Documentation
How does it work?
Write your notes in Markdown and store them in .md files. Recall will scan your workspace folders, identify possible “flashcards” and test you using spaced repetition.
Notes:
To see your note rendered as HTML, simply use the Markdown Preview functionality of VS Code.
For more information on Markdown, please visit Markdown Guide.
How are flashcards identified
Recall scans all the .md files in you workspace and identifies flashcards using the following algorithm
at header level
any header marks a start of a new flashcard
any blank line marks a start of a new page (of that flashcard)
at bullet point level (when enabled)
any first level bullet point with of the configured type is a start of a new card
the bullet point is split using the line divider string (e.g. ::), configurable
the newline at the end of the bullet point also serves as a page divider
if a flashcard has at least two pages, it will be put in the testing queue
To better understand the above, look at the practical examples in the demo folder.
For example, the above screenshot comes from this paragraph
Note: If you want to prevent a page break, put a single space on the line. Make sure you have the files.trimTrailingWhitespace set to false, otherwise it will get removed when saving the file.
Testing yourself
To really train your memory, you should do a daily review of your cards. To run it, simply run the “Recall: Flashcards Test” command in VSCode (easiest way is to press Ctrl-P to see list of commands and start typing the name).
A webview will appear show you the first page of the first card to review. You then have the following options/hotkeys:
Space- expands the card and shows one more page (repeat until the card is fully expanded)Enter- mark the card as rememberedS- mark the card as hard (Struggled)F- mark the card as forgottenN- go to the next card without logging any resultA- archive card - card won't show for review anymore, this needs to be confirmed byEnter
Recall level
Every card has a so called “recall level” which indicates how well you remember it. It also represents the number of days after which it will be queued for review again.
All cards start with a default recall level of 1 which means the card will be queued for review the next day. If you mark it as remembered, the recall level will be doubled. the recall level will be doubled. If you mark it as forgotten, the recall level will go back to 1.
More details on the algorithm can be find here
Reverse testing
Sometimes, you would like to test yourself in both directions. Best example of this is learning a vocabulary. Recall has this capability since version 0.8.0. Here’s how it works:
The “+” sign indicates that you want bi-directional testing in the current file. Now, once you archive the “el norte” card (signalling to Recall that you know it well enough in one direction) it will start showing in reverse, i.e. the front page will be “north” and the back page will be “el norte”. The card’s badge will also indicate that.
Configuration
You can find the configuration options in Settings (Extensions -> Recall). Pay extra attention to these configuration options
Default Level - defines at which level to detect flashcards
ignore - no detection will be done
header - any header starts a new card
bullet - any “-“ bullet starts a new card (as well as a header)
asterisk - any “*” bullet starts a new card (as well as a header)
Line Divider - delimiter used in single line cards (use any string that works for you - colon, semicolon, slash, etc.), can even be multi-character sequence
New Card Limit - defines the maximum number of new cards allowed in one review session
defaults to 0, i.e. unlimited
you can use it to better pace the learning, e.g. when you add a large number of new cards at the same time
Hide badges - as card are detected in the text, the editor adds a small badge with the card’s state (NEW, GOOD, HARD, etc.) depending on the review results. This setting gives you the option to hide certain states if you wish, or even all of them.
If you need to override the default level of card recognition for a particular file, you can use the YAML header.
For example, to tell Recall that individual bullets should be recognized in the file, simply add the following sequence at the beginning of your .md file.
Styling
VSCode allows extensions to provide own CSS styling to the Markdown preview. It has several built in plugins the provide additional styling. If you are experiencing differences between how your notes are rendered in the Markdown preview vs. how they are rendered as cards, you can experiment with applying additional styles from other extensions. To do that, simply run the Recall: Enable CSS styles from other markdown preview extensions command and select which additional styles should be applied.
For example, if you are using inline math formulas, you will probably need to enable the vscode.markdown-math/notebook-out/katex.min.css style.
Statistics
This was added in v0.13.1. Running the Recall: Show card statistics will open a new JSON document showing the number of cards in each file, their statuses and whether or not they are due for review. This can help investigate issues with cards not showing up for review and is primarily intended for troubleshooting.
Import from Mochi
To import export files from Mochi, you can run the Recall: Convert Mochi archive command, select a .mochi archive and for every deck in it a separate new Markdown file will be created. You can then save it anywhere you want.
If your Mochi deck contains media files (e.g. pictures), they will not be extracted. To extract them, simply unzip the .mochi file into a media subfolder and correct the links in the Markdown files accordingly. You can find all the links easily by searching for “@media”.
In certain cases, the import fails. This is due to a bug in the EDN parser library and would be difficult to correct. As a workaround, it is possible to unzip the .mochi file, convert the data.edn file to JSON using this tool, open the JSON file in VSCode and run the “Recall: Convert Mochi JSON” command.
Last updated
