Usenet 1994 January/usenetsourcesnewsgroupsinfomagicjanuary1994.iso/sources/misc/volume23/texindex/part01

Text File | 1991-09-23 | 7.8 KB | 317 lines

Newsgroups: comp.sources.misc
From: [e-mail] (Thomas Koenig)
Subject:  v23i006:  texindex - generate index for LaTeX', Part01/01
Message-ID: <[e-mail]>
X-Md4-Signature: fbb6fd82a6b8a8afe622068d4812596b
Date: Sun, 22 Sep 1991 20:12:18 GMT
Approved: [e-mail]

Submitted-by: [e-mail] (Thomas Koenig)
Posting-number: Volume 23, Issue 6
Archive-name: texindex/part01
Environment: UNIX

This is texindex, a little utility for formatting LaTeX - generated
*.idx (indexentry) - files into *.ind - files ready for inclusion
as an index.

It is written as a UNIX shell script, using the utilities expand, sed, sort,
uniq and awk (old style).

It has the following features:

      - parts enclosed in brackets [...] will be enclosed in \verb+...+
    in the *.ind - file

      - entries consisting of several words will be rotated.  If, for
        example, you have \index{foo bar} in your *.tex file,
    you'll find both "\item foo bar <page>" and
    "\item bar \subitem foo <page>" in your *.ind - file.
    This can be suppressed by inserting tildes between the words,
    as in \index{foo~bar}.

      - Commas can be used to get deeper in the \item hierarchy.

Credits: The idea and some of the implementation comes from Aho, Weinberger
and Kernighan: 'The Awk Programming Language'.

I place this in the public domain; do with it whatever you please.

Thomas Koenig
[e-mail], [e-mail],
[e-mail] (for now)

#---------------------------------- cut here ----------------------------------
# This is a shell archive.  Remove anything before this line,
# then unpack it by saving it in a file and typing "sh file".
#
# Wrapped by <[e-mail]> on Thu Sep 19 21:16:45 1991
#
# This archive contains:
#    README        texindex    ix.collapse    ix.extract
#    ix.format    ix.genkey    ix.polish    ix.rotate
#
# Error checking via wc(1) will be performed.

unset LANG

echo x - README
cat >README <<'@EOF'
This is texindex, a little utility for formatting LaTeX - generated
*.idx (indexentry) - files into *.ind - files ready for inclusion
as an index.

It is written as a UNIX shell script, using the utilities expand, sed, sort,
uniq and awk (old style).

It has the following features:

      - parts enclosed in brackets [...] will be enclosed in \verb+...+
    in the *.ind - file

      - entries consisting of several words will be rotated.  If, for
        example, you have \index{foo bar} in your *.tex file,
    you'll find both "\item foo bar <page>" and
    "\item bar \subitem foo <page>" in your *.ind - file.
    This can be suppressed by inserting tildes between the words,
    as in \index{foo~bar}.

      - Commas can be used to get deeper in the \item hierarchy.

Credits: The idea and some of the implementation comes from Aho, Weinberger
and Kernighan: 'The Awk Programming Language'.

I place this in the public domain; do with it whatever you please.

--
Thomas Koenig
[e-mail], [e-mail],
[e-mail] (for now)
@EOF
set `wc -lwc <README`
if test $1$2$3 != 301671068
then
    echo ERROR: wc results of README are $* should be 30 167 1068
fi

chmod 644 README

echo x - texindex
cat >texindex <<'@EOF'
expand $1.idx |
sed -f ix.extract |
sort -t    +0 -1 +1n -2 |
uniq |
awk -f ix.collapse |
awk -f ix.rotate |
sed -f ix.genkey |
sort -f -d |
awk -f ix.format |
sed -f ix.polish > $1.ind
@EOF
set `wc -lwc <texindex`
if test $1$2$3 != 1041183
then
    echo ERROR: wc results of texindex are $* should be 10 41 183
fi

chmod 755 texindex

echo x - ix.collapse
cat >ix.collapse <<'@EOF'
# ix.collapse : awk - script
# kollabiere Zahlenlisten fuer identische Begriffe
# Eingabeformat :
# string1 \t num
# string2 \t num
# string2 \t num
# ...
# Ausgabeformat:
# string1 \t num
# string2 \t num, num
BEGIN    {
        altwort = "ASDF!@#$*&%$IU11@"
        FS = OFS = "\t"
    }
$1 != altwort {
        if (NR > 1)
            printf "\n"
        altwort = $1
        printf "%s\t%s", $1, $2
        next
    }
    {
        printf " %s",$2
    }
END    {
        if (NR > 1)
            printf "\n"
    }
@EOF
set `wc -lwc <ix.collapse`
if test $1$2$3 != 2884434
then
    echo ERROR: wc results of ix.collapse are $* should be 28 84 434
fi

