Tools

The jED editor is a full screen, context sensitive, screen editor designed specifically for ease of use and learning and is the pre- ferred editing tool for the jBASE-operating environment and suited to the editing of jBASE BASIC programs. The features provided in jED will be familiar to users of many other editors. It incorporates many powerful facilities for manipulating text and data, and contains all the features that programmers have come to expect of a good editor.

jED has full screen access to jBASE file records and UNIX files - a feature not provided by other editors. The Command key- strokes are fully configurable by each user, as is the keyboard. Therefore, jED is customizable to mimic many of the operations of other editors and provide a familiar environment for new users. Keyboard and Command  independence make jED  the most versatile and powerful editing tool available for all jBASE-editing requirements

jED

jED is a robust, terminal independent screen-editing tool.

JED Command Syntax

JED Filename Item (Options JED Item )

Editor Screen

Sample jED editor is given below.

*File PROGS, Record cust_rep.b Insert 10:45:17 Command->
001
002
003
-------------------------------- End Of Record --------------------------------

The screen is divided into three sections:

• Editor status line at the top
• Command line
• Data editing area, which fills the rest of the screen.

Execution Commands

Command Line

From Edit mode, press the <Esc> key from keyboard to invoke the command line.

Invoking jED

Call the jED editor from the UNIX or Windows Command line.

Command Syntax

jed pathname {pathname..}
jed {DICT} filename{,filesection} {record-list} {(options)}}

If you simply issue the Command jED, the editor will open at the last file in use with the cursor positioned wherever it was when the last edit session was closed. In other words, you can continue from where you last left off. If the file referenced by pathname does not exist the Command, jed pathname, will either open an existing file or create a new one. The edit window displays the contents of the file. If you specify a list file, the editor will present the next file as you finish with each one.

When the editor is supplied with the name of a file resident in a database (such as a j-file), it scans the rest of the Command line looking for a list of record keys. If no record keys were specified, the jED editor will prompt for a list, else it will edit the list of record keys one after the other.

Command Line Options

Options available when executing the jed Command are as follows:

EXAMPLE

jed test.b

jed test.b (B5,2)

jed invoices.b subs.c

jed BP menu1.b menu1.1.b

jed ORDERS 0012753 0032779 (R

SSELECT ORDERS WITH CUST.NAME = "UPA"
>JED ORDERS

jed -F BP \*

The record STXFER.b in file BP is opened for editing. A tab stop is set at column 10 for use in this session.

jed -F BP STXFER.b \(T10

The jED editor has two different modes:

• Command mode for entering editor Commands, and
• Edit mode for entering or modifying data. The current mode is displayed at the top of the screen.

Command Mode

When the editor is invoked, the record or text file is displayed, and the user is placed in input mode with the cursor at the input position.

To change to Command mode simply press the <Esc> key on the keyboard. The cursor now moves to the top portion of the screen and the editor awaits input of a Command. Once a valid Command has been executed, control passes back to the Edit mode if appropriate.

Edit Mode

Use edit mode when entering or modifying data. This is the default mode for an editor session.

Keyboard control sequences are available to perform a variety of functions such as cursor positioning, scrolling and marking text for a subsequent action.

Some Command line operations are also available from keyboard control sequences.

Keyboard Personalization

The jED editor allows a considerable number of functions and Commands to be performed whilst in edit mode, mostly by com- bining the <Ctrl> key and one other key.

Most keys have a default value (which can be reset using the E option when invoking jED). These can be reconfigured for each Command. The keystroke sequence can be chosen to suit the keyboard, the installation environment or personal preference.

The keystroke environment is usually be set up by modifying the UNIX terminfo file parameters. The default editor Commands can also be overriden by configuring the .jedrc file.

jED Default Key Commands

The default keystroke sequences available from jED are shown below. If the system administrator has reconfigured these for a particular port, they can be re-assigned by using the E option when starting a jED session. The execution of a Command is relative to the current cursor position.

Key	Function
<F1>	Scrolls the screen up one line.
<F2>	Scrolls the screen down one line.
<F3>	Scrolls the screen up half a page.
<F4>	Scrolls the screen down half a page.
<F5>	Scrolls the screen up one page.
<F6>	Scrolls the screen down one page.
<F7>	Displays the first page of the record or file.
<F8>	Displays the last page of the record or file.
<F9>	Pressing <F9> when the cursor is positioned on a line of source code that begins a structured statement (IF, BEGIN CASE etc.), will cause the editor to locate the closing statement for the structure. If the cursor line is an IF statement then the editor will attempt to find the END statement that closes this structure. If there is no matching END statement then the editor will display a message to this effect.
<F10>	The <F10> key is complement of the <F9> key. Therefore, if the cursor is positioned on an END statement, then the editor will attempt to find the start of the structure that it is currently ter- minating. If the END has been orphaned (it matches no structure), then the editor will display a message to this effect.
<Ctrl A>/<Home>	Moves cursor to start of the current line.
<Ctrl E>/<End>	Moves the cursor to the end of the current line.
Left Arrow	Moves the cursor one character position to the left.
Right arrow	Moves the cursor one-character position to the right.
Up arrow	Moves the cursor to the previous line.
Down arrow	Moves the cursor to the following line.
<Tab>	Moves the cursor to the start of the next tab position on the line.
<Shift Tab>	Moves the cursor to the previous tab position on the line.
<Esc>	Moves the cursor to the Command LINE.
<Ctrl W>	Deletes from the cursor to the end of the word, including the fol- lowing whitespace characters.
<Ctrl K>	Clears text to the end of the line. If the cursor is situated at the end of the text line, then this command will join the following line with the current line.
<Back Space>	Performs a destructive backspace.
<Delete>	Deletes the character under the current cursor position.
<Ctrl D>	Deletes the current line. By default, this key must be pressed twice to delete the line. This is to avoid accidental deletion by users famil- iar with vi. To override, place "set delete-line = ^d" in the .jedrc file.
<Ctrl G>	Sets the start or end position for marking a block of text. The first <ctrl g> will mark the start of a block or mark a single line. The second <ctrl g> with the cursor on a different line will mark a com- plete block. The block can be unmarked by pressing <ctrl g> a third time.
<Ctrl L>	Inserts a blank line below the current line and positions the cursor on it.
<Ctrl N>	Locates the next occurrence of an earlier located string.
<Ctrl O>/<Insert>	Toggles between the Overwrite and Insert data entry modes.
<Ctrl P>	Locates the previous occurrence of an earlier located string.
<Ctrl R>	Redisplays the screen and discards the most recent updates (since the last carriage return).
<Ctrl T>	Copies the character at the corresponding cursor position on the line above the current line.
<Ctrl V>	Performs jbase BASIC program indentations on the current screen window.
<Ctrl X>	Exits the current record without writing away any updates. If the record has been changed within the current editing session then the editor will ask confirmation to exit the modified record.
<Ctrl ]>	Inserts the field value delimiter character.
<Ctrl \>	Inserts the field sub-value delimiter character.
<Enter>	Opens a new line. Any characters on the current line after the current cursor position are moved to the start of the new line.

Command Line Operations

To enter the Command line from the jED edit mode, press the <Esc> key, or one that has been reconfigured to perform the same action.

Leaving the Editor

There are several options available to exit a file or a current editing session. It can be:

• Deleted
• Stored in its latest form, under the current editing session
• Stored, as it existed before the edit session began.

Abandon, Edit and Start a New Session

The E Command will abandon the current edit (you will be asked to verify leaving a changed record) and edit the specified record(s).

Command Syntax

E unixfile
E filename record

If the form filename record is used, then the filename should be the name of a jBASE file. You can also specify the pathname of a standard UNIX file with the unixfile form.

NOTE: Wildcard characters such as asterisk (*) are not expanded by the E Command, and that you must use the jBASE file name again, even if you are currently editing a record from within that file.

Delete File Or Record

The Command Syntax is as follows:

FD {options}

Options can be K, T and O. See the Command Options topic for details.

This Command deletes the file or record and releases any lock set. Before it does so, the user is prompted for confirmation. The edit session then terminates, or continues with the next record if this choice is in effect.

Exit And Update

The Command Syntax is as follows:

FI {options} {unixCommand}

FI writes the updated version of the file or record back to disk and releases any lock set. The edit session then terminates, or continues with the next record, if this choice is in effect.

Options are B, K, R and T. See Command Options for details.

Unix Command specifies a UNIX Command to be executed on exiting the editor.

Exit and Discard

Command Syntax

EX {options}

EX leaves the file or record as it was at the start of the session, and releases any lock set. If updates have been made you will be prompted for confirmation before the updates are discarded. The edit session then terminates, or continues with the next record, if this choice is in effect.

Options are K, T and O. See Command Options for details.

Update without Exit COMMAND SYNTAX

FS {options} {unixCommand}

This Command writes the updated file or record to disk, then returns to the editing session at the point where it left off. options are B and R. See Command Options for details.

unixCommand specifies a UNIX Command to be executed on exiting the editor.

Display Record In Hexadecimal

The Command Syntax is as follows:

HX 
HEX

This Command acts as a toggle such that each iteration of the Command turns the hexadecimal display on or off depending upon its previous state. The HX (or HEX) Command is only used for display, the record is not stored as it appears in hexa- decimal.

Command Options

Options	Description
R	Specifies that, after the file has been written to disk, it should be executed. Additional parameters can be added to this option and passed to the program. The editor issues the Command filename {para- meters} to execute the program. Note that the .b suffix is removed. This option is particularly useful to jBASE BASIC programmers.
K or T option	Specifies that if the editor was working from a list of records, the list should be discarded and that the editor should exit directly to the shell (or to the calling process).
O	Specifies that the confirmation request normally issued with the FD and EX Commands should be suppressed.

Examples

The following command helps you to exit the record and write it to disk. If in the middle of editing a list of records, it abandons the list and terminates the editing session.

FIK

The following command helps you to delete the current record being edited. The normal confirmation of this action is not given.

FDO

Locating Strings

The editor allows the user to search and locate any string held in the body of the text being edited. There is also a keystroke Command sequence (default <Ctrl N>) to allow the user to find the next occurrence of the string used in the previous locate Command.

The locate Command Syntax is as follows:

L{nnn}dstring{doption}

Options	Description
Nnn	The numeric value of the number of lines to search from the cursor position. If omit- ted, the search continues to the end of the file or record. If this optional parameter has been specified then all occurrences of the string will be located over the specified num- ber of lines. If only a single occurrence is found then the cursor is placed at this point in the file. If multiple occurrences of the string are found then each one is listed below the editing screen.
D	The delimiter used to enclose the string to be located, which can be any character that does not form part of the string.
String	The string to locate.
Option	Can be one or more of the following:
F	Specifies that the search is to begin at the start of the file or record.
C	Performs a case insensitive search, else the search defaults to match the cases as provided in the string.

Examples

The L/report command helps you to search the record from the current position for the string "report" and halt at the first occurrence found with the cursor at the start.

The L9 FORM command helps you to search the next nine lines and locate all occurrences of the string "FORM".

The L/STARS/F command helps you to search from the first line of the file to find the first occurrence of the string "STARS". This line is placed at the top of the screen.

The L/acropolis/C command helps you to locate the first occurrence of the string "acropolis" with the letters in upper or lower case.

Replacing Strings

The editor allows the user to replace any occurrence of a string on any line with another from the Command line. This is in addi- tion to the overwrite mode.

The Command Syntax is as follows:

R{U}{nnn}dstring1dstring2{doption}

Option	Description
U	Replaces ALL occurrences of string1 with string2 on the current line only.
nnn	A numeric value for the number of lines, starting from the current one, over which to per- form the replace operation. If this optional parameter is specified and more than a single occurrence of string1 is found then all replacements are listed beneath the current edit- ing screen. d is the delimiter character used to separate the string values. It can be any character not in either of the strings.
string1	The string that is to be replaced.
string2	The replacement string, which can be shorter or longer than the original.

Options can be one or more of the following:

Option	Description
F	executes the replace Command from the first line of the file or record.
*	Replaces ALL occurrences of string1 with string2 on the current line.
Nnnnumeric	value for the number of times to repeat the replace operation on the current line.

