the array. Both numrows and numcols are expressed as integers.

Further, the elements of the array each have the default specification as for single Unibase by DMAC variables. Should one need a different size, then ":nnn:" may be appended to the numcols item where nnn is the size in bytes of the variable.

In particular, an array defined as follows: define array buffnam[1][1:47]

would replace the definition of a 47 character buffer in the Tartan Data Entry language.

Return to Questions
 

Yes, it does. It is used to put information into arrays using the following syntax:

build (arrayname[rownum][colnum]) using literal.
variable.
control function.
expression.

where rownum and colnum are numeric literals. Expression is used to mean a statement consisting of a variety of items which will be stored in the array item one right after another.

Return to Questions

For field edits, an asterisk in parentheses (*) denotes the current field.

Return to Questions
 

The control function <record> (not <records>, which is something different) returns the current record number.

Return to Questions
 

Yes, by using the verb "createidx" to create an index with no field. The syntax follows:

createidx idxnam with keylen idxlen else . .

Example: createidx zipix with keylen 10 else . .

Records (and their associated data file names) will then be added to this index using the insert, append, and include verbs.

Return to Questions

Yes, by using the verb "when [not] auto". It is used to test the current condition of the auto on/off switch.

Return to Questions
 

Because the environment variable UBRW=Y is in the unibase.ini file in the bin directory. When set to Y, the last record being entered (not verified) is removed from the file if the operator terminates/interrupts the batch before entering a keystroke. This happens even if a field edit has been run on the record.

Return to Questions
 

Yes. The verb "return". It is similar to the "exit' verb for a subroutine, but may be used anywhere within the subroutine to return to the calling statement.

Return to Questions
 

Yes, Unibase by DMAC does allow read/write of ascii text files with the AID language. These text files must be in the "text" sub-directory under the ETROOT. They may NOT have an extension in the name. Sub-directories may exist under the "text" sub-directory.

Return to Questions
 

Yes, by using the verb "define text". It is used to assign a name to a textfile. The textfile is described with the actual name of the file in the text sub-directory and parameter that give record length, key length, key offset, and whether the record is fixed or variable length. Variable length records must be terminated with newline (carriage return/line feed for dos based systems). Syntax:

define text txtnam with dirnam/filnam (rl;kl;ko;rt).

where txtnam is the name as it is used in AID language code, dirnam/filnam is the path and text file name under the text sub-directory, and the items in parenthesis are numbers for record length, key length, key offset, and record type. Key offset means how many characters from the beginning of the record does the key start. That means if the key offset is zero, the key starts at character 1; if the key offset is 10, the key starts at character 11. A record type of "0" means the records are fixed length. A record type of "1" means the records are variable length.

The text file does not have to have a key. However, if a key is present, the text file records must be in ascending order based on the key.

Return to Questions
 

Use the verb "put". It is used to replace the current data record in a text file with the contents of a specified array.

Put txtnam from arraynam.

Text files with keys must not have the key value changed.

Return to Questions
 

Yes. The verb "concatenate" is used to concatenate one variable holding a string to the end of another variable holding a string. The following code demonstrates the syntax and results.

Declare var1 var2.
Move "Hello " to var2.
Move "world." to var1.
Concatenate var1 to var2.
Show <loc2> var2.

The screen will display "Hello world." at line 2.

There is a special case of concatenate where the second item is the variable name associated with a text file. In the define text statement, the text file associated with the variable name has a directory name only, followed by a slash. For example:

define text outdata with cust/(100;0;0;0).

Then, the following code will expand the text file description to include a file name, not just a sub-directory name.

move "payroll" to filnam.
Concatenate filename to outdata|nm.

Then, when the text file "outdata" is opened, a complete path to a file is correctly specified.

Return to Questions
 

Yes, by setting the environment variable UBDUPT=Y in the unibase.ini file in the bin directory. When set to Y, the operator will be positioned at the first position of a duplicating field with the auto function turned off.

Return to Questions
 

Yes, environment variable UBAUTOT=Y in the unibase.ini file in the bin directory. When set to Y, it affects operator forward and back operations in a variety of ways.

If AUTO is off, a record forward will position the operator at the first keying position of the next record in the batch. Auto will remain off so that the operator may be placed into a field that has AUTO functions defined. (Without this environment variable, this operation places the operator at the last keying position of the current record.)

If AUTO is on, a record back, field back, or cursor back will position the operator accordingly, even if fields with AUTO functions precede the operator's current position. AUTO will be turned off if AUTO fields are present.

In particular, a record back that has intervening AUTO fields will turn auto off and place the operator at the first keying position of the record. If the operator is at the first keying position, Unibase will move to the first keying position of the prior record. If no AUTO fields exists, the AUTO function will remain on.

A field back into an AUTO field will turn the AUTO function off and position the operator at the first character of the prior field.

A cursor back into an AUTO field will turn the AUTO function off and position the operator at the last character of the prior field.

Return to Questions
 

Go into Record Format. From there to Edit Check Boxes, then to Range Check. Select your options.

Return to Questions
 

Contact your IT personnel. This is a hardware problem.

Return to Questions
 

You need to go to Utilities Menu and from there to Recover Corrupted Files. Follow the on-screen prompts.

Return to Questions
 

You need to rewrite your record format.

Return to Questions
 

Yes. You can go into Tape Utilities from the Utilities Menu and follow the on-screen prompts.

Return to Questions
 

This is a hardware problem. Contact your IT supervisor.

Return to Questions
 

Check your device table in Spooler Maintenance for correct spelling and to verify device exists. If device is there and spelled correctly in the standard job, check to see if output program exists and is spelled correctly.

Return to Questions
 

In order to change the attributes back you have to change each field in Edit Check Boxes.

Return to Questions
 

In the Miscellaneous Directory there is a file - "ETPASSWD" - you can edit this to see all passwords.

Return to Questions
 

Rename your existing to "opst.aid", then rename the "opst.fmr" file to "opst.aid" and it will print. Don't forget to change the names back.

Return to Questions
 

If you are using a WinNT server, please read the following:

Log into Unibase by DMAC.

Exit out the back door: [A] Advanced Processing, [E] System Functions, [F] OS Access, {FLDREL}, {REL}.

 Type: set | more and press enter. This lists your environment variables.

 There should be an environment variable: COMPUTERNAME=computername.

NOTE: If the computername is Unibase, you have failed to generate a unique computername. If two workstations have the same computername, one of them must change.

Return to Questions
 

Go back to Keyboard Mapping. Where you want your symbol to be, type "/nnn" where decimal code for that symbol, found in manual (Vol. 4, Appendix E).

Return to Questions
 

This happens when files from an earlier version have not been converted to the upgraded version.

Return to Questions
 

Reboot into Windows and run Unibase from a DOS Shell. It should work fine.

Return to Questions
 

Yes, there is a "hotkey" that will work with both data entry and image entry. If the operator presses the {FLD} key, releases it, then presses the letter "E", both data entry and image entry will execute the field edit program. The "hotkey"can be programmed to one key by using the code "FL E" when mapping the keyboard.

Return to Questions
 

Yes, by using the verbs "enable editkey" and "when editkey". The programmer must set usage of the editkey with "enable editkey". Then "when editkey" may be used to check if the operator has pressed the "hotkey" combination to access the field edit.

Return to Questions
 

Yes, you can "double key" a field in data entry as long as you have selected the appropriate entry in the record format check box edits. If "U" is chosen under "Rcd/Rcd Auto:", then once the operator has keyed that field in data entry, the contents of the field are blanked out, the cursor goes to the beginning of the field, and the operator re-enters the field. The data does not appear while entering it the second time until the operator exits the field.

If the field is keyed identically the second time, control goes to the next field to be keyed. If the second keying does not match the first keying, the message "Double Key Verification Error - Enter Original Field Again" appears at the top of the screen and the operator tries again. The field must be keyed identically twice in a row before it is accepted.

Return to Questions

Yes, Unibase by DMAC supports both pre-field and post-field edits in data entry and image entry. In addition, a field may have both a pre-field edit and a post-field edit. If a field has the pre-field edit enabled, then Unibase by DMAC executes the field edit before the operator is allowed to key in the field.

To enable a pre-field edit, use "S" (for Start of field) on the Fld Edit checkbox in record format creation/change. To enable both a pre-field and post-field edit, use "B" on the Fld Edit checkbox in record format creation/change.

The verb "when prefield" is for use in field edits only. It is used to execute instructions based on whether the current field is the specified pre-field. "when prefield" has features of "when field".

