npm - @thi.ng/hiccup-html-parse - Versions diffs - 0.3.9 → 0.3.10 - Mend

@thi.ng/hiccup-html-parse 0.3.9 → 0.3.10

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.

Files changed (3) hide show

package/CHANGELOG.md CHANGED Viewed

@@ -1,6 +1,6 @@
 # Change Log
-- **Last updated**: 2023-12-09T19:12:03Z
+- **Last updated**: 2023-12-11T10:07:09Z
 - **Generator**: [thi.ng/monopub](https://thi.ng/monopub)
 All notable changes to this project will be documented in this file.

package/index.js CHANGED Viewed

@@ -2,10 +2,7 @@ import { DEFAULT, defmulti } from "@thi.ng/defmulti/defmulti";
 import { defContext } from "@thi.ng/parse/context";
 import { defGrammar } from "@thi.ng/parse/grammar";
 import { unescapeEntities } from "@thi.ng/strings/entities";
-// HTML parse grammar rules (see: thi.ng/parse readme for details)
-// playground URL:
-// https://demo.thi.ng/umbrella/parse-playground/#l9oD3G5vZGU6ICc8JyEgKDxjb21tZW50PiB8IDxjZGF0YV9lbD4gfCA8dm9pZF9lbD4gfCA8ZWw-KSA7CmVsOiA8bmFtZT4gPGF0dHJpYj4qICg8ZWxfYm9keT4gfCA8ZWxfY2xvc2U-ISApIDsKZWxfYm9keTogPFdTMD4gJz4nISAoPGJvZHk-IHwgPG5vZGU-KSogIjwvIiEgPG5hbWU-ISA8V1MwPiAnPichID0-IGhvaXN0IDsKZWxfY2xvc2U6IDxXUzA-ICIvPiIhIDsKbmFtZTogW0EtWmEtejAtOV86XC1dKyA9PiBqb2luIDsKYXR0cmliOiA8V1MxPiA8bmFtZT4gPGF0dHZhbD4_IDsKYXR0dmFsOiAnPSchICg8dmFsPiB8IDxhbHRfdmFsPiB8IDxlbXB0eT4gfCA8YWx0X2VtcHR5PikgOwp2YWw6ICciJyEgLig_KyciJyEpID0-IGpvaW4gOwphbHRfdmFsOiAnXCcnISAuKD8rJ1wnJyEpID0-IGpvaW4gOwplbXB0eTogJyInICciJyA7CmFsdF9lbXB0eTogJ1wnJyEgJ1wnJyEgOwpib2R5OiAuKD8tJzwnISkgPT4gam9pbiA7Cgp2b2lkX2VsOiA8dm9pZF9uYW1lPiA8YXR0cmliPiogPFdTMD4gJy8nPyEgJz4nISA7CnZvaWRfbmFtZTogKCJhcmVhIiB8ICJiYXNlIiB8ICJiciIgfCAiY29sIiB8ICJlbWJlZCIgfCAiaHIiIHwgImltZyIgfCAiaW5wdXQiIHwgImxpbmsiIHwgIm1ldGEiIHwgInNvdXJjZSIgfCAidHJhY2siIHwgIndiciIpIDsKCmNkYXRhX2VsOiA8Y2RhdGFfbmFtZT4gPGF0dHJpYj4qICc-JyEgPGNkYXRhX2JvZHk-IDsKY2RhdGFfbmFtZTogKCJzY3JpcHQiIHwgInN0eWxlIikgOwpjZGF0YV9ib2R5OiAuKD8tPGNkYXRhX2Nsb3NlPiEpIDxjZGF0YV9jbG9zZT4hID0-IGpvaW4gOwpjZGF0YV9jbG9zZTogIjwvIiEgPGNkYXRhX25hbWU-ISA8V1MwPiAnPichIDsKCmRvY3R5cGU6ICI8ISIhICgiZG9jdHlwZSIgfCAiRE9DVFlQRSIpISA8V1MxPiAuKD8rJz4nISkgPFdTMD4gPT4gam9pbiA7CmNvbW1lbnQ6ICIhLS0iISAuKD8rIi0tPiIhKSA9PiBqb2luIDsKCm1haW46IDxTVEFSVD4gPGRvY3R5cGU-PyA8bm9kZT4rIDxFTkQ-IDukbWFpbtoBMjwhZG9jdHlwZSBodG1sPgo8aHRtbCBsYW5nPSJlbiI-CjxoZWFkPgogIDwhLS0gPGlnbm9yZT48L2lnbm9yZT4gLS0-CiAgPHNjcmlwdCBsYW5nPSJqYXZhc2NyaXB0Ij4KY29uc29sZS5sb2coIjwvIisic2NyaXB0PiIpOwogIDwvc2NyaXB0PgogIDxzdHlsZT4KYm9keSB7IG1hcmdpbjogMDsgfQogIDwvc3R5bGU-CjwvaGVhZD4KPGJvZHk-CiAgPGRpdiBpZD0iZm9vIiBib29sIGRhdGEteHl6PSIiIGVtcHR5PScnPgogICAgPGEgaHJlZj0iI2JhciI-YmF6IDxiPmJvbGQ8L2I-PC9hPjxici8-CiAgPC9kaXY-CjwvYm9keT4KPC9odG1sPqCgoKA
-export const lang = defGrammar(`
+const lang = defGrammar(`
 node: '<'! (<comment> | <cdata_el> | <void_el> | <el>) ;
 el: <name> <attrib>* (<el_body> | <el_close>! ) ;
 el_body: <WS0> '>'! (<body> | <node>)* "</"! <name>! <WS0> '>'! => hoist ;
@@ -32,143 +29,122 @@ comment: "!--"! .(?+"-->"!) => join ;
 main: <START> <doctype>? <node>+ <END> ;
 `);
-/**
- * Creates a parser context for given source string and calls the main parser
- * rule. Returns result object, incl. the context for further inspection and
- * transformation.
- *
- * @param src
- * @param opts
- */
-export const parseRaw = (src, opts) => {
-    const ctx = defContext(src, opts);
-    return { result: lang.rules.main(ctx), ctx };
+const parseRaw = (src, opts) => {
+  const ctx = defContext(src, opts);
+  return { result: lang.rules.main(ctx), ctx };
 };
-/**
- * Trims given HTML source string and attempts to parse it into a collection of
- * elements in thi.ng/hiccup format, using provided options to transform, clean
- * or filter elements.
- *
- * @param src
- * @param opts
- */
-export const parseHtml = (src, opts) => {
-    if (!src)
-        return { type: "success", result: [] };
-    opts = {
-        debug: false,
-        collapse: true,
-        unescape: true,
-        maxDepth: 128,
-        ...opts,
+const parseHtml = (src, opts) => {
+  if (!src)
+    return { type: "success", result: [] };
+  opts = {
+    debug: false,
+    collapse: true,
+    unescape: true,
+    maxDepth: 128,
+    ...opts
+  };
+  try {
+    const { result, ctx } = parseRaw(src.trim(), {
+      debug: opts.debug,
+      maxDepth: opts.maxDepth
+    });
+    const loc = {
+      offset: ctx.state.p,
+      line: ctx.state.l,
+      column: ctx.state.c
     };
-    try {
-        const { result, ctx } = parseRaw(src.trim(), {
-            debug: opts.debug,
-            maxDepth: opts.maxDepth,
-        });
-        const loc = {
-            offset: ctx.state.p,
-            line: ctx.state.l,
-            column: ctx.state.c,
-        };
-        if (result) {
-            const acc = [];
-            transformScope(ctx.root, opts, acc);
-            return {
-                type: ctx.done ? "success" : "partial",
-                result: acc,
-                loc,
-            };
-        }
-        else {
-            return { type: "fail", loc };
-        }
-    }
-    catch (e) {
-        return { type: "error", err: e };
+    if (result) {
+      const acc = [];
+      transformScope(ctx.root, opts, acc);
+      return {
+        type: ctx.done ? "success" : "partial",
