Online Amateur Radio Community Wiki

User Tools

Site Tools

Trace: • xrouter • manpages • parsing
packet:xrouter:manpages:parsing

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
Next revision		Previous revision
packet:xrouter:manpages:parsing [2025/04/19 18:30] – m0mzfpacket:xrouter:manpages:parsing [2025/04/20 07:28] (current) – m0mzf
Line 1: Line 1:
-====== G8PZT MAN page parsing script ======= +====== Script to parse Xrouter's MAN and HLP files ======= 
-Does what it says on the tin :) +<file awk parse-pzt-manhlp.sh>
- +
-<file>+
 #!/bin/bash #!/bin/bash
 ################################## ##################################
-# by Jason M0MZF +# by Jason M0MZF (not a programmer!) 
-# bash / awk / hammer / nail etc :)+# bash / awk / hammer / nail etc.
 # License - MIT. Crack on people. # License - MIT. Crack on people.
 #  # 
Line 15: Line 13:
 # The intention is to parse all MAN / HLP files within the folders and # The intention is to parse all MAN / HLP files within the folders and
 # write them with appropriate formatting to files which can then be # write them with appropriate formatting to files which can then be
-# copypasta'd directly into the wiki.+# pasted directly into the wiki.
 # #
 # This could also be done with groff > HTML > pandoc > ssml but pandoc's # This could also be done with groff > HTML > pandoc > ssml but pandoc's
 # output format for SSML doesn't sort out proper ====headers==== and I  # output format for SSML doesn't sort out proper ====headers==== and I 
-# don't know Lua. Yet. +# don't know Lua. Yet. Maybe something like this with a custom output formatter: 
-# +# cat ${manpage} | groff -Thtml -P -l -mmandoc 2>/dev/null | pandoc -f html -t dokuwiki -o "$manpage".txt 
 +# But when all you've got is awk, everything looks like a record / field... ;)
 # #
 ################################## ##################################
Line 42: Line 41:
 # 20250418 - Implemented MAN page parsing # 20250418 - Implemented MAN page parsing
 # 20250419 - Implemented HLP page parsing # 20250419 - Implemented HLP page parsing
-# 20250419 - Tidy up, more awk less bash, remove .MAN / .HLP+# 20250419 - Tidy up, more awk less bash, remove .MAN / .HLP from outputted headers
 ################################## ##################################
  
Line 51: Line 50:
 OUTPUTDIR="$BASEPATH/docuwiki-$DATE" OUTPUTDIR="$BASEPATH/docuwiki-$DATE"
  
-# Colourise output+# Handy functions
 echoRed () { echoRed () {
  echo -e "\e[1;31m----$1----\e[0m"  echo -e "\e[1;31m----$1----\e[0m"
Line 58: Line 57:
  echo -e "\e[1;32m----$1----\e[0m"  echo -e "\e[1;32m----$1----\e[0m"
 } }
- 
 checkRoot () { checkRoot () {
- if [[ $UID -eq 0 ]];  + if [[ $UID -eq 0 ]]; then 
- then + echoRed "Don't run this as root please"
- echoError "This script must NOT be run as root!"+
  exit 1  exit 1
  fi  fi
 } }
  
-awkParseMan=' +# Use awk to: 
-{+#  strip out comment lines and remove any <CR> from <CR><LF> lines 
 +#  turn the MAN page header into a code block, it contains a revision date 
 +#  find every subsequent MAN page header and turn it into a docuwiki header and 
 +#   terminate the previous code block before the header, then 
 +#   create a new code block after the header, thus encapsulating the subsection 
 +# (the final encapsulation is done using "echo" in the bash script below) 
 +awkParseMan='{
  if (NR==1 || NR==2) # For the first two lines  if (NR==1 || NR==2) # For the first two lines
  {  {
  gsub("\r", "") # remove all carriage return chars  gsub("\r", "") # remove all carriage return chars
- if (/^;/ || NF==0) {next} # skip comment or empty line + if (/^;/ || NF==0) {next} # skip the subsequent print function for comment or empty lines 
- print "<code>" $0 # and start a ssml code block+ print "<code>" $0 # annd output the line prepended with a code block start
  }  }
  
