New Eiffel Reverse-Engineering Tool 
Author Message
 New Eiffel Reverse-Engineering Tool


|
| I have developed the Eiffel Analyzer (EIFAN) freeware tool
| to reverse-engineer Eiffel source code into Booch object-oriented
| class diagrams.  Here is a copy of the README file, which describes
| the tool, and how to install it.  Hope someone finds it useful.
|
|       Bill

The Booch method (or the OMT or Unified methods) is of little interest
to me because of what the BON book writes about them.

In addition, I have found that the original documentation in the
several books by Grady Booch about his method contained a very
complicated graphical notation which was not presented in a logical
manner or expressed in clear writing.  Hence I have found the Booch
method, in particular, to be as amibiguous as are entity relationship
diagrams.  Perhaps that is why the Unified method stuff is being
produced by Rational and Lockheed Martin together, ie, Rational
supplies the ideas and Lockheed Martin supplies the ghost writers and
technical proofreaders.

(It is also my understanding that none of those methods is based on
second order predicate logic, as is Business Object Notation [BON].)

I would encourage your efforts along the line of BON.

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~

CEC Services, 2080 Kipling St, Lakewood, CO  80215-1502   USA
Voice: 303.231.9437;  Facsimile: .231.9438;  Data:  .231.9434  
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~



Sat, 02 May 1998 03:00:00 GMT  
 New Eiffel Reverse-Engineering Tool

Quote:

>|
>| I have developed the Eiffel Analyzer (EIFAN) freeware tool
>| to reverse-engineer Eiffel source code into Booch object-oriented
>| class diagrams...

>The Booch method (or the OMT or Unified methods) is of little interest
>to me...

>(It is also my understanding that none of those methods is based on
>second order predicate logic, as is Business Object Notation [BON].)

>I would encourage your efforts along the line of BON.

I would have to second this. Translating Eiffel to Booch loses so
much information and keeps information that is of lesser value.

Try to move to BON. Booch and OMT emphasize object relationships,
and in particular "data" relationships. BON rightly emphasizes
*behavior* which is what is most desired for OOD in *any* language.

--

(503) 264-9309, FAX: (503) 264-3375

"Successful projects... are not afraid continually to
 improve their abstractions..." -Grady Booch, Object Success



Sat, 02 May 1998 03:00:00 GMT  
 New Eiffel Reverse-Engineering Tool

Quote:

> I simply wanted a quick way to analyze a collection of Eiffel code
> that was written by another engineer who did not use any formal
> design methodology.  My first impulse was to create some diagrams
> by hand.  But, after thinking about it, I thought it would be nice
> to write a program to do it for me.  After poking around the newsgroups
> and web, I located the TROOPER Eiffel parser (written in Eiffel),
> the 'graphplace' graph-layout program, and the Object Domain CASE
> tool.  The author of Object Domain (Dirk Veemeersch) was kind enough to
> document the format of the files used to generate Booch diagrams, so...
> with a little bit of glue, tape, etc., I was able to construct EIFAN with a
> modest amout of work.  It serves the purpose for what it is intended,
> namely to help you analyze code written by someone else, so I thought
> I'd place it in the public domain in case others may find it useful.

Thanks for taking the time and effort. Even the 'modest' (to use
your estimate) amount of effort to integrate the components
is helpful to those of us interested in Booch modelling of
Eiffel classes.

Also, your message announcing the product, complete with
details and links to the other components, was most helpful.
I have downloaded all the parts and hope to start using it
ASAP.

As to BON (Better Object Notation), I have EiffelCase and use it
on occasion, but I am not sure how well it cooperates with
preserving changes made by EiffelBuild and EiffelBench, and
don't want to risk blowing away something when jumping between
tools. Also, I have trouble with modelling anything using
a specification language where the equality relation "a = b"
is unpredictable (i.e. value vs. reference is not captured
in the specification language). For these reasons (and the
overwhelming market preference for Booch) I am leaning
towards Booch notation for my projects, and again I appreciate
the 'early evaluation' made possible through EIFAN.