+        result: acc,
+        loc
+      };
+    } else {
+      return { type: "fail", loc };
     }
+  } catch (e) {
+    return { type: "error", err: e };
+  }
 };
-/**
- * Recursive depth-first transformation function to process the parse tree (this
- * is where the actual conversion to hiccup format happens).
- *
- * @remarks
- * The dispatch values for the various implementations here correspond to the
- * above grammar rules.
- *
- * @internal
- */
-const transformScope = defmulti((x) => x.id, { cdata_el: "el", void_el: "el" }, {
+const transformScope = defmulti(
+  (x) => x.id,
+  { cdata_el: "el", void_el: "el" },
+  {
     [DEFAULT]: (scope) => {
-        throw new Error(`missing impl for scope ID: ${scope.id}`);
+      throw new Error(`missing impl for scope ID: ${scope.id}`);
     },
     // root node of the parse tree
     root: ({ children }, opts, acc) => {
-        if (!children)
-            return;
-        children = children[0].children;
-        if (opts.doctype && children?.[0]) {
-            acc.push(["!DOCTYPE", children[0].result]);
-        }
-        for (let x of children[1].children)
-            transformScope(x, opts, acc);
+      if (!children)
+        return;
+      children = children[0].children;
+      if (opts.doctype && children?.[0]) {
+        acc.push(["!DOCTYPE", children[0].result]);
+      }
+      for (let x of children[1].children)
+        transformScope(x, opts, acc);
     },
     node: ({ children }, opts, acc) => {
-        transformScope(children[0], opts, acc);
+      transformScope(children[0], opts, acc);
     },
     comment: ({ result }, opts, acc) => {
-        if (opts.comments)
-            acc.push(["__COMMENT__", result.trim()]);
+      if (opts.comments)
+        acc.push(["__COMMENT__", result.trim()]);
     },
     // element node transformer, collects & filters attributes/children
     // adds resulting hiccup element to accumulator array
     el: ({ children }, opts, acc) => {
-        const [name, { children: $attribs }, body] = children;
-        if (opts.ignoreElements?.includes(name.result))
-            return;
-        const attribs = {};
-        const el = [name.result, attribs];
-        if ($attribs) {
-            for (let a of $attribs) {
-                const name = a.children[0].result;
-                if (opts.dataAttribs === false && name.startsWith("data-"))
-                    continue;
-                if (opts.ignoreAttribs?.includes(name))
-                    continue;
-                if (a.children[1].children) {
-                    const val = a.children[1].children[0].result;
-                    if (val != null)
-                        attribs[name] = unescapeEntities(val);
-                }
-                else {
-                    attribs[name] = true;
-                }
-            }
+      const [name, { children: $attribs }, body] = children;
+      if (opts.ignoreElements?.includes(name.result))
+        return;
+      const attribs = {};
+      const el = [name.result, attribs];
+      if ($attribs) {
+        for (let a of $attribs) {
+          const name2 = a.children[0].result;
+          if (opts.dataAttribs === false && name2.startsWith("data-"))
+            continue;
+          if (opts.ignoreAttribs?.includes(name2))
+            continue;
+          if (a.children[1].children) {
+            const val = a.children[1].children[0].result;
+            if (val != null)
+              attribs[name2] = unescapeEntities(val);
+          } else {
+            attribs[name2] = true;
+          }
         }
-        if (body) {
-            if (body.result) {
-                el.push(body.result.trim());
-            }
-            else if (body.children) {
-                for (let x of body.children)
-                    transformScope(x, opts, el);
-            }
+      }
+      if (body) {
+        if (body.result) {
+          el.push(body.result.trim());
+        } else if (body.children) {
+          for (let x of body.children)
+            transformScope(x, opts, el);
         }
-        const result = opts.tx ? opts.tx(el) : el;
-        if (result != null)
-            acc.push(result);
+      }
+      const result = opts.tx ? opts.tx(el) : el;
+      if (result != null)
+        acc.push(result);
     },
     // plain text transform (by default only resolves HTML entities)
     body: ({ result }, opts, acc) => {
-        if (!opts.whitespace && /^\s+$/.test(result))
-            return;
-        if (opts.collapse)
-            result = result.replace(/\s+/gm, " ");
-        if (opts.unescape)
-            result = unescapeEntities(result);
-        result = opts.txBody ? opts.txBody(result) : result;
-        if (result != null)
-            acc.push(result);
-    },
-});
+      if (!opts.whitespace && /^\s+$/.test(result))
+        return;
+      if (opts.collapse)
+        result = result.replace(/\s+/gm, " ");
+      if (opts.unescape)
+        result = unescapeEntities(result);
+      result = opts.txBody ? opts.txBody(result) : result;
+      if (result != null)
+        acc.push(result);
+    }
+  }
+);
+export {
+  lang,
+  parseHtml,
+  parseRaw
+};