Return to Questions
 

Yes. Configuration names in the unibase.ini file in the bin directory may be referenced by standard jobs. To place a configuration name in a standard job, choose create or change from the standard job menu and key the name of the standard job. Press [FLDREL] twice, then press [REL] to get past the "Record Format Assignment". Press [FLDREL] 5 times to see "Environment Name:". Enter the configuration name in the unibase.ini file that the standard job should reference. This allows environment variables (including DMACI) to be set differently for different standard jobs.

Environment variable UBINI, when set to Y in the unibase.ini file in the bin directory, will edit the unibase.ini file immediately after entering the configuration name in the standard job.

Return to Questions
 

You use the verb "open iwindow" in field edits to open a window in image entry to display other images. Its parameters are a number for the window and a string that gives the upper left x and y corner and lower right x and y corner of the window in pixels. The pixel number cannot exceed 9999. If the lower right corner is set 9999,9999 then the window will fill the remainder of the screen. The syntax is as follows:

open iwindow number At winloc.
variable

An example:

move "0400,0120,1000,0750" to winloc.
open iwindow 3 at winloc.

Return to Questions
 

You use the verb "clear iwindow" in field edits to remove the image display from an open image window. Its only parameter is the windows number:

clear iwindow number.
variable.

Return to Questions
 

You use the verb "close iwindow" in field edits to close an open image window. Its only parameter is the windows number:

close iwindow number.
variable.

Return to Questions
 

Yes. In field edits, the verb "showsnip" displays an image (or a snippet of an image) in an open window. The parameters are the full path of the image, the zone of the image, the zoom factor, the window number, and a variable to hold the result. Syntax is:

showsnip box of img page pnum in iwindow inum with zoom gives ans.

Where box is the upper left x and y and lower right x and y in pixels of the snippet of the image to be displayed in the window; img is the full path of the tif file being used; pnum is the page number of the image (pnum is 1 for non-multi-page tifs); num is the number of the open window; zoom is the zoom factor which ranges from -9 to +9 and describes how much larger or smaller the snippet should be from normal (0); and ans holds any errors. (Zero means the snippet displayed correctly.) The errors are positional. A units position error means a problem with the window parameter. A hundreds position error means a problem with the zoom parameter.

Return to Questions
 

Yes. The verb "pan" in field edits moves an image in a window up, down, left, or right. Syntax:

Pan window# up pixels.
Pan window# down pixels.
Pan window# left pixels.
Pan window# right pixels.

where window# is the number of the window to do "panning"; up, down, left, right are the panning directions, and pixels is the number of pixels to pan the image.

Return to Questions
 

By using the control function <imgcur #>. It gives the number of the current pointer to the list of possible images for the current batch on channel #. The value of <imgcur #> may be moved to a variable or a value may be moved to the control function <imgcur #>.

Return to Questions
 

Yes. Set the environment variable UBPR=Y in the unibase.ini file in the bin directory. When set, it allows the function key [RESET] to clear pause messages.

Return to Questions
 

Yes. Set the environment variable UBPF=Y in the unibase.ini file in the bin directory. When set, it allows the function key [FLD REL] to clear pause messages.

Return to Questions
 

Set the environment variable SHOWBC=Y in the unibase.ini file in the bin directory. When output is being done and this environment variable is set, Unibase by DMAC displays on line 2, the batch name, the number of records output for the batch, and the total number of records output for all batches.

Return to Questions
 

The problem is the spaces between "spooler" and "\mntlock". Unibase.ini had spaces after the ETSPOOL definition. None of the path entries should have spaces after the definition.

Return to Questions
 

At a dos prompt type NET USE LPT1:\\PSERVER\HPIIISI - Where PSERVER is the name of the print server and HPIIISI is the printer name. Then output device PRN in Unibase. However, the path on PRN should be changed to LPT1 to match the port in the NET USE command. The line should be put into the .bat file attached to the Unibase icons.

* This example came from the book "Microsoft Windows NT Workstation Resource Kit", page 206.

Return to Questions
 

You can either run a logon script for each workstation or have each workstation run a batch file on their own after logging onto the network.

Example batch file: NET TIME \\SERVERNAME /SET /YES

Return to Questions
 

First, create a device for each filename you wish to output to; device type should be COMM - example: FILE 1 path = %ETROOT%/tmp/file1

Second, in the unibase.ini file, set TYPE## = each device name you created - example: TYPE01=file1

Third, in your output program, at "when start" you should call a bat file you have created to delete the files you are outputting to (so that each output is not appended to the previous output).

Then, all you need do in your program is "type loc##" for each TYPE# set in the unibase.ini.

Return to Questions
 

In unibase.ini in the bin directory, edit DMACI factor V to HIGH1.

Return to Questions
 

The chipset, not the board manufacturer, is what's important here. Since the list of chipsets is extensive, please call DMAC Hotline Support at 540-667-4695 for verification of compatibility.

Return to Questions
 

**Be advised that if the server needs to be re-booted all other users must log out of Unibase.**

Unibase should be installed from a workstation as a peer-to-peer server with the server having a "network" drive letter mapped to it. This will create a unibase.ini file with a UBLINCO set for peer-to-peer and all environment variables with paths (ETROOT, ETBIN, etc.) using that drive letter as part of the path. This data is under the [COMMON] heading in the unibase.ini file.

The unibase.ini should be edited with a second area, labeled with whatever is relevant/memorable. Perhaps we should use [SRVRUSER] so DMAC knows what is going on. Under this area, put in a UBLINCO=6663 for a standalone server, and include all the environment variables that have paths from the [COMMON] area. Only these paths will have to be labeled with the C: drive.

The server user will have to have environment variable UNICFG=SRVRUSER set before calling up Unibase. This can be done with a bat file, or perhaps in the autoexec file or perhaps in the bat file for the icon (menu95.bat or de95.bat).

Return to Questions
 

Yes. For a complete explanation, see manual (Volume 2).

Return to Questions
 

No, it will not. However, if the operator presses {CTRL}{F10}, this tells Unibase to change the gender as it decodes the image. This takes effect on the next image to be decoded and will show up about two records later. It does not affect any images which have already been decoded because they are stored in memory and retrieved from memory if the operator backs up a record.

Return to Questions
 

This can be done for a single image by choosing [H] Image Processing from the main menu, then choosing [C] Image Processing Functions, then choosing [E] Change Image File. Enter the name of the image, press {FLDREL} over all items except Photometric where you enter 1. This process will create a new image whose name is the same as the old image except that the first character is replaced with an exclamation point.

Return to Questions
 

You can have more users than licenses so long as each user logs out of Unibase before another logs in.

Return to Questions
 

If one user aborts out of Unibase the comname will still have that user logged in if you are at the maximum number of users, no one else will be able to log in. To correct this you'll need to zap the user who aborted. Have someone go in Unibase and go to Utilities, Unlock File Locks, zap operator and follow on-screen prompts. This should make it possible for another user to log in. If you are using a version prior to 7.46, you can zap a single operator through the Applications Menu by following the instructions on the screen.

Return to Questions
 

You can restore deleted/lost files from your backup.

Return to Questions
 

Add O29=Y to unibase.ini in the bin directory.

Return to Questions
 

First, you have to find out the name of the printer device for this UNIX. The command line function "Ipstat -t" will give a printer status and also the name of the device. The device name goes in the path line of the Unibase device table.

Second, the device should have the name of a print server defined when it is created. Usually, this name is Ipserver, which is a c program that has to be compiled for each UNIX system. (Source code is provided in the Unibase bin directory).

Third, the device being output to must be spooled. This is one of the parameters when you first set up the device in the device table. From then on, any output to this device goes to a file named spxxxxxx where the xxxxxx is a number starting at 000000. The file is also created in the $ETSPOOL directory.

The device must be online before requests are sent to the print spooler. Also, unispool must be running. It is best to initiate unispool from the console since unispool will shut down anytime the workstation is shut down. 

SPOOLOCK in the $ETSPOOL directory indicates that unispool is running.

Return to Questions
 

You attempted to access a channel via the 'key' control function. The channel number indicated was not assigned to an index. Assign the channel number to an index and it will work.

Return to Questions
 

The 'key' control function was used to access a channel that had not been opened. Open the index to a channel, then it will work.

Return to Questions
 

This message appears when you enter a file in verify mode when all records in the file have been verified. Press {FIELD REL} and try opening another file.

Return to Questions
 

The character you entered was non alphabetic, but the field type only recognizes alphabetic (alpha) characters. Press the {FIELD REL} key to proceed. Enter an alphabetic character and it will accept it.

Return to Questions
 

The APPEND failed either because the associated standard job no longer exists or the record formats named in the job do not exist. If the standard job no longer exists, you'll need to recreate the standard job. If the record formats do not exist, you'll need to recreate the record formats.