Wed, 06 May 1998 03:00:00 GMT  
 New Eiffel Reverse-Engineering Tool

Quote:

> I simply wanted a quick way to analyze a collection of Eiffel code
> that was written by another engineer who did not use any formal
> design methodology.  My first impulse was to create some diagrams
> by hand.  But, after thinking about it, I thought it would be nice
> to write a program to do it for me.  After poking around the newsgroups
> and web, I located the TROOPER Eiffel parser (written in Eiffel),
> the 'graphplace' graph-layout program, and the Object Domain CASE
> tool.  The author of Object Domain (Dirk Veemeersch) was kind enough to
> document the format of the files used to generate Booch diagrams, so...
> with a little bit of glue, tape, etc., I was able to construct EIFAN with a
> modest amout of work.  It serves the purpose for what it is intended,
> namely to help you analyze code written by someone else, so I thought
> I'd place it in the public domain in case others may find it useful.

Thanks for taking the time and effort. Even the 'modest' (to use
your estimate) amount of effort to integrate the components
is helpful to those of us interested in Booch modelling of
Eiffel classes.

Also, your message announcing the product, complete with
details and links to the other components, was most helpful.
I have downloaded all the parts and hope to start using it
ASAP.

As to BON (Better Object Notation), I have EiffelCase and use it
on occasion, but I am not sure how well it cooperates with
preserving changes made by EiffelBuild and EiffelBench, and
don't want to risk blowing away something when jumping between
tools. Also, I have trouble with modelling anything using
a specification language where the equality relation "a = b"
is unpredictable (i.e. value vs. reference is not captured
in the specification language). For these reasons (and the
overwhelming market preference for Booch) I am leaning
towards Booch notation for my projects, and again I appreciate
the 'early evaluation' made possible through EIFAN.



Wed, 06 May 1998 03:00:00 GMT  
 New Eiffel Reverse-Engineering Tool
For those of you who may have missed the first EIFAN posting,
here it is again.

======================================================================

I have developed the Eiffel Analyzer (EIFAN) freeware tool
to reverse-engineer Eiffel source code into Booch object-oriented
class diagrams.  Here is a copy of the README file, which describes
the tool, and how to install it.  Hope someone finds it useful.

        Bill

-------------------------------------------------------------

                EIFAN - Eiffel Analyzer

    Reverse-engineer Eiffel-3 source files
    into object-oriented class diagrams.

Release:        1.0     November 1995

Author:         Bill O'Keefe

Introduction:
=============

EIFAN is a tool used to reverse-engineer Eiffel source code into Booch
object-oriented class diagrams.  It provides a way to analyze the
overall structure of a set of Eiffel classes without having to
purchase an expensive CASE tool.

EIFAN depends upon several other public domain packages.
Details are provided below.

If you find this program useful, or have any questions/problems
with it, please let me know.

Contents:
========

EIFAN is delivered in a compressed, tar archive, and is available
via ftp at

        ftp://ftp.shore.net/members/bokeefe/eifan/eifan-1.0.tar.Z

To unpack this file, use:

        zcat eifan-1.0.tar.Z | tar xvf -

After unpacking the EIFAN distribution, you will find the following files:

        README          This file
        COPYING         Copy of GNU General Public License, version 2
        Makefile        top-level makefile
        eifgraph/       Contains EIFGRAPH source files
        eifdiag/        Contains diagram generating tools

License:
=======

--| Copyright (C) 1995 Bill O'Keefe.
--|
--| This program is free software; you can redistribute it and/or modify
--| it under the terms of the GNU General Public License as published by
--| the Free Software Foundation; either version 2 of the License, or
--| at your option) any later version.
--|
--| This program is distributed in the hope that it will be useful,
--| but WITHOUT ANY WARRANTY; without even the implied warranty of
--| MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
--| GNU General Public License for more details.
--|
--| You should have received a copy of the GNU General Public License
--| along with this program; if not, write to the Free Software
--| Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
--|
--| This program is  based  in part on  TROOPER (Truly Reusable OO
--| Compiler Compiler) written by Jon  Avotins and Glenn Maughan and
--| associated source classes and files, which is distributed free
--| of charge by Monash University, Software Development Department,
--| 900 Danenong Road, Caulfield East, Victoria 3145 Australia,  
--| which may  be obtained by anonymous ftp from the WWW site
--| http://insect.sd.monash.edu.au/~trooper/
--|----------------------------------------------------------------

Dependencies
============

1. TROOPER
----------
The main component of EIFAN is the EIFGRAPH tool.  This tool is
implemented using the Truly Reusable Object Oriented Compiler
Compiler (TROOPER), written by Jon Avotins and Glenn Maughan.  
You need to install TROOPER before you can install EIFGRAPH.
It is available via anonymous FTP at:

    ftp://hornet.sd.monash.edu.au/pub/trooper/trooper.tar.Z

There is also a TROOPER home page at:

    http://insect.sd.monash.edu.au/~jona/trooper.html

TROOPER is based upon the ISE v3.2.8, and should work with
later versions of ISE Eiffel as well.  In fact, I got it to work
with an earlier version (ISE v3.2.3b).  However, I had to manually
apply the 'extended_parse_patch' to my copy of the ISE 'parse' library.
Refer to the TROOPER README files for installation instructions.
It runs much faster if you 'finalize' it.

