htmltoman

Tokyo Cabinet的Tokyo Cabinet 是一个DBM的实现。这里的数据库由一系列key-value对的记录构成。key和value都可以是任意长度的字节序列,既可以是二进制也可以是字符

#! /usr/bin/awk -f

function strip(text){
  gsub("^ *<[a-zA-Z0-9]*[^>]*>", "", text)
  gsub("</[a-zA-Z0-9]*> *$", "", text)
  return text
}

function unescape(text){
  gsub("&lt;", "<", text)
  gsub("&gt;", ">", text)
  gsub("&quot;", "\"", text)
  gsub("&amp;", "\\&", text)
  gsub("-", "\\-", text)
  return text
}

BEGIN {
  date = strftime("%Y-%m-%d")
  printf(".TH \"%s\" %d \"%s\" \"%s\" \"%s\"\n\n", "INTRO", 3, date, "Man Page", "Tokyo Cabinet")
}

/ *<h[1-3] *[^>]*>.*<\/h[1-3]> *$/ {
  text = $0
  text = strip(text)
  text = unescape(text)
  text = toupper(text)
  printf("\n")
  printf(".SH %s\n", text)
}

/ *<p *[^>]*>.*<\/p> *$/ {
  text = $0
  text = strip(text)
  text = gensub("<code *[^>]*>([^<]*)</code>", "\\\\fB\\1\\\\fR", "g", text)
  text = gensub("<var *[^>]*>([^<]*)</var>", "\\\\fI\\1\\\\fR", "g", text)
  gsub("<[^>]*>", "", text)
  text = unescape(text)
  printf(".PP\n")
  printf("%s\n", text)
}

/ *<dl *[^>]*> *$/ {
  printf(".PP\n")
  printf(".RS\n")
}
/ *<\/dl> *$/ {
  printf(".RE\n")
}
/ *<dt *[^>]*>.*<\/dt> *$/ {
  text = $0
  text = strip(text)
  text = gensub("<var *[^>]*>([^<]*)</var>", "\\\\fI\\1\\\\fB", "g", text)
  gsub("<[^>]*>", "", text)
  gsub("[\\||\\[|\\]]", "\\fR&\\fB", text)
  text = unescape(text)
  printf(".br\n")
  printf("\\fB%s\\fR\n", text)
}
/ *<dd *[^>]*>.*<\/dd> *$/ {
  text = $0
  text = strip(text)
  text = gensub("<code *[^>]*>([^<]*)</code>", "\\\\fB\\1\\\\fR", "g", text)
  text = gensub("<var *[^>]*>([^<]*)</var>", "\\\\fI\\1\\\\fR", "g", text)
  gsub("<[^>]*>", "", text)
  text = unescape(text)
  printf(".RS\n")
  printf("%s\n", text)
  printf(".RE\n")
}

/ *<ul *[^>]*> *$/ {
  printf(".PP\n")
  printf(".RS\n")
}
/ *<\/ul> *$/ {
  printf(".RE\n")
}
/ *<li *[^>]*>.*<\/li> *$/ {
  text = $0
  text = strip(text)
  text = gensub("<code *[^>]*>(.*)</code>", "\\\\fB\\1\\\\fR", "g", text)
  text = gensub("<var *[^>]*>([^<]*)</var>", "\\\\fI\\1\\\\fR", "g", text)
  gsub("<[^>]*>", "", text)
  text = unescape(text)
  printf("%s\n", text)
  printf(".br\n")
}

END {
  printf("\n")
  printf(".SH SEE ALSO\n")
  printf(".PP\n")
  printf(".BR tcutil (3),\n")
  printf(".BR tchdb (3),\n")
  printf(".BR tcbdb (3)\n")
  printf(".BR tcfdb (3)\n")
  printf(".PP\n")
  printf("Please see\n")
  printf(".I http://tokyocabinet.sourceforge.net/spex-en.html\n")
  printf("for detail.\n")
}