Examples

Command	Description
R/ABC/DEF	Replaces the first occurrence (reading from the left) of ABC in the current line with DEF.
R9/*/!	Replace on the next 9 lines, the first occurrence on the line of "*" with a "!". The changed lines are displayed before moving on.
RU9/*/!	Replace any occurrence of "*" with "!" over nine lines (the current line and the next 8).
R999//*/F	Starting at the first line place a "*" character on every line; all modified lines are shown before returning to the original line.
R/^/AM/*	All occurrences of the "^" character on the line are replaced with "AM".
R9/*//	Removes (replaces with null) the first occurrence of "*" on the next nine lines.
R/x//10	Removes the first 10 "x" characters on the current line.

Copying, Pasting and Cutting Blocks of Text

The editor allows the user to copy or move blocks of text from one location to another within the current record being edited. It is also possible to copy from another UNIX file or jBASE record. You must mark or highlight a block before moving or copying it. Marked lines have their line numbers replaced by the characters ++++.

Marking Text

Text can be marked whilst in edit mode by using the appropriate keystroke Command (default <Ctrl G>) to mark the start and end of the block.

To highlight a block, move the cursor to the first line to and press <Ctrl G> (or the reassigned ones). Move the cursor to the last line to and again press the <Ctrl G> the start and end lines can be marked in any order.

To cancel the marked text, press <Ctrl G> again, which removes the markers.

Once the text is marked, position the cursor on the line to which the text is to be copied or moved before invoking the Com- mand line or key sequence required.

Copying Marked Text

Once marked, text can be copied by moving the cursor to the target line, entering Command mode, then using the copy Com- mands provided. Use the CB Command to copy text to the line before the current line. To copy to the line following the cur- rent line, use the CA Command.

The Syntax for both Commands is the same:

CB {nn}
CA{nn}

The optional nn parameter is a numeric value that gives the number of copies of the marked text to transfer. This is par- ticularly useful for the creation of a large quantity of repetitive text.

Moving Highlighted Text

Commands used to move highlighted text are MB to move to the line before the current one, and MA to move to the line fol- lowing the current one.

The Syntax for both Commands is the same:

MB

The text is deleted from the original position. It is not valid to move text within the highlighted block.

Merging Text From Another Record

It is possible to merge current data using jED from any file or record by using the Command:

MERGE

You should use the following command sequence. Position the cursor one line above the desired position of the merged text.

You can spawn a new editor session using the ! Command (Detailed later).

Sample command is !jed record or any other valid jed syntax.

This executes another editing session, placing the current session into the background.

Mark the block of text you wish to merge, and then from the Command line, issue the MERGE Command. The newly spawned editing session will be exited and control will be passed back to the original edit session. The merged text will then be copied into the record before the current line.

Deleting Marked Text

The Command DB deletes the marked text. The position of the cursor or portion of the record being displayed has no effect on the action.

jBASE BASIC Line Indentation

The jED editor has the capability of formatting lines of jBASE BASIC program code with appropriate indentation and so makes it more readable. The Commands available and their Syntax are described below.

Command	Description
BI {nn}	Formats the entire record as jBASE BASIC code by adding indentations in appropriate places. The value nn gives the number of space characters per indentation (maximum 20), and defaults to three if omitted.
BION{nn}	Turns on the automatic indentation for jBASE BASIC source code. Equivalent to using the B option with the jed Command. The value nn gives the number of space characters per indentation, and defaults to the value used with the B option, or the value used in the last BI Command.
BIONA {nn}	This Command is the same as the BION Command, except that an alternative form of indentation is used for the CASE state- ment. It is equivalent to using the A option with the jed Command when opening an editing session.
BIOF{F}	Turns off the automatic indentation for jBASE BASIC source code. It is equivalent to not using an indent option when opening an editing session.

Miscellaneous Commands

Command	Description
DE{nnn}	Deletes the number of lines specified by nnn, starting from the current cursor position. If nnn is omitted it defaults to a value of one line.
S?	Displays in bytes the size of the record being edited; It includes field delimiter marks in the body of the record.
!{Command}	Executes Command - Can be any valid UNIX or jBASE Command.
!!	Re-executes the Command specified in the most recent ! Command executed.
U{nn}	Scrolls the screen up by nn lines. If omitted, nn defaults to one line.
D{nn}	Scrolls the screen down by nn lines. If omitted, nn defaults to one line.
I {nn}	Inserts nn blank lines after the line holding the cursor. If omitted, nn defaults to one line.
nn	Positions the cursor on line nn, which is positioned at the top of the screen if the number of remaining lines still allows a screen’s worth of data to be displayed.
IN	Equivalent to the <F10> key.
IP	Equivalent to the <F9> key.
?	Displays the main help screen menu.

Changing jED Command Keys

The keystrokes used for jED editor Commands are configured using the UNIX terminfo terminal characteristic database.

Terminfo For Altering Jed Keystrokes

Terminfo is a UNIX database that describes the capabilities of terminals and their keyboards. Terminal capabilities are defined for how operations are performed, any padding requirements, and the initialization sequences required for each function.  The terminfo system is comprehensively documented within the standard UNIX documentation.

The terminfo data is used by utilities such as vi and jED to allow them to work on entirely different terminals without needing to set up or change parameters for each one.

The terminfo data can usually be found in the /usr/lib/terminfo directory. Terminfo entries consist of a number of fields delimited by a comma.

Embedded whitespace characters are ignored.

The first line of each description gives one or more names (separated by a | character) by which the terminal is known. Names should not contain space characters and at least the first 14 characters should be unique. The first name in the list is normally the shortest and is used when defining the terminal to UNIX, e.g. in setting the TERM environment variable.

The terminal name is followed by a list of capabilities that describe the functionality available with it. There are three types of terminfo definitions:

Booleans - Indicate what features of the terminfo system the particular terminal supports such as: margin; color; erase; tabs.

Numerics - Indicate magnitudes such as numbers of columns per line, numbers of lines in the display.

Strings

Sample strings are cursor, italics, carriage return, and keyboard definitions.

The jED editor is affected mainly by the definitions of keystrokes in the strings section. If the terminfo definition for your ter- minal does not define the keyboard sequences for the jED editor (F1 - F10 keys, Cursor keys, etc.), you may change the defin- ition yourself like this:

# TERM=myterm ; export TERM 
# infocmp >termdef.myterm
# vi termdef.myterm
........add the new 
keystrokes and write back 
the new record # tic 
termdef.myterm

NOTE: Many systems need to have super user permissions to execute the tic command.

Although terminfo is documented extensively, it can become quite complex. JBASE consultants would be pleased to help you to define terminfo entries for terminals. However, in most circumstances, the jkeys program will provide all functionality required.

The ED editor provided with jBASE is a limited version of the familiar (but now superseded) ED editors.

You are strongly advised to use the jED editor in preference to ED - we have not included any operating instructions for the ED editor.

If you are only familiar with the old style ED editor, continue to use it in your accustomed fashion - but please find a few minutes to review the operation of jED. You will find that the transition from ED to jED is very simple and you will be amply rewarded by adopting the much more efficient and friendly environment provided by jED.

The jsh command invokes jSHELL - the jBASE shell. It can be invoked as your login shell by using the normal system

administration software supplied with the platform, either via *.bat files (Windows) or *.profiles (UNIX).

jSHELL has been designed to ease migration from older systems, and to overcome some of the differences between various platform command line environments. The more primitive features seen on some older platforms (such as the “dot” com- mand stacker) have been replaced with easier to use and equivalents that are more functional.

The most noticeable difference between jSHELL and other command line shells, such as the UNIX Korn shell (ksh), is that com- mand line arguments such as “*” and “?” are not expanded by the shell but passed directly to the command that has been invoked. In same manner, quoted strings (such as “quoted string”) are passed directly to the command with quotes intact. This enables query language statements such as:

SSELECT file = “[SPROUT]” BY *A1 –

The above command helps you to be issued directly from jSHELL. If the same command were issued from the ksh Prompt, it would have to be issued as:

SSELECT file = \”[SPROUT]\” BY \*A1 –

The above command helps you to avoid the quotes being removed and the “*” being expanded by the Korn shell.

Beyond this convenient feature, jSHELL also offers many significant advantages over traditional shells and is easier to use. Some of the main features of the jsh are:

• Easily customised command line prompt
• If required commands can be passed to other shells Easy command recall.
• Easily identified special prompt when a select list is active
• Long program names are automatically translated to their new names on SVR3.2 systems. Run jCL programs directly from jBASE hashed files.
• Command type-ahead is supported.

Command Syntax

jsh - -c command -s shell -p Prompt

Using JSH

To use this tutorial you will need to be logged in to your UNIX system and positioned at the shell prompt. If your user account has not been configured to run jSHELL by default, execute it and complete the following:

exec jsh –

The default jSHELL Prompt should now appear:

jsh user cwd -->

User is your login name and cwd is your current working directory. For this exercise, we will assume that your login name was jBASE and that your current working directory is the home directory for jBASE. In this case, your Prompt will look like this:

jsh jBASE ~ -->

The tilde character (~) is a shorthand method of referring to your home directory. The shell expands this character to the full path name of your login home directory before executing commands. If you had changed to a sub-directory called “source” your Prompt would now look like this:

jsh jBASE ~ --> cd source 
jsh jBASE ~/source -->

You can change the primary and/or secondary default Prompts by using the following commands:

jsh jBASE ~ -->set jps1 newPrompt 
jsh jBASE ~ -->set jps2 newPrompt

newPrompt is a string defining the new Prompts. The string can contain terminal control characters such as a bell character by specifying special character sequences in the newPrompt string. The character sequences allowed are:

The shell operates in two distinct modes each having a separate function:

• Command mode is used to issue all commands to jsh itself
• Operating mode is used to issue all commands to the system.

There is only one command available in the current implementation of jsh - the / command. This character introduces a search string to jsh. The search string is compared against every command in your command history and if a match is found, the command is recalled as the current command, just as if you had typed it in again. Pressing the escape key on your key- board normally enters command mode.

If you include the $%m sequence when you configure the Prompt, the Prompt will change to indicate whether you are in the shell command mode. For example, if the Prompt has otherwise been left in its default state, the following sequence will locate the last cd command in your command history.

jsh ~ --><Esc>
jsh ~ (Cmnd) -->/cd 
jsh ~ -->cd source

Two other keystrokes within jsh allow you to recall up to 50 previous commands. They are:

• <Ctrl P> - Go to previous command
• <Ctrl N> - Go to next command

Using these two keystrokes, you can retrace your commands by stepping backwards or forwards one command at a time. If you need to change any part of a command line, it is very easy as the jsh supports command line editing by using a subset of the jED editor keys.

In particular, you can use the right and left arrow keys to move the cursor to any position in the current command string. The jsh is configured for editing in insert mode by default. This means that any characters you type will be inserted just before the current cursor position. Use the backspace key to delete the previous character and the <Delete> key to delete the character directly under the cursor. Try recalling a previous command and experiment with the editing keys.

The jsh can be placed into overwrite editing mode by pressing <Ctrl O>. In this mode, all typed characters will replace the char- acter under the cursor.

The following table lists all the editing commands:

jSH Emulation Modes

If you are already familiar with operating UNIX under other shells, jsh will allow you to work in the environment with which  you are most comfortable. You can switch between the various emulation modes in jsh by using the function keys:

Some terminals may not support these function keys. If your terminal does not support the F1, F2 and F3 function keys, the emulation mode can be modified by the jsh internal command, jshelltype. The command syn- tax is as follows.

jshelltype shell

jshell can be one of the following:

• jsh pre-processes Meta characters like the asterisk (*), as expected by legacy systems.
• sh native system shell On UNIX depends of SHELL on NT/Win95 CMD.exe.
• msh mixed shell. pre-processes Meta characters as a combination of jsh and sh.
• msh Mapped Sequences

Options specified on the command line will have the leading bracket escaped. For example:

CT File1 Record1 (X becomes CT File1 Record1 \(X

Any asterisk used as part of a record specification is escaped. For example:

CT File1 * becomes CT File1 \*

Quotes used in a selection criteria specification are escaped

LIST File1 WITH A1 = “XYZ]” becomes LIST File1 WITH \*A1 = \”XYZ]\”

Other Meta characters are untouched so that pipes, etc. can be invoked:

The following describes the new jBASE 4.1 mechanisms for controlling certain profiling tools:

This profiling mechanism allows versatility in where the output logs are stored and hot spot profiling in C and jBASE BASIC code. The functionality can be applied dynamically to currently executing applications. It may be used to trace branch func- tions (INPUT, EXECUTE, CALL and RETURN instruction), to trace memory allocations (subsequently using jfatty to look for memory leaks), and general profiling activity.

