| import lmql | |
| from dataclasses import dataclass | |
| from typing import List | |
| # Data Structures | |
| @dataclass | |
| class Task: | |
| task: str | |
| # result: str |
| // OpenAI function call guide: https://platform.openai.com/docs/guides/gpt/function-calling | |
| // JSON Schema reference: https://json-schema.org/understanding-json-schema/ | |
| monaco.languages.json.jsonDefaults.setDiagnosticsOptions({ | |
| validate: true, | |
| schemas: [ | |
| { | |
| uri: "http://myserver/foo-schema.json", | |
| fileMatch: ["*"], | |
| schema: { |
(Edit: I've started collating all of my ChatGPT deep dives and explorations at the following new gist: https://gist.github.com/0xdevalias/4ac297ee3f794c17d0997b4673a2f160)
Exploring ChatGPT API's for exporting all history as markdown, etc.
List the history of past conversations:
| { | |
| "$schema": "http://json-schema.org/draft-07/schema#", | |
| "$id": "http://example.com/my-custom-draft-07/schema#", | |
| "title": "Core schema meta-schema", | |
| "allOf": [{ "$ref": "#/definitions/schema" }], | |
| "$comment": "Add constraints here that should only apply to root schemas and not sub-schemas", | |
| "definitions": { | |
| "schema": { |
| #!/bin/bash | |
| # This script takes a PDF or list of PDFs and outputs a file(s) | |
| # named <file>_scanned.pdf that looks like it has been scanned | |
| # | |
| # Requires imagemagic and popper to be installed (brew install imagemagick poppler) | |
| # | |
| # Accepts: a list of files | |
| # Usage: ./<scriptfilename> pdf1.pdf pdf2.pdf | |
| # | |
| # To use as a macOS automator quick action you need to: |
| ObjC.import('AppKit'); | |
| function pboardTypes() { | |
| return ObjC.deepUnwrap( | |
| $.NSPasteboard.generalPasteboard.pasteboardItems.js[0].types | |
| ); | |
| } | |
| function pboardUnpacked(strType) { |
Unless you are using Safari on OSX, most browsers will have some kind of free plugin that you can use to export the browser's history. So that's probably the easiest way. The harder way, which seems to be what Safari wants is a bit more hacky but it will also work for other browsers. Turns out that most of them, including Safari, have their history saved in some kind of sqlite database file somewhere in your home directory.
The OSX Finder cheats a little bit and doesn't show us all the files that actually exist on our drive. It tries to protect us from ourselves by hiding some system and application-specific files. You can work around this by either using the terminal (my preferred method) or by using the Cmd+Shft+G in Finder.
Once you locate the file containing the browser's history, copy it to make a backup just in case we screw up.
