Catalogue from Excel¶

Many libraries join the Ofelia project with a collection already typed into an Excel spreadsheet (an in-house id, a title, an author, sometimes an ISBN). Excel cataloging offers four tools to make use of that file:

Check a file — BibliOfelia annotates your spreadsheet with what the online databases know about each book, without changing anything in the catalog. Ideal before a migration, to measure the quality of the file and fix it by hand.
Import into the catalog — BibliOfelia turns a list of ISBNs into records and copies, all at once.
Export the catalog — BibliOfelia hands your whole collection back as a spreadsheet, one row per copy.
Update copies — you send that corrected spreadsheet back, and BibliOfelia applies your fixes to books already catalogued, without ever creating new ones.

Librarians only

Excel cataloging lives in the Advanced menu, available to librarians and administrators.

Open Excel cataloging¶

From the Advanced menu, Inventory section, click Excel cataloging.

The page shows four boxes: Check a file, Import into BibliOfelia, Export the catalog and Update copies.

Check a file¶

Use this to review a spreadsheet without touching the catalog.

Your file must be an .xlsx whose first row contains at least these four columns (case and accents are tolerated):

Column	Content
`ID`	your in-house id (kept as is)
`TITLE`	the book title
`AUTHOR`	the author(s)
`ISBN`	the full ISBN (10 or 13 digits)

Incomplete or wrong ISBN

The ISBN search only accepts a valid ISBN (10 or 13 digits). An incomplete or wrong ISBN is flagged ISBN_INVALID and cannot be used to find the book by ISBN — that is precisely the worst case. The TITLE and AUTHOR then save the day, through the title + author search: take care with those two columns.

In the Check a file box, choose your file then click Start verification.

BibliOfelia queries OpenLibrary, Google Books, the BNF and the BNE, first by ISBN, then by title + author. Processing runs in the background: expect about 10 minutes per 300 rows.

When the job is finished, click Download the annotated file. You get your original spreadsheet back, enriched with extra columns:

TITLE_FOUND_BY_ISBN, AUTHOR_FOUND_BY_ISBN, SOURCE_BY_ISBN — what the ISBN made it possible to find;
ISBN_FOUND_BY_TA, TITLE_FOUND_BY_TA, AUTHOR_FOUND_BY_TA — what the title + author search found;
CONFIDENCE — a 0-to-100 score for how reliable the match is.

Read the colours

Cells with a low confidence score appear in orange: those are the rows to review by hand. An ISBN_FOUND_BY_TA that differs from your ISBN often signals a typo in the original file.

The check writes nothing to the catalog: you can run it as many times as you need.

Import into the catalog¶

Use this to actually create the records and copies from a list of ISBNs.

Your .xlsx file must contain at least one ISBN column. Every other column is optional: add only the ones you have, in any order.

Column	Content
`ISBN`	required
`LOCATION`	the location code (otherwise the copy is created without a location)
`CLASSIFICATION`	the name of an existing classification (the old name `CATEGORY` is still read)
`TITLE`	the record title
`AUTHOR`	the author(s), separated by semicolons
`TYPE`	the document type (Book, Comic / manga, Magazine, Newspaper, Audio CD, Other)
`EDITOR`	the publisher
`YEAR`	the publication year
`LANGUAGE`	the language code (fr, en, es…)
`TAGS`	keywords separated by commas
`EXTERNAL_CODE`	another library's code already on the book
`PROVENANCE`	the code or the name of an existing source
`CLASSIFICATION_CODE`	the classification code (the old name `CATEGORY_ABBR` is still read)
`CONDITION`	the copy condition (New, Good, Worn, Damaged)

In the Import into BibliOfelia box, choose your file then click Import into the catalog.

Each ISBN becomes a record and a copy. If an ISBN is already present in the catalog, BibliOfelia does not recreate the record: it simply adds a copy to the existing one.

A filled-in column replaces the record's information

If you add one of the columns above (title, author, publisher…) and the cell is filled in, its value overwrites the matching field of the record — even if the record already exists. An empty cell changes nothing: the information already in place is kept. For author and tags, the file's list replaces the existing one (it is not added to it). A value that is not recognised for TYPE or CONDITION, or a year that is not a number, is ignored and reported in the batch warnings.

The import creates a cataloging batch: once the job is finished, click View the imported batch to open it, or find it again under Cataloging by scan, exactly like a batch scanned with the camera.

