@mdgate/ppt 0.1.0

package/README.md

@@ -0,0 +1,26 @@
+# @mdgate/ppt
+
+Convert legacy binary PowerPoint presentations to Markdown. Outputs GitHub-Flavored Markdown. Works in Node, Edge, and
+browsers. No native addons.
+
+Handles: `.ppt`, `.pps`, `.pot` (binary PowerPoint)
+
+## Usage
+
+```ts
+import { toMarkdown } from '@mdgate/ppt';
+
+const markdown = await toMarkdown(bytes);
+```
+
+Compose with other converters:
+
+```ts
+import { create } from '@mdgate/core';
+import { ppt } from '@mdgate/ppt';
+
+const convert = create([ppt()]);
+```
+
+Part of [mdgate converters](https://github.com/mdgate/converters); install
+`@mdgate/converters` for every format at once.

package/dist/index.d.ts

@@ -0,0 +1,2 @@
+export { ppt } from './ppt.js';
+export { toMarkdown } from './to-markdown.js';

package/dist/index.js

@@ -0,0 +1 @@
+import{detectOleDoc as w8}from"@mdgate/containers";import{documentToMarkdown as B8}from"@mdgate/document";import{fileExtension as X8}from"@mdgate/utils";import{CompoundFile as s,getU32 as X,MAX_ENTRY_BYTES as t,MAX_RECORD_DEPTH as n,MAX_RECORDS as p,readOleStream as d}from"@mdgate/containers";import{ConvertError as P}from"@mdgate/core";import{inlinesAreEmpty as e,inlinesToPlainText as z8,PLAIN as G8,stylesEqual as H8}from"@mdgate/document";import{AssetSink as J8,decodeBlip as K8,fbseBlip as Q8,flushList as L,rebaseEmphasis as V8,recordAt as D,resolveDelta as W8}from"@mdgate/office-common";import{cleanText as v,debug as o,warn as Z8}from"@mdgate/utils";import{getU16 as U,getU32 as R}from"@mdgate/containers";import{debug as f}from"@mdgate/utils";function x(){return{paragraphs:[],chars:[]}}function u(){return{bullet:void 0,bold:void 0,italic:void 0}}function b(H,J){let z=x(),G=0,Q=0;while(Q<=J){let K=R(H,G);if(K===void 0)break;let V=U(H,G+4);if(V===void 0)break;G+=6;let W=y(H,G);if(W===void 0)return f("unparseable paragraph style run; keeping styling parsed so far"),z;if(G=W.next,z.paragraphs.push({count:K,depth:V,bullet:W.bullet}),Q+=K,K===0)break}Q=0;while(Q<=J){let K=R(H,G);if(K===void 0)break;G+=4;let V=l(H,G);if(V===void 0){f("unparseable character style run; dropping remaining styling");break}if(G=V.next,z.chars.push({count:K,bold:V.bold,italic:V.italic}),Q+=K,K===0)break}return z}function y(H,J){let z=J,G=R(H,z);if(G===void 0)return;z+=4;let Q;if((G&15)!==0){let K=U(H,z);if(K===void 0)return;if((G&1)!==0)Q=(K&1)!==0;z+=2}if((G&128)!==0)z+=2;if((G&16)!==0)z+=2;if((G&64)!==0)z+=2;if((G&32)!==0)z+=4;if((G&2048)!==0)z+=2;if((G&4096)!==0)z+=2;if((G&8192)!==0)z+=2;if((G&16384)!==0)z+=2;if((G&256)!==0)z+=2;if((G&1024)!==0)z+=2;if((G&32768)!==0)z+=2;if((G&1048576)!==0){let K=U(H,z);if(K===void 0)return;z+=2+K*4}if((G&65536)!==0)z+=2;if((G&917504)!==0)z+=2;if((G&2097152)!==0)z+=2;if(z>H.length)return;return{bullet:Q,next:z}}function l(H,J){let z=J,G=R(H,z);if(G===void 0)return;z+=4;let Q,K;if((G&65535)!==0){let V=U(H,z);if(V===void 0)return;if((G&1)!==0)Q=(V&1)!==0;if((G&2)!==0)K=(V&2)!==0;z+=2}if((G&65536)!==0)z+=2;if((G&2097152)!==0)z+=2;if((G&4194304)!==0)z+=2;if((G&8388608)!==0)z+=2;if((G&131072)!==0)z+=2;if((G&262144)!==0)z+=4;if((G&524288)!==0)z+=2;if(z>H.length)return;return{bold:Q,italic:K,next:z}}function c(H,J){let z=U(H,0);if(z===void 0)return[];let G=2,Q=[],K=Math.min(z,10);for(let V=0;V<K;V+=1){if(J>=5)G+=2;let W=y(H,G);if(W===void 0)break;G=W.next;let Z=l(H,G);if(Z===void 0)break;G=Z.next,Q.push({bullet:W.bullet,bold:Z.bold,italic:Z.italic})}return Q}function r(H){let J;try{J=s.open(H)}catch(V){if(V instanceof P&&V.code!=="malformed")throw V;let W=V instanceof P&&V.detail!