Return to Questions
 

The APPEND failed either because the associated standard job no longer exists or the record formats named in the job do not exist. If the standard job no longer exists, you'll need to recreate the standard job. If the record formats do not exist, you'll need to recreate the record formats.

Return to Questions
 

Check to see if your archive device is online. If it is, you could have a hardware error. Contact your IT department for further assistance to this problem.

Return to Questions
 

Check to see if your archive device is online. If it is, you could have a hardware error. Contact your IT department for further assistance to this problem.

Return to Questions
 

AUTO has been disabled within a field edit program.

Return to Questions
 

Spaces, hyphens, asterisks and question marks cannot be used as part of a datafile name. Remove those characters and it will save the file.

Return to Questions
 

A bypass instruction was encountered that referenced an unopened channel. Open the channel and it should work.

Return to Questions
 

A bypass instruction was encountered that referenced a channel assigned to an index. Either remove the bypass verb from the program; or open the file(s) separately.

Return to Questions
 

The keystroke macro being called does not exist in the cmacro directory beneath ETROOT. Check to see if the macro is located in another directory, and, if so, copy it to the cmacro directory under ETROOT. Also see if it has been accidentally deleted and recreate if it has.

Return to Questions
 

The file that contains the numeric exception information does not exist. This file is, or should be, in the ETROOT/misc directory. The file is named numexcp. Else the file does not have proper permissions for access (i.e., the alpha character you typed is not recognized as being one of the exceptions).

Return to Questions
 

The <loc nn> referenced in a type statement does not have an associated environment variable (TYPE01 etc.), or the environment variable is set to a non-existent device.

Return to Questions
 

A.

Assuming you are using a DOS client to connect to the network:
1) Make new directory called binold and copy the contents of bin into it.
2) Insert floppy into drive a: and type a:install.
3) Press Field Release at welcome screen.
4) Press Field Release at dealer screen.
5) Press Field Release at installation program message.
6) Press Field Release at 2^nd installation program message.
7) Press Field Release at 3^rd installation program message.
8) Select "Unibase by DMAC server section", then press Field Release.
9) Select YES and press Field Release when it asks if this is correct.
10) Press Field Release at installation program message.
11) Select proper server configuration and press Field Release.
12) Select YES and press Field Release when it asks if this is correct.
13) Select "Update Installation" and press Field Release.
14) Select YES and press Field Release when it asks if this is correct.
15) Be sure a backup has been taken and select yes if one has, then press Field Release.
16) Select YES and press Field Release when it asks if this is correct.
17) Select install all executables (all others are optional for a 7.4x to 7.4x install) then press Field Release.
18) Select YES and press Field Release when it asks if this is correct.
19) Select the drive of the last install of Unibase by DMAC.
20) Key the directory of the last install of Unibase by DMAC.
21) Select YES and press Field Release when it asks if this is correct.
22) Feed remaining diskettes until install is complete.

Return to Questions
 

1) Log into Unibase by DMAC
2) Exit out the back door: [A] Advanced Processing, [E] System Functions, [F] OS Access, {FLDREL}, {REL}
3) Type: set | more and press enter. This lists your environment variables.
4) There should be an environment variable: COMPUTERNAME=computername.

NOTE: If the computername is Unibase, you have failed to generate a unique computername. If two workstations have the same computername, one of them must change.

Return to Questions
 

Many customers would like to use a SCSI tape drive (instead of a Pertec card) with the tape utilities menu in Unibase by DMAC. With the proliferation of operating systems and new hardware, this gets to be a delicate task. The following hardware/software items and procedure will work. Enhancements to tapespol are coming to allow 32bit operating system usage. If you find other hardware/software combinations that work, please let us know so we can update our instructions for other customers.

HARDWARE: Chi Corporation SCSI tape drive
Adaptec SCSI board

SOFTWARE: Chi Corporation OUTRIGHT - 16bit/DOS version
Adaptec 16 bit API manager (apsi8dos.sys)
DOS 6.22 workstation

PROCEDURE TO INSTALL TAPE DRIVE
The first step is to get the OUTRIGHT software to work. This involves configuring it for your setup. The outright.cfg file which is created will be copied to the \unibase\bin sub-directory.

Install OUTRIGHT per Chi Corporation's instructions. The tapedrive must be turned on BEFORE the workstation is turned on or the operating system can't find the configured device.

Execute OR. An Event Log screen will come up. Click on Config or type alt-c to bring up a small configuration window. Choose the third item "Configure API/SCSI". Enter your SCSI address. (All SCSI devices have a physical number. The lower the number, the higher the priority for the device. Our test system used the number 4 for the tape drive.) Adaptec Option 1 needs to be marked with an "X". Do this by pressing the space bar at the option. Then click on OK. At this point you should run "Test Adapter" to determine if OUTRIGHT sees the tape device. You should also write/read from the tape using OUTRIGHT to determine that the setup works.

PROCEDURE TO USE TAPE DRIVE WITH Unibase by DMAC
Edit your "unibase.ini" file and insert the environment variable that tells Unibase it will be using a Chi Corporation tape drive. From the main menu:

[A] Advanced Processing, [E] System Functions, [H] Edit Unibase.ini File
In the [COMMON] area insert the line
UBTPCL=y
Save the file and exit. (F10-w; F10-x)

Copy the following files from the OR directory to the \unibase\bin sub-directory:
lca.dmd
lca.sgl
outright.cfg
api.exe renamed to tdriver.exe
You may also need the version of tapespol dated 10-19-98 that is 133,456 bytes in length

Return to Questions
 

DMAC converts jobs from other data entry systems provided the relevant data can be written as a flat ASCII file. (All printable characters, no binary.)

The Nixdorf DPEX system can output its libraries any way you want if you give it a standard job with an output program. DMAC gives the customer a system library tape with standard jobs and output programs that creates flat ASCII files for each library. To make it easier to create these files, DMAC also provides command macros for each library and a macro procedure to call each command macro.

However, there are many variations of the Nixdorf DPEX system and sometimes the tape will not read in correctly. All is not lost. The output programs are short enough (and many of them can be copied with slight variations) so that the Nixdorf programmer can easily key them in and still create tapes for conversion.

Instructions for doing so follow.

Please key in the following program. Name it AIDFLD.
DECLARE FLd.
!START
MOVE ' ' TO FLD.
WHEN BATCH OUTPUT '!! '<BATCH>'FLD' (Note space after exclamation point)
<SKIP 40>
!TAPE ADD 1 TO FLD.
IF (FLD) = 40' ' GOTO !T1.
OUTPUT (FLD).
!T1 IF FLD = 10 RELEASE AT END
GOTO !FINI.
GOTO !TAPE.
RELEASE AT END GOTO !FINI.
!FINI OUTPUT <EOF>.
NOTE.

Then, copy the program 5 times giving each copy one of the following names. Change the three characters on line 4 that are shown in boldface above to the three characters associated with each different output name.

AIDFIL FIL
AIDEXT EXT
AIDSRT SRT
AIDOUT OUT
AIDMAC MAC

These programs output the editor language programs with an identifier that tells DMAC which library the program is from. The standard job library, record format library and command macro library each require different output programs.

The output program for record formats follows. Please name it AIDREC.

NOTE-OUTPUT RECFRMT FOR USE BY AID.
WHEN BATCH OUTPUT '!!' 8' '<BATCH>
<SKIP 111>.
OUTPUT <FMT><ALL>.
RELEASE AT END GOTO !FINI.
!FINI OUTPUT <EOF>.
NOTE.

The output program for command macros follows. Please name it AIDKEY.

NOTE-OUTPUT COMMAND MACRO FOR AID
USE.
WHEN BATCH OUTPUT '!!' 8' ' <BATCH>
<SKIP 81>.
OUTPUT <ALL>.
RELEASE AT END GOTO !FINI.
!FINI OUTPUT <EOF>.
NOTE.

There are two possible output programs for standard jobs. Please choose the one you need. If you are on DPEX I or II, then the longer program is needed to convert 1 digit format numbers to 2 digit format numbers. If you are on DPEX V, then the shorter program is used since two digit format numbers are already being used. However, it's best to check your current standard jobs. The initial releases of DPEX V did NOT have 2 digit format numbers and your system may operate this way. In either case, name the output program AIDSTD.