After installing TROOPER, it's probably a good idea to run a few
Eiffel source files through it, to verify the installation.
You don't need the TROOPER objects or executable to build EIFGRAPH,
since EIFGRAPH simply references the required TROOPER source files from
its Ace file.  Thus, if you're low on disk space, delete the TROOPER
'EIFGEN' directory after you verify its installation.

2. Graphplace
-------------
The 'graphplace' program is a wonderful tool developed by
Jos van Eijndhoven, and is used to layout the nodes of the generated
class diagrams.  It can be downloaded from a number of sites, including:

  ftp://ftp.funet.fi/pub/sci/math/misc/programs/graphplace/graphplace.t...

Refer to the 'graphplace' README for installation instructions.

The EIFAN distribution contains a slightly-modified version of the
'graphplace' postscript prologue (graphpro.ps), which it installs
in the 'lib' installation directory.  The change limits the maximum size
of node labels to 10 point text.  You can make other changes to this
file as described in the 'graphplace' man page.

3. PostScript Previewer
-----------------------
In order to preview the generated PostScript object-oriented class diagrams,
you will need a PostScript previewer.  By default, EIFAN is configured to
use GNU GhostScript.  If you don't already have GhostScript, you can get
it from:

    ftp://ftp.cs.wisc.edu:/pub/ghost/gnu/ghostscript*NNNN*

where NNNN is the latest version number.  If you have a different previewer,
then simply edit the 'graph2ps' script (described below) to use it.

4. Object Domain
----------------
Object Domain is a shareware CASE tool, written by Dirk Veemeersch, that
runs on a PC under Windows.  It produces nice looking Booch diagrams, and
supports reverse-engineering of C++ source code.  This is accomplished
through its 'C++ Analyzer', which generates Object Domain Analyzer (.oda)
files.  These files can then be imported into Object Domain.  

The EIFAN tool can also generate '.oda' files from Eiffel source code.
If you have a Windows/PC, you might want to give it a try.
The diagrams are more aesthetically pleasing than the PostScript
class diagrams, although they both contain the same information.

You can get Object Domain at a number of sites, including:

    http://www.acs.oakland.edu/oak/SimTel/win3/pgmtools.html

    Select 'dom107.zip'         (or a later version)

Installation:
============
Once all the dependencies have been obtained and installed, you are ready to
install EIFAN.  EIFAN consists of two major components, which are stored in
separate subdirectories:

    eifgraph/           Contains EIFGRAPH source files (Eiffel)

        Makefile
        Ace
        eifgraph.1      man page
        Tool/           main cluster
        constructs/     Parsing constructs
        semantics/      Parsing semantics
        emisc/          Miscellaneous Eiffel classes

    eifdiag/            Contains diagram-generating tools
                                (C and shell scripts)
        Makefile
        eifan           Eiffel Analyzer script
        graph2ps        Generate & preview PostScript class diagrams
        graph2oda.c     Generate Object Domain Analyzer class diagrams
        list.c          linked list module source
        list.h          linked list header file
        eifan.1         man pages
        graph2ps.1              
        graph2oda.1

Check the 'Ace' file to make sure it reflects your Eiffel configuration.
I didn't use any precompiled libraries, so you might want to change
it to use your pre-compiled EiffelBase library.  Note that the 'Ace'
file excludes a number of TROOPER classes, and replaces them with
modified versions.  Also, the 'TROOPER' environment variable, which
is defined as part of the TROOPER installation, is used by the 'Ace' file.

Before starting the build, review the makefiles, and make any changes to
suit your environment.  I can't vouch for the portability of the
makefiles.  However, they are pretty simply, so you shouldn't have a
problem getting them work on your system.  If your version of 'make'
supports the '-n' switch, it might be a good idea to use it to see
exactly what actions will be performed.

You should edit the top-level makefile to reflect the desired
installation directory.  The default is '/usr/local'.  This setting is
passed down to the sub-makefiles, so you don't have to edit them.