==void 0?V.detail:V instanceof Error?V.message:String(V);throw P.malformed(`not an OLE2 compound file: ${W}`)}let z=d(J,"PowerPoint Document"),G=a(J,"Current User")??new Uint8Array;if(X(G,12)===4090610911)throw P.encrypted();let Q=new g;if(!Q.parseSlides(z,G))Z8("ppt persist directory unusable; recovering text in raw stream order"),Q=new g,Q.recovering=!0,Q.walk(z),Q.endSegment(void 0);if(Q.encrypted)throw P.encrypted();let K=$8(J);return{blocks:Q.intoBlocks(),notes:[],assets:K}}function a(H,J){try{return d(H,J)}catch{return}}function $8(H){let J=a(H,"Pictures");if(J===void 0)return[];let z=new J8,G=0,Q=0;for(;;){let K=D(J,G);if(K===void 0)break;let[V,W,Z]=K;if(G+=8+Z.length,Q+=1,Q>1e5)break;let $=t,j=W===61447?Q8(Z,$):K8(V,W,Z,$);if(j!==void 0)z.add(j.mediaType,`pictures/${Q}.${j.extension}`,j.bytes);else o(`skipping unsupported Pictures record 0x${W.toString(16).padStart(4,"0")}`)}return z.assets}function*A(H){let J=0;for(;;){let z=D(H,J);if(z===void 0)return;J+=8+z[2].length,yield z}}function q8(H){let J=H.length&-2;return new TextDecoder("utf-16le").decode(H.subarray(0,J))}function j8(H){return new TextDecoder("latin1").decode(H)}class g{segments=[];current=[];currentIsNotes=!1;listRun=[];pending;masters=[];activeMaster=0;shapeCounter=0;encrypted=!1;recovering=!1;records=0;parseSlides(H,J){let z=F8(H,J);if(z===void 0)return!1;if(this.masters=Y8(z.masterList,z.persist,H),this.walkSlideList(z.slideList,z.persist,H,!1,1006),z.notesList!==void 0)this.walkSlideList(z.notesList,z.persist,H,!0,1008);return!0}walkSlideList(H,J,z,G,Q){let K;for(let[W,Z,$]of A(H))if(Z===1011){let j=this.finishSlide(K,J,z,Q,G);this.endSegment(j),this.currentIsNotes=G;let _=X($,0);if(K=_===void 0?void 0:[_,X($,12)??0],!G)this.selectMaster(K!==void 0?K[0]:void 0,J,z)}else this.record(W,Z,$);let V=this.finishSlide(K,J,z,Q,G);this.endSegment(V)}finishSlide(H,J,z,G,Q){if(H===void 0)return;let[K,V]=H,W=Q?void 0:V!==0?V:void 0,Z=J.get(K);if(Z!==void 0){let $=D(z,Z);if($!==void 0&&$[1]===G){let j=$[2];if(Q){for(let[,_,M]of A(j))if(_===1009){let F=X(M,0);W=F!==void 0&&F!==0?F:void 0;break}}this.walk(j)}}return W}endSegment(H){if(this.flushShape(),L(this.current,this.listRun),this.current.length>0)this.segments.push({blocks:this.current,id:H,isNotes:this.currentIsNotes}),this.current=[]}intoBlocks(){this.endSegment(void 0);let H=[],J=[];for(let Q of this.segments)if(Q.isNotes)J.push({id:Q.id,blocks:Q.blocks});else H.push({id:Q.id,blocks:Q.blocks});let z=J.map(()=>!1),G=[];for(let Q of H){G.push(...Q.blocks);for(let K=0;K<J.length;K+=1)if(!z[K]&&Q.id!==void 0&&J[K].id===Q.id)z[K]=!0,G.push({type:"blockQuote",blocks:J[K].blocks}),J[K].blocks=[]}for(let Q=0;Q<J.length;Q+=1)if(!z[Q]&&J[Q].blocks.length>0)G.push({type:"blockQuote",blocks:J[Q].blocks});return G}selectMaster(H,J,z){let G=0;if(H!==void 0){let Q=J.get(H);if(Q!==void 0){let K=D(z,Q);if(K!==void 0&&K[1]===1006){for(let[,V,W]of A(K[2]))if(V===1007){let Z=X(W,12);if(Z!==void 0){let $=this.masters.findIndex(([j])=>j===Z);if($>=0)G=$}break}}}}this.activeMaster=G}walk(H){let J=[{buf:H,pos:0}];while(J.length>0){let z=J[J.length-1],G=D(z.buf,z.pos);if(G===void 0){J.pop();continue}let[Q,K,V]=G;if(z.pos+=8+V.length,this.chargeRecord(),(Q&15)!==15){this.atom(K,V);continue}if(K===12052){this.encrypted=!0;continue}if(K===1008&&this.recovering){let W=!1;for(let[,Z,$]of A(V))if(Z===1009){let j=X($,0);W=j!==void 0&&(j&2147483648)!==0;break}if(!W)this.endSegment(void 0),this.currentIsNotes=!0,this.walk(V),this.endSegment(void 0),this.currentIsNotes=!1;continue}if(K===1008||K===1016||K===4041)continue;if(K===4080&&Q>>>4!==0)continue;if(J.length>=n)throw P.resourceLimit("max_record_depth",`record nesting exceeds ${n}`);J.push({buf:V,pos:0})}}record(H,J,z){if(this.chargeRecord(),(H&15)===15)if(J===12052)this.encrypted=!0;else if(J===1008||J===1016||J===4041);else if(J===4080&&H>>>4!==0);else this.walk(z);else this.atom(J,z)}chargeRecord(){if(this.records+=1,this.records>p)throw P.resourceLimit("max_records",`record stream exceeds ${p} records`)}atom(H,J){switch(H){case 3999:{this.flushShape(),this.pending={txType:J[0]??1,text:"",styles:void 0};break}case 4000:{this.pushText(q8(J));break}case 4008:{this.pushText(j8(J));break}case 4001:{if(this.pending!