Started a project to put together the more intricate correlations that make the Vaadin Flow framework tick.

Vaadin Flow Cheat sheets and flow diagrams

First one is the navigation life cycle:

Since Maven Central changed the search I had to fix my helper script. Time to try out Babashka.

#!/usr/bin/env bb
(defn artifact [{:keys [g a v latestVersion]}]
  (str g ":" a ":" (or v latestVersion)))

(defn query [q]
  (let [query-params {:start 0 :rows 50}]
    (if (str/includes? q ":")
      (let [[g a] (str/split q #":" 2)]
        (assoc query-params
               :q (str "g:" g " AND " "a:" a)
               :core "gav"))
      (assoc query-params :q q))))

(defn request [q]
  (-> (org.httpkit.client/get
        "https://search.maven.org/solrsearch/select"
        {:content-type "application/json"
         :query-params (query q)})
      deref
      :body
      (cheshire.core/parse-string true)
      :response
      :docs))

(assert (= 1 (count *command-line-args*)))

(let [[q] *command-line-args*]
  (->> (request q)
       (map artifact)
       (run! prn)))

Same as with Groovy beforehand:

(defmacro map-of
  [& xs]
  `(hash-map ~@(mapcat (juxt keyword identity) xs)))

(let [a 1
      b 2]
  (println (map-of a b)))
; {:b 2, :a 1}

From time to time I have the need to get text from an image, when someone shares a screen-shot, but not the URL in it or when I want to copy code from a stopped youtube video. So ideally I want to select a region of my screen and turn that into text.

First install gocr and flameshot. Then create a script, that runs gocr and opens the resulting text (in my case I want to open a terminal with vim). E.g. ~/bin/ocr:

#!/bin/sh
rxvt -e sh -c "gocr -i \"$1\" | vim -"

Then create a "desktop" entry for flameshot to pick up. ~/.local/share/applications/ocr.desktop:

[Desktop Entry]
Version=1.0
Type=Application
Terminal=false
Exec=.../bin/ocr %f
Name=OCR
Categories=Graphics;

(Note, that the path there is absolute, or you have to figure out, what works for you... I couldn't be bothered to look into this)

Then start flameshot gui (I have that on my Print Screen-key), select the region to OCR, press CTRL-o (or click the button to open the screenshot with an application), then pick the OCR-Application from the desktop entry, which then opens the text.