As mentioned above, the 'graph2ps' script uses GNU GhostScript to preview
the PostScript class diagrams.  If you have a different previewer, or
don't have one at all, you'll need to edit the following lines
in 'graph2ps':

    # PostScript Viewer
    #
    # Uncomment the one you use, or add your own
    #
    #PSVIEW="psview"
    PSVIEW="gs -q"

If everything looks okay, type:

        make all

from the top-level directory.  If everything was configured properly,
this will produce a 'finalized' version of EIFGRAPH in the
'EIFGEN/F_code/eifgraph' directory, with a symbolic link to it in the
'eifgraph' directory.  It should also produce the 'graph2oda' executable
in 'eifdiag'.

The first time the EIFGRAPH tool is run, it will generate an Eiffel
lexical analyzer, and save it to disk under the name 'eiffel_analyzer'.
This takes some time.  On subsequent runs, 'eiffel_analyzer' will
be retrieved from disk.  Generate 'eiffel_analyzer' by running
EIFGRAPH on one of the delivered Eiffel source files, e.g.,

        cd eifgraph
        eifgraph -v Tool/eifgraph.e

The 'EIFLEXER' environment variable should be set to the directory where
'eiffel_analyzer' is stored, which is normally '/usr/local/lib'
(unless you changed the installation directory in the top-level makefile).

If everything seems okay, then type:

        make install

from the top-level directory to copy the executables, man pages, and
other supporting files to the installation directories.  Again, make sure
you install a 'finalized' version of EIFGRAPH for the best performance.

There are other targets in the makefiles which you might find useful.
Take a look and see.

Examples:
========
To generate a PostScript class diagram:

        eifan [-P] file.e

This will create 'file.graph' and 'file.ps', then preview the
PostScript graph via GhostScript (or your configured viewer)

If you want a pretty Booch diagram, you'll have to get and
install Object Domain on a Windows 3.x PC.  Then,

        eifan -O file.e

and bring 'file.oda' to your PC, and import into Object Domain.

There are various ...

read more »



Fri, 08 May 1998 03:00:00 GMT  
 New Eiffel Reverse-Engineering Tool
Bill,

I just sit right in front of my first EIFAN-generated
Booch diagramm.

Installation of EIFAN only got some problems due to
cc, which didn't accept the BOOLEAN definitions in
list.h. Q&D I changed all occurrences of TRUE to 1 and
all of FALSE to 0 (in graph2oda.c).
We work with ISE Eiffel V3.3.7, so the TROOPER patches
on extended_parse refuted the index-clause changes.
Postscript-generation put a little bug in the header,

Quote:
>>showpage, press <return> to continue<<

!PS-Adobe-3.0 EPSF-3.0

must be changed to

%!PS-Adobe-3.0 EPSF-3.0

ACE-file analysis would be a *great* extension to EIFAN.

Thanks for your efforts,

Mario

--------------------------------------------------------------------------------
Mario Winter    FernUniversitaet Hagen     Phone:  +49 2331 987 2129
                Praktische Informatik III  Fax:    +49 2331 987 317
                58084 Hagen, Germany       E-mail:

--------------------------------------------------------------------------------



Sat, 09 May 1998 03:00:00 GMT  
 New Eiffel Reverse-Engineering Tool
Whoops!  I distributed an ISE copyrighted file by mistake
in EIFAN Release 1.0.  I've corrected the problem in Release 1.0A.
Here is an updated README.  Sorry.

        Bill

========================================================================

                EIFAN - Eiffel Analyzer

    Reverse-engineer Eiffel-3 source files
    into object-oriented class diagrams.

Release:        1.0A    November 22, 1995

Author:         Bill O'Keefe

What's Changed Since Release 1.0
================================

Nothing, really.  In Release 1.0, I distributed a slightly modified copy
of an ISE copyrighted file by mistake.  You will now have to make the
change to the 'construct.e' file in the TROOPER 'extended_parse'
directory yourself.  You can either place the edited copy in the
'eifgraph/constructs' directory, or edit the 'Ace' file to no longer
exclude 'construct.e' from the 'parse' cluster.  If you grabbed
EIFAN 1.0, please do not redistribute that file; delete it, and make
the change yourself :-)

The change is to simply remove (or comment out) the call to
'expected_found_error' in 'parse_child'.  

                ******* IMPORTANT**********
If you fail to make this change, the program won't work.

Introduction:
=============

EIFAN is a tool used to reverse-engineer Eiffel source code into Booch
object-oriented class diagrams.  It provides a way to analyze the
overall structure of a set of Eiffel classes without having to
purchase an expensive CASE tool.

