path: root/einfo

#!/bin/sh

# ===========================================================================
#
#                            PUBLIC DOMAIN NOTICE
#            National Center for Biotechnology Information (NCBI)
#
#  This software/database is a "United States Government Work" under the
#  terms of the United States Copyright Act.  It was written as part of
#  the author's official duties as a United States Government employee and
#  thus cannot be copyrighted.  This software/database is freely available
#  to the public for use. The National Library of Medicine and the U.S.
#  Government do not place any restriction on its use or reproduction.
#  We would, however, appreciate having the NCBI and the author cited in
#  any work or product based on this material.
#
#  Although all reasonable efforts have been taken to ensure the accuracy
#  and reliability of the software and data, the NLM and the U.S.
#  Government do not and cannot warrant the performance or results that
#  may be obtained by using this software or data. The NLM and the U.S.
#  Government disclaim all warranties, express or implied, including
#  warranties of performance, merchantability or fitness for any particular
#  purpose.
#
# ===========================================================================
#
# File Name:  einfo
#
# Author:  Jonathan Kans, Aaron Ucko
#
# Version Creation Date:   04/04/2020
#
# ==========================================================================

pth=$( dirname "$0" )

case ":$PATH:" in
  *:"$pth":* )
    ;;
  * )
    PATH="$PATH:$pth"
    export PATH
    ;;
esac

# conditionally execute original Perl implementation

PERL=""

internal=false
while [ "$#" -ne 0 ]
do
  case "$1" in
    -internal )
      internal=true
      shift
      ;;
    -newmode )
      USE_NEW_EDIRECT=1
      shift
      ;;
    -oldmode )
      USE_NEW_EDIRECT=0
      shift
      ;;
    * )
      break
      ;;
  esac
done
if [ "$internal" = true ]
then
  set _ -internal "$@"
  shift
fi

if [ ! -f "$pth"/ecommon.sh ]
then
  USE_NEW_EDIRECT=false
fi

case "${USE_NEW_EDIRECT}" in
  [FfNn]* | 0 | [Oo][Ff][Ff] )
    # set PERL path if using old EDirect
    PERL=perl
    case "$( uname -s )" in
      CYGWIN_NT* )
        # Use a negative match here because the shell treats 0 as success.
        if perl -e 'exit $^O !~ /^MSWin/'; then
           pth=$( cygpath -w "$pth" )
        fi
        ;;
      Darwin )
        PERL="/usr/bin/perl"
        ;;
    esac
    ;;
  "" | * )
    ;;
esac

if [ -n "${PERL}" ]
then
  exec "${PERL}" "$pth"/edirect.pl -info "$@"
  exit 0
fi

# handle common flags - dot command is equivalent of "source"

. "$pth"/ecommon.sh

# help texts

PrintHelp() {

  echo "einfo $version"
  echo ""
  sfx=""
  if [ "$external" = true ]
  then
    sfx=" - external"
  elif [ "$internal" = true ]
  then
    sfx=" - internal"
  fi
  echo "$( uname -s ) - $( uname -m )${sfx}"
  cat << "EOF"

Database Selection

  -dbs       Print all database names
  -db        Database name (or "all")

Data Summaries

  -fields    Print field names
  -links     Print link names

Field Example

  <Field>
    <Name>ALL</Name>
    <FullName>All Fields</FullName>
    <Description>All terms from all searchable fields</Description>
    <TermCount>245340803</TermCount>
    <IsDate>N</IsDate>
    <IsNumerical>N</IsNumerical>
    <SingleToken>N</SingleToken>
    <Hierarchy>N</Hierarchy>
    <IsHidden>N</IsHidden>
    <IsTruncatable>Y</IsTruncatable>
    <IsRangable>N</IsRangable>
  </Field>

Link Example

  <Link>
    <Name>pubmed_protein</Name>
    <Menu>Protein Links</Menu>
    <Description>Published protein sequences</Description>
    <DbTo>protein</DbTo>
  </Link>
  <Link>
    <Name>pubmed_protein_refseq</Name>
    <Menu>Protein (RefSeq) Links</Menu>
    <Description>Link to Protein RefSeqs</Description>
    <DbTo>protein</DbTo>
  </Link>

EOF
}

PrintErrors() {

  echo "einfo $version"
  cat << "EOF"