chmod 644 ix.collapse

echo x - ix.extract
cat >ix.extract <<'@EOF'
# ix.extract : sed - script, um die Daten aus einem LaTeX - *.idx - File
# rauszuholen.
# Eingabeformat:
# \indexentry{Begriff}{Seitenzahl}
# Ausgabeformat:
# Begriff \t Seitenzahl
s/\\indexentry{\(.*\)}{\(.*\)}/\1    \2/
@EOF
set `wc -lwc <ix.extract`
if test $1$2$3 != 730219
then
    echo ERROR: wc results of ix.extract are $* should be 7 30 219
fi

chmod 644 ix.extract

echo x - ix.format
cat >ix.format <<'@EOF'
# ix.format : awk - script
# entfernt Sortierschluessel, fuegt \item etc. ein
BEGIN    {
        FS = OFS = "\t"
        printf "\\begin{theindex}"
    }
    {
        n = split($2,feld,", ")
        if (n > 3)
        {
            for (i=4; i <=n; i++)
                feld[3] = feld[3] ", " feld[i]
            n = 3
        }
        for (i = 1; feld[i]==altfeld[i] && i<=n ; i++)
            ;
        for (k=i; k <=n; k++)
        {
            printf "\n\\"
            for (j=1; j<k; j++)
                printf "sub"
            printf "item %s ",feld[k]
        }
        nzahl = split($3,zahlfeld," ")
        for (j=1; j<nzahl; j++)
            printf"%s, ",zahlfeld[j]
        printf "%s",zahlfeld[nzahl]
        for (j=1; j<=i; j++)
            altfeld[i] = feld[i]
    }
END    {
        printf "\n\\end{theindex}\n"
    }
@EOF
set `wc -lwc <ix.format`
if test $1$2$3 != 33102635
then
    echo ERROR: wc results of ix.format are $* should be 33 102 635
fi

chmod 644 ix.format

echo x - ix.genkey
cat >ix.genkey <<'@EOF'
# ix.genkey : Generierung eines Sortierschluessels ohne Sonderzeichen
# Eingabe: string1 string2 ... \t num num ...
# Ausgabe: string1 string2 ... (ohne S.Z.) \t string1 string2 ... \t num num ..
#
# Ersetze global die Tilde durch Spaces
s/~/ /g
# kopiere die Zeile in den Zwischenpuffer
h
# alles hinter \t loeschen
s/    .*$//
# zum Sortieren loesche alle nicht - alphanumerischen Zeichen
s/[^a-zA-Z0-9 ]//g
# falls dabei eine leere Zeile rauskommt, setze ein Space rein
s/^$/ /
# kopiere die urspruengliche Zeile aus dem Puffer hinter den Speicher
G
# ersetze newline zwischen den Zeilen durch \t
s/\n/    /
# Falls ein Komma am Ende des Strings steht, loesche es
s/,    /    /
@EOF
set `wc -lwc <ix.genkey`
if test $1$2$3 != 20117669
then
    echo ERROR: wc results of ix.genkey are $* should be 20 117 669
fi

chmod 644 ix.genkey

echo x - ix.polish
cat >ix.polish <<'@EOF'
# ix.polish : sed - script zum letzten Nachpolieren der Eintraege
# \item etc. einruecken, eckige Klammern in \verb - Anweisungen abaendern
#
# Packe vor jedes \item ein \indexspace
/\\item/i\
  \\indexspace
# ruecke \items ein
/^\\item/s/^/  /
# ruecke \subitems etwas mehr ein
/^\\subitem/s/^/    /
# ruecke \subsubitems noch etwas mehr ein
/^\\subsubitem/s/^/      /
# aendere [ in \verb+
s/\[/\\verb+/g
# aendere ] in +
s/\]/+/g
@EOF
set `wc -lwc <ix.polish`
if test $1$2$3 != 1667433
then
    echo ERROR: wc results of ix.polish are $* should be 16 67 433
fi

chmod 644 ix.polish

echo x - ix.rotate
cat >ix.rotate <<'@EOF'
# ix.rotate: awk - script zur Rotation von Index - Begriffen
# Eingabe:
# string1 string2 string3 ... \t num num ...
# Ausgabe:
# string1 string2 string3 ... \t num num ...
# string2 string3 ... , string1 \t num num ...
# string3 ..., string1 string2 \t num num ...
#
BEGIN    {
        FS = OFS = "\t"
    }
    {
        print $1, $2 # Drucke die unrotierte Form
        for (i=1; (j=index(substr($1, i+1)," "))> 0;)
        {
            i += j # Finde jeden Blank und rotiere um ihn
            printf "%s, %s\t%s\n",substr($1,i+1),substr($1,1,i-1),$2
        }
    }
@EOF
set `wc -lwc <ix.rotate`
if test $1$2$3 != 1993513
then
    echo ERROR: wc results of ix.rotate are $* should be 19 93 513
fi

chmod 644 ix.rotate

exit 0

exit 0 # Just in case...
--
Kent Landfield                   INTERNET: [e-mail]
Sterling Software, IMD           UUCP:     uunet!sparky!kent
Phone:    [phone]         FAX:      [phone]
Please send comp.sources.misc-related mail to [e-mail].