NOTE-OUTPUT STANDARD JOBS FOR AID.
DECLARE A1 A2 A3 A4 A5 A6 A7 A8 A9 A0.
NOTE - 1 DIGIT FORMATS.
WHEN BATCH OUTPUT '!! '<BATCH>
<SKIP 118>.
WHEN FMT 2 GOTO !P2.
OUTPUT '0'<FMT><ALL><SKIP 118>.
RELEASE AT END GOTO !FINI.
!P2 MOVE ' ' TO A1. MOVE ' ' TO A2.
MOVE ' ' TO A3. MOVE ' ' TO A4.
MOVE ' ' TO A5. MOVE ' ' TO A6.
MOVE ' ' TO A7. MOVE ' ' TO A8.
MOVE ' ' TO A9. MOVE ' ' TO A0.
IF (2)# ' ' MOVE '0' TO A1.
IF (4)# ' ' MOVE '0' TO A2.
IF (6)# ' ' MOVE '0' TO A3.
IF (8)# ' ' MOVE '0' TO A4.
IF (10)# ' ' MOVE '0' TO A5.
IF (12)# ' ' MOVE '0' TO A6.
IF (14)# ' ' MOVE '0' TO A7.
IF (16)# ' ' MOVE '0' TO A8.
IF (18)# ' ' MOVE '0' TO A9.
IF (20)# ' ' MOVE '0' TO A0.
IF (2) = '0' MOVE TO A1.
IF (4) = '0' MOVE TO A2.
IF (6) = '0' MOVE TO A3.
IF (8) = '0' MOVE TO A4.
IF (10) = '0' MOVE TO A5.
IF (12) = '0' MOVE TO A6.
IF (14) = '0' MOVE TO A7.
IF (16) = '0' MOVE TO A8.
IF (18) = '0' MOVE TO A9.
IF (20) = '0' MOVE TO A0.
OUTPUT '02' (1) A1 (2) (3) A2
(4) (5) A3 (6) (7) A4 (8) (9)
A5 (10) (11) A6 (12) (13)A7 (14) (15)
A8 (16) (17) A9 (18) (19) 20 (20) (21) (22).
RELEASE AT END GOTO !FINI.
!FINI OUTPUT <EOF>.
NOTE: END
NOTE: OUTPUT STANDARD JOBS FOR AID.
NOTE: 2 DIGIT FORMATS.
WHEN BATCH OUTPUT '!!' <BATCH>
<SKIP 118>. WHEN FMT 2 GOTO !P2. OUTPUT <FMT><ALL><SKIP 118>.RELEASE AT END GOTO !FINI.
!FINI OUTPUT <EOF>.
!P2
OUTPUT '02' (1) (2) (3) (4) (5)
(6) (7) (8) (9) (10) (11) (12)
(13) (14) (15) (16) (17) (18)
(19) (20) (21) (22).
RELEASE AT END GOTO !FINI. NOTE: END OF AIDSTD.

Now we need to create a standard job for each library. Again, this is fairly simple as quite a few of them are copies of the first one entered with minor changes. Each standard job has the same name as the output program.

Please key in the following standard job. Name it AIDFLE.

Standard Job Name: AIDFLD
File Name Protection: **********
Record Format Names: (none)
Output Device: T1 (Or any other tape device)
Record Size: 0040
Block Size: 0040
Fixed Length: Y
Blocked Records: N
Character Counts: 
Pad/Skip Character:
Code Set:
Output Program: AIDFLD
File Protection: (none)

Copy this program 5 times. Give each copy one of the following names and change the output program to the same name as the standard job name.

AIDFIL AIDFIL
AIDEXT AIDEXT
AIDSRT AIDSRT
AIDOUT AIDOUT
AIDMAC AIDMAC

The standard job to output the record format library follows. Name it AIDREC.

Standard Job Name: AIDREC
File Name Protection: **********
Record Format Names: (none) (Or any other tape device)
Output Device: T1
Record Size: 0111
Block Size: 0111
Fixed Length: Y
Blocked Records: N
Character Count:
Pad/Skip Character:
Code Set:
Output Program: AIDREC
File Protection: (none)

There are several flavors of record formats in the DPEX system. If you get maximum record length exceeded when record formats are output, increase the record size and block size by 1 and increase the <skip 111> in the output program by 1. Keep doing this until you no longer get the error message. DMAC will be able to figure out minor differences with your help.

The standard job to output the command macro library follows. Name it AIDKEY.

Standard Job Name: AIDKEY
File Name Protection: **********
Record Format Names: (none)
Output Device: T1 (Or any other tape device)
Record Size: 0081
Block Size: 0081
Fixed Length: Y
Blocked Records: N
Character Counts: 
Pad/Skip Character:
Code Set:
Output Program: AIDKEY
File Protection: (none)

The standard job to output the standard job library follows. Name it AIDSTD.

Standard Job Name: AIDSTD
File Name Protection: **********
Record Format Names: (none)
Output Device: T1 (Or any other tape device)
Record Size: 0118
Block Size: 0118
Fixed Length: Y
Blocked Records: N
Character Counts: 
Pad/Skip Character:
Code Set:
Output Program: AIDSTD
File Protection: (none)

The next step is to create a command macro that will be copied and changed 8 times to output all the other libraries.

The command macro to output the fielded library follows. Name it AIDFLD.

HL HL D

C
H A I
D F L
D FR ** *
** * ** * KY
RL RL RL
HL HL HL

Copy this command macro 8 times. Use the following 8 names. Note that the boldface C changes for each library and the boldface letters FLD also change for each library. In some of the command macros, the letter H in italics (after the boldface c) may have to change to the letter G to get to the selection for printing the library.

AIDFIL D FIL
AIDEXT I EXT
AIDSRT F SRT
AIDOUT E OUT
AIDMAC H MAC
AIDREC B G REC
AIDKEY G G KEY
AIDSTD A G STD

You will also need an additional command macro named AIDTERM as shown below.

** * ** * ** *
HL HL HL
** * ** * ** *

The final program is a macro procedure program which is given below. This is the controlling program to output the various libraries in a flat ASCII format.

!SCREEN SHOW <TOP>
MOVE ' ' TO $VAR02. ADD 0 TO $VAR02.
SHOW <LOC 3.15> 'AID CONVERSIONS'
<LOC 4> '1. STANDARD JOBS-1 FILE'
<LOC 5> '2. RECORD FORMAT-1 FILE'
<LOC 6> '3. FIELD EDITS'
<LOC 7> '4. FILE EDITS'
<LOC 8> '5. OUTPUT PROGRAMS'
<LOC 9> '6. SORT PROGRAMS'
<LOC 10> '7. EXTERNALS'
<LOC 11> '8. MACRO PROCEDURES'
<LOC 12> '9. KEYSTROKE MACROS-1 FILE'
<LOC 13> '0.TERMINATE'.

ACCEPT <LOC 14,25> $VAR02
ELSE GOTO !SCREEN.
IF $VAR02> 9 PAUSE 'INVALID SELECTION';
GOTO !SCREEN.
GOTO !TAPE.
!TERM EXECUTE AIDTERM.
STOP.
!TAPE
IF $VAR02 = 0 GOTO !TERM.
IF $VAR02 = 1 EXECUTE AIDSTD.
IF $VAR02 = 2 EXECUTE AIDREC.
IF $VAR02 = 3 EXECUTE AIDFLD.
IF $VAR02 = 4 EXECUTE AIDFIL.
IF $VAR02 = 5 EXECUTE AIDOUT.
IF $VAR02 = 6 EXECUTE AIDSRT.
IF $VAR02 = 7 EXECUTE AIDEXT.
IF $VAR02 = 8 EXECUTE AIDMAC.
IF $VAR02 = 9 EXECUTE AIDKEY.

GOTO !SCREEN.

NOTE: END OF AID MACRO PROCEDURE.

Once all these programs are keyed in, proceed with the instructions that came with the magnetic tape to pull tapes of your libraries for DMAC to convert.

Return to Questions
 

Your problem is that OPID isn't stored in upper case. When you add a user to Novell (either with nwadmin or netadmin) be sure to type the name in all uppercase. Then when you want to reference the Login_Name environment variable that Novell provides, the result will be in all upper case.

OPID Syntax: SET OPID=LOGIN_NAME

This line goes in the Novell system login script.

Return to Questions
 

TeraForm's OCR/ICR recognition product produces a flat text file that identifies the record format and field for the "ICR'ed" data and also gives co-ordinates for the zone on the image that will display the data which was "ICR'ed". This text file can be converted into a Unibase by DMAC data file by use of the program ASCII2DF.

The Unibase record formats and standard job must already be present for ASCII2DF to operate.

A sample text file produced by TeraForm is shown below. The file is named EOB.TXT and when the name of the file is used as an argument to ASCII2DF, a Unibase file named EOB is created. Also, a sub-directory named EOB is created under the IMAGE sub-directory for Unibase and all the necessary tif files for the batch EOB are placed there. (It is possible to have data for more than one batch in a .TXT file. ASCII2DF will create a sub-directory under IMAGE for each filename, as well as creating the datafile.).

