#!/bin/sh #] #] ********************* #] $ bash "$d_strings"'strings header.sh' # www.BillHowell.ca 15May2025 initial [chrCde, chrs] from QNial 'strings header.ndf' # view in text editor, using constant-width font (eg courier), tabWidth = 3 #24************************24 #24************************24 # List of operators, generated with : # $ grep "^#]" "$d_strings"'strings header.sh' | sed "s/^#\]/ /" >"$d_strings"'strings header.sh'' TblOfCont.txt' # #08********08 # Setup # done by [Fileops, strings].sh # source "$d_bin"'standard header.sh' #08********08 # code #] 08********08 #] [chr, str] basics # see "$d_bin""Standard header.sh" # date_ymdhm=$(date +"%y%m%d %kh%Mm") # becho - echo to [screen, logfile] # beval - echo command to [screen, logfile] then eval it # for extreme examples of beval : # $ find "/home/bill/PROJECTS/bin/" -maxdepth 1 -type f -name "*.sh" | tr \\n \\0 | xargs -0 -IFILE grep --with-filename --line-number "beval" "FILE" # backup computer [to,from] TravelUSB.sh:29 # encrypt-open.sh:34 calc(){ awk "BEGIN { print "$*" }"; } chr_apo="'" chr_quote='"' chr_tab=" " chrs_symbols="[a-zA-Z0-9_]" chrs_fname="[A-Za-z0-9._-]" # sed use as, eg \([$chrL_path]+\) # [] - will be a problem!! sometimes used in my fNams 24Sep2023 I added it in, untested chrL_fNam="A-Za-z0-9._-()<>[]&%#" chrL_path="$chrL_fNam"'\/' # usually for website management, I start only with base path, worry about bookmarks later chrL_fNamWithBookmk="A-Za-z0-9._-()<>&%#" chrL_pathWithBookmk="$chrL_fNamWithBookmk"'\/' #+-----+ #] [grep, sed]*[Search, Replace]ChrsAsStrL - for general text inputs, eg fileReads #] [apo, qot] removed from lists below for pragmatism: DON'T use!! #] assumes that chrs have not already been escaped, #] if so - this isn't needed for most (some can't escape) # used with : # 'QNial check user[, WkSpace]Optrs.ndf' : str_check[Grep, Sed][Search, Replace]Chrs # 'file_ops.ndf' : str[Grep, Sed][Search, Replace]_escapeProblemChrs # 29Oct2021 newline chr \ not included yet (affects ALL!), apo is very special # problematic chrs for sedSearch= ^$\&'",*.-[]{}\|/ won't do ^$&\ for now, just '",*.[]{}-/ # not a chrSedBad?: `/ # chrRgxpSrchBadEscL='+-.,?*&\[]-{}' # 03Jan2026 not tested yet chrRgxpSrchBadBadL="'"'"' # dunno chrGrepSrchBadEscL='^.,*$+' # 03Jan2026 not sure if escape solves problem? chrGrepSrchBadBadL="'"'"&\[]-{}' chrGrepRplcBadEscL='^$,*.' # [apo, qot] removed for pragmatism chrGrepRplcBadBadL="'"'"' # 03Jan2026 not sure if escape solves problem? chrSedSrchBadEscL='&,.*\$^[]-+' # can use if escaped chrSedSrchBadBadL="'"'"{}|' # 03Jan2026 not sure if escape solves problem? chrSedRplcBadEscL='&,.*\$^[]-' chrSedRplcBadBadL="'"'"&\/' # https://stackoverflow.com/questions/407523/escape-a-string-for-a-sed-replace-pattern # Ben Blank # see "$d_SysMaint"'Linux/sed notes.txt' 02Aug2023 # https://unix.stackexchange.com/questions/32907/what-characters-do-i-need-to-escape-when-using-sed-in-a-sh-script # Here is a bash function for converting paths for use with SED: # #Escape path for use with sed # function sedPath { path=$((echo $1|sed -r 's/([\$\.