[caffeine-hx] r736 committed - Metadata to tag mapping

[caffe...@googlecode.com]

Revision: 736
Author: damon...@gmail.com
Date: Sun Aug 5 08:29:04 2012
Log: Metadata to tag mapping


http://code.google.com/p/caffeine-hx/source/detail?r=736

Modified:
/trunk/projects/chxdoc/src/chxdoc/DocProcessor.hx

=======================================
--- /trunk/projects/chxdoc/src/chxdoc/DocProcessor.hx Fri Aug 3 20:43:29
2012
+++ /trunk/projects/chxdoc/src/chxdoc/DocProcessor.hx Sun Aug 5 08:29:04
2012
@@ -68,14 +68,55 @@
};
if(doc != null)
this.doc = doc.split("\r\n").join("\n").split("\r").join("\n");
- /*
- <meta>
- <m n="values">
- <e>-1</e>
- <e>100</e>
- </m>
- </meta>
- */
+ this.meta = meta;
+ }
+
+ public static function process(pkg : PackageContext, ctx: Ctx, doc :
String, meta : Xml) : DocsContext
+ {
+ if( (doc == null || doc.length == 0) && (meta == null |
| !ChxDocMain.config.showMeta) )
+ return null;
+ var p = new DocProcessor(pkg, ctx, doc, meta);
+ return p.convert();
+ }
+
+ /**
+ **/
+ function convert() : DocsContext {
+ if(doc != null) {
+ // trim stars
+ doc = ~/^([ \t]*)\*+/gm.replace(doc, "$1");
+ doc = ~/\**[ \t]*$/gm.replace(doc, "");
+
+ var parts = doTags(doc.split("\n"));
+
+ var ch = doCodeBlockShortcut(parts.join("\n"));
+
+ // separate into paragraphs
+ parts = ~/\n[ \t]*\n/g.split(ch.parsed);
+ if( parts.length == 1 )
+ doc = parts[0];
+ else
+ doc = Lambda.map(parts,function(x) {
return "<p>"+StringTools.trim(x)+"</p>"; }).join("\n");
+
+ // put back code parts
+ var i = 0;
+ for( c in ch.codeBlocks )
+ ch.parsed = ch.parsed.split("##__code__"+(i++)+"##").join(c);
+ docCtx.comments = ch.parsed;
+
+ // since tags are parsed bottom->up, we will reverse all the
+ // arrays just so docs reflect the same order
+ docCtx.authors.reverse();
+ docCtx.params.reverse();
+ docCtx.requires.reverse();
+ docCtx.returns.reverse();
+ docCtx.see.reverse();
+ docCtx.throws.reverse();
+ docCtx.todos.reverse();
+ docCtx.typeParams.reverse();
+ }
+
+ /// <meta><m n="values"><e>-1</e><e>100</e></m></meta>
if(meta != null && ChxDocMain.config.showMeta) {
var fast = new Fast(Xml.parse(Std.string(meta)).firstElement());
for(m in fast.nodes.m) {
@@ -91,63 +132,60 @@
res.value += ",";
res.value += e.innerData;
}
+ var recognized = false;
+ // merge metadata with same names as doc tags
if(ChxDocMain.config.mergeMeta) {
-
+ recognized = true;
switch(res.name) {
case "author":
- this.docCtx.authors.push(res.value);
+ docCtx.authors.push(doEmbeddedTags(res.value));
continue;
+ case "deprecated":
+ docCtx.deprecated = true;
+ if(docCtx.deprecatedMsg == null)
+ docCtx.deprecatedMsg = doEmbeddedTags(res.value);
+ else if(docCtx.deprecatedMsg != res.value)
+ docCtx.deprecatedMsg += " " + doEmbeddedTags(res.value);
+ //case "param":
+ case "requires":
+ docCtx.requires.push(doEmbeddedTags(res.value));
+ case "return", "returns":
+ docCtx.returns.push(doEmbeddedTags(res.value));
+ case "see":
+ docCtx.see.push(doEmbeddedTags(res.value));
+ case "since":
+ docCtx.since.push(doEmbeddedTags(res.value));
+ case "throw", "throws":
+ var p = res.value.split(" ");
+ var e = p.shift();
+ docCtx.throws.push( {
+ name : e,
+ uri : pkg.rootRelative + (StringTools.replace(e,".","/")) +
ChxDocMain.config.htmlFileExtension,
+ desc : doEmbeddedTags(p.join(" ")),
+ });
+ case "private":
+ docCtx.forcePrivate = true;
+ case "todo":
+ var msg = doEmbeddedTags(res.value);
+ ChxDocMain.registerTodo(pkg, ctx, msg);
+ if(ChxDocMain.config.showTodoTags)
+ docCtx.todos.push(msg);
+ case "type":
+ var p = res.value.split(" ");
+ docCtx.typeParams.push({
+ arg : p.shift(),
+ desc : doEmbeddedTags(p.join(" "))
+ });
+ case "version":
+ docCtx.version.push(doEmbeddedTags(res.value));
+ default:
+ recognized = false;
}
}
- docCtx.meta.push(res);
+ if(!recognized)
+ docCtx.meta.push(res);
}
}
- }
-
- public static function process(pkg : PackageContext, ctx: Ctx, doc :
String, meta : Xml) : DocsContext
- {
- if( (doc == null || doc.length == 0) && (meta == null |
| !ChxDocMain.config.showMeta) )
- return null;
- var p = new DocProcessor(pkg, ctx, doc, meta);
- return p.convert();
- }
-
- /**
- **/
- function convert() : DocsContext {
- if(doc == null)
- return docCtx;
- // trim stars
- doc = ~/^([ \t]*)\*+/gm.replace(doc, "$1");
- doc = ~/\**[ \t]*$/gm.replace(doc, "");
-
- var parts = doTags(doc.split("\n"));
-
- var ch = doCodeBlockShortcut(parts.join("\n"));
-
- // separate into paragraphs
- parts = ~/\n[ \t]*\n/g.split(ch.parsed);
- if( parts.length == 1 )
- doc = parts[0];
- else
- doc = Lambda.map(parts,function(x) {
return "<p>"+StringTools.trim(x)+"</p>"; }).join("\n");
-
- // put back code parts
- var i = 0;
- for( c in ch.codeBlocks )
- ch.parsed = ch.parsed.split("##__code__"+(i++)+"##").join(c);
- docCtx.comments = ch.parsed;
-
- // since tags are parsed bottom->up, we will reverse all the
- // arrays just so docs reflect the same order
- docCtx.authors.reverse();
- docCtx.params.reverse();
- docCtx.requires.reverse();
- docCtx.returns.reverse();
- docCtx.see.reverse();
- docCtx.throws.reverse();
- docCtx.todos.reverse();
- docCtx.typeParams.reverse();
return docCtx;
}