Batch Name: EOB Batch Name; Image sub-dir name
Form ID: 10 Standard Job Name: 10
Image File: c:\unibase\Image\eob1.tif loc of image; used until next img
Format: 1 Record format Name: 10-1
Field ID: 2 Field into which data is placed
Field Name: Provider_Name not used by ASCII2DF
Coords: 1610 238 244 32 zone coordinates where data found
Data Type: Text not used by ASCII2DF
Data: INC data ICR'ed by TeraForm
Field ID: 4 Repeated for as many fields in
Field Name: Provider_ID Record format. Note that
Coords: 1615 275 194 31 fields are even numbers.
Data Type: Text
Data: 161387862
Format: 2 Record Format Name: 10-2
Field ID: 2
Field Name: Last_Name
Coords: 139 645 157 32
Data Type: Text
Data: FISCHER
Field ID: 4
Field Name: First_Name
Coords: 402 646 111 32
Data Type: Text
Data: JOYCE
Format: 3 Record Format 10-3/may be repeated
Field ID: 2
Field Name: Date_of_Service
Coords: 689 644 177 34
Data Type: Text
Data: 01/13/95
Field ID: 4
Field Name: Expense
Coords: 1349 644 111 32
Data Type: Text-Numeric

Data: 25.85
Field ID: 6
Field Name: Copay
Coords: 1968 642 241 29
Data Type: Text-Numeric
Data: 
Field ID: 8
Field Name: Benefit
Coords: 2268 646 88 30
Data Type: Text-Numeric
Data: 5.63
Format: 4
Field ID: 2
Field Name: Expense_Subtotal
Coords: 1349 718 111 31
Data Type: Text-Numeric
Data: 25.85
Field ID: 4
Field Name: Copay_Subtotal
Coords: 1968 711 241 41
Data Type: Test-Numeric
Data: 
Field ID: 6
Field Name: Benefit_Subtotal
Coords: 2268 719 89 30
Data Type: Text-Numeric
Data: 5.63
Format 2,3,4 may be repeated
Image File: c:\unibase\image\eob2.tif Next image used
Format: 1 Record Format 10-1
Field ID: 2 followed by formats 2,3,4
Field Name: Provider_Name as needed.
Coords: 1613 196 245 32
Data Type: Text
Data: I INC
Field ID: 4
Field Name: Provider_ID
Coords: 1619 233 192 32
Data Type: Text
Data: 161387862

When possible, use the Field Names shown in the .txt file as tags for the fields when creating the record formats. This makes it a little easier to match up items. In the check box edits, mark 'Y' for Image Zone Data (the very last checkbox) for all the keying fields. Do not mark Image Zone Data 'Y' for the hidden fields. When the check box edits go to the second screen for the zones on the fields to be keyed, data must be entered for the Primary Display Zone or else an error message, "Display Zone not defined" is displayed and none of the image check box edits are saved. Since the TeraForm coordinates are used to display the zones, the data is irrelevant for all fields except one. Enter the Primary Display Zone as 0000,0000/0010,0010. Mark the Load Zone Data field as 'Y'.

When you are in Image Entry, the size of the top window where the zones are displayed is determined by the size of the largest Primary Display Zone. So at least one field in at least one record format for the standard job to be corrected/verified must be large enough to display all the zones defined by the TeraForm coordinates in the batch. Setting one of the Primary Display Zones as 0000,0000/400,300 allocates about half the screen to the image area. If this is too big or too small, it can be adjusted accordingly in the record format.

To actually convert a TeraForm .txt file to a Unibase datafile, put the .txt file off the Unibase root. Then Type: ascii2df filename.txt
Note that the .txt has to be present.

Return to Questions
 

On some computers, the CMOS setup for the bios has two options: "load best performance" and "load optimal performance". Choose optimal performance if exiting from a dos program or bat file locks up and does not return you to the desktop.

Return to Questions
 

From the system console screen at the console, type the following command:

SPOOL<spooler number>TO QUEUE<print queue name>

When SPOOL 0 queue_name, any output to lanpr1 will go to spool 0
When SPOOL 1 queue_name, any output to lanpr2 will go to spool 1
.
.
.
When SPOOL n queue_name, any output to lanrx will go to spool n

How to set up printer for Novell v3.x:

To create a print queue:
1. Log on to the NetWare file server as the SUPERVISOR and execute PCONSOLE.
2. Select the Print Queue Information option from the menu and press <ENTER>. If no queues have been assigned, the list will be empty.
3. Press <INSERT> and enter the new queue name, then press <ENTER>.
Queue names may be up to 47 characters long. Spaces and the following characters are NOT allowed in queue names: * [ ] / ? < > , " : ;
4. Press <ESC> to return to the Print Queue Information menu.

To create a print server:
1. Log on to the NetWare file server as the SUPERVISOR and execute PCONSOLE.
2. Select Printer Server Information and press <ENTER>. A list of all defined print servers is displayed. If you haven't yet defined a print server, the list is empty.
3. Press <INSERT>. PCONSOLE prompts you to enter the name of the print server. The name can contain up to 47 characters, but it's more practical to use a name that's brief and easy to remember - you'll need to use it every time to start the PSERVER module.
4. Select Printer Configuration. The printers are attached to the print server locally while those attached to the workstation are remote.
5. Press <ESC>. Select Queues Serviced by printer. Associate a printer with a print queue. Reference: "Learning Printing with Netware". Wave Technologies

How to set printer for Novell v4.x:

3 steps to Printer Setup:
1. Create print queue
2. Create the printer
3. Create the printer server

To create a print server:

1. Logon as supervisor to the v.4x server. The easiest way to begin the setup is to start Windows 3.x or higher and execute NWADMIN located in the public directory. You could use PCONSOLE but the new utilities seem to be a little easier.
2. Select Object/Create or select the container and press insert. You'll be presented with a dialog box asking for the type of object you want to create. Choose Print Queue and press enter. You then need to define the print queue name and the print volume (usually the SYS volume).

Creating the printer:

1. This is basically the same as setting up the print queue. Choose the container where the printer will be stored by selecting it with your mouse and pressing insert or choose Create from the Object menu.
2. Choose printer and give it a descriptive name. Click on "define additional properties" and choose Create.
3. Click on the Assignments tab located on the right hand side. Choose the queue that you created earlier.

Creating the print server:

1. As before select the container that will store the print server with your mouse and press insert or choose Create from the Object menu. Choose print server.
2. Give the server a descriptive name, click on the "define additional properties" check box and choose Create.
3. Click on the assignments tab located on the right. Select the printer that you just created by choosing "Add".

Reference: "CNE Study Guide for NetWare 4.1" Clarke, David James

Return to Questions
 

Click on field to highlight, then click on scissors in blue box ( zone boxes will disappear), then click on paper clip, then click OK.

Return to Questions
 

If you type "texwrite/?" you will get a copy of instructions displayed on screen. While the information you get is correct, the following list of instructions is a cookbook procedure you can do:

1. Install a copy of Unibase by DMAC. Let's assume it's on a standalone pc.
2. Copy the texwrite.exe file into the \unibase\bin subdirectory.
3. Copy the ezc8.0 record formats you want to convert to the \unibase\sql directory. This directory is generally empty and is as good a work area as any.
4. Change to the \unibase\sql directory. Type the following command:
\unibase\bin\texwrite R * temp
5. This should make a file named temp that holds a text only version of the EZC8.0 record formats.
6. Move the temp file to the \unibase\tmp directory.
7. Log into Unibase and go to the recfmt directory by typing:
[A] Advanced Processing
[D] File Definition
[B] Record Format
[K] Read/Write Library
Read or Write (R/W):R
Enter Device Name: TEMP
REL to proceed
Unibase displays the standard jobs being read in
REL to proceed when requested at bottom of screen.

You should now have your standard jobs converted to Unibase. From the standard job directory you may do a change on any of the standard jobs to see the batch protection, the record format names and linkages, and the field and fileedit names.

Programs such as fldedits, fileedits, outputs, sorts, and macro procedures should be copied from the EZC8.0 libraries to the corresponding Unibase directories. Only the files with a .et extension need to be copies. Then log into Unibase and parse these programs as you did in EZC8.0.

Return to Questions
 

You need to upgrade the memory for your output computers to 8mg.

Return to Questions
 

If file protection is set in Standard Job as, for example, ABC/, then the IDC file should be in a ABC subdirectory. You need to check and make sure this is so, and if not, make the changes accordingly.

