gsheetplus.table — io.github.mlimotte/gsheetplus 0.9.2

High-level table reading and data parsing for Google Sheets. Builds on gsheetplus.core (raw cell reads) and gsheetplus.cell (type conversion).

->kebab-keyword clj

(->kebab-keyword s)

Convert a column header string to a kebab-case keyword. Drops parenthetical annotations: 'First Name (optional)' → :first-name

blocks clj

(blocks raw-data block-titles)

Split raw-data into contiguous blocks identified by a marker string in column A. block-titles is a seq of expected marker strings in order. Rows before the first block are discarded. The title row is the first row of each returned block.

drop-after-stop-row clj

(drop-after-stop-row raw)

Truncate rows at the first row where column A = "/STOP".

fill-down clj

(fill-down keys-to-fill required? records)

For each key in keys-to-fill, propagate non-blank values from the previous record to any record with a blank value for that key. required? — when true, throws if a fill key has no value to propagate from.

headers-from-row clj

(headers-from-row headers-raw
                  &
                  [{:keys [column-name-fn keep-leading-blanks]
                    :or {column-name-fn ->kebab-keyword}}])

Parse a raw header row (vector of string-or-nil) into a vector of keywords. Options: :column-name-fn - fn to convert header strings (default ->kebab-keyword) :keep-leading-blanks - when true, preserve leading nil headers for alignment

info-params clj

(info-params raw-data)

Extract leading Name-Value pairs from a vec-vec sheet (as from raw-data). Reads rows until the first blank in column A. Returns [info-map remaining-rows start-row-idx].

prune-start-rows clj

(prune-start-rows raw)

Drop rows up to and including the first row where column A = "/START".

raw-data clj

(raw-data service spreadsheet-id sheet-title-or-range)

Read the sheet as vec-of-vecs, truncating at /STOP if present.

read-as-vec-vec clj

(read-as-vec-vec service spreadsheet-id sheet-title-or-range)

Read sheet-title-or-range from spreadsheet-id as a vector of vectors of Clojure values (cell->clj applied to each cell).

read-single-table clj

(read-single-table service gsheet-url options)

(read-single-table service spreadsheet-id sheet options)

Read a Google sheet as a sequence of records (maps keyed by header keywords). sheet can be a sheet title string, a numeric sheet-id, or a full gsheet URL.

Options: :column-name-fn - header string → keyword (default ->kebab-keyword) :prune-start? - drop rows before /START marker :drop-rows - drop the first N rows :filter-fn - keep only records matching this predicate :row-idx? - add ::core/row-idx (1-based) to each record :stop-on-blank-row? - stop at the first all-blank record

Returns sequence with :headers metadata. Throws on duplicate headers.

read-single-table-with-info clj

(read-single-table-with-info service
                             spreadsheet-id
                             sheet-name
                             &
                             [{:keys [prune-start? filter-fn row-idx?
                                      column-name-fn]
                               :or {column-name-fn ->kebab-keyword}
                               :as options}])

Read a sheet that has leading Name-Value info params followed by a table. Returns [info-map records].

Options: :prune-start?, :filter-fn, :row-idx?, :column-name-fn

row->record clj

(row->record headers
             row
             &
             [{:keys [remove-blank-keys] :or {remove-blank-keys true}}])

Map a header vector to a row vector. By default removes nil-key and blank-value entries. Options: :remove-blank-keys - when false, keeps all entries including blanks (default true)

split-sections clj

(split-sections raw headers {:keys [section-fn subsection-fn record-filter-fn]})

Group a vec-vec into sections (and optionally subsections). Options: :section-fn - (record, raw-row) → section name string or nil :subsection-fn - (record, raw-row) → subsection name string or nil (optional) :record-filter-fn - (record) → truthy to keep record (optional)

split-tables clj

(split-tables raw)

Parse a vec-vec where non-blank column-A values are section names. The row immediately after each section name is the header row. Returns a map of kebab-keyword section names → sequences of records.