🚀 go-pugleaf

I'm the original author of mkid, and I have resumed development and
maintenace.  To bring you up-to-date on what mkid is and where it's
going, I have appended the README, NEWS and TODO files.

The code is fairly stable, but it's alpha because there are still
portability problems.  I only have access to Linux, so I need help
from others running different systems.

Since I don't have upload access to an ftp archive site, I must email
the release to you.  It's about 247 Kb tarred/gzipped/uuencoded.  If
you can offer an anonymous ftp site for the alpha releases, that would
be great.

Please respond via email.

Thanks,

--gkm

------------------------------------------------------------------------------
`mkid' README

`mkid' is a simple, fast, high-capacity, language-independent
identifier database tool.  Actually, the term `identifier' is too
limiting--`mkid' stores tokens, be they program identifiers of any
form, literal numbers, or words of human-readable text.  Database
queries can be issued from the command-line, or from within emacs,
serving as an augmented tags facility.

`mkid' was originally written by Greg McGary <gkm@magilla.cichlid.com>
and posted to comp.sources.unix in September 1987.  It was then
maintained and enhanced by a loose knit group of programmers on the
Internet led by Tom Horsley <Tom.Horsley@mail.hcsc.com>.  Tom released
`mkid2' on comp.sources.unix in March, 1991.  Since then, Greg McGary
has resumed maintenance and is releasing an improved version 3 under GPL.

Version 3 is an interim release.  Version 4 will follow in the coming
months and include a cscope clone plus other improvements.

See file `COPYING' for copying conditions.
See file `INSTALL' for compilation and installation instructions.
See file `NEWS' for a list of major changes in the current release.
See file `THANKS' for a list of contributors.
See file `TODO' for planned enhancements.

Send bug reports to <gkm@magilla.cichlid.com>.  Please describe your
problem adequately: your input, what you expected, what you got, and
why it is wrong.  Diffs are welcome, but please include a description
of the problem as well, since this is sometimes difficult to infer.
------------------------------------------------------------------------------
`mkid' NEWS - User visible changes.

Version 3.0 - February 1995, by Greg McGary

* `mkid' uses a new efficient algorithm for building the database.  The old
algorithm was O(n^2) for space and would exhaust memory or lead to page
thrashing on very large programs.  The new algorithm is linear with the
size of the input for both time and space.  It is now possible to build
very large `mkid' databases covering many thousands of files totalling tens
of megabytes and containing hundreds of thousands of unique tokens.

* The database header is now byte-order and word-size independent.  ID
files may be shared across networks of dissimilar architectures.

Version 2.0 - March 1991, by Tom Horsley


Version 1.0 - September 1987, by Greg McGary

* Initial release to comp.sources.unix
------------------------------------------------------------------------------
`mkid' TODO - Future directions

* Bring all code into conformance with GNU project coding standards.
  - Use command-line arguments, not the name of the program in argv[0]
    to determine lid's behavior
  - Use getopt_long () to parse command-line arguments.

* Portability
  - Fix filenames.c to handle non-UNIX file names properly.
  - Fix sizeof (int) == sizeof (long) assumptions.
  - The configure script tests for lots of things, but the code doesn't
    use all of the results that it should.

* lid
  - change the name of the program to something a bit more
    descriptive, e.g., "findid".
  - use "long long" (if the compiler supports it) integer comparisons
    in find_number().
  - generalize the command-line interface for different behaviors.
    User passes a command-format string for operating on the results
    of a query:
    %t = token, %f = crunched file list, %F = uncrunched file list
    Recommend the following aliases to emulate old behavior:
      alias lid="findid -e 'echo %t %f'"
      alias eid="findid -e 'vi +/\\<%t\\>/ %F'" # for vi users
      alias gid="findid -e 'egrep \\<%t\\> %F'"

* mkid & lid
  - store & retrieve floating point literals

* mkid
  - add a recursive file tree walker, so a complete list of file names
    need not be supplied.
  - Store attributes with symbols: type, func, variable struct-tag
    enum-tag, keyword, etc.  Store caller/callee relationships between
    identifiers.  Indicate which files have defns vs. uses.  This info
    is needed to support a cscope interface.

* cscope
  - add a cscope work-alike query interface (this is being done by
    Maureen Lecuona <lecuona@paul.rutgers.edu>)

* grep
  - Hack GNU grep to use a mkid database for hints about which files to
    search. 
------------------------------------------------------------------------------