diff options
| -rw-r--r-- | Makefile | 10 | ||||
| -rw-r--r-- | README.md | 3 | ||||
| -rw-r--r-- | adapter/format/md2html.nim | 380 | ||||
| -rw-r--r-- | res/mime.types | 1 | ||||
| -rw-r--r-- | src/config/config.nim | 35 |
5 files changed, 413 insertions, 16 deletions
diff --git a/Makefile b/Makefile index 7b13deb1..ba29138f 100644 --- a/Makefile +++ b/Makefile @@ -45,7 +45,8 @@ all: $(OUTDIR_BIN)/cha $(OUTDIR_BIN)/mancha $(OUTDIR_CGI_BIN)/http \ $(OUTDIR_CGI_BIN)/cha-finger $(OUTDIR_CGI_BIN)/about \ $(OUTDIR_CGI_BIN)/data $(OUTDIR_CGI_BIN)/file $(OUTDIR_CGI_BIN)/ftp \ $(OUTDIR_CGI_BIN)/man $(OUTDIR_CGI_BIN)/spartan \ - $(OUTDIR_LIBEXEC)/urldec $(OUTDIR_LIBEXEC)/urlenc + $(OUTDIR_LIBEXEC)/urldec $(OUTDIR_LIBEXEC)/urlenc \ + $(OUTDIR_LIBEXEC)/md2html $(OUTDIR_BIN)/cha: lib/libquickjs.a src/*.nim src/**/*.nim src/**/*.c res/* \ res/**/* res/map/idna_gen.nim @@ -81,6 +82,11 @@ $(OUTDIR_LIBEXEC)/gopher2html: adapter/format/gopher2html.nim \ $(NIMC) $(FLAGS) --nimcache:"$(OBJDIR)/$(TARGET)/gopher2html" \ -o:"$(OUTDIR_LIBEXEC)/gopher2html" adapter/format/gopher2html.nim +$(OUTDIR_LIBEXEC)/md2html: adapter/format/md2html.nim + @mkdir -p "$(OUTDIR_LIBEXEC)" + $(NIMC) $(FLAGS) --nimcache:"$(OBJDIR)/$(TARGET)/md2html" \ + -o:"$(OUTDIR_LIBEXEC)/md2html" adapter/format/md2html.nim + GMIFETCH_CFLAGS = -Wall -Wextra -std=c89 -pedantic -lcrypto -lssl -g -O3 $(OUTDIR_CGI_BIN)/gmifetch: adapter/protocol/gmifetch.c @mkdir -p "$(OUTDIR_CGI_BIN)" @@ -216,6 +222,7 @@ install: install -m755 "$(OUTDIR_CGI_BIN)/ftp" $(LIBEXECDIR_CHAWAN)/cgi-bin install -m755 "$(OUTDIR_CGI_BIN)/gopher" $(LIBEXECDIR_CHAWAN)/cgi-bin install -m755 "$(OUTDIR_LIBEXEC)/gopher2html" $(LIBEXECDIR_CHAWAN) + install -m755 "$(OUTDIR_LIBEXEC)/md2html" $(LIBEXECDIR_CHAWAN) install -m755 "$(OUTDIR_LIBEXEC)/gmi2html" $(LIBEXECDIR_CHAWAN) install -m755 "$(OUTDIR_CGI_BIN)/gmifetch" $(LIBEXECDIR_CHAWAN)/cgi-bin install -m755 "$(OUTDIR_CGI_BIN)/cha-finger" $(LIBEXECDIR_CHAWAN)/cgi-bin @@ -253,6 +260,7 @@ uninstall: rm -f $(LIBEXECDIR_CHAWAN)/cgi-bin/spartan rmdir $(LIBEXECDIR_CHAWAN)/cgi-bin || true rm -f $(LIBEXECDIR_CHAWAN)/gopher2html + rm -f $(LIBEXECDIR_CHAWAN)/md2html rm -f $(LIBEXECDIR_CHAWAN)/gmi2html rm -f $(LIBEXECDIR_CHAWAN)/urldec rm -f $(LIBEXECDIR_CHAWAN)/urlenc diff --git a/README.md b/README.md index 696ed4ec..11facc87 100644 --- a/README.md +++ b/README.md @@ -111,7 +111,8 @@ the configuration; please consult [doc/config.md](doc/config.md) for details.) ### Can I view Markdown files using Chawan? -[Yes.](doc/mailcap.md) +Yes; Chawan now has a built-in markdown converter. If you don't like it, you +can always [replace it](doc/mailcap.md) with e.g. pandoc. ### Why write another web browser? diff --git a/adapter/format/md2html.nim b/adapter/format/md2html.nim new file mode 100644 index 00000000..1d68e7b9 --- /dev/null +++ b/adapter/format/md2html.nim @@ -0,0 +1,380 @@ +import std/strutils + +proc toggle[T](s: var set[T], t: T): bool = + result = t notin s + if result: + s.incl(t) + else: + s.excl(t) + +type BracketState = enum + bsNone, bsInBracketRef, bsInBracket, bsAfterBracket, bsInParen, bsInImage + +proc getId(line: openArray[char]): string = + result = "" + var i = 0 + var bs = bsNone + var escape = false + while i < line.len: + let c = line[i] + if bs == bsInParen: + if escape: + escape = false + inc i + continue + if c == ')': + bs = bsNone + elif c == '\\': + escape = true + inc i + continue + case c + of 'A'..'Z': result &= char(int(c) - int('A') + int('a')) + of 'a'..'z', '-', '_', '.': result &= c + of ' ': result &= '-' + of '[': + if bs != bsNone: + bs = bsInBracket + of ']': + if bs == bsInBracket: + bs = bsAfterBracket + of '(': + if bs == bsAfterBracket: + bs = bsInParen + else: discard + inc i + +type InlineState = enum + isItalic, isBold, isCode, isComment + +const AsciiWhitespace = {' ', '\t', '\n', '\r'} +proc parseInline(line: openArray[char]) = + var state: set[InlineState] = {} + var bs = bsNone + var i = 0 + var bracketChars = "" + var quote = false + var image = false + template append(s: untyped) = + if bs in {bsInBracketRef, bsInBracket}: + bracketChars &= s + else: + stdout.write(s) + while i < line.len: + let c = line[i] + if bs == bsAfterBracket and c != '(': + stdout.write("[" & bracketChars & "]") + bracketChars = "" + bs = bsNone + image = false + if quote: + append c + elif isComment in state: + if i + 2 < line.len and line.toOpenArray(i, i + 2) == "-->": + state.excl(isComment) + append "-->" + i += 2 + else: + append c + elif isCode in state: + case c + of '<': append "<" + of '>': append ">" + of '"': append """ + of '\'': append "'" + of '`': + append "</CODE>" + state.excl(isCode) + else: append c + elif c == '\\': + quote = true + elif c == '*' or c == '_' and (i == 0 or line[i - 1] in AsciiWhitespace): + if i + 1 < line.len and line[i + 1] == c: + if state.toggle(isBold): + append "<B>" + else: + append "</B>" + inc i + else: + if state.toggle(isItalic): + stdout.write("<I>") + else: + stdout.write("</I>") + elif c == '`': + state.incl(isCode) + append "<CODE>" + elif c == '!' and bs == bsNone and i + 1 < line.len and line[i + 1] == '[': + image = true + elif c == '[' and bs == bsNone: + bs = bsInBracket + if i + 1 < line.len and line[i + 1] == '^': + inc i + bs = bsInBracketRef + elif c == ']' and bs == bsInBracketRef: + let id = bracketChars.getId() + stdout.write("<A HREF='#" & id & "'>" & bracketChars & "</A>") + bracketChars = "" + elif c == ']' and bs == bsInBracket: + bs = bsAfterBracket + elif c == '(' and bs == bsAfterBracket: + if image: + stdout.write("<IMG SRC='") + else: + stdout.write("<A HREF='") + bs = bsInParen + elif c == ')' and bs == bsInParen: + if image: + stdout.write("' ALT='" & bracketChars & "'>") + else: + stdout.write("'>" & bracketChars & "</A>") + image = false + bracketChars = "" + bs = bsNone + elif c == '\'' and bs == bsInParen: + stdout.write("'") + elif i + 4 < line.len and line.toOpenArray(i, i + 3) == "<!--": + append "<!--" + i += 3 + state.incl(isComment) + else: + append c + inc i + if bracketChars != "": + stdout.write(bracketChars) + if isBold in state: + stdout.write("</B>") + if isItalic in state: + stdout.write("</I>") + +proc parseHash(line: openArray[char]): bool = + var n = -1 + for i, c in line: + if line[i] != '#': + if line[i] != ' ': + return false + n = i + 1 + break + if n == -1: + return false + n = min(n, 6) + let L = n + var H = line.high + for i in countdown(line.high, L): + if line[i] != '#': + if line[i] != ' ': + break + H = i - 1 + break + H = max(L - 1, H) + let id = line.toOpenArray(L, H).getId() + stdout.write("<H" & $n & " id='" & id & "'>") + line.toOpenArray(L, H).parseInline() + stdout.write("</H" & $n & ">\n") + return true + +type ListType = enum + ltOl, ltUl + +proc getListDepth(line: string): tuple[depth, len: int, ol: ListType] = + var depth = 0 + for i, c in line: + if c == '\t': + depth += 8 + elif c == ' ': + inc depth + elif c == '*': + let i = i + 1 + if i < line.len and line[i] in {'\t', ' '}: + return (depth, i, ltUl) + break + elif c in {'0'..'9'}: + let i = i + 1 + if i < line.len and line[i] == '.': + let i = i + 1 + if i < line.len and line[i] in {'\t', ' '}: + return (depth, i, ltOl) + break + else: + break + return (-1, -1, ltUl) + +proc matchHTMLPreStart(line: string): bool = + var tagn = "" + for i, c in line: + if i == 0: + if c != '<': + return false + continue + if c in {' ', '\t', '>'}: + break + if c notin {'A'..'Z', 'a'..'z'}: + return false + tagn &= c.toLowerAscii() + return tagn in ["pre", "script", "style", "textarea"] + +proc matchHTMLPreEnd(line: string): bool = + var tagn = "" + for i, c in line: + if i == 0: + if c != '<': + return false + continue + if i == 1: + if c != '/': + return false + continue + if c in {' ', '\t', '>'}: + break + if c notin {'A'..'Z', 'a'..'z'}: + return false + tagn &= c.toLowerAscii() + return tagn in ["pre", "script", "style", "textarea"] + +type + BlockType = enum + btNone, btPar, btList, btPre, btHTML, btHTMLPre, btComment + + ParseState = object + blockType: BlockType + blockData: string + listDepth: int + lists: seq[ListType] + hasp: bool + reprocess: bool + +proc pushList(state: var ParseState, t: ListType) = + case t + of ltOl: stdout.write("<OL>\n<LI>") + of ltUl: stdout.write("<UL>\n<LI>") + state.lists.add(t) + +proc popList(state: var ParseState) = + case state.lists.pop() + of ltOl: stdout.write("</OL>\n") + of ltUl: stdout.write("</UL>\n") + +proc parseNone(state: var ParseState, line: string) = + if line == "": + discard + elif line[0] == '#' and line.toOpenArray(1, line.high).parseHash(): + discard + elif line.startsWith("<!--"): + state.blockType = btComment + state.reprocess = true + elif line[0] == '<' and line.find('>') == line.high: + state.blockType = if line.matchHTMLPreStart(): btHTMLPre else: btHTML + state.reprocess = true + elif line.len >= 3 and line.startsWith("```"): + state.blockType = btPre + stdout.write("<PRE>") + elif (let (n, len, t) = line.getListDepth(); n != -1): + state.blockType = btList + state.listDepth = n + state.hasp = false + state.pushList(t) + state.blockData = line.substr(len) & "\n" + else: + state.blockType = btPar + state.hasp = true + stdout.write("<P>\n") + state.reprocess = true + +proc parsePre(state: var ParseState, line: string) = + if line.startsWith("```"): + state.blockType = btNone + stdout.write("</PRE>\n") + else: + stdout.write(line & "\n") + +proc parseList(state: var ParseState, line: string) = + if line == "": + state.blockData.parseInline() + state.blockData = "" + while state.lists.len > 0: + state.popList() + state.blockType = btNone + elif (let (n, len, t) = line.getListDepth(); n != -1): + state.blockData.parseInline() + state.blockData = "" + if n < state.listDepth: + if state.lists.len > 0: + state.popList() + else: + state.pushList(t) + elif n > state.listDepth: + state.pushList(t) + stdout.write("<LI>") + state.listDepth = n + state.blockData = line.substr(len) & "\n" + else: + state.blockData &= line & "\n" + +proc parsePar(state: var