==void 0)this.pending.styles=b(J,this.pending.text.length);break}case 4051:{o("ppt hyperlink records present; targets are not resolved");break}default:break}}pushText(H){if(this.pending!==void 0)this.pending.text+=H;else this.pending={txType:1,text:H,styles:void 0}}flushShape(){let H=this.pending;if(this.pending=void 0,H===void 0||H.text.length===0)return;this.shapeCounter+=1;let J=this.shapeCounter,z=H.txType===0||H.txType===6,G=H.styles??x(),Q=this.masters[this.activeMaster]?.[1].get(H.txType)??[],K=(q)=>Q[q]??u(),V=0,W=G.chars[0],Z=W?.count??Number.MAX_SAFE_INTEGER,$=0,j=G.paragraphs[0],_=j?.count??Number.MAX_SAFE_INTEGER,M=[],F=[],Y="",w={...G8},S=(q)=>q!==void 0?[q.depth,q.bullet]:[0,void 0];for(let q of H.text){let C=K(S(j)[0]),O={bold:W?.bold??C.bold??!1,italic:W?.italic??C.italic??!1,strike:!1,code:!1};if(q==="\r"){if(Y.length>0){let E=v(Y);if(Y="",E.length>0)F.push({type:"text",text:E,style:{...w}})}let[B,I]=S(j);M.push([F,B,I]),F=[]}else if(q==="\v"){if(Y.length>0){let B=v(Y);if(Y="",B.length>0)F.push({type:"text",text:B,style:{...w}})}F.push({type:"lineBreak"})}else{if(!H8(O,w)&&Y.length>0){let B=v(Y);if(Y="",B.length>0)F.push({type:"text",text:B,style:{...w}})}w=O,Y+=q}let N=q.length;if(Z=Math.max(0,Z-N),Z===0)V+=1,W=G.chars[V],Z=W?.count??Number.MAX_SAFE_INTEGER;if(_=Math.max(0,_-N),_===0)$+=1,j=G.paragraphs[$],_=j?.count??Number.MAX_SAFE_INTEGER}if(Y.length>0){let q=v(Y);if(q.length>0)F.push({type:"text",text:q,style:{...w}})}if(F.length>0){let[q,C]=S(j);M.push([F,q,C])}for(let[q,C,O]of M){if(e(q)){L(this.current,this.listRun);continue}if(z){L(this.current,this.listRun);let B=K(C),I={bold:B.bold,italic:B.italic,strike:void 0,code:void 0};V8(q,W8(I));let E=z8(q);this.current.push({type:"heading",level:2,anchor:E,content:q});continue}if(O??K(C).bullet??!1)this.listRun.push({level:C,key:{instance:J,marker:"bullet"},number:0,label:void 0,blocks:[{type:"paragraph",inlines:q}]});else L(this.current,this.listRun),this.current.push({type:"paragraph",inlines:q})}}}function F8(H,J){let z=new Map,G,Q=X(J,16);if(Q===void 0)return;let K=Q;for(let M=0;M<100;M+=1){if(K===0)break;let F=D(H,K);if(F===void 0)return;let Y=F[1],w=F[2];if(Y!==4085)return;if(G===void 0)G=X(w,16);let S=X(w,12);if(S===void 0)return;let q=D(H,S);if(q!==void 0&&q[1]===6002){let O=q[2],N=0;while(N+4<=O.length){let B=X(O,N);if(B===void 0)return;let I=B&1048575,E=B>>>20;N+=4;for(let k=0;k<E;k+=1){let T=X(O,N);if(T===void 0)return;let m=I+k>>>0;if(!z.has(m))z.set(m,T);N+=4}}}let C=X(w,8);if(C===void 0)return;if(C===K)break;K=C}if(G===void 0)return;let V=z.get(G);if(V===void 0)return;let W=D(H,V);if(W===void 0||W[1]!==1000)return;let Z=W[2],$,j,_;for(let[M,F,Y]of A(Z)){if(F!==4080)continue;let w=M>>>4;if(w===0&&$===void 0)$=Y;else if(w===2&&j===void 0)j=Y;else if(w===1&&_===void 0)_=Y}if($===void 0)return;return{persist:z,slideList:$,notesList:j,masterList:_}}function i(H){let J=new Map;for(let[z,G,Q]of A(H))if(G===4003){let K=z>>>4;if(!J.has(K))J.set(K,c(Q,K))}return J}function Y8(H,J,z){let G=[];if(H!==void 0)for(let[,Q,K]of A(H)){if(Q!==1011)continue;let V=X(K,0),W=X(K,12);if(V===void 0||W===void 0)continue;let Z=J.get(V);if(Z===void 0)continue;let $=D(z,Z);if($!==void 0&&$[1]===1016)G.push([W,i($[2])])}if(G.length===0){let Q=[...J.values()].sort((K,V)=>K-V);for(let K of Q){let V=D(z,K);if(V!==void 0&&V[1]===1016)G.push([0,i(V[2])])}}return G}var _8=new Set(["ppt","pps","pot"]);function h(){return{id:"ppt",sniff(H,J){if(w8(H)==="ppt")return 2;if(J?.path!==void 0&&_8.has(X8(J.path)??""))return 1;return 0},convert(H){return{markdown:B8(r(H))}}}}import{create as C8}from"@mdgate/core";var M8=C8([h()]);function D8(H,J){return M8(H,J)}export{D8 as toMarkdown,h as ppt};

