diff options
Diffstat (limited to 'aux/emparse')
| -rwxr-xr-x | aux/emparse | 163 |
1 files changed, 163 insertions, 0 deletions
diff --git a/aux/emparse b/aux/emparse new file mode 100755 index 0000000..d6073f3 --- /dev/null +++ b/aux/emparse @@ -0,0 +1,163 @@ +#!/bin/awk -f + +# aux/emparse -- parse em source + +function beginblock(name) { + if (name == "nl") printf "<ol>\n" + else printf "<%s>\n", name +} + +function endblock(name) { + if (name == "nl") printf "</ol>\n" + else printf "</%s>\n", name +} + +function newblock(name) { + beginblock(name) + openblock = name + openitem = 0 + expectblock = 0 +} + +function breakblock() { + if (openitem) { + printf "</li>\n" + while (itemlevel-- > 1) + endblock(openblock) + itemlevel = 1 + } + if (opendef) printf "</dd>\n" + if (openblock) endblock(openblock) + openitem = 0 + opendef = 0 + openblock = 0 + expectblock = 1 +} + +function heading(level, line) { + sub("^=* ", "", line) + sub(" =*$", "", line) + printf "<h%d>%s</h%d>\n", level, line, level + # should inline formatting be supported in headings? +} + +function item(level, type, line) { + if (openitem) printf "</li>" + openitem = 1 + if (level > itemlevel) printf "<ul>\n" + if (level < itemlevel) printf "</ul>\n" + itemlevel = level + if (type == "ul") { + sub("^ +- ", "") + printf "<li>" + } + if (type == "ol") { + match($0, "[0-9a-z]+\.") + v = substr($0, RSTART, RLENGTH-1) + sub("^ +[0-9a-z]+\. ", "") + printf "<li value=\"%s\">", v + } + if (type == "nl") { + match($0, "\\[[0-9a-z]+\\]") + v = substr($0, RSTART+1, RLENGTH-2) + sub("^ +\\[[0-9a-z]+\\] ", "") + printf "<li value=\"%s\" id=\"ref%s\">", v, v + } +} + +function term(line, t) { # t is a local variable + if (opendef) printf "</dd>" + opendef = 1 + t = $0; d = $0 + sub("^ ", "", t) + sub(": .*$", "", t) + sub("^ [^:]+: ", "") + printf "<dt>%s</dt><dd>", t +} + +function ref(v) { + if (ENVIRON["ref" v] != "") + return "(<a href=\"" ENVIRON["ref" v] "\">" linktext "</a>)" + else + return "[<a href=\"#ref" v "\">" v "</a>]" +} + +function format(line) { + n = split(line, w, "[ ]") + for (i = 0; i <= n; i++) { + if (w[i] == "") continue; + + if (match(w[i], "^\\[[0-9a-z]+\\]$")) + w[i] = ref(substr(w[i], 2, RLENGTH-2)) + else if (match(w[i], "^\[[0-9a-z]+\][.,:;?!]$")) + w[i] = ref(substr(w[i], 2, RLENGTH-3)) substr(w[i], RLENGTH) + + if (match(w[i], "^`")) + w[i] = "<tt>" substr(w[i], 2) + else if (match(w[i], "^\\*")) + w[i] = "<i>" substr(w[i], 2) + else if (match(w[i], "^_")) + w[i] = "<b>" substr(w[i], 2) + + if (match(w[i], "`$")) + w[i] = substr(w[i], 1, RSTART-1) "</tt>" + else if (match(w[i], "\\*$")) + w[i] = substr(w[i], 1, RSTART-1) "</i>" + else if (match(w[i], "_$")) + w[i] = substr(w[i], 1, RSTART-1) "</b>" + else if (match(w[i], "`[.,:;?!]$")) + w[i] = substr(w[i], 1, RSTART-1) "</tt>" substr(w[i], RSTART+1) + else if (match(w[i], "\\*\.$")) + w[i] = substr(w[i], 1, RSTART-1) "</i>" substr(w[i], RSTART+1) + else if (match(w[i], "_[.,:;?!]$")) + w[i] = substr(w[i], 1, RSTART-1) "</b>" substr(w[i], RSTART+1) + + printf "%s", w[i] + if (i < n) printf " " + } + printf "\n" +} + +BEGIN { + expectblock = 1 + itemlevel = 1 + linktext = ENVIRON["linktext"] + if (!linktext) linktext = "link" +} +END { breakblock() } + +/^$/ { breakblock(); getline } +expectblock && /^ / { newblock("table") } +expectblock && /^ - / { newblock("ul") } +expectblock && /^ [0-9a-z]+\. / { newblock("ol") } +expectblock && /^ \[[0-9a-z]\]+ / { newblock("nl") } +expectblock && /^ .*: / { newblock("dl") } +expectblock && /^ / { newblock("pre"); sub("^ ", "") } +expectblock && /^---$/ { expectblock = 0; printf "<hr/>\n"; next } +expectblock && /^= .* =$/ { heading(1, $0); next } +expectblock && /^== .* ==$/ { heading(3, $0); next } +expectblock && /^=== .* ===$/ { heading(3, $0); next } +expectblock && /^==== .* ====$/ { heading(4, $0); next } +expectblock && /^===== .* =====$/ { heading(5, $0); next } +expectblock && /^====== .* ======$/ { heading(6, $0); next } +expectblock { newblock("p") } + +block = "ul" && /^ - / { item(1, "ul", line) } +block = "ol" && /^ [0-9a-z]+\. / { item(1, "ol", line) } +block = "nl" && /^ \[[0-9a-z]\]+ / { item(1, "nl", line) } +block = "dl" && /^ .*: / { term(line) } + +block = "ul" && /^ -/ { item(2, "ul", line) } +block = "ol" && /^ [0-9a-z]+\./ { item(2, "ol", line) } +block = "ul" && /^ -/ { item(3, "ul", line) } +block = "ol" && /^ [0-9a-z]+\./ { item(3, "ol", line) } +block = "ul" && /^ -/ { item(4, "ul", line) } +block = "ol" && /^ [0-9a-z]+\./ { item(4, "ol", line) } +block = "ul" && /^ -/ { item(5, "ul", line) } +block = "ol" && /^ [0-9a-z]+\./ { item(5, "ol", line) } +block = "ul" && /^ -/ { item(6, "ul", line) } +block = "ol" && /^ [0-9a-z]+\./ { item(6, "ol", line) } + +block = "pre" && /^ / { sub("^ ", ""); printf "%s\n", $0; next } + +{ format($0) } |