#!/bin/bash

# Description : A jiffy to search CTAN file list
# Author      : Anh K. Huynh <[email protected]>
# Version     : $Id: ctan 82 2009-09-30 01:51:59Z pi $
# License     : LPPL (any version)
# Homepage    : CTAN:/support/ctan.tools
# Document    : self documented, try `$0 about`
# Requirements: bash, lynx, wget, tar, bzip2,
#               sed, sort, awk, (e)grep, cut
# TODO        : fancy get options (code/interface)

export PATH=$PATH:$HOME/bin

CTAN_MIRROR=http://mirror.ctan.org/
CTAN_ARCHIVE=http://ctan.org/tex-archive          # to be updated
CTAN_CONTRIB=$CTAN_ARCHIVE/macros/latex/contrib/  # to be updated
CTAN_FILES=$CTAN_ARCHIVE/FILES.byname             # to be updated
VIETTUG_CACHEDIR=http://viettug.org/ctan.tools/
DESTDIR=$HOME/.ctan/
TOOLS="lynx wget tar bzip2 sed sort awk grep"
HAVE_LZMA=0

# functions

msg()
{
 echo -e ":: $*"
}

# check for executable tool

ctan_check()
{
 type $1 2>/dev/null 1>&2
 if [ $? -gt 0 ]; then
   ERROR=1
   msg "not found '$1'"
   return 1
 fi
}

# check for all tools

ctan_check_all()
{
 msg "checking dependencies..."
 ERROR=0
 for tool in $TOOLS; do
   ctan_check $tool
 done
 if [ $ERROR = 1 ]; then
   msg "error occured. exit 1"
   exit 1
 fi
 msg "tools found: $TOOLS"

 ctan_check lzma
 if [ $? = 0 ]; then
   HAVE_LZMA=1
 fi

 return 0
}

# mirror detection
# see also http://viettug.org/blogs/show/310

ctan_mirror_select()
{
 msg "Hello, this is ctan_mirror_select"
 src="$1"
 msg "found $src"
 while true; do
   location="`wget --max-redirect=0 $src -O - 2>&1 1>/dev/null| grep Location: 2>/dev/null`"
   if [ $? -eq 0 ]; then
     src="`echo $location | cut -d' ' -f2`"
     msg "found $src"
   else
     break
   fi
 done
 CTAN_ARCHIVE="$src"
 CTAN_CONTRIB="$CTAN_ARCHIVE/macros/latex/contrib/"
 CTAN_FILES="$CTAN_ARCHIVE/FILES.byname"
 msg "mirror detected: $CTAN_ARCHIVE"
}

# show information after update

ctan_info()
{
   msg "* `wc -l $DESTDIR/packages.txt |awk '{print $1}'` package(s) found"
   msg "* `wc -l $DESTDIR/files.txt |awk '{print $1}'` file(s) found"
}

# update database

ctan_update()
{
 if [ "x$1" = "x--first-time" ]; then
   shift
   msg "This maybe the first time you use the tool."
   msg "The directory ~/.ctan will be created"
 fi

 mkdir -pv $DESTDIR

 # using the cache file http://kyanh.net/ctan.tools/cache.tbz

 if [[ ( "x$1" == "x" ) || ( "x$1" == "x--use-cache" ) ]]; then
   shift
   msg "update by using cache file..."

   if [ $HAVE_LZMA = 1 ]; then
     msg "getting $VIETTUG_CACHEDIR/ctan.tools/cache.tlzma..."
     wget $VIETTUG_CACHEDIR/cache.tlzma -O $DESTDIR/cache.tlzma

     msg "uncompressing $DESTDIR/cache.tlzma..."
     lzma --decompress < $DESTDIR/cache.tlzma | tar xvf - -C $DESTDIR
   else
     msg "getting $VIETTUG_CACHEDIR/ctan.tools/cache.tbz..."
     wget $VIETTUG_CACHEDIR/cache.tbz -O $DESTDIR/cache.tbz

     msg "uncompressing $DESTDIR/cache.tbz..."
     tar xfjv $DESTDIR/cache.tbz -C $DESTDIR
   fi

   msg "now remove some big files"
   rm -fv $DESTDIR/{dump_files.txt,cache.tbz,cache.tlzma}

   ctan_info

   return 0
 fi

 # updating by downloading database from CTAN
 # previous version uses a search result returned by CTAN
 # since 1.5.0, Jim Hefferon suggest to use FILES.byname
 # see also (ctan_joiner)