EIFAN depends upon several other public domain packages.
Details are provided below.

If you find this program useful, or have any questions/problems
with it, please let me know.

Contents:
========

EIFAN is delivered in a compressed, tar archive, and is available
via ftp at

        ftp://ftp.shore.net/members/bokeefe/eifan/eifan-1.0A.tar.Z

To unpack this file, use:

        zcat eifan-1.0A.tar.Z | tar xvf -

After unpacking the EIFAN distribution, you will find the following files:

        README          This file
        COPYING         Copy of GNU General Public License, version 2
        Makefile        top-level makefile
        eifgraph/       Contains EIFGRAPH source files
        eifdiag/        Contains diagram generating tools

License:
=======

--| Copyright (C) 1995 Bill O'Keefe.
--|
--| This program is free software; you can redistribute it and/or modify
--| it under the terms of the GNU General Public License as published by
--| the Free Software Foundation; either version 2 of the License, or
--| at your option) any later version.
--|
--| This program is distributed in the hope that it will be useful,
--| but WITHOUT ANY WARRANTY; without even the implied warranty of
--| MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
--| GNU General Public License for more details.
--|
--| You should have received a copy of the GNU General Public License
--| along with this program; if not, write to the Free Software
--| Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
--|
--| This program is  based  in part on  TROOPER (Truly Reusable OO
--| Compiler Compiler) written by Jon  Avotins and Glenn Maughan and
--| associated source classes and files, which is distributed free
--| of charge by Monash University, Software Development Department,
--| 900 Danenong Road, Caulfield East, Victoria 3145 Australia,  
--| which may  be obtained by anonymous ftp from the WWW site
--| http://insect.sd.monash.edu.au/~trooper/
--|----------------------------------------------------------------

Dependencies
============

1. TROOPER
----------
The main component of EIFAN is the EIFGRAPH tool.  This tool is
implemented using the Truly Reusable Object Oriented Compiler
Compiler (TROOPER), written by Jon Avotins and Glenn Maughan.  
You need to install TROOPER before you can install EIFGRAPH.
It is available via anonymous FTP at:

    ftp://hornet.sd.monash.edu.au/pub/trooper/trooper.tar.Z

There is also a TROOPER home page at:

    http://insect.sd.monash.edu.au/~jona/trooper.html

TROOPER is based upon the ISE v3.2.8, and should work with
later versions of ISE Eiffel as well.  In fact, I got it to work
with an earlier version (ISE v3.2.3b).  However, I had to manually
apply the 'extended_parse_patch' to my copy of the ISE 'parse' library.
Refer to the TROOPER README files for installation instructions.
It runs much faster if you 'finalize' it.

After installing TROOPER, it's probably a good idea to run a few
Eiffel source files through it, to verify the installation.
You don't need the TROOPER objects or executable to build EIFGRAPH,
since EIFGRAPH simply references the required TROOPER source files from
its Ace file.  Thus, if you're low on disk space, delete the TROOPER
'EIFGEN' directory after you verify its installation.

2. Graphplace
-------------
The 'graphplace' program is a wonderful tool developed by
Jos van Eijndhoven, and is used to layout the nodes of the generated
class diagrams.  It can be downloaded from a number of sites, including:

  ftp://ftp.funet.fi/pub/sci/math/misc/programs/graphplace/graphplace.t...

Refer to the 'graphplace' README for installation instructions.

The EIFAN distribution contains a slightly-modified version of the
'graphplace' PostScript prologue (graphpro.ps), which it installs
in the 'lib' installation directory.  The change limits the maximum size
of node labels to 10 point text.  You can make other changes to this
file as described in the 'graphplace' man page.

3. PostScript Previewer
-----------------------
In order to preview the generated PostScript object-oriented class diagrams,
you will need a PostScript previewer.  By default, EIFAN is configured to
use GNU GhostScript.  If you don't already have GhostScript, you can get
it from:

    ftp://ftp.cs.wisc.edu:/pub/ghost/gnu/ghostscript*NNNN*

where NNNN is the latest version number.  If you have a different previewer,
then simply edit the 'graph2ps' script (described below) to use it.

4. Object Domain
----------------
Object Domain is a shareware CASE tool, written by Dirk Veemeersch, that
runs on a PC under Windows.  It produces nice looking Booch diagrams, and
supports reverse-engineering of C++ source code.  This is accomplished
through its 'C++ Analyzer', which generates Object Domain Analyzer (.oda)
files.  These files can then be imported into Object Domain.  

