mirror of
https://github.com/chubin/cheat.sh.git
synced 2026-06-20 13:16:44 +02:00
Merge pull request #215 from ErezBinyamin/master
Restructure for text parsing. Faster, more stable, more maintainable
This commit is contained in:
Igor Chubin
GitHub
+119
-94
@@ -8,148 +8,173 @@
|
||||
# oeis <sequence ID> <language>
|
||||
# oeis <val_a, val_b, val_c, ...>
|
||||
oeis() (
|
||||
local URL='https://oeis.org'
|
||||
local URL='https://oeis.org/search?q='
|
||||
local TMP=/tmp/oeis
|
||||
local DOC=/tmp/oeis/doc.html
|
||||
local MAX_TERMS=10
|
||||
local MAX_TERMS_LONG=30
|
||||
local MAX_TERMS_SHORT=10
|
||||
mkdir -p $TMP
|
||||
# -- get_desc --
|
||||
# @return print description of OEIS sequence
|
||||
get_desc() {
|
||||
grep -A 1 '<td valign=top align=left>' $DOC \
|
||||
| sed '/<td valign=top align=left>/d; /--/d; s/^[ \t]*//; s/<[^>]*>//g;' \
|
||||
| sed 's/ / /g; s/\&/\&/g; s/>/>/g; s/</</g; s/"/"/g'
|
||||
return $?
|
||||
}
|
||||
# -- get_seq --
|
||||
# @param MAX_TERMS
|
||||
# @return Print the first MAX_TERMS terms of a sequence
|
||||
get_seq() {
|
||||
local MAX_TERMS=${1}
|
||||
grep -o '<tt>.*, .*[0-9]</tt>' $DOC \
|
||||
| sed 's/<[^>]*>//g' \
|
||||
| grep -v '[a-z]' \
|
||||
| grep -v ':' \
|
||||
| cut -d ',' -f 1-${MAX_TERMS}
|
||||
return $?
|
||||
}
|
||||
# -- parse_code --
|
||||
# @param GREP_REGEX
|
||||
# @return Code snippet that corresponds to GREP_REGEX
|
||||
parse_code() {
|
||||
local GREP_REGEX="${1}"
|
||||
cat $DOC \
|
||||
| tr '\n' '`' \
|
||||
| grep -o "${GREP_REGEX}" \
|
||||
| tr '`' '\n' \
|
||||
| sed 's/^[ \t]*//; s/<[^>]*>//g; /^\s*$/d;' \
|
||||
| sed 's/ / /g; s/\&/\&/g; s/>/>/g; s/</</g; s/"/"/g'
|
||||
return $?
|
||||
}
|
||||
rm -f ${TMP}/authors ${TMP}/bibliograpy ${TMP}/section $TMP/code_snippet
|
||||
# -- MAIN --
|
||||
# Search sequence by ID (optional language arg)
|
||||
# . oeis <SEQ_ID>
|
||||
# . oeis <SEQ_ID> <LANGUAGE>
|
||||
# . oeis <LANGUAGE> <SEQ_ID>
|
||||
# . oeis <SEQ_ID> <SECTION>
|
||||
# . oeis <SECTION> <SEQ_ID>
|
||||
isNum='^[0-9]+$'
|
||||
if [ $# -lt 3 ] && [[ ${1:1} =~ $isNum || ${2:1} =~ $isNum || ${1} =~ $isNum || ${2} =~ $isNum ]] && ! echo $1 | grep -q '[0-9]' || ! echo $2 | grep -q '[0-9]'
|
||||
# Search for specific sequence (and potentially language or :SECTION (list)
|
||||
if [ $# -ge 1 ] \
|
||||
&& [[ $(echo $1 | tr -d 'aA') =~ $isNum || $(echo $2 | tr -d 'aA') =~ $isNum ]] \
|
||||
&& [[ ! $(echo $1 | tr -d 'aA') =~ $isNum || ! $(echo $2 | tr -d 'aA') =~ $isNum ]]
|
||||
then
|
||||
# Arg-Parse ID, Generate URL
|
||||
