Move your collection in and out of Glasshouse.

Glasshouse reads and writes plain CSV files: the kind that open in Numbers, Sheets, or Excel. You can export everything you have recorded, bring in a spreadsheet you have kept for years, or move over from Orchid Album. This page walks through all of it.

Start on the web for big files. The import runs the same everywhere, but the web app at glasshouse.celerycity.app gives a large preview more room. The phone app's own import screen points here for big spreadsheets. Sign in with the same account you use on your phone and everything you import syncs to every device.

Open Settings and choose “Export collection”. In the app's words: “Everything you've recorded, as two plain CSV files — they open in Numbers, Sheets, or Excel, and Glasshouse can read its own export back in full.”

• Plants: every plant in every state, one row each, covering the garden, wishlist, and graveyard, saved as glasshouse-plants-YYYY-MM-DD.csv.
• Care history: every logged watering, feeding, repotting, treatment, and observation, saved as glasshouse-care-YYYY-MM-DD.csv.

Both files round-trip. Re-importing the plants file is recognized as a Glasshouse export and read back column for column; re-importing the care file adds only entries that are not already in your log. The files carry references to your photos, not the pictures themselves, so they stay small and your photo library stays where it is.

Open Settings, choose “Import a spreadsheet”, and press “Download the template” (“Every supported column with two example rows, ready to fill in.”). The file arrives as glasshouse-import-template.csv and is, by construction, a file the importer accepts: fill it in, import it, done.

Only name is required. Every other column is optional, and a blank cell simply imports as nothing recorded. The same table lives inside the app under “Supported columns”. As the app notes there: “The alternate spellings apply when importing a spreadsheet that is not a Glasshouse export; a Glasshouse file is read by its exact column headers.”

As the import screen puts it: “Bring in a CSV file — Glasshouse and English-language Orchid Album exports are recognized automatically, and other spreadsheets are matched by their column names.” You never pick a format; the file's header row decides.

By default an import only adds new plants; rows that match something you already have are left untouched. To bring changes from a spreadsheet onto existing plants, turn on “Update matching plants” on the preview screen. Its own description is the whole contract:

“Matches by collection number first, then by name (ignoring case, punctuation, spacing, and × versus x); Orchid Album files match their own rows. Blank cells never overwrite saved details. Off leaves matching rows untouched.”

With the toggle on, the app briefly shows “Checking your care log…” while it confirms which care entries already exist, then previews exactly four outcomes before anything is written:

• “N new plants will be added.”
• “N plants will be updated.” Each one gets a card naming its basis and its changes, like “Matched by collection number · Color, Price”.
• “N rows already match what is saved and will not change.”
• “N rows will be skipped.” Every skipped row is listed with its reason.

Some things an update never does: it never touches a matched plant's photos, never moves a plant between the garden, wishlist, and graveyard, and never rewrites a matched plant's collection number. Blank cells in your file never erase saved details.

If two of your plants share a name, the row is skipped rather than guessed at, with a line like: “Row 5 · Dendrobium kingianum: two plants share this name; add a collection number to the row to say which one.” Give the row that plant's collection number and run the file again.

Export a CSV from Orchid Album using its English-language export, then import it like any other file. Glasshouse recognizes the format automatically and says so: “Found 3 plants in orchid-album-export.csv — recognized as an Orchid Album export.”

What comes across:

• Your plants, with their names, genus, source, acquisition date, purchase price, awards, parentage, fragrance notes, and notes.
• Dated log entries become real care history: “2 dated log entries will come across as care history.”
• Watering schedules stay readable: “1 watering schedule found — their details will stay with the plants that use them.”

One thing cannot come across, and the app says it plainly: “Orchid Album CSV files do not contain photos. The plants and their written history will still be imported.”

Re-importing the same Orchid Album file is safe: its rows match their own earlier imports, so running the file twice never creates copies.

A plain import reports what was held back as counts in its summary. With “Update matching plants” on, the preview and the summary go further and name each skipped row with its reason, in the app's own words. Here is what each reason means and how to fix it; some appear only in specific modes, noted with each.

The preview says there is nothing to change

That is update-on-match confirming everything in the file is already saved: “Nothing to write. Everything in this file already matches your garden.” A re-import that changes nothing reports exactly that, rather than pretending to update.

My file is seen but nothing imports

Check the preview's column mapping: the one column that must be matched is the plant's name. If your name column is called something the table above does not list, rename the header (any of the alternate spellings works) or paste your data into the template.

A column of mine went to the wrong place, or nowhere

The preview shows exactly how every header was matched, and what will not be imported. Headers are matched case-insensitively, and spaces, hyphens, and underscores are treated alike. If a match is wrong, cancel, rename the header to the column name or an alternate spelling from the table, and choose the file again. The template's headers always match.

What happens to my collection numbers?

Numbers in your file are kept. Rows without one get the next free number assigned automatically. Gaps and fractional splits like 20.1 are fine, and an update never rewrites the number of a plant it matched.

Is it safe to run the same import twice?

Yes. A plain re-import holds back rows whose collection number is already taken and flags matching names as possible duplicates, so nothing is created twice by accident. Orchid Album files go further: their rows carry their own identity, so a repeat of the same file is recognized outright. And with “Update matching plants” on, changes land on the same plants rather than creating copies.

Do photos travel in these files?

No. CSVs carry references to photos, never the pictures themselves. Your photos stay in your photo library, and Orchid Album files do not contain photos at all.

Can an import move a plant to the graveyard or wishlist?

An update never moves an existing plant between the garden, wishlist, and graveyard. New rows land in the state their lifecycle column names, so an old export restores wanted and gone plants where they belong.

What about rows marked wanted?

A row whose lifecycle is wanted lands on the wishlist. Wanted plants hold no collection number until you mark them acquired.

Where does the imported data live?

In your account, like everything else you record: import on the web and the plants are on your phone when you open it. The privacy policy describes exactly what is stored.