 msg "update will get about 8MB from CTAN and may take very long time."
 read -n1 -p":: Are you sure you wanna do this (y/N)? " reply
 echo ""
 if [ ! "x$reply" = "xy" ]; then
   msg "Update was cancelled."
   return 0
 fi

 ctan_mirror_select $CTAN_MIRROR

 msg "searching all files. please wait as this is a heavy task"
 lynx -dump $CTAN_FILES > $DESTDIR/tmp0.txt
 # cp $HOME/projects/ctan.tools/FILES.byname $DESTDIR/tmp0.txt
 if [ $? -gt 0 ]; then
   msg "failed to download from CTAN. exit 1"
   rm -fv $DESTDIR/tmp0.txt
   exit 1
 fi

 msg "getting list of LaTeX packages..."
 ctan_joiner $DESTDIR/tmp0.txt \
   |sed -e 's# ##g' \
   > $DESTDIR/files.txt
 pkg_filter='|macros/latex/contrib/'
 grep $pkg_filter $DESTDIR/files.txt \
   |grep -v 'info/' \
   |grep -E '\.(zip|sty|dtx)' \
   |sed -e 's#/[^/]*$##g' \
   |sed -e "s#$pkg_filter#|#g" \
   |sort -t'|' -uk3 \
   > $DESTDIR/packages.txt
 msg "removing temporary file..."
 rm -fv $DESTDIR/tmp0.txt

 msg "compressing the results..."
 cd $DESTDIR

 msg "creating tbz archive..."
 tar -cvj -f .cache.tbz {files,packages}.txt
 mv .cache.tbz cache.tbz

 if [ $HAVE_LZMA = 1 ]; then
   msg "creating lzma archive..."
   tar -cv --use-compress-program lzma -f .cache.tlzma {files,packages}.txt
   mv .cache.tlzma cache.tlzma
 fi

 ctan_info

 if [[ "x$1" == "x--keep-only-cache" ]]; then
   msg "removing all temporary files. Only cache is kept"
   rm -fv $DESTDIR/*.txt
 fi
}

# search something

ctan_search()
{
 [ -d $DESTDIR/ ] || ctan_update --first-time --use-cache
 if [ "x$1" = "x-file" ]; then
   shift
   grep $* $DESTDIR/files.txt \
     |grep -v obsolete \
     |awk -F '|' '{printf("%s | %s\t| %s\n",$1, $2, $3)}'
 else # search package name
   grep $* $DESTDIR/packages.txt \
     |grep -v obsolete \
     |awk -F '|' '{printf("%s | %s\t| %s\n",$1, $2, $3)}'
 fi
 if [ $? -gt 0 ]; then
   msg "no package/file found"
   return 1
 fi
}

# download files

ctan_get()
{
 RETFOUND=$DESTDIR/tmp2.txt
 RETDOWN=$DESTDIR/tmp.txt
 ctan_search $* > $RETFOUND

 if [ $? -gt 0 ]; then
   msg "no package matches '$*'"
   return
 fi
 if [ "x$1" = "x-file" ]; then
   shift
   ctan_grep_type='file'
 else
   ctan_grep_type='pkg'
 fi

 cat $RETFOUND
 awk -F'|' '{print $3}' $RETFOUND > $RETDOWN

 if [ "$ctan_grep_type" = "file" ]; then
   read -n 1 -p ":: Are you sure you want to download (y/N)? " download
   echo ''
   if [[ "x$download" == "xy" ]]; then
     ctan_mirror_select $CTAN_MIRROR
     for pkg in `cat $RETDOWN`; do
       msg "downloading $pkg..."
       wget -nv -c "$CTAN_ARCHIVE/$pkg";
     done
   fi
 else
   read -n 1 -p "Are you sure you want to download (y/N)? " download
   echo ''
   if [[ "x$download" == "xy" ]]; then
     ctan_mirror_select $CTAN_MIRROR
     for pkg in `cat $RETDOWN`; do
       msg "downloading ${pkg}.zip..."
       wget -nv -c "$CTAN_CONTRIB/${pkg}.zip";
     done
   fi
 fi
}