The EIFAN tool can also generate '.oda' files from Eiffel source code.
If you have a Windows/PC, you might want to give it a try.
The diagrams are more aesthetically pleasing than the PostScript
class diagrams, although they both contain the same information.

You can get Object Domain at a number of sites, including:

    http://www.acs.oakland.edu/oak/SimTel/win3/pgmtools.html

    Select 'dom107.zip'         (or a later version)

Installation:
============
Once all the dependencies have been obtained and installed, you are ready to
install EIFAN.  EIFAN consists of two major components, which are stored in
separate subdirectories:

    eifgraph/           Contains EIFGRAPH source files (Eiffel)

        Makefile
        Ace
        eifgraph.1      man page
        Tool/           main cluster
        constructs/     Parsing constructs
        semantics/      Parsing semantics
        emisc/          Miscellaneous Eiffel classes

    eifdiag/            Contains diagram-generating tools
                                (C and shell scripts)
        Makefile
        eifan           Eiffel Analyzer script
        graph2ps        Generate & preview PostScript class diagrams
        graph2oda.c     Generate Object Domain Analyzer class diagrams
        list.c          linked list module source
        list.h          linked list header file
        eifan.1         man pages
        graph2ps.1              
        graph2oda.1

Check the 'Ace' file to make sure it reflects your Eiffel configuration.
I didn't use any precompiled libraries, so you might want to change
it to use your pre-compiled EiffelBase library.  Note that the 'Ace'
file excludes a number of TROOPER classes, and replaces them with
modified versions.  Also, the 'TROOPER' environment variable, which
is defined as part of the TROOPER installation, is used by the 'Ace' file.

Before starting the build, review the makefiles, and make any changes to
suit your environment.  I can't vouch for the portability of the
makefiles.  However, they are pretty simply, so you shouldn't have a
problem getting them work on your system.  If your version of 'make'
supports the '-n' switch, it might be a good idea to use it to see
exactly what actions will be performed.

You should edit the top-level makefile to reflect the desired
installation directory.  The default is '/usr/local'.  This setting is
passed down to the sub-makefiles, so you don't have to edit them.

As mentioned above, the 'graph2ps' script uses GNU GhostScript to preview
the PostScript class diagrams.  If you have a different previewer, or
don't have one at all, you'll need to edit the following lines
in 'graph2ps':

    # PostScript Viewer
    #
    # Uncomment the one you use, or add your own
    #
    #PSVIEW="psview"
    PSVIEW="gs -q"

If everything looks okay, type:

        make all

from the top-level directory.  If everything was configured properly,
this will produce a 'finalized' version of EIFGRAPH in the
'EIFGEN/F_code/eifgraph' directory, with a symbolic link to it in the
'eifgraph' directory.  It should also produce the 'graph2oda' executable
in 'eifdiag'.

The first time the EIFGRAPH tool is run, it will generate an Eiffel
lexical analyzer, and save it to disk under the name 'eiffel_analyzer'.
This takes some time.  On subsequent runs, 'eiffel_analyzer' will
be retrieved from disk.  Generate 'eiffel_analyzer' by running
EIFGRAPH on one of the delivered Eiffel source files, e.g.,

        cd eifgraph
        eifgraph -v Tool/eifgraph.e

The 'EIFLEXER' environment variable should be set to the directory where
'eiffel_analyzer' is stored, which is normally '/usr/local/lib'
(unless you changed the installation directory in the top-level makefile).

If everything seems okay, then type:

        make install

from the top-level directory to copy the executables, man pages, and
other supporting files to the installation directories.  Again, make sure
you install a 'finalized' version of EIFGRAPH for ...

read more »



Sun, 10 May 1998 03:00:00 GMT  
 
 [ 7 post ] 

 Relevant Pages 

1. EIFAN - New Eiffel Reverse-Engineering Tool

2. new reverse engineering tool

3. new version of FORTRAN reverse engineering tool

4. PYUT : new Python UML Tool (next : reverse engineering)

5. Re-Engineering/Reverse Engineering Tool

6. Re-Engineering/Reverse Engineering Tool

7. Documentations and Reverse Engineering Tools

8. VW reverse engineering tool exists?

9. Reverse Engineering Tools

10. Wanted - Reverse Engineering Tools / CASE that handles Rexx

11. Announce: Understand for Ada 1.3B67 - reverse engineering tool for Ada 83/95

12. Ada Reverse Engineering Tool Sought

 

 
Powered by phpBB® Forum Software