\*\/\[\\^])/\\\1/g'|sed 's/[]]/\[]]/g')>&1) } # user2428118 #] sedChrL_legend - chrL of special chrs that have to be encoded for sed #] sed[x]To[y] - sedExpressions to [, de]-code chrSedL between [human, encoded] formats # view in text editor, using constant-width font (eg courier), tabWidth = 3 # # sedChrL_legend='sedChrL_legend : the 2-letter codes for each chr are enclosed between zz # mnemonics : # & a@ = ampersand # '" [s, d]q =[single (apo), double] quote # , c@ = comma # . p@ = period # * s@ = star (asterix) # /\ [f, b]s = forward, backward] slash # $ d@ = dollar sign # ^ u@ = up (carat) # () r[l, r] = parenthesis [left, right] # [] s[l, r] = square bracket [left, right] #* {} c[l, r] = curly braces (brackets) [left, right] # - m@ = minus sign #* | v@ = vertical line' #* \t tb = tab must NOT be encoded for strP!!!???? #* as per chrSedMisL, these characters are NOT encoded yet (problematic) chrSedL='&",.*/\$^()[]-{}|' chrSedUnknown='!@#%+' # Apo = chrSedCtrl apostrophe chr (') # Npo = chrSedCtrl non-apostrophe chrs # sedXpr = expression following sed cmd, that defines text line conversions # chrSedCtlInn = sedSearch problematic chrs, which must be converted to work in a sedXpr # chrSedCtlNot = sedSearch problematic chrs, not yet included in my [func, test]s # the following sedXpr cannot be used directly, as 'sed "$sedXpr". # Copy full text following 'sed ', or use sedChrsEscSearch # They are shown here for clarity, and echo comments for tests, log files for future reference, etc #chrSedApoInc='"'"' # chrSeq to include in apo-enclosed expression to include apo # chrSedApoInc must be [created by strSedCtrl_get_strOrigin, done manually], cannot use as variable chrSedCtlApo="'" chrSedCtlNpo='",*.[]{}-/' chrSedCtlNot='^$&\' # 16May2025 not yet included in my [func, test]s sedXprApoFwd="s|[']|[']|" sedXprApoRev="s|\['\]|'|" sedXprNpoFwd='s|[/]|[/]|g;s|\"|\\"|g;s|\,|\\,|g;s|[-]|[-]|g;s|[*]|[*]|g;s|[{]|[{]|g;s|[}]|[}]|g;s#[|]#[|]#;s|\.|\\.|g;s|[]]||g;s|[[]||' sedXprNpoRev='s|\[/\]|/|g;s|\\"|"|g;s|\\,|,|g;s|\\.|.|g;s|\[-\]|-|g;s|\[{\]|{|g;s|\[}\]|}|g;s|[*]|*|g;s#\[|\]#|#;s|[[]||g;s|[]]||g' # 16May2025 tests # 11h18m 16May25$ echo 'Sam'"'"'s "best" test' | sed "s|[']\"[']\"[']|'|" # Sam's "best" test # >> crap, it worked, or seems to!?? why? wait for tests to confirm # 11h23m 16May25$ echo 'Sam$chrSedApoInc "'"best" test' | sed "s|[']\"[']\"[']|'|" # > # >> nyet, chrSedApoInc must be [created by strSedCtrl_get_strOrigin, done manually], cannot use as variable # 11h28m 16May25$ echo "$chrSedCtlNpo" | sed 's|\[|[|;s|\]|]|;s|\"|"|;s|\,|,|;s|\.|.|;s|\*|*|;s|[{]|{|;s|[}]|}|;s|\-|-|;s#[|]#|#' # ",*.