package/dist/ppt.d.ts

@@ -0,0 +1,2 @@
+import type { Converter } from '@mdgate/core';
+export declare function ppt(): Converter;

package/dist/to-markdown.d.ts

@@ -0,0 +1,6 @@
+import { type ConvertHint } from '@mdgate/core';
+/**
+ * Convert a single file's bytes to Markdown with the ppt converter.
+ * `hint.path` is only a sniff hint; it is never read from disk.
+ */
+export declare function toMarkdown(bytes: Uint8Array, hint?: ConvertHint): Promise<string>;

package/package.json

@@ -0,0 +1,43 @@
+{
+  "name": "@mdgate/ppt",
+  "version": "0.1.0",
+  "description": "mdgate legacy PowerPoint converter (binary .ppt)",
+  "type": "module",
+  "license": "MIT",
+  "sideEffects": false,
+  "exports": {
+    ".": {
+      "types": "./dist/index.d.ts",
+      "import": "./dist/index.js"
+    }
+  },
+  "main": "./dist/index.js",
+  "types": "./dist/index.d.ts",
+  "files": [
+    "dist/**/*.js",
+    "dist/**/*.d.ts"
+  ],
+  "scripts": {
+    "build": "bun ../../scripts/build-package.ts",
+    "prepublishOnly": "bun run build"
+  },
+  "dependencies": {
+    "@mdgate/containers": "0.1.0",
+    "@mdgate/core": "0.1.3",
+    "@mdgate/document": "0.1.0",
+    "@mdgate/office-common": "0.1.0",
+    "@mdgate/utils": "0.1.2"
+  },
+  "publishConfig": {
+    "access": "public"
+  },
+  "engines": {
+    "node": ">=20"
+  },
+  "keywords": [
+    "mdgate",
+    "markdown",
+    "ppt",
+    "powerpoint"
+  ]
+}