if echo ${1^^} | grep -q '[B-Z]'
|
||||
if [[ $(echo $1 | tr -d 'aA') =~ $isNum ]]
|
||||
then
|
||||
ID=${2^^}
|
||||
LANGUAGE=$1
|
||||
else
|
||||
ID=${1^^}
|
||||
LANGUAGE=$2
|
||||
SECTION=$2
|
||||
else
|
||||
ID=${2^^}
|
||||
SECTION=$1
|
||||
fi
|
||||
[[ ${ID:0:1} == 'A' ]] && ID=${ID:1}
|
||||
ID=$(bc <<< "$ID")
|
||||
ID="A$(printf '%06d' ${ID})"
|
||||
URL+="/${ID}"
|
||||
URL+="id:${ID}&fmt=text"
|
||||
curl $URL 2>/dev/null > $DOC
|
||||
# Print Code Sample
|
||||
if [[ ${LANGUAGE^^} == ':LIST' ]]
|
||||
# :list available language code_snippets
|
||||
if [[ ${SECTION^^} == ':LIST' || ${SECTION^^} == ':PROG' ]]
|
||||
then
|
||||
rm -f ${TMP}/list
|
||||
grep -q 'MAPLE' $DOC && printf 'maple\n' >> $TMP/list
|
||||
grep -q 'MATHEMATICA' $DOC && printf 'mathematica\n' >> $TMP/list
|
||||
parse_code 'PROG.*CROSSREFS' \
|
||||
| grep -o '^(.*)' \
|
||||
| sed 's/ .*//g' \
|
||||
| tr -d '()' \
|
||||
| sort -u >> $TMP/list
|
||||
[ $(wc -c < $TMP/list) -ne 0 ] && cat ${TMP}/list || printf 'No code snippets available.\n'
|
||||
grep -q '%p' $DOC && echo 'maple' >> $TMP/section
|
||||
grep -q '%t' $DOC && echo 'mathematica' >> $TMP/section
|
||||
grep '%o' $DOC \
|
||||
| grep "${ID} (" \
|
||||
| sed "s/^.*${ID} (//; s/).*//" \
|
||||
| awk 'NF == 1' \
|
||||
>> $TMP/section
|
||||
[[ -f $TMP/section && $(wc -c < $TMP/section) -ne 0 ]] \
|
||||
&& cat ${TMP}/section | sort -u \
|
||||
|| printf 'No code snippets available.\n'
|
||||
return 0
|
||||
fi
|
||||
# Print ID, description, and sequence
|
||||
# Print ID
|
||||
printf "ID: ${ID}\n"
|
||||
get_desc
|
||||
printf '\n'
|
||||
get_seq ${MAX_TERMS}
|
||||
# Print Description (%N)
|
||||
grep '%N' $DOC | sed "s/^.*${ID} //"
|
||||
printf '\n'
|
||||
# Print Sequence (Three sections %S %T nd %U)
|
||||
grep '%S' $DOC | sed "s/^.*${ID} //" | tr -d '\n' > $TMP/seq
|
||||
grep '%T' $DOC | sed "s/^.*${ID} //" | tr -d '\n' >> $TMP/seq
|
||||
grep '%U' $DOC | sed "s/^.*${ID} //" | tr -d '\n' >> $TMP/seq
|
||||
cat $TMP/seq \
|
||||
| cut -d ',' -f 1-${MAX_TERMS_LONG} \
|
||||
| sed 's/,/, /g; s/$/ .../'
|
||||
# Generate code snippet (%p, %t, %o) (maple, mathematica, prog sections)
|
||||
if [ $# -gt 1 ]
|
||||
then
|
||||
if [[ ${LANGUAGE^^} == 'MAPLE' ]] && grep -q 'MAPLE' $DOC
|
||||
printf "\n\n"
|
||||
# MAPLE section (%p)
|
||||
if [[ ${SECTION^^} == 'MAPLE' ]] && grep -q '%p' $DOC
|
||||
then
|
||||
GREP_REGEX='MAPLE.*CROSSREFS'
|
||||
grep -q 'PROG' $DOC && GREP_REGEX='MAPLE.*PROG'
|
||||
grep -q 'MATHEMATICA' $DOC && GREP_REGEX='MAPLE.*MATHEMATICA'
|
||||