Invoking Profiling

The basic way to invoke the functionality is through the JDIAG environment variable. Here follows a quick example as run from the Unix shell.

% export JDIAG=branch=on:output=filename.out
% MYPROGRAM

In this example, two commands are specified delimited by a colon (There can be any number of commands delimited). In this example, 'branch=on' means to turn on the tracing of branch instructions and write the trace output to the file 'file- name.out'.

It is also possible to run these commands against running programs. The same example as above but applied dynamically to port 23 would be:

% jprof 23 branch=on:output=filename.out

On most platforms the effect should be immediate but due to a bug in RedHat Linux and the limitation of Windows, the effect might take up to 60 seconds before it starts working.

You can obtain a short help screen as a quick reference reminder by the following:

% JDIAG=help WHO
JDIAG=option{:option{:option ...}} 
option can be one of ...
profile={off|short|long|user|jcover}
 output={stdout|stderr|tmp|filename{,refresh_mins} 
memory={off|on|verify}
branch={off|on} 
help
1 greg

Although the JDIAG environment variable name is case sensitive, the commands contained in its value are not e.g. profile=ON is the same as ProFILE=On

A list of all the available commands follow, below which is listed an example of dynamic usage to test for memory leaks in jBASE.

As an example, let us assume that a program is running and gradually consuming more and more memory. You want to invest- igate where the memory is being leaked.

Procedure:

1. Find the port number from where it is running.

   jsh --> WHERE
   Port	Device	Account	PID	Command
   
   0	6	greg	10941	MYPROG
   
   *1	8	greg	10949	WHERE
   

2. Start dynamically tracing the memory allocations.

   jsh->jprof 0 
   memory=on:output=MYPROG.output 
   Message sent successfully to port 0

   The above command could take up to 60 seconds to invoke. Run MYPROG for 3-4 minutes to accumulate some statistics. The MYPROG.file output will increase in size as the trace begins.

3. Stop the trace.

   jsh->jprof 0 memory=off

   Message sent successfully to port 0.

   Wait 60 seconds to take effect

4. Run the jfatty command against file created file jsh->jfatty MYPROG.output

   Remaining: 00005785: 100 bytes at 0x08115108 at test9C.c,6(test9.b,6)
   Remaining: 00005795: 100 bytes at 0x08115148 at test9C.c,6(test9.b,6) 
   Remaining: 00005805: 100 bytes at 0x08115188 at test9C.c,6(test9.b,6) 

   You can see where the memory is leaking.

This is an example of hot-spot user profiling, the idea of which is that you place some function calls in either your jBASE BASIC code or your C code and let the profiling tools measure the CPU used between these calls.

You can call function JBASEUserProfile, where n is an integer between 0 and 10. A value of 0 means profiling ticks are discarded. A value of 1 through 10 means each CPU clock tick is stored in a 10 element array so we can ascertain how many CPU ticks arrived for that period. Consider the following example jBASE BASIC code ...

INCLUDE JBASIC.h
OPEN "FB1" TO DSCB ELSE DEBUG 
S1 = SYSTEM(12)
FOR I = 1 TO 100000
JBASEUserProfile(1)
READU rec FROM DSCB,"x" ELSE NULL
JBASEUserProfile(2) 
RELEASE
JBASEUserProfile(0) 
NEXT I
S2 = SYSTEM(12)
PRINT "ELAPSED = ":(S2-S1)/1000

Compile normally with BASIC and CATALOG, or use the jcompile command. When it is run set the JDIAG variable as follows (assuming Unix):

JDIAG=profile=user PROGRAMNAME 
ELAPSED = 3.66
PROFILE:User Profile 1 , 30 ticks 83.33 %
PROFILE:User Profile 2 , 5 ticks 13.89 %

The output shows you that 83.33% of the CPU time is spent between the JBASEUserProfile(1) and JBASEUserProfile(2) function calls, and that 18.39% of the CPU time is spent between the JBASEUserProfile(2) and the JBASEUserProfile(0) function calls, thus proving that the RELEASE statement is consuming large amounts of CPU.

The function calls can equally be used in C code.

To ensure more accuracy, the program should be run for at least a few seconds in order to accumulate as many CPU ticks as possible. Anything less than a second will increase the margin of error

In particular, developers might like to notice the hot-spot user defined profiling which allows C code to be modified and easy verification of where CPU time is being spent within a function.

You can add checkpoints into your jBASE BASIC or C code. At the end of the test program, it tells you how many ticks and what percentage of CPU was spent between these checkpoints something very valuable when trying to determine, for example, " There is a big JediOpen function that takes a huge amount of CPU, but where inside the JediOpen() does it take the most?" Using the profiler, in this example, you can tell it is not JediOpen, but rather JediClose.

The char utility displays a character conversion table, providing decimal, hexadecimal and octal representations of the ANSI character

Encrypt

The Encrypt utility provides simple encryption/decryption of files using a user specified key.

Encrypt –k key File > EncryptedFile Encrypt –d –k key EncryptedFile > File

ERRMSG

The errmsg tool displays the platform related error message text when used with an error message number.

errmsg 13

HAD

The had utility displays any file in a hexadecimal format. A hexadecimal offset can be specified from which to start the display.

had –offset
 

The jcomp utility provides a mechanism to compare two Hash files or directories.

jcomp -Options FileName (Options

The jcover programs provide a mechanism to generate statistical information on the coverage of an application during a test run. It can be used to calculate the percentage of code that was executed, percentage not executed, what sources were not used in files and so on.

The usage of jcover depends upon much of the existing profiling tools. Therefore this document should be in conjunction with existing documentation available for profiling tools. The jcover utility can be used to great advantage with the jkeyauto utility. Using both of them together it is possible to build up highly automated scripts that will automatically test new releases of software and ensure that a known percentage of your entire application gets executed as well as ensuring that all paths through your application are executed and tested.

Using jcover involves the three key steps:

1. Run the application and record the information.
2. Collate the recorded information.
3. Report the jcover information.

Recording the Information

This phase involves running your application in a certain manner and allowing jBASE to record the statistics about the lines of code that were executed whilst the application was run. There are two ways of telling jBASE to record this information

Use the -JC option to the program. For example:

% MYAPPLICATION -JC COMMANDARG ETC ETC

In the above example, when the application ‘MYAPPLICATION’ terminates a file called ‘jprof’ will be written to the current working directory which contains all the profiling information for this application.

Set the JBCPROFILE environment variable. For example:

% JBCPROFILE=3
% export JBCPROFILE
% MYAPPLICATION COMMANDARG ETC ETC

In this second invocation jBASE will create a file in the format jprof_mmmmm_nnn, one file for each program that is executed where mmmm is the process id and nnn is the PERFORM/EXECUTE level. In this way you can generate multiple copies of pro- filing information which will automatically include all executed procs and programs. Each file will be stored in the current work- ing directory as of when the application was loaded.

Remember to unset the JBCPROFILE environment variable when you are finished otherwise you will find a very large number of the profiling files cluttering up your directory. It might be useful to run your application through a jkeyauto script. In this manner you can accurately repeat a test case through your application and repeatedly use the jcover tools to ensure as much as possible percentage of your application is executed.

At the end of the recording phase you will have one or more profiling files. These profiling files can then be used in the second stage of jcover profiling. Remember that as UNIX files they can be easily moved and renamed. For example you could run a number of test cases using the -JC option and each time rename the generated ‘jprof’ file to something else, such as jprof_a, jprof_b and so on. Then in the next phase you can supply multiply file name to be manipulated.

Collating the Information

Once you have recorded one or more profiling files as shown in the previous section, you are ready to collate all the inform- ation. This is the function of the jcover program. The jcover program is run as shown below.

jcover {-fFilename {-fFileName ...}} {-oOutputFile} {-ddelim} {-e} {-h|?} {-u} {-x} {jprof
{jprof ...}}

Command	Description
-ddelim	Change the delimiter in output records from '*'
-e	Write out verbose details of all lines of code executed
-fName	Name of file to extract source from
-h | -?	Help screen output
-oOutputFile	Output file name
-u	Write out details of unused source code in the supplied input files
-x	Write out verbose details of all lines of code NOT executed
prof	Profile name

By default it will take the input Unix file ‘jprof’, extract some statistical information about it, and write it back out to a newly- created file called jcover_nnn where nnn is your port number. It will also create suitable dictionary items.

The default information stored will be a summary of the percentages of all source items executed. You can use the -e, -u and -x options to create additional statistical information.

The -oOutputFile option allows you to specify an alternative output file to jcover_nnn (where nnn is your port number). If this option is not used, we default to jcover_nnn and create the file if necessary. Whatever file name you choose, both the DICT and the DATA section will be cleared and suitable DICT items written away for later usage.

Once this phase is complete, you will have an output file ready for use with the final phase.

Reporting the Information

This third phase requires the use of the file created in the second phase. For convenience we will call it jcover_23, but in fact it can be any file name used as the output of the jcover program.

There are basically 3 ways of using the information created in file jcover_23 Using jQL with the default DICT macros

To show a summary of percentages of coverages use:

LIST jcover_23 STATS

To show a summary of general information, such as times, dates, files used and so on. Use:

LIST jcover_23 GENINFO

If the -e option was used; you can display a list of all the lines of source code executed with:

LIST jcover_23 EXEC

If the -x option was used, you can display a list of all the lines of source code not executed with:

LIST jcover_23 NOTEXEC

If the -u option was used; you can display a list of all source items that didn’t have any code executed during the application execution

LIST jcover_23 NOTUSED

Using jQL with your own command line

This involves simply using a jQL statement with something other than the supplied macros. For example you could:

LIST jcover_23 WITH S.ITEMID EQ “MAR]” S.ITEMID

Using a jBASE BASIC program

Refer to the layout of the output file and the dictionaries to write your own jBC program to further report on the information available.

Caveats

At present, jcover is available on UNIX platforms only.

When you are running profiling for jcover your application will run at 2-20 times slower, so this type of profiling isn't suitable for performance measurement at all. If you use this against batch jobs, you may find this performance degradation is too much. You will additionally use around 0.5 to 5 Mb extra memory per application while running this jcover profiling.

The profiling can only detect whole lines of code executed, or not executed. In the following line of code: READ record FROM filevar, itemid ELSE record = "xxx" ; GOSUB subname

The jcover profiling will only record the fact that the READ statement was executed -- it cannot be used to determine if the code following the ELSE was executed or not. If you want to test for this, then the application will need to be changed so that the ELSE code is on a separate line.

Layout of the JCOVER Output File

The records written to the jcover output file have different formats depending upon what they are showing. The formats are identified by a prefix on the record key. Note that in the following examples a * character is used to delimit certain fields. This is the default delimiter character. Remember if the -d option was used to jcover, then this delimiter will be different.

E*itemid*filename*lineno

These items show which lines of code have been executed. They are only generated if the -e option is used on jcover. There will be one item for each different line of code executed during the application execution.

Item Id	The 4 fields show the record type (E), the item id, the source file name and the line number that were executed.
<1>	The actual line of source code taken from the source 'itemid' in file 'filename'

F*itemid*filename

Used internally to keep track of what items in what files have been accessed as part of the jcover execution Item Id. The 3 fields show the record type (F) , the item id and the source file name that were executed.

G*descr

These are general information items.

Item Id The 2 fields show the record type (G) and then a unique identifier, not really of any use.

<1>	Textual description of the general information
<2>	The actual general information.
<3>	The order in which the fields are to be displayed using 'LIST filename GENINFO'

S*itemid

These items show what item has been executed and are used to build up a list of statistical analysis of the executed item. Item Id The 2 fields show the record type (S) and the item id that were executed.

