DataMuseum.dk

Presents historical artifacts from the history of:

DKUUG/EUUG Conference tapes

This is an automatic "excavation" of a thematic subset of
artifacts from Datamuseum.dk's BitArchive.

See our Wiki for more about DKUUG/EUUG Conference tapes

Excavated with: AutoArchaeologist - Free & Open Source Software.


top - metrics - download
Index: I T

⟦2fc21a205⟧ TextFile

    Length: 5844 (0x16d4)
    Types: TextFile
    Names: »INSTALL«

Derivation

└─⟦52210d11f⟧ Bits:30007239 EUUGD2: TeX 3 1992-12
    └─⟦c319c2751⟧ »unix3.0/TeX3.0.tar.Z« 
        └─⟦036c765ac⟧ 
            └─⟦this⟧ »TeX3.0/TeXcontrib/tib/INSTALL« 
└─⟦060c9c824⟧ Bits:30007080 DKUUG TeX 2/12/89
    └─⟦this⟧ »./tex82/TeXcontrib/tib/INSTALL« 
└─⟦52210d11f⟧ Bits:30007239 EUUGD2: TeX 3 1992-12
    └─⟦63303ae94⟧ »unix3.14/TeX3.14.tar.Z« 
        └─⟦c58930e5c⟧ 
            └─⟦this⟧ »TeX3.14/TeXcontrib/tib/INSTALL« 

TextFile

Tib installation

Tape: 9 track, 1600 6250 bpi, variable record length, 
      unlabelled, block size 512 8000
      ANSI (DEC File-11), labelled "tib", block size 2048

Source language: C

List of files attached at end.

There are four directories needed,
    1. source listing directory,
    2. macro directory,
    3. documentation directory,
    4. index directory.

It is suggested these be subdirectories of a directory `tib'.
(Alternately they can all be the same directory.)

1. Dump the tape into the source listing directory.
       a. Dump the first file with the unix instruction
            dd if=<tape unit> of=dump
          where <tape unit> refers to the device on which the tape is mounted.
          Do *not* rewind the tape.
       b. Edit this file "dump", replacing all "tapeunit" by the tape unit.
       c. Make "dump" executable, and execute it.  It should dump the tape.
       d. The final file vitaMake has its name at the top.  Take a look.
   OR
       a. Dump the single file with the unix instruction
            dd if=<tape unit> of=abc ibs=8000
          This creates a file abc (size ~630K) which is a shell script for 
          writing all the tib files.  It uses the unix commands sh, echo, ls,
          sed, chmod.
       b. Enter
            sh abc
          to write the tib files.
       c. Discard files abc and 0.

2. With a text editor, change (once in each appropriate file except as noted):
       a. tibdoc.tex --- A at top, B at top, C at top,
       b. tib.man    --- A (5 times), B, C, E, F,
       c. intro.tib  --- A (3 times),
       d. tib.h      --- B (2 times), C, F,
       e. Makefile   --- D (line 3).

   A. string AAAAA to name of documentation directory (absolute path),
   B. string BBBBB to name of macro directory (absolute path),
   C. string CCCCC to name of common-words file (absolute path),
      (see 8 below, `common' is supplied common-words file),
   D. string DDDDD to name of system bin directory (absolute path),
   E. string EEEEE to name of source directory (absolute path),
   F. string FFFFF to name of system index file (absolute path)
      (see 12 below).
      (Note: CCCCC and FFFFF are files; the others are directories.)

3. Make certain paths to temporary files in  #define  statements
   are correct in tib.h (they are set to /tmp/)  (there are
   three of them).

4. Make certain paths to system descriptions in  #include  statements
   are correct in tibargs.c (#include <sys/types.h>, #include <sys/stat.h>).

5. Move the following files to the documentation directory:
   tibdoc.tex, intro.tib, test.tex, test.ref, Styles.

6. Create tib.cat (`make manual') and install tib.man, tib.cat in
   manual directories.  If desired, move tib.man and/or tib.cat to 
   documentation directory.

7. Move all files with extensions tib, ttz, ttx, to the macro directory.

8. Move common to the index directory.

9. Compile and link the programs.  Single command `make' should do it.
   4 executables should result: tib, tibdex, tiblist, tiblook
   (tiblook uses unix getopt; see instructions at beginning of source).

10. Test
       a. Make a test directory.  To it copy INDEX, test.tex and test.ref---
          the last two from the documentation directory.
       b. Run
                         tib test.tex
          This uses INDEX and should create a TeX file.
          (If you get a "INDEX not up to date" message when you test tib, it 
          is because tib is comparing the last modify dates of test.ref and 
          INDEX (with chkindex). Just put INDEX in and out of a text editor 
          (touch) to give it a new date.)
          Run (plain) TeX on the output file as usual.
       c. Run
                         tiblook
          Enter derham when it asks for you to enter something.
          This also uses INDEX.
       d. Run
                         tiblist test.ref
          Run plain TeX on the output.
       e. Rename INDEX.  Run
                         tibdex test.ref
          The output of tibdex is a new INDEX which should be
          identical to the one supplied for testing purposes.
       f. If everything works, the directory can be discarded.

11. Install the programs.  Command `make install' should do it.

12. Change directory to index directory.  Install system reference
    file.  Apply tibdex to this reference file.  If no system reference
    file is to be used, create dummy index by applying tibdex to empty file.

13. Clean up as desired (delete object files?).

14. Advertise <documentation directory>/intro.tib to TeX users.

15. Comments:
       a. The siam format styles (siamd.tib, siamn.tib, siamdl.tib, 
          siamnl.tib) contain options on fonts (SIAM specifies smaller 
          type for bibliographies).  Check if you want to modify.
       b. On some machines, the stack can overflow in locate.c from `refs'.
          I believe making `refs' a global variable should not cause problems.
          Put the line `struct reftype refs[MAXREFS]...' after the 
          `static struct reftype' declaration.
       c. In numo.tib and numos.tib, the font ammi10 is used.  For some
          sites, this should be cmmi10.
       d. Files ieeabb.ttz and ieefll.ttz are included, but not documented.
          They consist of journal names and abbreviations for ieee journals.
          They can be installed in the macro subdirectory and made available
          with either the -i flag on the call to tib or by modifying .tib 
          files.
       e. A small unix utility named tibabb is included, but not documented.
          It should be made executable and called as `tibabb <word>'.  
          It will exhibit all entries in the journal file(s) containing 
          that word, so that the appropriate definition code can be used.  
          It should be modified to local use.