Import and export

Import and export are in Settings → Data & Migration, under Bookmarks data. Both work on the Active Source. An import goes into the source you are currently using, and an export saves that source.

Import a file

1. Click Import… and choose a file. The extension accepts .html, .htm and .csv files.
2. Read the summary, for example Found 120 bookmarks in 8 folders. For CSV files, it also tells you how many rows were skipped because they had no valid URL.
3. Choose the folder the bookmarks should go into. If you have chosen a root folder, it is selected by default.
4. Optionally, type a name in Optional: import into a new subfolder to keep the import together in one folder.
5. Click Import here. When the import finishes, you see a result such as Imported 118 bookmarks and 8 folders. 2 duplicates skipped.

If an item cannot be saved, the rest of the import still goes ahead, and the result tells you how many items failed and why.

From a browser: HTML

Most browsers can export their bookmarks as an HTML file, in a format they all share. The extension reads that file and brings in:

• folders and the bookmarks inside them, with their nesting
• bookmark titles and URLs
• the date each item was added, when the file includes it

The import leaves out separators and entries that have no URL. A bookmark with no title is named after its site, for example example.com.

From Raindrop or Pocket: CSV

The extension works out which service a CSV file came from by looking at the names in its first (header) row:

The header row has…	Read as	Where the bookmarks go
a folder column	Raindrop	Raindrop’s folder paths, such as Parent/Child, become nested folders.
a time_added or status column	Pocket	Everything goes into one folder named Pocket.
only a url column	Other	Everything goes into one folder named after the file.

Only the url, title and (for Raindrop) folder columns are used. Rows whose URL does not start with http:// or https:// are skipped. A file with no url column contains nothing to import.

Folders with the same name

When an imported folder has the same name as a folder already in that spot (ignoring upper and lower case), the two are combined: the incoming bookmarks go into the existing folder. This works the same way at every level.

Duplicates

A duplicate is an incoming bookmark whose URL is already saved in the folder it would go into. When comparing URLs, the extension ignores differences in upper and lower case in the https:// part and the site name, and treats a default port as no port. Everything after the site name must match exactly, so #/inbox and #/sent are two different bookmarks. Two copies of the same URL inside the imported file are not treated as duplicates.

If there are duplicates, a dialog opens before anything is written. You can decide for all of them at once:

• Skip all: keep the bookmark that is already there and leave out the incoming one.
• Replace all: keep one bookmark, and give it the incoming title.
• Keep both for all: add the incoming bookmark next to the existing one.

Or click Review one by one to decide for each bookmark with Skip, Replace or Keep both. Tick Apply my choice to the remaining conflicts to use your answer for the rest.

Cancel import leaves your bookmarks exactly as they were, because nothing is saved until you choose.

Export

Click Export… and choose what to export:

• Dashboard folder: only the root folder the dashboard shows. It is saved as bookmarks-<folder-name>.html. This option is only available once you have chosen a root folder.
• Everything: the whole Active Source, saved as bookmarks.html.

The file uses the standard HTML bookmarks format, so any browser can import it. It contains folders, bookmark titles and URLs, and the dates items were added.

An HTML export is not a backup of a Vault

When the Active Source is a Vault, the export only includes what the HTML format can hold. It leaves out notes, *.assets/ folders, and the files that record the order you arranged things in. To keep a complete copy of a Vault, see Back up your vault.