parse_code "${GREP_REGEX}" \
|
||||
| sed 's/MAPLE/(MAPLE)/; /MATHEMATICA/d; /PROG/d; /CROSSREFS/d' \
|
||||
> ${TMP}/code_snippet
|
||||
elif [[ ${LANGUAGE^^} == 'MATHEMATICA' ]] && grep -q 'MATHEMATICA' $DOC
|
||||
grep '%p' $DOC | sed "s/^.*${ID} //" > $TMP/code_snippet
|
||||
# MATHEMATICA section (%t)
|
||||
elif [[ ${SECTION^^} == 'MATHEMATICA' ]] && grep -q '%t' $DOC
|
||||
then
|
||||
grep '%t' $DOC | sed "s/^.*${ID} //" > $TMP/code_snippet
|
||||
# PROG section (%o)
|
||||
elif grep -qi '%o' $DOC && grep -qi $SECTION $DOC
|
||||
then
|
||||
GREP_REGEX='MATHEMATICA.*CROSSREFS'
|
||||
grep -q 'PROG' $DOC && GREP_REGEX='MATHEMATICA.*PROG'
|
||||
parse_code "${GREP_REGEX}" \
|
||||
| sed 's/MATHEMATICA/(MATHEMATICA)/; /PROG/d; /CROSSREFS/d' \
|
||||
> ${TMP}/code_snippet
|
||||
else
|
||||
# PROG section contains more code samples (Non Mathematica or Maple)
|
||||
parse_code 'PROG.*CROSSREFS' \
|
||||
| sed '/PROG/d; /CROSSREFS/d' \
|
||||
> ${TMP}/prog
|
||||
# Print out code sample for specified language
|
||||
rm -f ${TMP}/code_snippet
|
||||
awk -v tgt="${LANGUAGE^^}" -F'[()]' '/^\(/{f=(tgt==toupper($2))} f' ${TMP}/prog > ${TMP}/code_snippet
|
||||
grep '%o' $DOC \
|
||||
| sed "s/%o ${ID} //" \
|
||||
| awk -v tgt="${SECTION^^}" -F'[()]' '{act=$2} sub(/^\([^()]+\) */,""){f=(tgt==toupper(act))} f' \
|
||||
> ${TMP}/code_snippet
|
||||
fi
|
||||
# Print code snippet with 4-space indent to enable colorization
|
||||
if [ $(wc -c < $TMP/code_snippet) -ne 0 ]
|
||||
if [[ -f $TMP/code_snippet && $(wc -c < $TMP/code_snippet) -ne 0 ]]
|
||||
then
|
||||
printf "${LANGUAGE}"
|
||||
# Get authors
|
||||
cat ${TMP}/code_snippet \
|
||||
| grep -o ' _[A-Z].* [A-Z].*_, [A-Z].*[0-9]' \
|
||||
| sort -u \
|
||||
> ${TMP}/authors
|
||||
i=1
|
||||
# Replace authors with numbers
|
||||
while read author
|
||||
do
|
||||
author=$(<<<"$author" sed 's/[]\\\*\(\.[]/\\&/g')
|
||||
sed -i "s|${author}|[${i}]|" ${TMP}/code_snippet
|
||||
echo "[${i}] [${author}]" | tr -d '_' >> ${TMP}/bibliograpy
|
||||
let i++
|
||||
done <${TMP}/authors
|
||||
# Print snippet
|
||||
cat ${TMP}/code_snippet \
|
||||
| sed "s/(${LANGUAGE^^})/\n/; s/(${LANGUAGE})/\n/;" \
|
||||
| sed 's/^/ /'
|
||||
else
|
||||
printf "${LANGUAGE^^} unavailable. Use :list to view available languages.\n"
|
||||
printf "${SECTION^^} unavailable. Use :list to view available languages.\n"
|
||||
fi
|
||||
fi
|
||||
# Search unknown sequence
|
||||
else
|
||||
elif [ $# -gt 1 ] && ! echo $@ | grep -q -e [a-z] -e [A-Z]
|
||||
then
|
||||
# Build URL
|
||||