Line 80: Line 83:
  if (/^[A-Z]/) # If the line begins with a character  if (/^[A-Z]/) # If the line begins with a character
  {  {
- endcode="</code> **" # end previous code block + starthead="</code> **" # end previous code block 
- startcode="** <code>" # set bold and start code block+ endhead="** <code>" # set bold and start code block
  gsub("\r", "") # remove all carriage return chars  gsub("\r", "") # remove all carriage return chars
- print endcode $0 startcode  # and output the line+ print starthead $0 endhead  # and output the line
  }  }
  else # else for all other lines  else # else for all other lines
Line 94: Line 97:
 } }
 ' '
- +# Use awk to: 
-awkParseHlp=' +#  strip out comment lines (this is always line 1, sometime 2 and 3) and remove any <CR> from <CR><LF> lines 
-{ +#  insert a start code block in place of the now-empty line 1 
- header1="<code>"+# (the final encapsulation is done using "echo" in the bash script below) 
 +awkParseHlp=' { 
 + endhead="<code>"
  gsub("\r", "") # remove all carriage return chars  gsub("\r", "") # remove all carriage return chars
- if (NR==1) {print header1} # start code block on first line+ if (NR==1) {print endhead} # start code block on first line
  if (/^;/ || NF==0) {next} # skip comment / empty lines  if (/^;/ || NF==0) {next} # skip comment / empty lines
- print $0+ print $0 # output the refined line
 } }
 ' '
 +# Use awk to extract a section name from the directory structure
 awkSectionHeader=' awkSectionHeader='
- BEGIN { FS="/" } #Set field separator + BEGIN { FS="/" } # Set field separator to get section name from path 
-{ +{ # /path/looks/like/this/Section Header Name/ 
- header="=======" # create header formatting + header="=======" # create header formatting 
- print header $(NF-1) header # we need the penultimate field as that + print header $(NF-1) header # the penultimate field is the section name 
-} # is the folder name /it/looks/like/this/+}
 ' '
 +# Use awk to extract a name from the filename.extension
 awkFileHeader=' awkFileHeader='
- BEGIN { FS="." } # Set field separator + BEGIN { FS="." } # Set field separator to separate file extenstion 
-{ # to separate out the .MAN .HLP+{ # because we want the file name from FILENAME.MAN
  header="=====" # create header formatting  header="=====" # create header formatting
- print header $1 header # output first field with headers+ print header $1 header # the first field is the file name
 } }
 ' '
Line 127: Line 132:
  do  do
  # Get the penultimate field in file path, i.e. the section (folder) name  # Get the penultimate field in file path, i.e. the section (folder) name
- section=$(echo $folder | awk  -F/ '{print $(NF-1)}') + local section=$(echo $folder | awk  -F/ '{print $(NF-1)}') 
- # Prepend the section name as a docuWiki header+ # Format the section name as a docuWiki header
  echo "$folder" | awk "$awkSectionHeader" >> "${OUTPUTDIR}"/"$1"/"${section}".docuwiki  echo "$folder" | awk "$awkSectionHeader" >> "${OUTPUTDIR}"/"$1"/"${section}".docuwiki
  # Spit some stuff out to the shell  # Spit some stuff out to the shell
Line 136: Line 141:
  do  do
  # Get the last field in file path, i.e. file name  # Get the last field in file path, i.e. file name
- title=$(echo $file | awk -F/ '{print $NF}') + local title=$(echo $file | awk -F/ '{print $NF}') 
- # Prepend the file name as a docuwiki header + local outputpath="${OUTPUTDIR}"/"$1"/"${section}".docuwiki 
- echo "$title" | awk "$awkFileHeader" >> "${OUTPUTDIR}"/"$1"/"${section}".docuwiki+ # Format the file name as a docuwiki header 
 + echo "$title" | awk "$awkFileHeader" >> "$outputpath"
  case "$1" in  case "$1" in
  # For MAN files, after awk has done it's job we need to remove the last line; this last line breaks  # For MAN files, after awk has done it's job we need to remove the last line; this last line breaks
  # the following <code> statement and is just an EOF message, so we don't lose anything.  # the following <code> statement and is just an EOF message, so we don't lose anything.
- MANFILES) awk "$awkParseMan" "$file" | head -n -1 >> "${OUTPUTDIR}"/"$1"/"${section}".docuwiki + MANFILES) awk "$awkParseMan" "$file" | head -n -1 >> "$outputpath" 
- echo -e "</code>\n----" >> "${OUTPUTDIR}"/"$1"/"${section}".docuwiki+ echo -e "</code>\n----" >> "$outputpath"
  ;;  ;;
  # For HLP files we don't want to remove the last line because that truly is real content  # For HLP files we don't want to remove the last line because that truly is real content
- HLPFILES) awk "$awkParseHlp" "$file" >> "${OUTPUTDIR}"/"$1"/"${section}".docuwiki + HLPFILES) awk "$awkParseHlp" "$file" >> "$outputpath" 
- echo -e "</code>" >> "${OUTPUTDIR}"/"$1"/"${section}".docuwiki+ echo -e "</code>" >> "$outputpath"
  ;;  ;;
  esac  esac
packet/xrouter/manpages/parsing.1745087409.txt.gz · Last modified: 2025/04/19 18:30 by m0mzf