<1>	Item ID of the source executed
<2>	File name containing the executed source
<3>	Number of lines of source. This attribute will be the sum of attributes <4> and <5>.
<4>	Total number of comment lines in the source code.
<5>	Number of lines of code in the source that are executable. Blank lines and comment lines are not included in this count.
<6>	Number of lines of code executed by the application.
<7>	Number of lines of code executed by the application as a percentage of the total number of executable lines.
<8>	Number of lines of code not executed by the application
<9>	Number of lines of code not executed by the application as a percentage of the total number of executable lines.
<10>-<20>	Reserved for future use
<21>	Details of line 1 of the source code; a character 'C' shows line 1 is a comment line. A numeric value shows line 1 is execut- able code and has been executed this many times. A blank shows line 1 is executable code, but was not executed during the application execution.
<22>	Similar to <21> but for line 2 of the source code.
<23>	Similar to <21> but for line 3 of the source code. etc. for all lines in the source code.

U*itemid*filename

For every file name specified with the -ffilename parameter of jcover, there will be one of these items for every item in every file that was not used during the application execution. These items are only generated if the -u option is used on jcover. Item id's that show a non-source item will not be included and these are items that are appended with .o , .a , .so , .sl , .obj , .d or .d or items that begin with ! , $ or *.

Item Id The 3 fields show the record type (U) , the item id and the source file name that were not executed in any way what- soever.

X*itemid*filename*lineno

These items show which lines of code have not been executed. They are only generated if the -x option is used on jcover. There will be one item for each different line of code not executed during the application execution. Only the source files that have one or more line of code executed will be included here. Any source items not executed at all will not have a X* item but will be included in the U* items (assuming the -u option was used)

Item Id The 4 fields show the record type (X) , the item id, the source file name and the line number that were not executed, but the source did have one or more lines of code executed.

<1> - The actual line of source code taken from the source 'itemid' in file 'filename'

The jfb command is used to produce source code listings in a standard format for listing to the terminal or printer. The format of the command is as follows:

jfb -Options FileName Itemlist (Options

For example, to list the file batch.b in sub-directory source to the printer, indenting by four spaces per level and starting non- labeled code at 8 spaces from the left margin:

jfb -L4,2 -V source/batch.b

The Windows platform does not provide a command line search program like the Unix find command. This is a limited form of the Unix find command for the Windows platform.

jfind C:\MYDIR -name ".*FRED.*" -print

You can use the jgrep command to search for strings in jBASE files or directories.

jgrep -Options SearchString FileName (Options

Many commands are built-in to the cmd.exe shell and so cannot be executed like a like other executables. The following com mands have been provided to use in either jsh or cmd shell.

Use the jmsgbox utility to display a message box on Windows platforms. Use it with a console process and the buttons are not configurable.

Command Syntax

jmsgbox text

Example

PERFORM "jmsgbox    Test     Box"                  SETTING Result
IF       Result<1,1>=0        THEN    CRT      "Ok       was clicked"
IF Result<1,1>=1 THEN 

The jshow command can be used to find jBASE files or programs.

Command Syntax

jshow -Options Name {Name ...}

Where Options can be one of the following:

Syntax Elements

Name - Name of file, subroutine, program or dll/shared object

The jBASE profiling tools jprof enables developers to analyze applications to determine potential bottlenecks or trouble spots within the application code.

By default, no profiling is done in the program. Programs do not have to be compiled in any special manner to enable profiling for that program. All that is required is that the programs were not compiled with optimization, as this discards the debug information which is required for profiling.

The mechanism works by receiving a signal at every clock tick and keeping note of where the program was when the signal arrived. Thus, for the profiling to be accurate, the application must be run for a relatively long time. It will not show par- ticularly good results if, for example, a program executes in less than a second. Several minutes or longer is preferred.

Currently profiling is available only with jBASE on Unix (including Linux) platforms.

Enabling profiling

Profiling can either be enabled using the -JP option, which only profiles the root process, or via the JBCPROFILE environment variable, which profiles the root process as well as all EXECUTEd processes.

MAINPROG –JP

This command generates a profiling file called jprof_n in the current directory where n is the port number. Only MAINPROG and any CALLed subroutines are profiled, any EXECUTEd programs will not be profiled. However, if the CPU time spent execut- ing the actual EXECUTE statement is significant, that line will be included in the profiling statistics. Profiling terminates when the application stops or chains to another program.

JBCPROFILE=1 
MAINPROG

This command generates a different profiling file for each process executed in the form jprof_pid_n, where pid is the process id and n is an incrementing number starting at 0.

The profiling file generated will only contain information about user CPU time. The time spent in system calls, file I/O, and lines which do not accumulate more than a clock tick are not included in the profiling statistics.

Profile Reporting

You should use the jprof command to provide profile analysis of the jprof files generated by a program executed with the JP option.

jprof    - kfilename        {jprof    {_    nnn}}
jprof    - a        {jprof    {_    nnn}}
jprof    - o {-    v}    {jprof    {_    nnn}}
jprof    - s        {jprof    {_    nnn}}
jprof {-n{-u}} {-i} {-fFilename}} {jprof{_nnn}}

Where,

• -a - Display all ancillary information.
• -fName - Name of file to extract source from.
• -I - Sort by increasing ticks, rather than decreasing tick.
• -kKeyFile - Name of file to store keyboard INPUT, used by jkeyauto.
• -n - Subtotaled and sorted by source name
• -o - Display shared object usage
• -s - Display list of subroutines called
• -u - Sorted by CPU utilization
• -v - Verbose mode

jprof{_nnn} Profile name (default "jprof")

Example of Profiling

Imagine the source "test1.b" below has been edited into file BP, where BP is a directory. Notice the INCLUDE of another source file "test2.b".

OPEN "fb1" TO DSCB ELSE STOP 201,"fb1"
PRINT  "Phase  1 -  - start" S1  =   SYSTEM  (9)
FOR Id = 1 TO 100
Rec   =    " FOR I =  1 TO 100
Line   =    " FOR J =  1 TO 20
Line  := CHAR (SEQ ("A")+RND (26)) NEXT     J
Rec  = Line NEXT   I
WRITE  Rec ON DSCB,Id NEXT    Id
PRINT "Phase 1 - - end, CPU = ":SYSTEM (9)- S1
INCLUDE test2.b
PRINT C1:" records in file fb1" PRINT "End"

The program can be created normally with the following command:

BASIC BP test1.b CATALOG BP test1.b

By default, when the program is run, no profiling will take place. Run the program with the -JP switch to create a file "jprof":

test1 -JP

You can now examine the profile file with the "jprof" command, using the -f option to generate optional source code listings from the file BP.

jprof -f BP jprof

PROFILE REPORT

Profile of program test1 from profile jprof Page 1

Source        Line        Ticks    % Source
test2.b    8    166        32.93 READ Rec    FROM DSCB,Key EL
test1.b    9    160        31.74 Line :=    CHAR (SEQ ("A")+RND (
test1.b        11    128    25.39    Rec = Line
test2.b    7    28        5.55 WHILE    READNEXT Key DO
test1.b        10        9 1.78    NEXT J
test2.b        9        5    0.99 C1++
test1.b    13    3        0.59 WRITE    Rec ON DSCB,Id
test2.b        5    2    0.39    SELECT DSCB
test1.b        7    2    0.39    Line = "
test2.b 10    1    0.19    REPEAT        

The -i option sorts the output with incrementing Ticks counts. The -n option would additionally sort it by file name, so the "test1.b" entries will be displayed separately to the "test2.b" entries.

You can use the jfb command to produce source code listings in a standard format for listing to the terminal or printer. The format of the command is as follows:

jfb -Options FileName Itemlist (Options

For example, to list the file batch.b in sub-directory source to the printer, indenting by four spaces per level and starting non- labeled code at 8 spaces from the left margin:

jfb -L4,2 -V source/batch.b

The Windows platform does not provide a command line search program like the UNIX find command. This is a limited form of the UNIX find command for the Windows platform.

jfind C:\MYDIR -name ".*FRED.*" -print

You can use the jshow command to find jBASE files or programs.

Command Syntax

jshow -Options Name {Name ...} Where Options can be:

• -a - Display subroutine names in dll/shared object
• -c - Display compile time and source file
• -f - File name only search
• -h - Display this help screen
• -p - Program name only search
• -s - Subroutine name only search
• -v - Verbose mode

Syntax Elements

Name - Name of file, subroutine, program or dll/shared object

The jstart utility was originally provided to circumvent a problem with the Win95 CreateProcess API. The API fails to set the standard input handle to the new console if the standard input handle of the parent process is not a console device, i.e. a DATA statement has been used. However, jstart can be used to start up a process in the same window, in a new window or in the background. Not available on UNIX platforms.

jstart -Options Command

EXAMPLE

EXECUTE "jstart jsh -s jsh -"

This will start up the jSHELL in a new window, with a different port number. This works for both Win9x as well as NT.

jtic -Options DescriptionFile

Where Options can be one of the following:

By default, the program jtic will take a source description file called terminfo.src and assume it contains standard terminfo names. The output will be to a file called /usr/lib/terminfo/x/xyz, where x/xyz depends upon the terminal name as contained in the description file.

You can use any alternative description file to terminfo.src by specifying the description file name on the command line. You can specify an alternative output directory to /usr/lib/terminfo by amending the TERMINFO environment variable. However, when you run a BASIC program that accesses these definitions in an alternative directory, the TERMINFO variable needs to match that when the definition was compiled; by default the jtic program assumes the description file contains standard ter- minfo definitions: e.g.

cuu1=\E[1A, cols#80,

If you want to create a binary with the extended capabilities, use the -x option. Remember when running jtic you will prob- ably require root privileges to write to the /usr/lib/terminfo directory.

jtic Description File

To run jtic you need a description file to describe the capabilities. This file is very similar to that required by the UNIX tic com- mand.

Comment lines start with # as the first character of each line. Blank lines are also counted as comment lines.

Terminal names - This tells jtic the names of the terminal definition. It is a list of names delimited by | characters. The final field is treated as a comment. Following the terminal names will be the binary definitions, integer values and string values for those terminals. When another set of terminal names are defined, the current definitions are written to file and the current defin- itions nulled. E.g.

vt100|vt100am|prism-ans|Simple vt100 support

• Binary definition - These are just the name of the definition (For example, on).
• Integer value - The name of the definition followed by # then the value. E.g. cols#132
• String value - The name of the definition followed by = followed by the string value. jtic supports all the functionality of tic, such as defining characters 1 to 26 using ^A through ^Z, specials such as \E for escape, \s for space, \t for tab. It also supports the if/else/endif structure and logical/arithmetic operations supported by tic. E.g. if_prtr_letter=\E[02l,
• use=name - Resets the definition to that of a previously defined name in the definition; you can use this to create a terminfo definition for one or more names, and then base subsequent definitions on that. An example of using this can be found in the source $JBCRELEASEDIR/src/prism. E.g. use=ansi

The comments and terminal names must all start at column 1. The binaries, integers, strings and use=name must all have a lead- ing tab character. There can be more than one binary, integer or string on a line, and each definition should be delimited by a comma. When invoked with the -x option a terminfo binary file with the suffix _jbase is generated for the extended cap- abilities.

