heatherleaf / crossword-maker Goto Github PK

All clues that have been added are removed when a new word is added. This makes planning the crossword very difficult.

This is an extension of #13, where the ranking/probability of a word suggestion depends on the current words in the crossword. This could be done using e.g. word vector similarity.

The matching words are now shown in alphabetical order, and if there are too many matches there's a random selection. One could instead sort the words according to frequency, or some other ranking (e.g. similarity to a given theme). And let the ranking decide the random distribution.

This is probably difficult, but we could perhaps learn a model that suggests words (and selections) according to how easy it will be to complete the construction of the crossword. Perhaps this can be done by autogenerating millions of crosswords and learning which words made it easier to complete construction.

Possible export options:

• JSON
• PDF
• image

Note: this is probably very difficult and will not happen in a long time

We'd like to have a multilingual interface, perhaps using this framework: https://www.i18next.com

Add (command-line) code for extracting a lexicon from a corpus, or from statistics files (from https://spraakbanken.gu.se/en/resources)

In Scandinavian style crosswords the clues are written directly in the crossword. To support this we need to:

• be able to have two clues in the same cell (one horizontal and one vertical) (#8)
• automatically(?) decide which cell should contain the clue, e.g.:
  • vertical words: the cell immediately above
  • horizontal words: the cell immediately to the left (unless there is no cell there, in which case the one above)
  • if there is no cell above/to left, we can perhaps use the one to the left/above - but then we might end up in 3-4 clues in the same cell
• if a clue is not in the "normal" place we need arrows from the clue to the word (#7)

Currently we only handle mouse events, so you cannot use a touch-based device. We want to add support for touch-events: https://developer.mozilla.org/en-US/docs/Web/API/Touch_events

In most non-Scandinavian countries, a reference to a clue is given as a small number in the corner of a cell, and then the actual clues are listed below the crossword.

Sometimes the constructor wants to add a word that's not in the lexicon. We can do this e.g. by letting the user write the new word in the search filter. When the filter contains the exact amount of letters, an "override" button appears.

If we have a (horizontal) selection that ends in a (vertical) word, then it's also possible to let the selection continue downwards until the end of the (vertical) word. This can be a good alternative if there is no good word matching a selection.

There are two kinds of turns: ⤵︎ and ⤷, and the corresponding arrows have to be shown in the crossing cell.

We have to also take care if the user later extends the original selection horizontally, because then the turn should be removed.

If a clue is written in a cell, we want an arrow to point to which word it is referring to

A clue cell often contains two clues, usually split horizontally

If we make the lexicons smaller and thematic, we could let the user select several lexicons at the same time, instead of only one.

Change the layout to be like an American/British crossword:

• No clues within the crosswords
• Solid black cells where there are no letters
• Clues are numbered (across/down)
• List of clues beneath the crossword
• Add side-words as clues too (i.e., words that are automatically added by the engine)

When there are too many matching words, it currently shows a random selection of them. If you want to update the random selection, you can write a letter in the filter and then delete it again. Instead we can have a dedicated "update" button that shows a new random list.

Make it possible to add blocks between two cells. There are two alternatives:

• let the user double-click on a wall
• add the wall automatically if the user ends the selection just before an occupied cell

If we extend a word (e.g., HORSE --> HORSECART), then we should remove the old word.

Similar: if we have a turned word and add a straight word over it, the turned word should be removed.