[]{}-/ # >> perfect, category theory-like in one line... # 08********08 #] characters [special, unicode] often used ... : # IF flag_debug THEN write '+-----+' ; ENDIF ; # IF flag_debug THEN write 'characters [special, unicode] often used ...' ; ENDIF ; # preferred symbols - easiest to use text eiditor to manually replace in the source document # raw unpreferred # `· `• dot product # ` `∫ integral sign # I can't use this because it has strange QNial unicode properties - use backwards apostrophe instead : # ` `´ forward prime for RFp # special characters in text # `. period delimits keywords # 03Sep2016 - this format may be critical to some of my QNial code! (changes will cause problems with legacy programs) chrCde_tab=9 chrCde_newline=10 chrCde_carriagereturn=13 # not normally used with Unix/Linux chrCde_space=32 chrCde_quotation=34 chrCde_apostrophe=39 chrCde_apo="$chr_apostrophe" chrCde_quote="$chr_quotation" chrCde_semicolon=59 chrCde_backApo=96 # back-slanting apostrophe chrCde_underscore=95 # str_apo= string chr_apo # actual chrs - most should have ' EACH string ' (eachStr) - will do later chrs_alpha_caps='ABCDEFGHIJKLMNOPQRSTUVWXYZ' #(eachStr) chr_codes 65- 90 chrs_alpha_small='abcdefghijklmnopqrstuvwxyz' #(eachStr) chr_codes 97-122 chrs_alphabet="$chrs_alpha_caps$chrs_alpha_small" #(sortup) chr_codes [65-90,97-122] chrs_numbers='0123456789' #(eachStr) chr_codes 48-57 chrs_alphaNumeric="$chrs_alphabet$chrs_numbers" #(sortup link) chr_codes [48-57,65-90,97-122] chrs_operators='~!@#$%^&:+-*/[\][|]"' #(eachStr) chr_codes [126,33-41,etc...] # I had to "fix" several - won't work? chrs_paren_strt='([{' #(eachStr) chr_codes ??? chrs_paren_stop=')]}' #(eachStr) chr_codes ??? chrs_parenthesis='()[]{}<>' #(eachStr) chr_codes ??? chrs_SpcTab="$chrCde_space$chrCde_tab" #(eachStr) chr_codes [32,9] chrs_punctuation="$chrs_operators$chrs_SpcTab$chrs_parenth" #(eachStr) chr_codes [126,33-41,etc...] chrs_fnames="$chrs_alphaNumeric"'_ .,-' # for QNial chrs_arithmetic='*+-/=' #(eachStr) chr_codes [42,43,45,47,61] chrs_math_spaced='+-=' #(eachStr) chr_codes [42,43,45,47,61] # for "symbols translate.ndf -> symTranslate_string" chrs_lineType='#$%^' #(eachStr) chrs_writing="$chrs_alphaNumeric$chrs_punctuation$chrs_math_spaced$chrs_SpcTab" #(needSort) chrs_line="$chrCde_space$chrCde_tab$chrCde_newline" #(needSort) chrs_SpcTabCrNl="$chrCde_space$chrCde_tab$chrCde_newline$chrCde_carriagereturn" #(needSort) chrs_symbols_matlab="'``='"'"' #(???) str_SpcTab=' ' #+-----+ #] unicode >1 byte sequences chrs_Greek_caps='ΑΒΓΔΕΖΗΘΙΚΛΜΝΞΟΠΡΣΣΤΥΦΧΨΩ' chrs_Greek_small='αβγδεζηθικλμνξοπρςστυφχψω' chrs_Greek_all="$chrs_Greek_caps$chrs_Greek_small" # note that at least some unicode work! names_Greek_all=' ' chrs_calculus='∫∮∬∂Σ' chrs_symbols_QNial="$chrs_alphaNumeric$chrs_Greek_all"'_´' # a´=6 # aε='greek letter in str' # +-----+ #] chrs_QNialSymbols - res ipsa loquitor (not complete, but should be safe) # 14Aug2021 careful - use chrs_symbols_QNial with [strL_cutBy_chrInclL, etc]!!! chrs_QNialSymbols='0123456789_ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz' #+----+ #] Howell's Flat-Liner Notation (HFLN) - including unicode for QNial chrs_symbols_HFLN="$chrs_alphaNumeric$chrs_Greek_all$chrs_calculus"'_´' # unicodes_HFLN_1= EACH charrep