Extended Definitions

Access	Long name Description
@(-53)	if_slave_only Send data to slave printer only. No VDT display
@(-54)	if_crt_type Returns VDT terminal type
@(-55)	if_crt_graphics Returns a 1 if Regis graphics available
@(-59)	if_crt_cuprot Clear all unprotected fields
@(-27)	if_crt_132 Switch VDT to 132 column mode
@(-28)	if_crt_80 Switch VDT to 80 column mode
@(-29)	if_crt_dwide Display double wide characters
@(-30)	if_crt_swide Display single wide characters
@(-32)	if_crt_sron Set-up scrolling region. (Enter ? for help)
@(-33)	if_crt_sroff Reset scrolling region to normal
@(-47)	if_crt_udhdw Top half of double height line
@(-48)	if_crt_bdhdw Bottom half of double height line
@(-60)	if_prtr_executive Change paper size to executive
@(-61)	if_prtr_a4 Change paper size to A4
@(-62)	if_prtr_monarch Envelope type "Monarch"
@(-63)	if_prtr_comm10 Envelope type "Commercial 10"
@(-64)	if_prtr_interntldl Envelope type "International DL"
@(-65)	if_prtr_reset Reset printer defaults
@(-66)	if_prtr_envfeed Envelope feeder
@(-70)	if_prtr_letter Change paper size to letter
@(-71)	if_prtr_legal Change paper size to legal
@(-72)	if_prtr_chgcpy Change number of copies to print on Laser
@(-73)	if_prtr_cpwo1 Compressed print for Service WO form 1
@(-74)	if_prtr_spcol Start printing at specified column
@(-75)	if_prtr_sprow Start printing at specified row
@(-76)	if_prtr_utray Use Upper Tray
@(-77)	if_prtr_ltray Use Lower Tray
@(-78)	if_prtr_portrt Portrait orientation
@(-79)	if_prtr_land Landscape orientation
@(-80)	if_prtr_simplx Simplex binding
@(-81)	if_prtr_duplxl Duplex, long edge binding
@(-82)	if_prtr_duplxs Duplex, short edge binding
@(-83)	if_prtr_macro Call MACRO
@(-84)	if_prtr_setdef Set default ( Font size, HMI, VMI )
@(-85)	if_prtr_lpi2 2 lines per inch
@(-86)	if_prtr_lpi3 3 lines per inch
@(-87)	if_prtr_lpi4 4 lines per inch
@(-88)	if_prtr_lpi6 6 lines per inch
@(-89)	if_prtr_lpi8 8 lines per inch
@(-90)	if_prtr_lpi12 12 lines per inch
@(-91)	if_prtr_dwide Double wide mode
@(-92)	if_prtr_swide Single wide mode
@(-93)	if_prtr_96 96 column mode
@(-94)	if_prtr_pld 1/2 line down
@(-95)	if_prtr_plu 1/2 line up
@(-96)	if_prtr_suon Superscript mode
@(-97)	if_prtr_sbon Subscript mode
@(-98)	if_prtr_ssoff Superscript and subscript off
@(-99)	if_prtr_40 Double wide for 80 column mode (5 pitch)
@(-100)	if_prtr_48 Double wide for 96 column mode (6 pitch)
@(-101)	if_prtr_ff Top of form
@(-102)	if_prtr_80 80 column mode (10 pitch)
@(-103)	if_prtr_132 132 column mode (16 pitch)
@(-104)	if_prtr_bold Bold
@(-105)	if_prtr_ul Underline
@(-106)	if_prtr_norm Turn off bold and underline
@(-107)	if_prtr_hmi Set horizontal motion index to U-1
@(-108)	if_prtr_vmi Set vertical motion index to U-1
@(-109)	if_prtr_pson Proportional spacing on
@(-110)	if_prtr_psoff Proportional spacing off
@(-111)	if_prtr_1key Linefeed and backspace
@(-112)	if_prtr_2key Linefeed
@(-113)	if_prtr_3key Linefeed and space
@(-114)	if_prtr_4key Backspace
@(-115)	if_prtr_6key Space
@(-116)	if_prtr_7key Negative linefeed and backspace
@(-117)	if_prtr_8key Negative linefeed
@(-118)	if_prtr_9key Negative linefeed and space
@(-119)	if_prtr_cvd Coarse vertical distance ( 1 line = 1 inch)
@(-120)	if_prtr_mvd Medium vertical distance ( 1 line = 1/6 inch )
@(-121)	if_prtr_type Returns slave/printer type
@(-122)	if_prtr_fvd Fine vertical distance ( 1 line = 1/48 inch)
@(-123)	if_prtr_chd Coarse horizontal distance ( 1 space = 1 inch )
@(-124)	if_prtr_mhd Medium horizontal distance ( 1 space = 1/12 inch )
@(-125)	if_prtr_fhd Fine horizontal distance ( 1 space = 1/120 inch )
@(-126)	if_prtr_status Retrieve slave/printer device status
@(-220)	IF_PRTR_DUL_ON
@(-221)	IF_PRTR_DUL_OFF
@(-223)	IF_PRTR_STRIKE_ON
@(-224)	IF_PRTR_STRIKE_OFF
@(-225)	IF_PRTR_HARD_HYPHEN
@(-226)	IF_PRTR_HARD_BLANK
@(-230)	IF_PRTR_FONT_0
@(-231)	IF_PRTR_FONT_1
@(-232)	IF_PRTR_FONT_2
@(-233)	IF_PRTR_FONT_3
@(-234)	IF_PRTR_FONT_4
@(-235)	IF_PRTR_FONT_5
@(-236)	IF_PRTR_FONT_6
@(-237)	IF_PRTR_FONT_7
@(-238)	IF_PRTR_FONT_8
@(-239)	IF_PRTR_FONT_9

jBASE Extended terminfo Definitions Generation

To generate a terminfo extension file, using the jtic -x comamnd, setting any of the following terminfo strings references in lowercase. Ensure that the prt_video emulation flag is configured to true, for the required emulation.

Generate jBASE extension tic file   
ED   .         Mytic
TOP            
.            
001 # My   jBASE ext   ension   file   for vt220
002            vt220|VT220
003   if_   crt_      132=My132String,
004
.FI   if_   crt_      80=My80String,
Generate jBASE terminfo extension file.            
jtic -x Mytic            

This generates a jBASE extension entry for vt220 in the terminfo database. For example, /usr/lib/terminfo/v/vt220_ext_jbase Test terminfo produces correct result.

ED . TESTMYTIC.b
    TOP
    .
001    CRT    "Output    my    80    column    terminfo    command    ":@ (- 28)
002    CRT    "Output    my    132    column    terminfo    command    ":@ (- 27)
.FI
export TERM=vt220
export JBCEMULATE=ROS
jbc - Jo TESTMYTIC.b TESTMYTIC > MyFile

The keys utility displays the keyboard input. This program does not exist on the NT/Win95 platform but could be coded as fol- lows:

NOEXIT
LOOP                =            1
ECHO                            OFF
CRT        "Hit        a    key        :":
IN    ch    FOR    100    ELSE    NOEXIT    =    0
ECHO                            ON
WHILE                NOEXIT            DO
IF ch <> 127            THEN
CRT "." : CHAR (9): "0x0" :    ch        "MCDX"
END            ELSE
CRT CHAR (ch) : CHAR (9): "0x0"    :ch        "MCDX"
END
REPEAT

The libutils shared library is provided to enable users to code terminal independent code to handle keyboard input.

EXAMPLE

Include key definitions INCLUDE jCmdKeys.h
Initialize command key  strings CALL   JBASECommandInit
TimeOut = 150 ;* Set timeout value deciseconds ECHO      OFF
LOOP
Get Next Command Value CALL JBASECommandNext(RetNo, RetString, TimeOut)
RetNo should match numbers in include/header file
BEGIN CASE
CASE  RetNo =  cmd_ cursor_  up CRT   "CURSOR   UP"
CASE  RetNo =  cmd_ cursor_  down CRT   "CURSOR   DOWN"
CASE  RetNo =  cmd_ cursor_  left CRT   "CURSOR   LEFT"
CASE  RetNo =  cmd_ cursor_  right CRT   "CURSOR   RIGHT"
CASE RetNo = cmd_ alpha_ numeric CRT    "ALPHANUMERIC"
CASE RetNo = cmd_ timeout
END CASE
CRT "TIMEOUT"
BREAK
*   Output   the   actual   string   returned
   CRT   "RetString      :":OCONV   (RetString,"MCP.")
REPEAT               

The MW42 tool is designed to assist with diagnosing issues which relate to a gradual build up of resource usage over a period of time. It records information from a series of both jBASE and operating system tools, collecting data on such things as system configuration, process status, memory usage, disk space, state of jBASE daemons etc.

Command Syntax

Mw42 {-a –f –n –pPot –pp1-p2 –t -uUser –nn1 –nn2 > output.file}

Syntax Elements

Example

mw42 -f -a -t -n 30 1440 > mw42.out

This command produces a full listing for the application account with timestamps on every line without asking for input. Sample every 30 seconds for 1440 samples and pipe the output to mw42.out.

The @USERSTATS allows a program to retrieve miscellaneous information about itself. For example if a program wants to find out how many database I/O's it performed, the following command can be executed.

info1 = @USERSTATS
read1 = info1<19>
EXECUTE 'COUNT fb1 WITH *A1 EQ "x"'
info2 = @USERSTATS
read2 = info2<19>
PRINT "The COUNT command took ":(read2-read1):" READ's from the database"
PRINT "The COUNT command took ":(read2-read1):" READ's from the database"

The following definitions have been added to JBC.h file which defines the layout of data returned either through the @USERSTATS variable or by opening file SYSTEM(1027) and reading the items in like that.

EQUATE USER_PROC_PORT_NUMBER TO 1;* The port number
EQUATE USER_PROC_NUM_PROGRAMS TO 2;* Number of programs running in this port
EQUATE USER_PROC_START_TIME TO 3;* Time user started in UTC format
EQUATE  USER_PROC_PID TO 4  ;* Process ID
EQUATE USER_PROC_ACCOUNT TO 5;* Name of the account
EQUATE  USER_PROC_USER TO 6 ;* Name of the user
EQUATE USER_PROC_TERMINAL_JBASE TO 7;* Name of terminal according to jBASE
EQUATE USER_PROC_TERMINAL_OS TO 8;* Name of terminal as seen by OS
EQUATE USER_PROC_DATABASE TO 9;* Name of database connected to
EQUATE USER_PROC_TTY TO 10;* Name of TTY device
EQUATE USER_PROC_LANGUAGE TO 11;* Language
EQUATE USER_PROC_LISTENING_TIME TO 12;* Time in UTC the listening thread last worked
EQUATE USER_PROC_MEM_FREE TO 13;* Amount of memory in heap space free chain
EQUATE USER_PROC_MEM_USED TO 14;* Amount of heap space memory in use
EQUATE USER_PROC_THREAD_TYPE_INT TO 15;* Thread type expressed as an integer
EQUATE USER_PROC_THREAD_TYPE_TXT TO 16;* Thread type expressed as a text string
EQUATE  USER_PROC_LICENSE TO 17;* License counters
EQUATE USER_PROC_STATS_OPEN TO 18;* Number of OPEN's performed.
EQUATE USER_PROC_STATS_READ TO 19;* Number of READ's performed.
EQUATE USER_PROC_STATS_WRITE TO 20;* Number of WRITE's performed.
EQUATE USER_PROC_STATS_DELETE TO 21;* Number of DELETE's performed.
EQUATE USER_PROC_STATS_CLEARFILE TO 22;* Number of CLEARFILE's performed.
EQUATE USER_PROC_STATS_PERFORM TO 23;* Number of PERFORM's / EXECUTE's performed.
EQUATE USER_PROC_STATS_INPUT TO 24;* Number of INPUT's performed.
EQUATE  USER_PROC_UNUSED_1 TO 25;* Unused
EQUATE USER_PROC_OPEN_FILES_VIRTUAL TO 26;* Number of files application thinks open
EQUATE USER_PROC_OPEN_FILES_REAL TO 27 ;* Number of files really open by OS
EQUATE USER_PROC_USER_ROOT TO 28;* Application data set by @USER.ROOT
EQUATE USER_PROC_PROCESS_TXT TO 29;* Text string to identify process
EQUATE USER_PROC_PROGRAM TO 41;* Program name and command line arguments
EQUATE USER_PROC_LINE_NUMBER TO 42;* Line number currently being executed.
EQUATE USER_PROC_SOURCE_NAME TO 43;* Name of source currently being executed.
EQUATE  USER_PROC_UNUSED_2 TO 44;* Unused
EQUATE  USER_PROC_UNUSED_3 TO 45;* Unused
EQUATE  USER_PROC_STATUS_TXT TO 46;* Status of program as a readable text
EQUATE  USER_PROC_STATUS_INT TO 47;* Status of program as an integer
EQUATE  USER_PROC_CPU_USR TO 48;* User CPU time
EQUATE  USER_PROC_CPU_SYS TO 49;* System CPU time
EQUATE  USER_PROC_CPU_USR_CHILD TO 50;* User CPU time used by child processes
EQUATE  USER_PROC_CPU_SYS_CHILD TO 51;* System CPU time used by child processes
EQUATE  USER_PROC_USER_THREAD TO 52;* Application data set by @USER.THREAD

There are two sides to JIMI:

Event Driven Metrics (EDM), which are produced in real time

In jBASE release 4.1 onwards there is new functionality that allows you to gather information about application events in a non-real time manner. These statistics include information such as execution path length, database I/O operations, execution path branches (CALL and PERFORM) and so on.