Error Positive Controls

  nquire -url http://api.geonames.org/countryCode -lat 41.796 -lng "\-87.577"

  einfo -db NONE

  esearch -db pubmed -query "1233456789 [NONE]"

  elink -db pubmed -id 123456789 -related

  efetch -db pubmed -id 123456789 -format docsum

  efetch -db nuccore -id U1234567890 -format acc

EOF
}

# initialize specific flags

db=""
dbs=false
fields=false
links=false
test=false
repeats=1

# read command-line arguments

while [ $# -gt 0 ]
do
  case "$1" in
    -db )
      shift
      if [ $# -gt 0 ]
      then
        db="$1"
        shift
      else
        echo "ERROR: Missing -db argument" >&2
        exit 1
      fi
      ;;
    -dbs )
      dbs=true
      shift
      ;;
    -field | -fields )
      fields=true
      shift
      ;;
    -link | -links )
      links=true
      shift
      ;;
    -test | -tests )
      test=true
      shift
      ;;
    -repeat | -repeats )
      shift
      if [ $# -gt 0 ]
      then
        repeats="$1"
        shift
        if [ "$repeats" -lt 1 ] || [ "$repeats" -gt 20 ]
        then
          repeats=1
        fi
      fi
      ;;
    -h | -help | --help )
      PrintHelp
      exit 0
      ;;
    -error | -errors )
      PrintErrors
      exit 0
      ;;
    -* )
      ParseCommonArgs "$@"
      if [ "$argsConsumed" -gt 0 ]
      then
        shift "$argsConsumed"
      else
        echo "ERROR: Unrecognized option $1" >&2
        exit 1
      fi
      ;;
    * )
      # allows while loop to check for multiple flags
      break
      ;;
  esac
done

FinishSetup

# take database from dbase value (not expected for einfo) or -db argument

if [ -z "$dbase" ]
then
  dbase="$db"
fi

# -dbs

if [ "$dbs" = true ]
then
  res=$( RunWithCommonArgs nquire -get "$base" einfo.fcgi )
  echo "$res" |
  xtract -pattern DbList -sep "\n" -element DbName |
  sort -f

  exit 0
fi

# -db

if [ -n "$dbase" ]
then
  res=$( RunWithCommonArgs nquire -get "$base" einfo.fcgi -db "$dbase" -version "2.0" )

  # shortcut for fields

  if [ "$fields" = true ]
  then
    echo "$res" |
    xtract -pattern DbInfo -block Field -tab "\n" -element Name,FullName |
    sort -f
  fi

  # shortcut for links

  if [ "$links" = true ]
  then
    echo "$res" |
    xtract -pattern DbInfo -block Link -tab "\n" -element Name,Menu |
    sort -f
  fi

  # if neither -fields nor -links, print cleaned-up XML result, keeping original DOCTYPE line

  if [ "$fields" = false ] && [ "$links" = false ]
  then
    echo "$res" | transmute -format indent -doctype ""
  fi

  exit 0
fi

# -test (undocumented)

RunTests() {

  res=$(
    nquire -get https://eutils.ncbi.nlm.nih.gov/entrez/eutils esummary.fcgi -db pubmed -id 2539356 -version 2.0 | tr '\n' ' '
  )
  case "$res" in
    *"Nucleotide sequences required for Tn3 transposition immunity"* )
      ;;
    * )
      echo "FAIL: esummary.fcgi -db pubmed -id 2539356 -version 2.0" >&2
      ;;
  esac

  res=$(
    nquire -get https://eutils.ncbi.nlm.nih.gov/entrez/eutils esearch.fcgi -db pubmed -term "tn3 transposition immunity" | tr '\n' ' '
  )
  case "$res" in
    *"QueryTranslation>tn3"* )
      ;;
    * )
      echo "FAIL: esearch.fcgi -db pubmed -term \"tn3 transposition immunity\"" >&2
      ;;
  esac

  for sym in ATP6 ATP7B CBD DMD HFE PAH PRNP TTN
  do
    res=$( esearch -db gene -query "$sym [GENE]" -organism human |
           efetch -format docsum |
           xtract -pattern DocumentSummary -def "-" -lbl "${sym}" \
             -element NomenclatureSymbol Id Description CommonName )
    if [ -z "$res" ]
    then
      echo "FAIL: $sym" >&2
    fi
  done
}

if [ "$test" = true ]
then
  for i in $(seq 1 $repeats)
  do
    RunTests

    sleep 5
  done

  exit 0
fi

# warn on insufficient arguments

echo "ERROR: einfo requires either -db or -dbs on command line" >&2
exit 1