Return to Questions

No. Unibase will process them one after another, but not all at once.

Return to Questions
 

You have etroot set to the wrong drive letter. You must edit the unibase.ini file in the bin directory after determining what the correct drive is.

Return to Questions
 

Call the Unibase by DMAC Technical Support Hotline at 540-667-4695 during our regular business hours and we will walk you through the process.

Return to Questions
 

Check the second page of your Standard Job. The pad/skip is probably set as zero. Make the necessary change and it should be okay.

Return to Questions
 

In newer versions the graphic user interface (gui) for record format generation is for those users who prefer working with a mouse (where possible) instead of a keyboard. Unibase by DMAC will allow you to select the method of generation that you prefer.

There are four selections on the gui Record Format Maintenance Menu: Paint the Screen, Check Box Edits, Define Image Zones, and Paint Zone Display. You choose the item by double clicking on the appropriate icon with the left button on the mouse.

PAINT THE SCREEN:
Choosing this option brings up a window similar in style to many found on other applications. There is a button in the upper left corner which allows you to close Paint the Screen. If changes had been made to the record format, Paint the Screen will display an Exit Record Format window with the options to Save & Exit; Exit - No Save, or Cancel (which will bring you back into Paint the Screen). The 2^nd row is a series of pull down menus, followed by a third row of buttons. The bottom of the screen shows the name of the record format, the total bytes for the keying fields, and the row and column locations of the cursor. The "painting" area has scroll bars on the right side and the bottom. Paint the Screen opens with the cursor at row 1, column 1. There is also a double-headed arrow that indicates mouse position.

Actually creating a screen is very similar to the original screen painting function in Unibase. Use the arrow keys, the {FIELD REL} key, and the Page Up and Page Down keys to position the cursor. In addition, you can position the mouse, click the left mouse button, and the cursor will move to that position. Use the keyboard to enter tags (descriptions/instructions/field identifiers) and fields (data entry area) as needed. Fields are defined by using the underscore key. Once the screen has been started, various functions on the pull down menus or the button bar may speed up the process of finishing or changing the screen.

Pull Down Menus

The second row has five pull down menus labeled File, Edit, View, Options, and Help.

The File menu allows you to Save or Exit the Paint the Screen function. Again, if you choose to exit and changes have been made, the same Exit Record Format window that comes up when you click the close button comes up here.

The Edit menu allows you to Cut, Copy, or Paste, but is only active if you have marked lines for doing so.

The View menu is where you turn on/off display of Non-Display (hidden) fields.

The Option menu allows you to Set Key Order, toggle insert mode on/off (which may also be done by pressing the Insert key), enter non-display fields, and set tag attributes.

The Help menu is not implemented yet.

Button Bar

The next row has a series of buttons/icons that perform a variety of functions.
The first button saves the file with all changes that have been made. 

The next two buttons only operate if a field (or several fields) has been highlighted. You highlight a field by placing the mouse cursor in the field, clicking and holding the left-mouse button, and then moving the mouse slightly. The whole field will turn a different color. You can highlight several fields by holding down the left mouse button and moving the mouse over the fields.

The 2^nd button cuts a highlighted area. The highlighted area disappears from the screen and is placed in a clipboard area. If the highlighted area is on a line by itself, or covers several lines, then the lines the fields were on also disappear and the data below those lines is moved up.

The 3^rd button copies a highlighted area. The highlighted area is placed on the clipboard but the information does NOT disappear from the screen.

The 4^th button pastes an area that has been cut. Be aware that if an individual field has been cut out of a line that had other fields, then that field may be pasted anywhere on the screen where it will fit. If a full line has been cut/copied, placing that data on a line by itself means the column placement remains the same. If there is not enough room for the pasted data, an error message comes up and the operation is not performed. If data is pasted between other fields, the fields below the pasting area are moved down as needed.

The 5^th button toggles the display of hidden fields.

The 6^th button allows you to create non-display fields. The characters ND show up in the right-most position of the button bar. Any tags/fields created while ND is showing will NOT appear for a data entry operator even though they show up on this screen. Toggling to not display the hidden fields has no effect if you are in create non-display fields mode.

The 7^th button (with the keys picture), is used to set keying order. Clicking on it gives you instructions for setting keying order.

The 8^th button only has an effect if a single field is highlighted. It opens a window to the check box edits for that field, provided you are changing an existing record format, not creating a new one.

Buttons 9 through 12 operate on single lines. If button 9 (with centered lines picture) is clicked, the data in the line where the cursor is located is centered. (Pressing the {FLD} key, followed by the letter {C} or pressing the {LOC RTN} key has the same effect.)

If button 10 (with the scissors and solid lines picture) is clicked, the line where the cursor is positioned is completely deleted and all other lines are moved up. (Pressing the {FLD} key, then the {DELETE} key has the same effect.)

If button 11 (with the scissors and two-color lines picture) is clicked, then the data from the cursor position to the end of the line is deleted.

If button 12 (with the paste bottle picture) is clicked, then a blank line is inserted at the cursor position and all other lines below the original cursor line are moved down. (Pressing the {FLD} key, followed by the {INSERT} key has the same effect.)

The next two groups of five buttons are similar. They set the attributes for the tag and field data for whatever is next created. The codes N, U, R, H, B stand for Normal, Underline, Reverse, Highlight and Blink. (There are default colors for these attributes, but they may be changed for data entry from the Operator Login menu off the Main Menu of Unibase by DMAC.) The current setting for attributes is indicated by a red square in the upper left corner of appropriate letter.

CHECK BOX EDITS:
Choosing this option brings up the check box edits for the first field in the record format. It is overlaid by a window that shows the status of the Data Entry Full Screen Option. If the circle has a dot in the center, then the full Screen Option is enabled and an operator will see the entire record format displayed. If the circle does not have a dot in the center, the Full Screen Option is not enabled and fields will be presented to the operator one at a time for keying. Change the option by clicking on the circle. (It does not toggle.) Then click on the Continue box.

Similar to the Menu User Interface (mui) of check box edits, each field in the record format is shown one at a time on a full screen display with all the check box edit choices. There are scroll bars on the right hand side and at the bottom of the screen. If you click on the box in the upper right corner (with an up-pointing AND a down-pointing arrowhead), the display with shrink to about a third of its size and display one column of check boxes at a time. There are scroll bars to move the check boxes up and down or right and left. (Right and left are not necessary unless you shrink the display.)

Move from check box to check box by pressing the tab key or by clicking on a particular check box with the mouse.

If a check box edit has a number of choices, the current choice is highlighted when that check box is selected. Bring up a window of the other possible selections by clicking on the arrowhead to the right of the check box edit. Make a selection by clicking on it. It will appear in the check box display and the selection window will close.

If a check box can only be yes or no, a box with dotted lines appears around the check box edit when that check box is selected. If the circle to the left of a yes/no check box has a dot in it, then the choice is yes. If the circle is empty, the choice is no.

A few of the check box edits will need data entered from the keyboard.

The first five buttons at the top of the screen are used to move to other fields in the record format. Button 1 (left arrowhead with a vertical bar) takes you to the first field in the format. Button 2 (left arrowhead) takes you to the prior field in the format. Button 3 (number sign) opens up a window with the tag names of all the fields in the record format. Highlight the field you wish to go to, then click on the Go to Field Indicated box. Button 4 (right arrowhead) takes you to the next field in the format. Button 5 (vertical bar with right arrowhead) takes you to the last filed in the format. If any of the button icons are light gray instead of black, those functions are not available. (For example, if you are already on the first field, then the first two buttons are useless and so are shown in light gray.)

Many of the mui ways of proceeding to the next field still exist. Press {REC}{FWD} to go forward or press {REL}. Press {REC}{BACK} to back up to the prior field in the record format.

The 6^th button is used to duplicate check box edits from other fields. When you click on it a window opens up with the tag names of all the fields in the record format. Highlight the field whose check box edits you wish to duplicate, then click on the Dup Field Indicated box.

DEFINE IMAGE ZONES:
Choosing this option brings up a divided screen. The blank area on top is where the image zones will be drawn. The bottom area shows the data entry screen that was painted using the Paint the Screen option.

A sample image is needed to create the zones the operator will see while keying. So the first thing this option does is ask you for an image to display. A box labeled File, Open... appears which displays the names of the files in the unibase\image\ sub-directory. Choose an image that goes with the record format you are displaying by locating the image name in the leftmost box. Display the image either by double-clicking on its name or clicking on the OK button on the right-side of the File, Open...box. (You may also choose images from sub-directories under the image directory.) The File, Open... box will disappear and as much os the image as will fit displays in the view area at the top of the screen.