There are three basic ways of enabling EDM.

• Using the API to enable this functionality on the same port or another user’s port.
• Using the jprof command from the command line (you can enable this functionality for your own port or another user’s port)
• Using the JDIAG environment variable

The use of EDM means jBASE will generate a text file describing all sorts of information. However no attempt is made to per- form any analysis of this output, it is up to the user to decide what to do with the information.

EDM calling API

Two new functions have been created to turn on and turn off the Event Driven Metrics. These functions are called JIMITraceOn and JIMITraceOff and are prototyped in the jBASE supplied file jimiB.h.

JIMITraceOn(PortNumber , FileName , Options )

This function calls turns EDM on for a particular port.

Command	Description
PortNumber	Enter the port number you wish to turn on or -1 to denote your own port.
FileName	The name of a file where the details will be written to. On UNIX systems it can also be the name of a tty device (like /dev/pts/0) so you can monitor on a terminal the events occurring in an application. Also “stdout” and “stderr” are valid.
Options	Define here a list of options separated by commas which describe what details to gather. By default i.e. “” , just short details are gathered. The options are: DATABASE - All OPEN’s and database I/O (such as READ , WRITE) are traced OPEN - All OPEN’s are traced FILEIO - All database I/O (such as READ, WRITE) but not including OPEN are traced BRANCH - All CALL, RETURN and PERFORM/EXECUTE statements are traced CALL - All CALL and RETURN statements are traced EXECUTE - All PERFORM/EXECUTE statements are traced TRANSACTION - Details of committed transactions including before and after images are traced ALL - All the above options VERBOSE - By default summaries are displayed at the end of the session. Using verbose gives a detailed output of each option as it happens. HOLDIO - Holds before and after images in memory during the trace

JIMITraceOff(PortNumber)

This function call turns EDM off for a particular port.

PortNumber: Enter the port number you wish to turn off or -1 to denote your own port.

Once turned off, you can read in this information in a number of ways. You can use jBASE editors like jed or ED, use UNIX edit- ors like vi or you can read the file programmatically using OPEN and READ. Because the file might be quite substantial perhaps OPENSEQ and READSEQ would be more appropriate.

The return value from these functions are 0 for successful or non-zero for an error. You need to include jimiB.h in your applic- ation to get the relevant DEFC statements to support calling this function.

Once EDM is initiated you get a text oriented output sent to the file name specified

In a multi-thread environment, only the event details for that port are recorded even if multiple ports are running in different threads inside the same process.

JIMIBeforeImage() / JIMIAfterImage()

These functions provide before and after images respectively of any records which have been updated as part of the current Transaction context. These are only populated if the HOLDIO parameter is specified when starting the JIMI trace.

Unless the HOLDIO parameter is specified when the JIMI trace is started, this detail appears in the logfile as a single line;

COMMIT_BEGIN: WRITE_BEFORE_FILEIO: WRITE_AFTER_FILEIO:    
FILENAME=testfile_1 KEY=1 IMAGE=2 FILENAME=testfile_1 KEY=1 IMAGE=2
…..
WRITE_BEFORE_FILEIO:    FILENAME=testfile_10 KEY=10    IMAGE=11
WRITE_AFTER_FILEIO:    FILENAME=testfile_10 KEY=10    IMAGE=11
COMMIT_END:

EDM from the command line

You can use the jprof command to enable or disable this EDM functionality as shown below

jsh > jprof –j ON nnn filename options

Or

jsh -> jprof –j OFF nnn filename options Where

nnn is the port number to turn on or off. You can specify % which means the same port number you are running jprof from.

filename is the name of the output file as described above in the “FileName” parameter to the JIMITraceOn function.

options is one or more of the strings as described above in the “Options” parameter to the JIMITraceOn function

EDM using the JDIAG environment variable

The JDIAG environment variable encompasses a large number of options and the EDM has been added to this. To find a short summary of all the JDIAG options use the command below from a UNIX shell.

machinename- ~src/jbase4/4.1.3/jbase: JDIAG=HELP WHO JDIAG=option{:option{:option ...}}
option can  be one of ... profile=   {off|short|long|user|jcover|all}
filename=  {stdout|stderr|tmp|pathname {,refresh_  mins} (%p in pathname is substituted with process   id) (%t in pathname is substituted with time stamp) memory=     {off|on|verify}
branch= {off|on|verbose}
jimi=  {database|open|fileio|branch|call|perform|transaction|all|verbose} trace=env_ name {,env_ name..} help
3 greg
machinename-~src/jbase4/4.1.3/jbase:

So to trace all database and transaction information into a file called ‘trace_nnn_out’ where ‘nnn’ is the process id (hence you get a new file for each process) you would do this example below, which sets JDIAG, runs a command (WHO) and then exam- ines the resultant output file.

machinename-  ~src/jbase4/4.1.3/jbase:  export   JDIAG- G=jimi=database,transaction,verbose:filename=trace_%p_      out machinename- ~src/jbase4/4.1.3/jbase: ls  - l trace*
ls:  trace*: No  such file or directory Machinename-  ~src/jbase4/4.1.3/jbase:   WHO
3 trace
Machinename- ~src/jbase4/4.1.3/jbase: ls - l trace*
- rw- r- - r- - 1 trace trace  1045  Mar 2 18:04 trace_ 7338_ out Machinename-  ~src/jbase4/4.1.3/jbase:  cat   greg_  7338_ out TIMESTAMP:     time       1078250684
OPEN: FILENAME=/home/greg//4.0_ rels/jbc4.1.4.3/tmp/jBASEWORK TYPE=HASH4 STACK=1,0 SUCCESS=0SOURCE=jmainfunction.b,0 PATHLENGTH=121 TIME=2535235.970
OPEN: FILENAME=DICT   POINTER- FILE   STACK=1,0   SUCCESS=2
SOURCE=jmainfunction.b,0   PATHLENGTH=125   TIME=2535235.980
OPEN:  FILENAME=POINTER- FILE  STACK=1,0  SUCCESS=2  SOURCE=jmainfunction.b,0  PATHLENGTH=125 TIME=2535235.980
RELEASE:    FILENAME=/home/greg//4.0_ rels/jbc4.1.4.3/tmp/jBASEWORK    KEY=JBASE_ SET_ ROOT*3   TRANSACTION=NO    STACK=1,0    SUCCESS=0    SOURCE=jmainfunction.b,0  PATHLENGTH=129 TIME=2535235.980
DELETE: FILENAME=/home/greg//4.0_ rels/jbc4.1.4.3/tmp/jBASEWORK TRANSACTION=NO KEY=JBASE_ SET_ ROOT*3  STACK=1,0  SUCCESS=2  SOURCE=jmainfunction.b,0   PATHLENGTH=129 TIME=2535235.980
*
JIMI output created by user greg on port 3 at Tue Mar 2 18:04:44 2004
*
READS: 0
WRITES: 0
OPENS: 3
DELETES: 1
CLEARFILES: 0
RELEASES: 0
RELEASEFILES: 0
SELECTS: 0
TRANSENDS: 0
TRANSABORTS: 0
PERFORMS: 0
PATHLENGTHS: 173
CALLS: 2
OPENSEQS: 0
READSEQS: 0
WRITESEQS: 0
SLEEPS: 0
INPUTS: 0
STOPS: 1
Machinename- ~src/jbase4/4.1.3/jbase: unset JDIAG Machinename-~src/jbase4/4.1.3/jbase:

EDM Output

The output generated by calling JIMITraceOn is simple text layout which can be sent either to a genuine operating system file or on Unix systems, to a terminal. Each line is usually a single line with each detail delimited by a tab character to make for easy

parsing later. The use of the trace functions means a text file is created at the point specified by the user. Each line of text is usu- ally a single event and consists of a number of fields delimited by a tab character.

The output may contain blank lines and also comment lines that start with * or #

The following is a list of output generated by various options on the call to JIMITraceOn.

Standard output with no additional options specified

These will all appear as a summary when JIMI is turned OFF by various mechanisms e.g. using the JIMITraceOff API, or the user logging off, or the ‘jprof –j off’ command being executed.

READ: nnn
WRITES: nnn
OPENS: nnn
DELETES: nnn
CLEARFILES : nnn
RELEASES: nnn
RELEASEFILES: nnn
SELECTS: nnn
TRANSENDS: nnn
TRANSABORTS: nnn
PERFORMS: nnn
PATHLENGTHS: nnn
CALLS: nnn
OPENSEQS: nnn
READSEQS: nnn
WRITESEQS: nnn
SLEEPS: nnn
INPUTS: nnn
STOPS: nnn

Optional output with the DATABASE or FILEIO option

These appear as a summary when JIMI is turned OFF by various mechanisms. For example, using the JIMITraceOff API, or the user log- ging off, or the ‘jprof –j off’ command being executed.

TRANSTART:    NAME=XXX COUNT=nnn
TRANEND:    COUNT=nnn 
TRANSABORT:    COUNT=nnn
READ:    FILENAME=name_of_file    COUNT=nnn
READ:    FILENAME=name_of_file    COUNT=nnn
WRITE:    FILENAME=name_of_file    COUNT=nnn
DELETE:    FILENAME=name_of_file    COUNT=nnn
SELECT:    FILENAME=name_of_file    COUNT=nnn
CLEARFILE:    FILENAME=name_of_file    COUNT=nnn
RELEASE:    COUNT=nnn    
RELEASE:    FILENAME=name_of_file    COUNT=nnn
READSEQ:
WRITESEQ:    FILENAME=name_of_file
FILENAME=name_of_file    COUNT=nnn
COUNT=nnn

Optional output with either the DATABASE or FILEIO option and VERBOSE option

These appear before each statement is executed. Note that you should parse each field delimited by TAB, separately and not assume each field will be in any particular order.

Each line, as part of the **COMMON_INFO** as detailed later, has a SUCCESS=nn part and you should always be aware that if the value for nn is not 0, the operation failed and so some of the values might be incorrect. For example, during a READ, if SUCCESS=2, then the record didn’t exist and the SIZE=nn part is invalid.

TRANSTART:  NAME=XXX  SYNC=ON|OFF **COMMON_INFO** TRANEND: 
**COMMON_INFO**
TRANSABORT: **COMMON_INFO**
READ:
SIZE=nnn   FILENAME=name_ of_ file KEY=item_ id
**COMMON_INFO**   LOCK=YES|NO TRANSACTION=YES|NO
READ: TRANSACTION=YES|NO   FILENAME=name_ of_ file SIZE=nnn FIELD=nnn   KEY=item_ id
**COMMON_INFO**   LOCK=YES|NO
WRITE: FILENAME=name_ of_ file KEY=item_ id SIZE=nnn LOCK=YES|NO TRANSACTION=YES|NO **COMMON_INFO**
DELETE: FILENAME=name_ of_ file  KEY=item_ id TRANSACTION- N=YES|NO **COMMON_INFO**
SELECT: FILENAME=name_of_file  **COMMON_INFO** CLEARFILE: FILENAME=name_of_file  **COMMON_INFO** RELEASE: TRANSACTION=YES|NO **COMMON_INFO**
RELEASE:          TRANSACTION=YES|NO     FILENAME=name_of_file **COMMON_INFO**
RELEASE: TRANSACTION=YES|NO FILENAME=name_ of_ file KEY=item_ id
**COMMON_INFO**
READSEQ: FILENAME=name_ of_ file   SIZE=data_ size **COMMON_ INFO**
WRITESEQ:           FILENAME=name_ of_ file  SIZE=data_ size **COMMON_ INFO**

Optional output with either the DATABASE or OPEN option and VERBOSE option.

