From 84c74e5ce24d64b530640e30555765931a7f818b Mon Sep 17 00:00:00 2001
From: John Whitington <john@coherentgraphics.co.uk>
Date: Sun, 3 Apr 2022 11:31:36 +0200
Subject: [PATCH] UTF8 @B bookmarks on split

---
 Changes        | 4 ++++
 cpdfattach.ml  | 1 +
 cpdfcommand.ml | 4 ++--
 cpdfmanual.tex | 4 +++-
 4 files changed, 10 insertions(+), 3 deletions(-)

diff --git a/Changes b/Changes
index ebf1a76..0994d54 100644
--- a/Changes
+++ b/Changes
@@ -1,3 +1,7 @@
+2.6 (next version, to come)
+
+o Allow -utf8 with -split-bookmarks -o @B.pdf to produce UTF8 filenames
+
 2.5.1 (January 2022)
 
 o Fix a regression where standard fonts could not be chosen
diff --git a/cpdfattach.ml b/cpdfattach.ml
index 80e16ec..a2066c7 100644
--- a/cpdfattach.ml
+++ b/cpdfattach.ml
@@ -4,6 +4,7 @@ open Cpdferror
 
 (* Remove characters which might not make good filenames. *)
 let remove_unsafe_characters encoding s =
+  if encoding = Cpdfmetadata.UTF8 then Pdftext.utf8_of_pdfdocstring s else (* For @B bookmarks splitting. *)
   if encoding = Cpdfmetadata.Raw then s else
     let chars =
       lose
diff --git a/cpdfcommand.ml b/cpdfcommand.ml
index 6610241..3b68812 100644
--- a/cpdfcommand.ml
+++ b/cpdfcommand.ml
@@ -2,8 +2,8 @@
 let demo = false
 let noncomp = false
 let major_version = 2
-let minor_version = 5
-let version_date = "(patchlevel 1, 31st Jan 2022)"
+let minor_version = 6
+let version_date = "(devel, 3 April 2022)"
 
 open Pdfutil
 open Pdfio
diff --git a/cpdfmanual.tex b/cpdfmanual.tex
index bae7fd8..99f5f61 100644
--- a/cpdfmanual.tex
+++ b/cpdfmanual.tex
@@ -1026,7 +1026,7 @@ cpdf -merge in1.pdf [<range>] in2.pdf [<range>] [<more names/ranges>]
   \noindent\verb!cpdf -split in.pdf -o <format> [-chunk <chunksize>]!
 
   \vspace{1.5mm}
-  \noindent\verb!cpdf -split-bookmarks <level> in.pdf -o <format>!
+  \noindent\verb!cpdf -split-bookmarks <level> in.pdf [-utf8] -o <format>!
   \end{framed}
 
   \vspace{12mm}
@@ -1141,6 +1141,8 @@ one of the output files.
   \verb! / ? < > \ : * | " ^ + =!
   \end{framed}
 
+  To prevent this process, and convert bookmark names to UTF8 instead, add \texttt{-utf8} to the command.
+
 \section{Encrypting with Split and Split Bookmarks}
 \index{encryption}
 The encryption parameters described in Chapter \ref{encryption} may be added to the command line to encrypt each split PDF. Similarly, the \texttt{-recrypt} switch described in Chapter \ref{basicusage} may by given to re-encrypt each file with the existing encryption of the source PDF.