Filling in what's missing online

Do you only have ISBNs, with no title or author? Then run an enrichment on the batch to fetch the metadata online (OpenLibrary, Google Books, BnF…). The file's columns stay authoritative: enrichment only fills in what is still empty.

Export the catalog¶

Use this to get your whole collection back as a spreadsheet: to read it through, to keep an offline copy, or to prepare a bulk correction.

In the Export the catalog box, click Export the catalog. The file catalogue-YYYY-MM-DD.xlsx downloads straight away — there is nothing to wait for.

The spreadsheet holds one row per copy, not per title. A book you own in three copies takes three rows: that is expected, because location, condition, provenance and external code belong to the copy, not to the record.

Column	Content
`OFELIA_CODE`	the copy's Ofelia code (the barcode on the label)
`INTERNAL_ID`	the readable code printed next to the barcode (`OFL-…`)
`EXTERNAL_CODE`	another library's code already on the book
`ISBN`, `TITLE`, `AUTHOR`, `EDITOR`, `YEAR`, `LANGUAGE`	the record's details
`CLASSIFICATION`, `CLASSIFICATION_CODE`, `TYPE`, `TAGS`	the filing
`CONDITION`, `PROVENANCE`, `LOCATION`	the copy's details

This is the update file

The export columns are exactly the ones BibliOfelia can read back. Fix whatever you like in Excel, then send the same file through Update copies: nothing else to prepare.

Update copies¶

Use this to correct in bulk books already in the catalog: change locations after moving a shelf, mark a series as “Worn”, assign external codes, fix badly typed titles.

No book is ever created

This tool never creates a record or a copy. If a row points at a copy that does not exist, it is reported and set aside — never turned into a new book. So you can send an export back without any risk of duplicating your library.

Every row must say which copy it is about. The file must therefore hold at least one of these two columns:

Column	Content
`OFELIA_CODE`	the copy's Ofelia code — the `290…` barcode or the readable `OFL-…` code
`EXTERNAL_CODE`	another library's code on the book

If both columns are filled in

The Ofelia code is what identifies the copy, and the row's external code is applied to it. That is how external codes get assigned to many books at once: an OFELIA_CODE column to say which book, an EXTERNAL_CODE column with the code to put on it.

Every other import column is accepted and optional: TITLE, AUTHOR, CLASSIFICATION, CLASSIFICATION_CODE, TYPE, EDITOR, YEAR, LANGUAGE, TAGS, CONDITION, PROVENANCE, LOCATION and ISBN.

An empty cell erases nothing

A filled cell replaces the existing value; an empty cell leaves the value alone. So this tool cannot be used to clear a field — open the book's page for that. That is what lets you send back a whole export after fixing only two columns.

Pick your file, click Update the copies, then follow the job like an import. The detail page shows:

Copies changed — the rows that really changed something;
Rows with no change — the copy was found, but the file already said the same thing as the catalog;
Errors — the rows that were not applied, with a red banner and the details below.

Warning	What it means
`OFELIA_CODE_UNKNOWN`	no copy carries this Ofelia code — row skipped
`EXTERNAL_CODE_UNKNOWN`	no copy carries this external code — row skipped
`NO_KEY`	the row does not say which copy it is about
`EXTERNAL_CODE_DUPLICATE`	this external code is already on another book — not applied, the rest of the row is
`ISBN_CONFLICT`	this ISBN already belongs to another record — not applied, the rest is
`LOCATION_UNKNOWN`, `CATEGORY_UNKNOWN`, `PROVENANCE_UNKNOWN`	the value is not in your lists — ignored, the rest is applied

One record, several copies

Title, author and publisher belong to the record: fixing them on one copy's row fixes them for every copy of that book. Location, condition, provenance and external code only touch the copy on that row.

Track your jobs¶

At the bottom of the Excel cataloging page, the Recent jobs section lists your latest checks and imports. Click Details to track progress, download an annotated file or review the warnings row by row.

Good to know¶

Format and limits

Only .xlsx files are accepted (no .xls, .csv or .ods).
Maximum size: 5 MB, 10,000 rows.
For better ISBN coverage, a Google Books key can be configured by the administrator; without it, a quota may leave a few rows incomplete (column SOURCE_BY_ISBN = RATE_LIMITED). Re-run the next day: the quota resets every day.