OPEN:
INFO**   FILENAME=name_ of_ file   TYPE=DB2|ORACLE|J3|J4    **COMMON_
OPENSEQ:
INFO**   FILENAME=name_ of_ file   TYPE=SEQUENTIAL   **COMMON_

Optional output with the BRANCH or CALL option

These appear as a summary when JIMI is turned OFF by various mechanisms. For example, using the JIMITraceOff API, or the user log- ging off, or the ‘jprof –j off’ command being executed.

CALL: SUBNAME=nnn COUNT=nnn

Optional output with the BRANCH or PERFORM option

PERFORM: COMMAND=Command Line Details COUNT=nnn

Optional output with either the BRANCH or CALL option and VERBOSE options

These appear after each statement is executed.

CALL: SUBNAME=name_of_subroutine **COMMON_INFO**
RETURN: SUBNAME=name_ of_ subroutine PATHLENGTH=num_ instructions **COMMON_ INFO**

Optional output with either BRANCH or PERFORM option and VERBOSE option

PERFORM: COMMAND=Command Line Details **COMMON_INFO** 

Optional output with the TRANSACTION option.

This output appears irrespective of whether the VERBOSE option was used or not.

These appear following successful execution of a TRANSEND statement and detail in chronological order the WRITE’s and DELETE’s that make up the transaction.

Option	Description
COMMIT_BEGIN and COMMIT_END	Operations surround the entire transaction so just one set of these per transaction
WRITE_BEGIN and WRITE_END	Operations surround a WRITE operation inside a transaction so one set of these per WRITE
DELETE_BEGIN and DELETE_END	Operations surround a DELETE operation inside a transaction so one set of these per DELETE
WRITE_BEFORE_BEGIN and WRITE_BEFORE_END	Operations show any before image information associated with a WRITE
WRITE_AFTER_BEGIN and WRITE_AFTER_END	Operations show the after image information associated with a WRITE
DELETE_BEFORE_BEGIN and DELETE_BEFORE_END	Operations show any before image information associated with a DELETE. Consider the following lines of code and assume the file is empty to start off with. TRANSTART ELSE STOP WRITE “Blood bank information line 1” ON FILEVAR,”ID1” WRITE “Overwrite the blood bank information” ON FILEVAR,”ID1” DELETE FILEVAR,”ID1” TRANSEND ELSE STOP This would give the following output with the characters in bold type showing comments added
COMMIT_BEGIN	Shows the start of the transaction information
WRITE_BEGIN	FILENAME=fb1 KEY=ID1  the start of the first WRITE statement
WRITE_AFTER_BEGIN	The start of the AFTER image of the first WRITE statement
000001	Blood bank information line 1 Line 1 of the AFTER image of the first WRITE statement
WRITE_AFTER_END	The end of the AFTER image of the first WRITE statement
WRITE_END	The end of the first WRITE statement
WRITE_BEGIN	FILENAME=fb1 KEY=ID1  the start of the second WRITE statement
WRITE_BEFORE_BEGIN	The start of the BEFORE image of the second WRITE statement
000001	Blood bank information line 1 Line 1 of the BEFORE image of the first WRITE statement
WRITE_BEFORE_END	The end of the BEFORE image of the second WRITE statement
WRITE_AFTER_BEGIN	The start of the AFTER image of the second WRITE statement
000001	Overwrite the blood bank information Line 1 of the AFTER image of the second WRITE statement
WRITE_AFTER_END	The end of the AFTER image of the second WRITE statement
WRITE_END	The end of the second WRITE statement
DELETE_BEGIN	FILENAME=fb1 KEY=ID1  the start of the first (and only) DELETE statement
DELETE_BEFORE_BEGIN	The start of the BEFORE image of the first (and only) DELETE statement
000001	Overwrite the blood bank information Line 1 of the BEFORE image of the first (and only) DELETE statement
DELETE_BEFORE_END	The end of the BEFORE image of the first (and only) DELETE statement
DELETE_END	The end of the first (and only) DELETE statement
COMMIT_END	Shows the end of the transaction information

• **COMMON_INFO** refers to a number of statements that are common to the trace messages …
• STACK=3,2 - This shows the current PERFORM/EXECUTE level is 3 and that the CALL stack level is 2. In other words this is the 3rd concurrent program executing for the user and in the current running program the subroutine nest is 2 i.e. there is another subroutine below it. Note the main program counts as a subroutine.
• SOURCE=sourcename,line,sourcename,line,sourcename,line -  Shows where the trace message was called from and the stack of subroutines below it. Displayed to a maximum depth of 3 entries
• PATHLENGTH=nnn - Shows the current number of licensing instructions executed so far.
• SUCCESS=nn where 0 is a success for the operation and any other value is the error number for the failure. If a failure, other parts of the event might be incorrect.
• TIME=nnnnnn.nnn - Gives a decimal number in seconds since 1st Jan 1970 and so allows timings between different statements.

Source Quality Metrics (SQM), which are produced when programs are compiled

There are options to both the BASIC command and the jcompile command such that the quality of the source code compiled can be evaluated. On the whole jBASE merely generates the statistics during a compilation. It is up to the caller to then inter- pret the results as he/she sees fit. Use the (J) option to generate the SQM data to the BASIC command and the -JJ option to the jcompile command.

In either case the output will be a simple text file whose name is of the format “sourcename.sqm” where sourcename is the ori- ginal source name. The text file will be placed in the same file that the original source code was found.

The layout of the file is a multi-value file whose attributes are as follows. Use the text name given to each attribute rather than hard-coding an attribute number as the layout might change (found in jimiBh).

001 JBASE_SQM_VARIABLES

A multi-value list of all variable names local to the subroutine (or main program). In the case of dimensioned arrays the initial dimensions are given as defined e.g. FILEVAR(10,2)

002 JBASE_SQM_SUBROUTINES

A multi-value list of all subroutines called in the format SUBROUTINENAME<vm> Line Number. Note that a call such as “CALL @subname” will have “@subname” as the subroutine name allow you to differentiate between direct and indirect calls. 

005 JBASE_SQM_INSERT

A multi-value list of all the INSERT statements seen. The format of this is not accurately defined at the moment but will probably be in the format ‘NAME_OF_INSERT_FILE<sv>NUMBER_LINES_EQUATES<sv>NUMBER_LINES_CODE’

011 JBASE_SQM_SELECT

A multi-value list of all ‘SELECT’ statements seen where each value is the line number it is seen on.

012 JBASE_SQM_RELEASE

A multi-value list of all line numbers containing a ‘RELEASE’ statement.

013 JBASE_SQM_RELEASE_VAR

A multi-value list of all ‘RELEASE VARiable’ statements seen where each value is the line number it is seen on.

014 JBASE_SQM_TRANSACTION

A multi-value list of all transaction statements where each value is of the format ‘TRANSTART|TRANSEND|TRANSABORT<sv>line_number’

015 JBASE_SQM_GOTO

A list of all line numbers where a GOTO is seen.

016 JBASE_SQM_RETURNTO

A list of all line numbers where a ‘RETURN TO’ statement is seen.

019 to 049 RESERVED

050 JBASE_SQM_DATE

The date in internal format this SQM file was created

051 JBASE_SQM_TIME

The time in internal format this SQM file was created.

052 JBASE_SQM_JBASE_USER

The jBASE user name who compiled this SQM file.

053 JBASE_SQM_OS_USER

The operating system name of the user who compiled this SQM file.

CLEAR-ITEM-LOCKS

Use the CLEAR-ITEM-LOCKS command to clear a specific lock or all locks taken against a specific file. Use this command with jBASE hashed files only.

Command Syntax

CLEAR-ITEM-LOCKS filename
CLEAR-ITEM-LOCKS filename, itemname

Refer to the SHOW-ITEM-LOCKS command

If JEDI_NOSHMEM is set (i.e. JEDI_NOSHMEM=1) the CLEAR-ITEM-LOCKS will not function (?)

SHOW-ITEM-LOCKS

The SHOW-ITEM-LOCKS command displays details of locked items in jBASE-hashed files.

Command Syntax

SHOW-ITEM-LOCKS

The @USERSTATS allows a program to retrieve miscellaneous information about itself. For example if a program wants to find out how many database I/O's it performed, the following command can be executed.

info1 = @USERSTATS
read1 = info1<19>
EXECUTE 'COUNT fb1 WITH *A1 EQ "x"'
info2 = @USERSTATS
read2 = info2<19>
PRINT "The COUNT took ":(read2-read1):" READ's”

The following definitions have been added to JBC.h file which defines the layout of data returned either through the @USERSTATS variable or by opening file SYSTEM(1027) and reading the items in like that.

* Definitions for the data returned from @USERSTATS variable or from
* the record read from PROC file (using SYSTEM(1027) as file name)
*

EQUATE USER_PROC_PORT_NUMBER TO 1;* The port number
EQUATE USER_PROC_NUM_PROGRAMS TO 2;* Number of programs running in this port
EQUATE USER_PROC_START_TIME TO 3;* Time user started in UTC format
EQUATE USER_PROC_PID TO 4  ;* Process ID
EQUATE USER_PROC_ACCOUNT TO 5;* Name of the account
EQUATE USER_PROC_USER TO 6 ;* Name of the user
EQUATE USER_PROC_TERMINAL_JBASE TO 7;* Name of terminal according to jBASE
EQUATE USER_PROC_TERMINAL_OS TO 8;* Name of terminal as seen by OS
EQUATE USER_PROC_DATABASE TO 9;* Name of database connected to
EQUATE USER_PROC_TTY TO 10;* Name of TTY device
EQUATE USER_PROC_LANGUAGE TO 11;* Language
EQUATE USER_PROC_LISTENING_TIME TO 12;* Time in UTC the listening thread last worked
EQUATE USER_PROC_MEM_FREE TO 13;* Amount of memory in heap space free chain
EQUATE USER_PROC_MEM_USED TO 14;* Amount of heap space memory in use
EQUATE USER_PROC_THREAD_TYPE_INT TO 15;* Thread type expressed as an integer
EQUATE USER_PROC_THREAD_TYPE_TXT TO 16;* Thread type expressed as a text string
EQUATE USER_PROC_LICENSE TO 17;* License counters
EQUATE USER_PROC_STATS_OPEN TO 18;* Number of OPEN's performed.
EQUATE USER_PROC_STATS_READ TO 19;* Number of READ's performed.
EQUATE USER_PROC_STATS_WRITE TO 20;* Number of WRITE's performed.
EQUATE USER_PROC_STATS_DELETE TO 21;* Number of DELETE's performed.
EQUATE USER_PROC_STATS_CLEARFILE TO 22;* Number of CLEARFILE's performed.
EQUATE USER_PROC_STATS_PERFORM TO 23;* Number of PERFORM's / EXECUTE's performed.
EQUATE USER_PROC_STATS_INPUT TO 24;* Number of INPUT's performed.
EQUATE USER_PROC_UNUSED_1 TO 25;* Unused
EQUATE USER_PROC_OPEN_FILES_VIRTUAL TO 26           ;* Number of files application thinks open
EQUATE USER_PROC_OPEN_FILES_REAL TO 27 ;* Number of files really open by OS
EQUATE USER_PROC_USER_ROOT TO 28;* Application data set by @USER.ROOT
EQUATE USER_PROC_PROCESS_TXT TO 29;* Text string to identify process
EQUATE USER_PROC_PROGRAM TO 41;* Program name and command line arguments
EQUATE USER_PROC_LINE_NUMBER TO 42;* Line number currently being executed.
EQUATE USER_PROC_SOURCE_NAME TO 43;* Name of source currently being executed.
EQUATE USER_PROC_UNUSED_2 TO 44;* Unused
EQUATE USER_PROC_UNUSED_3 TO 45;* Unused
EQUATE USER_PROC_STATUS_TXT TO 46;* Status of program as a readable text
EQUATE USER_PROC_STATUS_INT TO 47;* Status of program as an integer
EQUATE USER_PROC_CPU_USR TO 48;* User CPU time
EQUATE USER_PROC_CPU_SYS TO 49;* System CPU time
EQUATE USER_PROC_CPU_USR_CHILD TO 50;* User CPU time used by child processes
EQUATE USER_PROC_CPU_SYS_CHILD TO 51;* System CPU time used by child processes
EQUATE USER_PROC_USER_THREAD TO 52;* Application data set by @USER.THREAD

BASE is supplied with a windows telnet server to enable users to telnet into a windows server. After connecting to the telnetd service and giving your password you will be logged into the NT system in a simple passthru mode and be sitting at a normal DOS command shell. You may however change the startup program from cmd.exe to any jBASE program including the jsh.exe (jSHELL).