ParseState, line: string) = + if line == "": + state.blockData.parseInline() + state.blockData = "" + state.blockType = btNone + elif line[0] == '<' and line.find('>') == line.high: + state.blockData.parseInline() + state.blockData = "" + if line.matchHTMLPreStart(): + state.blockType = btHTMLPre + else: + state.blockType = btHTML + state.reprocess = true + elif line.len >= 3 and line.startsWith("```"): + state.blockData.parseInline() + state.blockData = "" + state.blockType = btPre + state.hasp = false + stdout.write("<PRE>") + else: + state.blockData &= line & "\n" + +proc parseHTML(state: var ParseState, line: string) = + if state.hasp: + state.hasp = false + stdout.write("</P>\n") + if line == "": + state.blockData.parseInline() + state.blockData = "" + state.blockType = btNone + else: + state.blockData &= line & "\n" + +proc parseHTMLPre(state: var ParseState, line: string) = + if state.hasp: + state.hasp = false + stdout.write("</P>\n") + if line.matchHTMLPreEnd(): + stdout.write(state.blockData) + state.blockData = "" + state.blockType = btNone + else: + state.blockData &= line & "\n" + +proc parseComment(state: var ParseState, line: string) = + let i = line.find("-->") + if i != -1: + stdout.write(line.substr(0, i + 2)) + state.blockType = btNone + line.substr(i + 3).parseInline() + else: + stdout.write(line & "\n") + +proc main() = + var line: string + var state = ParseState(listDepth: -1) + while state.reprocess or stdin.readLine(line): + state.reprocess = false + case state.blockType + of btNone: state.parseNone(line) + of btPre: state.parsePre(line) + of btList: state.parseList(line) + of btPar: state.parsePar(line) + of btHTML: state.parseHTML(line) + of btHTMLPre: state.parseHTMLPre(line) + of btComment: state.parseComment(line) + state.blockData.parseInline() + +main() diff --git a/res/mime.types b/res/mime.types index cb3ad6cc..b4a9065f 100644 --- a/res/mime.types +++ b/res/mime.types @@ -8,3 +8,4 @@ application/xhtml+xml xhtml xhtm xht text/plain txt text/css css image/png png +text/markdown md diff --git a/src/config/config.nim b/src/config/config.nim index 59cdad50..96b78bb7 100644 --- a/src/config/config.nim +++ b/src/config/config.nim @@ -356,20 +356,9 @@ proc getMailcap*(config: Config): tuple[mailcap: Mailcap, errs: seq[string]] = let gopherPath = gopherPath0.unquote().get const geminiPath0 = ChaPath("${%CHA_LIBEXEC_DIR}/gmi2html") let geminiPath = geminiPath0.unquote().get - var mailcap = @[ - MailcapEntry( - mt: "text", - subt: "gopher", - cmd: gopherPath, - flags: {HTMLOUTPUT} - ), - MailcapEntry( - mt: "text", - subt: "gemini", - cmd: geminiPath, - flags: {HTMLOUTPUT} - ) - ] + const mdPath0 = ChaPath("${%CHA_LIBEXEC_DIR}/md2html") + let mdPath = mdPath0.unquote().get + var mailcap: Mailcap = @[] var errs: seq[string] var found = false for p in config.external.mailcap: @@ -381,6 +370,24 @@ proc getMailcap*(config: Config): tuple[mailcap: Mailcap, errs: seq[string]] = else: errs.add(res.error) found = true + mailcap.add(MailcapEntry( + mt: "text", + subt: "gopher", + cmd: gopherPath, + flags: {HTMLOUTPUT} + )) + mailcap.add(MailcapEntry( + mt: "text", + subt: "gemini", + cmd: geminiPath, + flags: {HTMLOUTPUT} + )) + mailcap.add(MailcapEntry( + mt: "text", + subt: "markdown", + cmd: mdPath, + flags: {HTMLOUTPUT} + )) if not found: mailcap.insert(MailcapEntry( mt: "*", |