package/package.json CHANGED Viewed

@@ -1,6 +1,6 @@
 {
   "name": "@thi.ng/hiccup-html-parse",
-  "version": "0.3.9",
+  "version": "0.3.10",
   "description": "Well-formed HTML parsing and customizable transformation to nested JS arrays in @thi.ng/hiccup format",
   "type": "module",
   "module": "./index.js",
@@ -24,7 +24,9 @@
   "author": "Karsten Schmidt (https://thi.ng)",
   "license": "Apache-2.0",
   "scripts": {
-    "build": "yarn clean && tsc --declaration",
+    "build": "yarn build:esbuild && yarn build:decl",
+    "build:decl": "tsc --declaration --emitDeclarationOnly",
+    "build:esbuild": "esbuild --format=esm --platform=neutral --target=es2022 --tsconfig=tsconfig.json --outdir=. src/**/*.ts",
     "clean": "rimraf --glob '*.js' '*.d.ts' '*.map' doc",
     "doc": "typedoc --excludePrivate --excludeInternal --out doc src/index.ts",
     "doc:ae": "mkdir -p .ae/doc .ae/temp && api-extractor run --local --verbose",
@@ -33,13 +35,14 @@
     "test": "bun test"
   },
   "dependencies": {
-    "@thi.ng/api": "^8.9.11",
-    "@thi.ng/defmulti": "^3.0.9",
-    "@thi.ng/parse": "^2.4.9",
-    "@thi.ng/strings": "^3.7.2"
+    "@thi.ng/api": "^8.9.12",
+    "@thi.ng/defmulti": "^3.0.10",
+    "@thi.ng/parse": "^2.4.10",
+    "@thi.ng/strings": "^3.7.3"
   },
   "devDependencies": {
     "@microsoft/api-extractor": "^7.38.3",
+    "esbuild": "^0.19.8",
     "rimraf": "^5.0.5",
     "tools": "^0.0.1",
     "typedoc": "^0.25.4",
@@ -82,5 +85,5 @@
     "status": "alpha",
     "year": 2023
   },
-  "gitHead": "25f2ac8ff795a432a930119661b364d4d93b59a0\n"
+  "gitHead": "5e7bafedfc3d53bc131469a28de31dd8e5b4a3ff\n"
 }