From f04d2c293ee8ee855fd28e7fa51e5789d8b43e5c Mon Sep 17 00:00:00 2001
From: rich <rich>
Date: Mon, 6 Dec 2004 17:40:50 +0000
Subject: [PATCH] Added the Csv.trim function. Cleaned up some of the
 documentation.

---
 csv.ml  | 57 ++++++++++++++++++++++++++++++++++++++++++++++++++-------
 csv.mli | 35 ++++++++++++++++++++++++++++++++---
 2 files changed, 82 insertions(+), 10 deletions(-)

diff --git a/csv.ml b/csv.ml
index b5c392b..93da2c4 100644
--- a/csv.ml
+++ b/csv.ml
@@ -1,6 +1,6 @@
 (* csv.ml - comma separated values parser
  *
- * $Id: csv.ml,v 1.1 2003-12-17 16:05:08 rich Exp $
+ * $Id: csv.ml,v 1.2 2004-12-06 17:40:50 rich Exp $
  *)
 
 (* The format of CSV files:
@@ -15,9 +15,12 @@
  * following is a quote: "", and that's all" is the CSV equivalent of
  * the following literal field: The following is a quote: ", and that's
  * all
+ *
+ * "0 is the quoted form of ASCII NUL.
  * 
  * CSV fields can also contain literal carriage return characters, if
- * they are quoted, eg: "This field is split over lines" represents a
+ * they are quoted, eg: "This field
+ * is split over lines" represents a
  * single field containing a \n.
  * 
  * Excel will only use the quoting format if a field contains a double
@@ -35,17 +38,17 @@
  * and often will, have different lengths). We then provide simple
  * functions to read the CSV file line-by-line, copy it out, or copy a
  * subset of it into a matrix.
- * 
- * For future work: According to the Text::CSV_XS manual page, "0 is a
- * valid encoding, within quoted fields, of the ASCII NUL character. In
- * Unix this character could, of course, be encoded directly in the
- * file.
  *)
 
 type t = string list list
 
 exception Bad_CSV_file of string
 
+let rec dropwhile f = function
+  | [] -> []
+  | x :: xs when f x -> dropwhile f xs
+  | xs -> xs
+
 let lines = List.length
 
 let columns csv =
@@ -161,6 +164,46 @@ let load filename =
   close_in chan;
   csv 
 
+let trim ?(top=true) ?(left=true) ?(right=true) ?(bottom=true) csv =
+  let rec empty_row = function
+    | [] -> true
+    | x :: xs when x <> "" -> false
+    | x :: xs -> empty_row xs
+  in
+  let csv = if top then dropwhile empty_row csv else csv in
+  let csv =
+    if right then
+      List.map (fun row ->
+		  let row = List.rev row in
+		  let row = dropwhile ((=) "") row in
+		  let row = List.rev row in
+		  row) csv
+    else csv in
+  let csv =
+    if bottom then (
+      let csv = List.rev csv in
+      let csv = dropwhile empty_row csv in
+      let csv = List.rev csv in
+      csv
+    ) else csv in
+
+  let empty_left_cell =
+    function [] -> true | x :: xs when x = "" -> true | _ -> false in
+  let empty_left_col =
+    List.fold_left (fun a row -> a && empty_left_cell row) true in
+  let remove_left_col =
+    List.map (function [] -> [] | x :: xs -> xs) in
+  let rec loop csv =
+    if empty_left_col csv then (
+      let csv = remove_left_col csv in
+      loop csv
+    ) else csv
+  in
+
+  let csv = if left then loop csv else csv in
+
+  csv
+
 (* Quote a single CSV field. *)
 let quote_field field =
   if String.contains field ',' ||
diff --git a/csv.mli b/csv.mli
index fd33ac3..dacb3dc 100644
--- a/csv.mli
+++ b/csv.mli
@@ -1,6 +1,6 @@
 (** csv.mli - comma separated values parser
   *
-  * $Id: csv.mli,v 1.1 2003-12-17 16:05:08 rich Exp $
+  * $Id: csv.mli,v 1.2 2004-12-06 17:40:50 rich Exp $
   *)
 
 type t = string list list
@@ -14,8 +14,9 @@ val lines : t -> int
 
 val columns : t -> int
 (** Work out the (maximum) number of columns in a CSV file. Note that each
-   line may be a different length, so this finds the one with the most
-   columns. *)
+  * line may be a different length, so this finds the one with the most
+  * columns.
+  *)
 
 val load_in : in_channel -> t
 (** Load a CSV file.
@@ -32,10 +33,38 @@ val load_rows : (string list -> unit) -> in_channel -> unit
   * this function is appropriate. It parses the input one row at a time and
   * calls your function once for each row.
   *
+  * Note that if you CSV file contains cells which have embedded
+  * line feeds, then it is non-trivial to parse these lines and
+  * pass them correctly to [load_rows].
+  *
   * @param f Callout function.
   * @param chan Input file stream.
   *)
 
+val trim : ?top:bool -> ?left:bool -> ?right:bool -> ?bottom:bool -> t -> t
+(** This takes a CSV file and trims empty cells.
+  *
+  * All four of the option arguments ([~top], [~left], [~right], [~bottom])
+  * default to [true].
+  *
+  * The exact behaviour is:
+  *
+  * [~right]: If true, remove any empty cells at the right hand end of
+  * any row.  The number of columns in the resulting CSV structure will
+  * not necessarily be the same for each row.
+  *
+  * [~top]: If true, remove any empty rows (no cells, or containing just empty
+  * cells) from the top of the CSV structure.
+  *
+  * [~bottom]: If true, remove any empty rows from the bottom of the
+  * CSV structure.
+  *
+  * [~left]: If true, remove any empty columns from the left of the
+  * CSV structure.  Note that [~left] and [~right] are quite different:
+  * [~left] considers the whole CSV structure, whereas [~right] considers
+  * each row in isolation.
+  *)
+
 val print : t -> unit
 (** Print string list list - same as [save_out stdout] *)
 
-- 
1.8.3.1