The left most button at the top of the screen is used to close this option. If changes had been made to the record format, Define Image Zones will display an Exit Record Format window with the options to Save & Exit; Exit - No Save, or Cancel (which will bring you back into Define Image Zones.)

The 2^nd row is a series of pull down menus, followed by a 3^rd row of buttons. The data entry screen area has scroll bars on the right side and the bottom.

Define Image Zones is entirely mouse driven so an arrow shows up that indicates mouse position. The mouse performs different functions depending upon which window it is operating in. It is used to draw zones when in the image window, highlight fields when in the data entry screen window, and perform specific functions when buttons are clicked.

The Unibase by DMAC mouse driven screen generator is extremely versatile in that zoning is sone in the fashion most comfortable to you. You may select a field, then zone it, or draw the zones, then attach them to a field. There are also a variety of ways to do these procedures, depending upon whether you prefer menus or buttons. Various steps for a particular procedure will be described. Remember that you may choose whatever step is most logical to do first for you.

Pull Down Menus

The 2^nd row has five pull down menus labeled File, Edit, View, Options, and Help.

The File menu allows you to Save or Exit the Define Image Zones function. Again, if you choose to exit and changes have been made, the same Exit Record Format window that comes up when you click the close button comes up here.

The Edit menu allows you to Cut Display Zone, Cut Image Zone, Cut Highlight Zone, or Copy All Zones. However, these operations only make sense if a field is active that is already zone.

The View menu can be used to scroll up, down, left, or right one "jump" at a time. It is also where you turn on/off the display of hidden fields.

The Options Menu lets you Attach Zones to Field or Define Zones for Next Field (which clears any zones on the screen.)

Drawing Zones

A zone is drawn on the image by holding down the left mouse button and dragging the mouse to the right and down. As you move the mouse, a rectangular box is drawn between the point where you first clicked the mouse and the current position of the mouse. (If you move the mouse left or up, no box shows up.) When the box is the desired size and shape, release the mouse button. In addition, the 8^th and 9^th buttons on the button bar now have color to indicate they can be used. (Before the first zone is drawn, buttons 8 through 11 are gray outlines only.)

This 1^st box is called the display zone. You may now draw an image zone inside the display zone just the way you drew the display zone. If the image zone is NOT completely inside the display zone, Unibase brings up a window title "Define Zones Error" that has two buttons. You may either clear the image zone by clicking on Cancel, or make the image zone the new display zone by clicking on OK. Notice that button 10 now has color to indicate that it is active. One more zone, the highlight zone, may be drawn inside the image zone. An error window giving different options comes up if the highlight zone is not active. (The colors of the boxes may vary, depending upon whether your image has a black background or a white background. But colors around buttons 9 through 11 should always be blue, green, and red.)

Selecting a Field

A field in the data entry screen is selected by putting the mouse cursor anywhere in the field (even the tag area) and double clicking the mouse. (If you double click on a tag only field, you get a window titled "Record Format Error" that tells you a tag must have a field to hold the image zones. Click on OK to try again. If you double click in an area where there is no field, you get a similar "Record Format Error" window.)

Attaching Zones to Fields

The crucial step in defining image zones is to attach the zones to the fields. It is possible to draw (mouse?) any number of zones, but unless you specifically attach the zones to a field, they will be lost.

If you select a field first, then draw the zones for that field, you must attach the zones to the field in one of two ways. Click on button 12 (the paper clip) and a window titled "Attaching Zones To Field" opens up. It has two choices. Click on OK to attach your zones to the highlighted field or click on Cancel to NOT attach and give yourself a chance to redraw the zones. The same window titled "Attaching Zones To Field" opens up with the same two choices. Once a field has zones attached to it, the first character of the field changes from a carat (^) to an asterisk (*) so that you know the field has been zoned.

Pull Down Menus

The second row has five pull down menus labeled file, Edit, View, Options, and Help.

The File menu allows you to Save or Exit the Paint the Screen function. Again, if you choose to exit and changes have been made, the same Exit Record Format window that comes up when you click the close button comes up here.

The Edit menu allows you to cut (remove) the display zone, image zone, or highlight zone, form the image view, but only works if there are zones showing on the image. The Edit menu also lets you copy all the zones displayed to any field. When you choose this options, only the zoned area is displayed, the rest of the image is cleared and replaced with a gray background. Copy the zones to a field by moving the mouse cursor to the desired field on the data entry screen and double clicking. A window titled "Attaching Zones To Field" opens up and three choices are displayed: OK to attach the zones to the field indicated, Abort to redefine the zones before attaching, or Cancel to select a different field.

The View menu is used to scroll the image area up, down, left, or right a small increment at a time. It is also where you turn on/off display of Non-Display (hidden) fields.

The Options menu allows you to attach the zones to a field or to define the zones for the next field (which clears the zones off the screen and puts the image back to its initial position.) If you choose "attach", only the zoned area is displayed, the rest of the image is cleared and replaced with a gray background. Attach the zones to a field by moving the mouse cursor to the desired field on the data entry screen and double clicking. A window titled "Attaching Zones To Field" opens up and the same three choices described above are displayed: OK, Abort, or Cancel.

The Help menu is not implemented yet.

To change the size of the image display area, position the mouse on the dividing line between the image display area and the data entry screen area.. The cursor will become a double-headed arrow to indicate that's where you are. Hold the left-button on the mouse and move the mouse up or down to reduce/enlarge the area devoted to the image display. Unibase has a minimum and maximum display that it will not let you exceed.

You may either draw a zone first and then attach it to the field of your choice, or select the field first, then draw your zone.

SELECT FIELD FIRST
Place the mouse cursor anywhere in the field to be zoned and double click to select the field. It will change color to show it has been selected.

PAINT ZONE DISPLAY
If changes had been made to the record format, Paint the Screen will display an Exit Record Format window with the options to Save & Exit; Exit - No Save, or Cancel (which will bring you back into Paint the Screen). The second row is a series of pull down menus, followed by a third row of buttons. The bottom of the screen shows the name of the record format, the total bytes for the keying fields, and the row and column locations of the cursor. The "painting" area has scroll bars on the right side and the bottom. Paint the Screen opens with the cursor at row 1, column 1. There is also a double-headed arrow that indicates mouse position.

Return to Questions
 

If you are running on a release of Unibase by DMAC lower than 7.42 you should make arrangements to upgrade as soon as possible. Also, you will want to examine the makeup of any of your programs that rely on date storage. For a further explanation, or if you have specific questions, please contact the Unibase by DMAC Technical Support Hotline. Our number is 540-667-4695.

Return to Questions
 

1. The File Input/Output Function [E] Read Unformatted File will do short reads now. That is, a new record will be started whenever a delimiter is encountered. The delimiter must be entered as a decimal number, usually "010" for dos-based systems.

2. A variable may now be defined to be an array. The syntax is - 
define array arrayname[numrows][numcols]

where numrows is the number of rows in the array and numcols is the number of columns in the array. Both numrows and numcols are expressed as integers.

Further, the elements of the array each have the default specification as for single Unibase by DMAC variables. Should one need a different size, then ":nnn" may be appended to the numcols item where nnn is the size in bytes of the variable.

In particular, an array defined as follows:
define array buffnam[1][1:47]

would replace the definition of a 47 character buffer in the Tartan Data Entry language.

3. For field edits, an asterisk in parentheses (*) denotes the current field.

4. The control function <record> (not <records>, which is something different) returns the current record number.

5. The environment variable UBRW (UniBase Record Written) has been added. When set to Y, the last record being entered (not verified) is removed from the file if the operator terminates/interrupts the batch before entering a keystroke. This happens even if a field edit has been run on the record.

6. Unibase by DMAC has implemented reading/writing ascii text files with the AID language. The these text files must be in the "text" sub-directory under the ETROOT. They may not have an extension in the name. Sub-directories may exist under the "text" sub-directory. Defining and accessing these files is done with several new verbs which you can access by asking for them in this technical support program.

7. Because DMAC has been converting customers where programs accessed fields using their keying order number instead of their field order number, an option has been added to store batch/file data in keying order sequence. This avoids having to change massive amounts of code where the field references were different.

The feature is implemented from the record format maintenance screen. When item "[D] Edit Check Boxes" is chosen, the third question is "Use Keying Order as Field Number: N". Default is "N". Change to "Y" to have data in batches stored in keying order instead of field number order.

CAUTION: Using this feature means that if you change keying order on a record format, any batches keyed prior to the change will not be usable in data entry.