#
# awk script to join lines in FILES.byname
# some server wraps long lines hence we need this script
#
ctan_joiner()
{
 awk '
   BEGIN {
     a = ""
   }
   {
     if ( $0 ~ /^[0-9]+/ ) {
       if (a != "") print a
       a=$0
     }
     else{
       a = (a $0)
     }
   }
   END {
     if ( $0 !~ /^[0-9]+/ ) {
       a = (a $0)
     }
     else {
       a = $0
     }
     if (a != "") print a
   }' \
   $*
}

ctan_about()
{
 msg "ABOUT"
 msg "\tThis jifty tool helps you to seach/download LaTeX packages listed at"
 msg "\t\t$CTAN_CONTRIB"
 msg "\tor search/download any files in $CTAN_ARCHIVE (or its mirror)."
 msg ""
 msg "\tThe tool requires following programs"
 msg "\t\t$TOOLS"
 msg
 msg "\tThis tool was written by Anh K. Huynh <[email protected]>."
 msg "VERSION"
 msg "\t1.0.0 2008/05/11: first version"
 msg "\t1.1.0 2008/05/15: search files (use cache file)"
 msg "\t1.2.0 2008/05/16: smaller cache. Thanks to Karl Berry <[email protected]>"
 msg "\t1.3.0 2008/05/23: everything is cached. Faster. New options. Thanks to texer."
 msg "\t1.4.0 2009/09/12: auto. select ctan mirror (suggested by Jim Hefferon)"
 msg "\t1.5.0 2009/09/13: load database from http://mirror.ctan.org/FILES.byname"
 ctan_usage
 msg "TODO"
 msg "\t* package information supported"
 msg "\t* search by package description,..."
 msg "\t* package build script"
 msg "THANKS to"
 msg "\t* Karl Berry <[email protected]>"
 msg "\t* Nguyen Van Hanh <[email protected]>"
 msg "\t* Jim Hefferon <[email protected]>"
 msg "LICENSE"
 msg "\tThis tool is published under LPPL."
 msg "BUGS"
 msg "\tPlease report to kyanh <[email protected]>."
}

ctan_usage()
{
 msg "USAGE"
 msg "\tctan about           : show all information about this tool"
 msg "\tctan usage           : show usage"
 msg "\tctan version         : show script version"
 msg
 msg "\tctan update          : update using $VIETTUG_CACHEDIR/cache.tbz (~1MB)"
 msg "\tctan update --direct : update directly from CTAN. You are going to download 8MB"
 msg "\tctan grep <string>   : search packages match <string>. grep ability is supported"
 msg "\tctan get  <string>   : download packages match <string> to working directory"
 msg
 msg "\tIf you want to search files:"
 msg
 msg "\tctan fgrep <string>  : search files."
 msg "\tctan fget  <string>  : download files match <string> to working directory"
 msg "\tctan grep -file <string>"
 msg "\tctan get  -file <string>"
 msg "EXAMPLES"
 msg "\tctan grep theorem    # search packages match 'theorem'"
 msg "\tctan grep ^n         # search packages srated by 'n'"
 msg "\tctan get  ^n         # download packages started by 'n'"
 msg ""
 msg "\tctan get -file contrib/ntheorem.zip"
 msg "\tctan fget      contrib/ntheorem.zip"
 msg "\tctan fgrep     ntheorem | grep zip"
}

ctan_arg()
{
 if [ "x$1" = "x-file" ]; then
   shift
 fi
 if [ "x$1" = "x" ]; then
   msg "missing parameter"
   exit 1
 fi
}

# check for requirements or not

not_check="xusage:xhelp:xabout:xdoc:xversion:xcheck:"
if ! `echo $not_check | grep -q "x$1"`; then
 ctan_check_all
fi

# main program

case "x$1" in
"xupdate")
 shift
 ctan_update $*
 ;;
"xfgrep")
 shift
 ctan_arg $*
 ctan_search -file $*
 ;;
"xgrep")
 shift
 ctan_arg $*
 ctan_search $*
 ;;
"xget")
 shift
 ctan_arg $*
 ctan_get $*
 ;;
"xinfo")
 shift
 ctan_info
 ;;
"xfget")
 shift
 ctan_arg $*
 ctan_get -file $*
 ;;
"x")  ctan_usage;;
"xusage") ctan_usage;;
"xhelp") ctan_usage;;
"xdoc") ctan_about;;
"xabout") ctan_about;;
"xversion") msg '$Id: ctan 82 2009-09-30 01:51:59Z pi $';;
"xcheck") ctan_check_all;;
*)
 msg "wrong parameter. please try 'ctan usage'"
 ;;
esac

You are viewing proxied material from ftp.icm.edu.pl. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.