URL+="/search?q=signed:$(echo $@ | tr -sc '[:digit:]-' ',')"
|
||||
URL+="signed:$(echo $@ | tr -sc '[:digit:]-' ',')&fmt=short"
|
||||
curl $URL 2>/dev/null > $DOC
|
||||
# Sequence IDs
|
||||
grep -o '=id:.*&' $DOC \
|
||||
| sed 's/=id://; s/&//' > $TMP/id
|
||||
# Descriptions
|
||||
get_desc > $TMP/desc
|
||||
# Sequences
|
||||
get_seq ${MAX_TERMS} > $TMP/seq
|
||||
# Print data for all
|
||||
grep -o '"/A[0-9][0-9][0-9][0-9][0-9][0-9]">A[0-9][0-9][0-9][0-9][0-9][0-9]' $DOC \
|
||||
| sed 's/.*>//' \
|
||||
> $TMP/id
|
||||
readarray -t ID < $TMP/id
|
||||
# Descriptions
|
||||
grep -A 1 '<td valign=top align=left>' $DOC \
|
||||
| sed '/--/d; s/<[^>]*>//g; /^\s*$/d; s/^[ \t]*//' \
|
||||
| sed 's/ / /g; s/\&/\&/g; s/>/>/g; s/</</g; s/"/"/g' \
|
||||
> $TMP/desc
|
||||
readarray -t DESC < $TMP/desc
|
||||
# Sequences
|
||||
grep 'style="color:black;font-size:120%' $DOC \
|
||||
| sed 's/<[^>]*>//g; s/^[ \t]*//' \
|
||||
| cut -d ',' -f 1-${MAX_TERMS_SHORT} \
|
||||
| sed 's/,/, /g; s/$/ .../' \
|
||||
> $TMP/seq
|
||||
readarray -t SEQ < $TMP/seq
|
||||
# Print all ID, DESC, SEQ
|
||||
for i in ${!ID[@]}
|
||||
do
|
||||
printf "${ID[$i]}: ${DESC[$i]}\n"
|
||||
printf "${SEQ[$i]}\n\n"
|
||||
done
|
||||
else
|
||||
printf "
|
||||
# oeis
|
||||
#
|
||||
# The On-Line Encyclopedia of Integer Sequences (OEIS),
|
||||
# also cited simply as Sloane's, is an online database of integer sequences.
|
||||
|
||||
# Find all possible OEIS sequences for some sequence (1,1,1,1...)
|
||||
curl cheat.sh/oeis/1+1+1+1
|
||||
|
||||
# Describe an OEIS sequence (A2)
|
||||
curl cheat.sh/oeis/A2
|
||||
|
||||
# Implementation of the A2 OEIS sequence in Python
|
||||
curl cheat.sh/oeis/A2/python
|
||||
|
||||
# List all available implementations of the A2 OEIS sequence
|
||||
curl cheat.sh/oeis/A2/:list
|
||||
"
|
||||
return 1
|
||||
fi
|
||||
grep 'results, too many to show. Please refine your search.' /tmp/oeis/doc.html | sed -e 's/<[^>]*>//g; s/^[ \t]*//'
|
||||
# Error statements
|
||||
grep 'results, too many to show. Please refine your search.' $DOC | sed -e 's/<[^>]*>//g; s/^[ \t]*//'
|
||||
grep -o 'Sorry, but the terms do not match anything in the table.' $DOC
|
||||
# print bibliography
|
||||
printf "\n\n"
|
||||
[ -f ${TMP}/bibliograpy ] && cat ${TMP}/bibliograpy
|
||||
# Print URL for user
|
||||
printf "\n[${URL}]\n" | rev | sed 's/,//' | rev
|
||||
printf "[${URL}]\n" \
|
||||
| rev \
|
||||
| sed 's/,//' \
|
||||
| rev \
|
||||
| sed 's/&.*/]/'
|
||||
)
|
||||
|
||||
oeis $@
|
||||
|
||||
Reference in New Issue
Block a user