8. DMAC runs on a variety of platforms. Some of those platforms are more useful than others in providing data about the number of users of the software. The platforms that do not provide this information require other solutions. Unibase by DMAC uses a file called "comname" that resides in the bin directory to track the users logged into Unibase. The users are tracked by their computernames, NOT but their Unibase operator id names.

If a user does not exit Unibase by using the exit options on the menu, then the user's computername is not cleared from the comname file. To keep comname cleaned up, DMAC has added a new program: "ntstat2.exe" which should be run on a regular basis when the fewest operators are logged into the system. Use "scheduler" for NT systems to automate execution of nstat2. (Use "cron" for UNIX systems to automate execution of UNSTAT2.)

NSTAT2 is a 32-bit application and must be in the sub-directory "bin32" under your ETROOT directory. It should also be run from the server. If users are accessing the NT server using dos with tcp/ip, then they should not execute ntstat2 from their workstations.

To get mor information about NTSTAT2, type "ntstat2/?" from the command line in the bin 32 sub-directory.

9. More sophisticated scanners will create ".IDC" files and put them in the "ETROOT\idc" sub-directory. Unibase Image Entry uses these files to determine which tiff images to use for a batch of image entry. Unibase assumes the images are in the "ETROOT\image" sub-directory unless the fourth parameter of the line entry in the ".IDC" file specifies a line number in the "imgpath.idx" file.

The entry in the "imgpath.idx" file is a path that precedes the tiff name in the ".IDC" file. (In other words, Unibase is NOT expecting the tiff name in the ".IDC" file to have a complete path name, just a partial path from the image sub-directory.)

However, some scanners can only put entries into an ".IDC" file with a complete path. To accommodate these scanners, put this same path as an entry in the "imgpath.idx" file. Then put a minus sign before the parameter in the ".IDC" line that specifies the number of the entry in the ".IDC" file.

Then Unibase Image Entry will "subtract" the contents of that line number from the tiff name given in the ". IDC" file before preceding the tiff name with the path in the "imgpath.idx" file. (Yes, it sounds redundant, but Unibase was not consulted before this particular scanner software was designed to put only complete path names into a text file.)

For example, the ".IDC" file for a particular batch might contain the line:
i:\custa\joba\tifnam.tif:2:01:-03:1:

and the corresponding entry in the "imgpath.idx" file will have the entry:
i:\custb\jobb

Unibase will look for the tiff as
i:\custb\jobb\tifnam.tif

instead of 
i:\custb\jobb\custa\joba\tifnam.tif

which is what Unibase would look for it the imgpath entry were a positive number instead of a negative number.

Note that the directory path may be changed with this technique PROVIDED the new directory path has exactly the same number of characters as the old directory path.

Return to Questions
 

A. Assuming you are using a Chi card, instructions are as follows:

Rename api.exe to tdriver.exe and put it in the bin directory.
Copy outright.cfg to unibase\bin directory.
Copy lca.dmd and lca.sgl to unibase\bin directory.

In the unibase.ini file (in the unibase\bin directory) insert the following line:
UBTPCL=y Also make sure that PATH has an = sign after it.

If you use SCSI cards, then the above still holds for 16bit DOS versions of Chi's outright software. However, you now need an aspi manager for the outright software to talk to. This is supplied by the company that supplies the scsi card. There should/may be dos and windows versions of the aspi manager. You'll need the dos version to go with the dos version of outright.

The above definitely works for a dos client, regardless of the server.

For any SCSI drive, tape drive MUST be turned on before workstation is turned on or operating system can't find the configured device and Chi's OR software (and consequently Unibase) won't find the tape drive.

For DOS

OR cannot find the tape drive in the booted dos mode of Win 95 with the dos aspi manager and definitely not in Win 95. This will be corrected when we do 32bit tapespol.

OR16 in win95 with aspi32 locks up when you try to configure adapter.
OR32 in win95 with aspi32 works fine.
OR16 ind dos with aspi8dos.sys for adaptec.

Call up OR first.
Configure. Alt-c brings up configuration menu.
Configure aspi/scsi. Have to put in number of your device.

Return to Questions
 

Chi Corporation SCSI tape drive with 16bit version OR (Outright) software.

Adaptec SCSI board with 16bit version of aspi manager. It is named aspi8dos.sys and must be loaded in memory.

 Tape drive MUST be turned on.

First you have to make a configuration file (outright.cfg) that works for your configuration.

Execute Chi's software named OR. You'll see an Event Log come up on your screen.

Either click on Config in the tool bar at the top or type alt-c and you will get a small menu. Choose the third item - Configure ASPI/SCSI.Enter the SCSI Address for the tape drive. This is the physical number assigned to each SCSI device in your system. The lower the number the higher the priority for the device. (Our test system happened to be "4".)

You MUST put an X in Adapter Option 1. (This is done by pressing the space bar.) Then click on OK to save the configuration. 

This creates the outright.cfg file that is put into the \unibase\bin\directory.

You should then run Test Adapter to see if the OR software sees the tape device. You may also want to try writing a file to the tape with the OR software.

Return to Questions
 

It sounds as if you installed the program on something other than a workstation (maybe the server?). Go to a workstation and install the program. If you still get this error message, call the Unibase by DMAC Technical Support Hotline at 540-667-4695.

Return to Questions
 

Go into your local printer and change the settings from portrait to landscape.

Return to Questions
 

You need to reparse the output program then try again.

Return to Questions
 

You set it for must complete but not must enter. That way an operator can skip the field, but if one keystroke is entered, then all keystrokes must be entered.

Return to Questions
 

Under a DOS client, Unibase version 7.46 requires at least 8 mg of memory. You'll need to increase the memory.

Return to Questions
 

It means you are having a hardware issue. If your are using Novell, run vrepair. If you are using DOS, run scandisk. If this does not help, call the Unibase by DMAC Technical Support Hotline at 540-667-4695 and we will try to help you or give you some answers.

Return to Questions
 

Check your output directory in the unibase.ini file. You probably have 0 bytes in this directory. You need to make the proper change to the number of bytes and the message should go away.

Return to Questions
 

You need to upgrade your version of Unibase to 7.46.

Return to Questions
 

You need to re-parse the program.

Return to Questions
 

The standard job name already exists in the ETROOT/stdjob directory.

Return to Questions
 

This message will display when execution of the 'Dump a File' function is interrupted or if the file is corrupt.

Return to Questions
 

This message will display when execution of the 'Dump Index' function is interrupted, or if a file within the index is corrupt.

Return to Questions

The file edit name does not exist or there is not enough memory to load the program.

Return to Questions
 

(nn=[operating system error code]). There is not enough memory to load the MENU program.

Return to Questions
 

There is not enough memory to load the sort program or the named program does not exist in sort directory.

Return to Questions
 

The terminal type defined in UBTERM (Unix Systems) does not exist in the ETBIN/termcap file.

Return to Questions
 

The device table file SPOOLDIR/devtable does not exist. Create the file by selecting: System Functions menu option [E] Spooler Maintenance.

Return to Questions
 

The record format(s) required by the standard job contains errors or is corrupted and cannot be read. Press {FIELD REL} to proceed. Recreate the record format and try again.

Return to Questions
 

The help files are missing from the ETROOT/help directory. Get the source diskettes or CD and copy all files with the .hlp extension to the ETROOT/help directory.

Return to Questions
 

The record format specifies that the field edit must be run when the current field is released. This error message is displayed if the field edit specified in the standard job does not exist. The field edit name is displayed where X's are shown. This error message is followed by the error message "Cannot Start Field Edit".

Return to Questions
 

This error message is displayed if the standard job or the job specifications for Data Entry menu option [B] Enter a File specify a format link that cannot be created. For example, the standard job may specify that format 1 links to format 2, but does not assign format 2. Press {RESET} or {FIELD REL} to return to the Data Entry menu. Check and correct the format numbers and links.

Return to Questions
 

When a data file is opened, Unibase by DMAC creates a lock file etroot/misc/xxxxx.lck to indicate that the data file is in use. This error message is displayed if Unibase by DMAC cannot create the lockfile for the data file you wish to open. Press {RESET} or {FIELD REL} to return to the Data Entry menu. Check and correct the format number and links.

Return to Questions
 

When Unibase by DMAC Main menu option [B] is selected, Unibase by DMAC reads the file ETROOT/misc/chkdig.et which contains the check digit calculating algorithms. This error message is displayed if the file does not exist or has been damaged. Remove the damaged file from the ETROOT/misc directory and re-program the check digits. Press {RESET} {FIELD REL} to return to the Unibase by DMAC Main Menu. The data entry supervisor can create or re-program the check digit file by selecting File Definition menu option [I] Check Digit.