chrs_symbols_HFLN # unicodes_HFLN_shapes= EACH (gage shape) unicodes_HFLN # unicodes_HFLN_2= (2 EACHRIGHT = unicodes_HFLN_shapes) sublist unicodes_HFLN # unicodes_HFLN_3= (3 EACHRIGHT = unicodes_HFLN_shapes) sublist unicodes_HFLN # # HFLN does not currently have 4-byte unicode symbol characters # chrs_HFLN_2= EACH char unicodes_HFLN_2 # chrs_HFLN_3= EACH char unicodes_HFLN_3 # # # non-HFLN symbols # chrs_non_HFLN='•·´∇∆≈∞≤≥∈∘' # chrs_bad='·' '' '.' # chrs_added='–“”→…’‘→…∗×−' # 07Feb2019 add new codes here!!! # chrs_all= link chrs_symbols_HFLN chrs_non_HFLN chrs_bad chrs_added # unicodes_ALL= EACH charrep chrs_ALL # unicodes_ALL_shapes= EACH (gage shape) unicodes_ALL # unicodes_ALL_2= (2 EACHRIGHT = unicodes_ALL_shapes) sublist unicodes_ALL # unicodes_ALL_3= (3 EACHRIGHT = unicodes_ALL_shapes) sublist unicodes_ALL # # # [grep, sed, regexpr]-related - defult use of `| separator chrs_sed_notSearch='$.*\/[]^|'"$chrCde_apo$chrCde_quote" chrs_sed_notReplace='&\/|'"$chrCde_apo$chrCde_quote" # # # 18Oct2021 needs work! chrCdes_commonSeparators="$chrCde_tab$chrCde_space$chrCde_quotation$chrCde_apo$chrCde_quote$chrCde_semicolon$chrCde_exclam$chrCde_at$chrCde_hash$chrCde_percent$chrCde_expon$chrCde_amp$chrCde_times$chrs_punctuation" # to see chr_codes : #qnial> mix EACHBOTH pair chrs_ALL (EACH charrep chrs_ALL) # https://en.wikipedia.org/wiki/List_of_Unicode_characters#Latin_script # List of Unicode characters # Code Glyph Decimal Octal HTML Description # # U+00A9 © 0169 0302 0251 © Copyright sign 0105 # U+00B0 ° 0176 0302 0260 ° Degree sign 0112 # 24May2018 15:07 WARNING!! >> I still have some problems with unicode # see also "/media/bill/SWAPPER/Qnial/code development/strings - unicode usage.txt" # qnial> charrep '•' # -30 -128 -94 # qnial> charrep '·' # -62 -73 bad character! # qnial> charrep '' # -18 -124 -87 # qnial> charrep '´' # -62 -76 # qnial> charrep '' # -18 -126 -108 bad character! # qnial> charrep '∫' # -30 -120 -85 # qnial> charrep '' # -18 -123 -99 bad character! # qnial> charrep '∮' # -30 -120 -82 # qnial> charrep '∬' # -30 -120 -84 # qnial> charrep '∂' # -30 -120 -126 # qnial> charrep 'Σ' # -50 -93 # qnial> charrep 'Ψ' # -50 -88 # qnial> charrep 'ζ' # -50 -74 # EACH char ( -62 EACHRIGHT pair (-1 * (tell 120)) ) # example chrSets : # echo "$chrs_QNialSymbols #includes sortup (chrs_alphaNumeric '_')) # echo "$chrs_symbols_QNial #includes '[chrs_alphaNumeric chrs_Greek_all '_' '´' chrs_Greek_[caps, small]) # echo "$chrs_HFLN_1$unicodes_ALL$chrs_sed_notSearch$$chrs_sed_notReplace" # echo "$chrs_HFLN_2$unicodes_ALL$unicodes_HFLN]_2$chrs_sed_notSearch$$chrs_sed_notReplace" # echo "$chrs_HFLN_3$unicodes_ALL$unicodes_HFLN]_3$chrs_sed_notSearch$$chrs_sed_notReplace" # echo "$chrs_alphaNumeric$chrCdes_commonSeparators$chrs_Greek_caps$chrs_Greek_caps_small" # echo "$mix EACHBOTH pair chrs_ALL (EACH charrep chrs_ALL) #08********08 # Procedures - To run : #08********08 # run - select one of options # $ bash "$d_strings"'strings header.sh' # enddoc