Account Naming Issues

Windows NT account names can exist in several name spaces. For example a Windows NT station in an Advanced Server domain has a local “Administrator” account and also has a corresponding “Administrator” account in its default domain.  jBASE TCP Remote Logon Services use the following rules to disambiguate account names:

If the account name is qualified (contains a backslash), the name preceding the backslash is first treated as a domain name, if there is no corresponding domain, then it is treated as a machine name. (Example: “MainDo- main\Administrator”).

If the account name is not qualified (does not contain a backslash), the name is first looked up on the local machine. If the account name is not found, it is then looked up in the default domain of the machine.

User Environment

When users logon, their environment will contain all system-wide environment variables that are set on the local system. They will not receive their normal user environment settings at this moment in time (the Win32 API does not provide this ability), however, this will change in a future release. To circumvent this omission in the Win32 API, the jBASE TCP Remote Logon Ser- vices automatically set the following environment variables:

USERDOMAIN	The domain name in which the user account is defined.
USERNAME	The account name of the user.
HOMEPATH	The path name of the home directory of the user. If the user’s home dir- ectory is a remote path, then this will contain the Universal Naming Con- vention (UNC) name of the user’s home directory.
HOMESHARE	Always set to NULL. See the comments on remote directories below.
HOMEDRIVE	If the user’s home directory is local, this contains the drive letter followed by a colon. If the home directory is remote, this is set to NULL.
JBC_TELNET_FLAG	This is set so that a user process can determine if it is being run via telnet session or via the console.

Because the remote user shares the drive map with all other users, it is not possible to automatically mount a remote user’s remotely named directory on its normal drive letter. However many sites may wish to establish conventions whereby remote users are allowed to use certain drive letters remotely. Further, other environment variables may need to be set at logon. Thus the jBASE TCP Remote Logon Services execute the file “remote.cmd” if present in the user’s home directory. If a remote user’s home directory is specified as a remote directory, the user’s initial directory will be “C:\”. If desired, this can be overridden in remote.cmd.

By default if you have a remote.cmd file, telnet process reads and parses it, and the first command it finds will be executed. This file is intended mainly to allow the user to set up the environment, and so this parsing is primitive, for example only the first command it finds is executed, so a remote.cmd created by the jBASE “iju” for instance this will be the  jsh. To disable this feature and have the NT command proceessor read the file, the registry setting ParseProfile should be set to zero.

You should therefore avoid using command such as cd and call. If supported by the telnet client, then the TERM variable will also be set.

Customization

NOTE: You can edit the registry by using Registry Editor (Regedit.exe or Regedt32.exe). If you use Registry Editor incorrectly, you can cause serious problems that may require you to reinstall your operating system. jBASE does not.

Presenting a banner to the remote user after logon

NOTE: If you change or create any of the registry entries below, you will need to stop and then restart the services with the new or changed values before those.

Create a registry entry of type REG_SZ with the name Banner under registry key:

HKEY_LOCAL_MACHINE\Software\JAC\jBASE Telnetd Server\CurrentVersion

In the "Value data:" string, the sequence "\n" generates an end of line output, to use the '\' characters in the banner string use two '\' characters in a row.

Example

Name Type Data
Banner REG_SZ Welcome to jBASE\nThe Multi-value database of the future\n

Presenting a banner to the remote user prior to logon

Create a registry entry of type REG_SZ with the name PreBanner under registry key:

HKEY_LOCAL_MACHINE\Software\JAC\jBASE Telnetd Server\CurrentVersion

In the "Value data:" string, the sequence "\n" generates an end of line output, to use the '\' characters in the banner string use two '\' characters in a row.

Example

Name Type Data
PreBanner REG_SZ jBASE telnet process\nPlease log into system\n

Changing the logon prompt

Create a registry entry of type REG_SZ with the name LogonPrompt under registry key:

HKEY_LOCAL_MACHINE\Software\JAC\jBASE Telnetd Server\CurrentVersion

Example

Name Type Data
LogonPrompt REG_SZ Windows User Name:

If this registry value is not present, the default logon prompt is "Account Name: ".

Changing the password prompt.

Create a registry entry of type REG_SZ with the name PasswordPrompt under registry key:

HKEY_LOCAL_MACHINE\Software\JAC\jBASE Telnetd Server\CurrentVersion

Example

Name Type Data
PasswordPrompt REG_SZ Windows Password:

If this registry value is not present, the default password prompt is "Password: ".

Changing the Failed logon message

Create a registry entry of type REG_SZ with the name FailedLogonMessage under registry key:

HKEY_LOCAL_MACHINE\Software\JAC\jBASE Telnetd Server\CurrentVersion

In the "Value data:" string, the sequence "\n" generates an end of line output, to use the '\' characters in the banner string use two '\' characters in a row.

Example

Name Type Data
FailedLogonMessage REG_SZ \nFailed logon...Valid "WINDOWS" USER/PASSWORD?\n

If this registry value is not present, the default message is:

"Logon failed: unknown user name, password or privilege incorrect."

Logging user logon/logoff messages in the registry

Create a registry entry of type REG_DWORD with the name LogEventLogon under registry key:

HKEY_LOCAL_MACHINE\Software\JAC\jBASE Telnetd Server\CurrentVersion

If this registry entry is present and set to 0, then normal logon and logoff events will not be logged into the registry. If this entry is either not present, or it is set to 1, then an entry will be written into the registry whenever a user logs on or off the system.

Changing the default command processor

Create a registry entry of type REG_SZ with the name CommandProcessor under registry key::

HKEY_LOCAL_MACHINE\Software\JAC\jBASE Telnetd Server\CurrentVersion

Example

Name Type Data
CommandProcessor REG_SZ C:\JBASE30\BIN\JSH.EXE -

This causes the jSHELL to start up and execute the login PROC in your MD if you have one.

By default, all services invoke CMD.EXE as the command processor. By adding the registry value you can override the com- mand processor used by all users. NOTE: if you override the command processor, then the automatic environment  setup using the REMOTE.CMD script will no longer be available.

Bump up priority while logging on

Create a registry entry of type REG_DWORD with the name IncreaseLogonPriority under registry key:

HKEY_LOCAL_MACHINE\Software\JAC\jBASE Telnetd Server\CurrentVersion

On loaded systems, the logon process can be slow. You can set this registry value to 1 if you want to increase the priority of the logon, AT THE EXPENSE OF OTHER PROCESSES ON THE SYSTEM.

If this registry entry is present and set to 1 then the priority of the logon process will be increased.

Changing default exit detection timeout.

Create a registry entry of type REG_DWORD with the name ExitDetectionTimeout under registry key:

HKEY_LOCAL_MACHINE\Software\JAC\jBASE Telnetd Server\CurrentVersion

Example

Name Type Data
ExitDetectionTimeout REG_DWORD 0x000001f4 (500)

Due to a technical issue related to pipes, the telnetd server has to “guess” when a command prompt is about to exit. This will be corrected in a later release. At those times when it this guess occurs, the telnetd server waits before doing a read of input. If this read timeout is too short, the exit detection doesn’t work, and so users need to type an extra input before the telnetd exit. (In other words, when the user types “exit” to CMD.EXE, the telnet client will not “hang up” until the user types an extra character.) Through internal experimentation we have found that a value of 400 milliseconds works well under most cir- cumstances. However some customers have found this inadequate, so we provide this value. The value should be set to as small a value as provides the desired behavior.

If this registry value is not present, the default value is 400 milliseconds "0x00000190 (400)".

Changing the number of logon attempts

Create a registry entry of type REG_DWORD with the name LogonAttempts under registry key:

HKEY_LOCAL_MACHINE\Software\JAC\jBASE Telnetd Server\CurrentVersion

Example

Name Type Data
LogonAttempts REG_DWORD 0x00000005 (5)

If this registry value is not present, the default value is 3 attempts "0x00000003 (3)".

Parsing the “remote.cmd” file

Create a registry entry of type REG_DWORD with the name ParseProfile under registry key:

HKEY_LOCAL_MACHINE\Software\JAC\jBASE Telnetd Server\CurrentVersion

Example

Name Type Data
ParseProfile REG_DWORD 0x00000000 (0)

By default the telnet service will read any remote.cmd file that it finds in a users home directory, and execute the first com- mand it finds. This is efficient and avoids having a command processor for each port, however, the commands that are sup- ported are restricted to setting environment variables (set) and echoing text (echo). If you want the NT command processor to parse the file then set the value to 0.

Change the telnet port (jBASE 4.1 only)

The telnet port used by jtelnetd is stored in the %windir%\system32\etc\services item. jBASE will use the port signified by jtel- net (if present) or telnet if no jtelnet is present. To change the port for jBASE telnet only you can add this entry, which allows another telnet client to run on the port defined by "telnet".

jtelnet 2323/tcp (This would start jBASE telnet process on port 2323)

Instead, you can change telnet itself if jBASE telnet is the only telnet process to be run.

telnet 2323/tcp (This would start jBASE telnet process on port 2323)

The format of the services file will be similar to:

# Copyright (c) 1993-1999 Microsoft Corp.
#
# This file contains port numbers for well-known services defined by IANA
#
# Format:
#
# <service name> <port number>/<protocol> [aliases...] [#<comment>]
#
echo 7/tcp
echo 7/udp
...
telnet 23/tcp

Troubleshooting or Technical Support

Event Log

The jBASE TCP Remote Logon Services report error messages to the Application Event Log. This log can be viewed using the Event Viewer application which can usually be launched by double clicking its icon in the Program Manager group: Admin- istrative Tools. Make sure the Application event log is selected. (Its entry in the Log menu should have a check mark beside it... if not, select it.)

All jBASE TCP Remote Logon Services entries begin with the tag “jBASE”. Most of the error messages are self explanatory. Any error codes mentioned are standard Windows NT error codes as returned by GetLastError(). Your VAR should be able to help you with any errors here.

On rare occasions you may have a service failure. These are logged by the Service Control Manager in the System Log.

List of known problems

Eventlog says: jBASE Command Starter: CreateProcess: 5

This means that an account with Administrator privileges tried to logon, but that account does not have sufficient rights to execute transcmd.exe. The error cannot be caught sooner as Administrator accounts have a large number of privileges that allow all other operations up to this point to succeed.

Service Manager can’t find .exe or Start failed: 2

Usually this error means that you’ve moved the software after you installed it. Services cannot be moved after installation as an absolute path name is stored. To correct the problem, remove, then reinstall the software.

Access denied when accessing a drive mounted with NET USE

Because Windows NT wasn’t initially designed with the idea of more than one interactive user logged on at the same time, oddities often occur when accessing remote drives via a drive letter in the shared drive map. Most users seem to have better results when accessing remote drives via the SUBST command. Accessing remote drives via UNC names also works and is recommended.

Login failure messages in the Event Log every minute

This is due to a bug in Compaq’s Insight Manager program. Compaq has provided a work-around: Upgrade to version 2.60b or newer of the Insight Manager product, then use the Control Panel for the Insight Manager product to disable telnet detec- tion. Compaq is working on a better solution.

Program exits immediately with no output

The most likely problem is that you have a needed DLL missing from your PATH. Because telnetd users need to see the error messages in a non-GUI form, we set things so that Windows NT will not let an error pop-up box occur for this error, but rather cause the error code to be returned to the calling program. In most instances this setting has the desired effect of letting a remote error see the error message, however, for reasons unknown to us, Microsoft has made the CMD.EXE suppress the error message for a missing DLL. This is particularly odd because it would be necessary for Microsoft to have written CMD.EXE to explicitly ignore that error.